''--'T H E W O R L D B a N K . i - - A a * _ hi1 nE Il Marcelo M. Giugale Olivier Lafourcade o e c Connie Luff 25426 December 2002 Marcelo M. Giugale t' UTC"' - lve4aorae t-" -I I-- The Economic Foundation Editors Marcelo M. Giugale n2P0 Olivier LafourcadeOU P UU b Connie Luff THE WORLD BANK WASHINGTON, DC Copyright C) 2003 The International Bank for Reconstruction and Development / THE WORLD BANK 1818 H Street, N.W. Washington, D.C 20433, USA All tights reserved Manufactured in the United States of America First printing December 2002 1 2 3 4 03 02 The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organ- izations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. The boundaries, col- ors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any )udgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. The World Bank encourages dissemina- tion of its work and will normally give permission to reproduce portions of the work promptly Permission to photocopy items for internal or personal use, for the internal or personal use of specific clients, or for educational classroom use is granted by the World Bank, provided that the appropriate fee is paid directly to the Copyright Clearance Center, Inc., 222 Rose- wood Drive, Danvers, Massachusetts 01923, USA; telephone 978-750-8400, fax 978-750- 4470. Please contact the Copyright Clearance Center before photocopying items. For permission to reprint individual articles or chapters, please fax a request with com- plete information to the Republication Department, Copyright Clearance Center, fax 978- 750-4470 All other queries on rights and licenses should be addressed to the Office of the Publisher, World Bank, at the address above or faxed to 202-522-2422. The cover illustration is Nacimiento de los Andes (1961) by Alejandro Obreg6n, reproduced courtesy of his son, Diego Obreg6n, Santa Fe de Bogota, Colombia. ISBN 0-8213-5348-9 Library of Congress Cataloging-zn-Publication Data has been appliedfor Contents Acknowledgments ........................... xxxi Preface ......................................xxxiii Editor Biographies ........................... xxxv Acronyms ........................... xxxvii Synthesis ........................... 1 I. Rationale and Organization . .......................... ....... 1 11. Central Messages. ............ ........................ ...2 111. The Agenda's Diagnoses and Policies ........ ............ 4 Achieving Fast and Sustainable Growth ...... ... .............. 4 Sharing the Fruits of Growth ........ ....................... 19 Building a Government of Quality .................. ........ 25 IV A Path to a New Colombia ........... ...... ................ 29 Table 1. Colombia: A Possible Priontization of Policies for the New Presidential Administration .. . 31 Part I Thematic Notes 1. Violence, Sustainable Peace, and Development . .35 I. Introduction ....................... .... ............... 35 II Violence Trends and Characteristics ..... ........... 35 III. Sources of Violence and Conflict ............ ................ . 39 Armed Conflict ... .................... ................. 40 Social Conflict ................................... .... 41 Drug Trade ................ 43 IV. Costs of Violence and Conflict ......... . .. ........... ....... 44 Economic Costs of Violence . 44 Social Costs of Violence ......... . ............ .......... 47 Political Costs of Violence ....... . .. ..................... 48 Environmental Costs of Violence ......... 49 iii iV COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE V. Toward Building Peace and Sustainable and Equitable Development ..... 49 Armed Conflict and Drug Trade ............................. 50 Underlying Social Conflicts ........1.. ..................... 51 Critical Elements of Strategic Approach ........................ 54 VI. Conclusion ..................................... 55 References ..........................5 ....................... 56 Figure 1. Murder Rate per 100,000 Population, 1993-94 . . 36 Figure 2. Murder Rate, 1946-98. ............. ................... 38 Figure 3. Cocaine Production, Homicide Rate, and Guerrilla Front ........ 43 2. Enabling Sustainable Growth ................................. 59 I. Introduction ............. ........ 59 II. Providing a Sustainable Macroeconomic and Fiscal Framework . ........ 60 Background ........................................ 60 Fiscal and Debt Sustainability ............................... 61 Debt Management ....................................... 61 Fiscal Adjustment Options ................................. 62 III. Ensuring a Healthy Financial System .......... .................. 65 IV. Creating a Business Environment to Foster Private Sector Development .............................................. 69 V. Improvement in Infrastructure and Public Services to Foster Competitiveness .............. 70 Energy Sector ........... ................ .............. 71 Telecommunications ...................................... 74 Transport .............................................. 76 Urban Development and Low-Income Housing .................. 78 Superintendency of Domestic Public Services ......... ........... 79 VI. Management of Natural Resources and Physical Environment .... ...... 81 Rural and Agricultural Development ................ .......... 81 Water Resources Management ............................... 83 Potable Water Supply, Sanitation and Wastewater Treatment .... ..... 84 Solid Waste Management .................................. 86 Disaster Management and Disaster Insurance Mechanisms .... ...... 86 VII. Conclusions .............................................. 88 3. Shared Growth, Poverty, and Inequality . ........................ 89 I. Introduction .............................................. 89 II. Colombla's Tradition of Economic and Social Progress ......... ...... 90 Economic Growth Was Instrumental to Poverty Reduction .......... 90 Persistent Social Progress ................................... 91 Generally Propoor Expansion of Social Expenditures in the 1990s ..... 92 III. Recent Setbacks ............................................ 94 Economic Recession ..................... ................ 94 CONTENTS V Increasing Macroeconomic Instability and Judicial Uncertainty .... . 95 The Growing Burden of Violence and Crime ......... ......... 96 Rising Income Inequality: Induced by Higher Wage Skill Premia and Eroding Potential Welfare Gains ...... ...... . .......... . 97 IV. Resulting Vulnerability ............................. ..... 100 V. Addressing the Emerging Challenges: Employment, the Efficient and Equitable Provision of Public Services, and Security ....... . ....... 102 Restore Economic Growth to Accelerate Job Creation and Reverse the Rise in Poverty ........... .... ................... 103 Improve the Equity and Efficiency of Public Social Services .105 Establish a Functional Social Safety Net .... ............. .. 109 VI. Conclusion ................................... ......... 111 Raising the Income of the Poor .............. .............. 111 Ensuring the Equitable and Efficient Provision of Basic Social Services ................ ..... ... 112 Reducing Exposure to Risk ................ ............ 112 Appendix ..... ....... .... .. 113 References ..................... ....... 128 Figure 1. Labor Earnings by Skill Level, Urban Colombia, 1999 . . 99 Figure 2. The Poverty Rate Under High and Low Growth Scenarios, 2001-10, Urban Colombia ..... ............ ...... 104 Figure 3. Share of Total Household Income Required to Close Poverty and Extreme Poverty Gaps ....................... .... 106 Table I Poverty and Inequality Indicators, Colombia 1978-99. .... .... 91 Table 2. Social Indicators, Urban Colombia 1978-99 . . 92 Table 3. Change in Social Service Coverage Rates, 1992-97 ......... 93 Table 4. Colombia's Volatility. ............... ................. 96 Table 5. Poverty Count for Different Subgroups of the Population, Urban Colombia 1978-99 (percent) ..................... 101 Table 6. Coverage and Expenditure for Social Services, Colombia 1992-97 (growth) .................. .............. 107 Table 7. Coverage Shortfalls and Targeting in Selected Social Programs . . 107 Table A. I Summary of Issues, Objectives, and Recommendations by Sectors ............... ........................ 113 Table A.2. Summary of Policy Handles that Impact Welfare in Three Dimensions: Income, Access to Social Service, and Risk Management .............................. 122 Table A.3. Definitions Associated with Table 1 . ............... 126 Table A.4. Inequality Decomposition, Urban Colombia ........... .... 127 4. The Demand for Governance and Quality of Government .... ...... 131 I Summary ............. .................... ............ 131 II. Background: A Dual Quality of Government ....... . ... ....... 132 vi COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE III. The Problem with the Coexistence of Strong and Weak Governance .... 136 IV. Strategizing to Strengthen Governance: Selectivity and Gradualism in Building High-Quality Government . . 138 Box 1 Budget Implementation and Control .... ......... 133 Box 2. Efforts to Improve Quality of Government in the United States and Mexico ............. 139 Box 3. Targeting Reforms to Plug Loopholes ............. 140 Box 4. Steps to Ensure Good Governance in Colombia ............. 141 Box 5. Focalizing Reforms ..... . .. ......................... .. 142 Box 6. Singapore's Experience with Phased-In Governance Strengthening .... 142 Part II Achieving Fast and Sustainable Growth 5. Macroeconomic and Fiscal Frameworks . . .145 I. The 1990s: From Rapid Growth to Recession. ... 147 II. Policies for Adjustment . . . ................................. 152 III. Main Issues and Options .............................. 157 Fiscal Sustainability . . . 157 Expenditures . . ................. 159 Net Worth and Illusory Adjustment . . .164 Revenues . ......................................... .. 165 The Central Bank . . . 167 IV. Conclusions ............ ... ........ 168 References ............... ........ 170 Figure 1. Central Government Revenues and Expenditure .151 Figure 2. Central Government Current Savings .151 Figure 3. Growing Deficits .152 Figure 4. Growth of Expenditures .160 Figure 5. Composition of Expenditures .166 Figure 6. Major Tax Reforms, 1990-2000 .......................... 166 Table 1 Increased Volatility in Colombia in the 1990s ...... .......... 147 Table 2. Growth in Central Government Spending Due to Legal and Constitutional Changes, 1990-98 (percent of GDP) ..... .... 149 Table 3. Two Medium-Term Scenarios .......... .................. 158 6. The Tax System ..173 I. Introduction .173 II. Background ...................... ...................... 173 III. The Tax System: Issues and Policy Recommendations .178 National Taxes .179 Subnational Taxes .......... ............................ 184 CONTENTS Vil Tax Administration .................. .. ......... 186 Appendix A Summary of Previous Tax Reforms ...... .. ......... 188 References . . .. .......... ................ ..... 191 Table 1. Colombia-Structure of Taxes . .......... . 174 Table 2. Colombia-National Taxes .......... . ............. ... 175 Table 3 Colombia-Taxes Collected by Departments (1999) .......... 176 Table 4 Comparative Composition of Tax Revenue (Average 1995-97) . 177 Table 5. Colombia-Central Government-Earmarked Taxes (1999) ..... 180 Table 6. Colombia-Taxes Collected by Municipalities (1999). ... 185 7. Public Debt Sustainability and Management .193 1. Summary .. ... ............... ................ .... . 193 II. Background ... . .................. ........... ..... .. 193 III Issues, Diagnosis, and Policy Recommendations . . 201 The Primary Balance and Debt Sustainability. 201 Vulnerability and Management Risks in the Debt Structure . 206 Key Government Liabilities and Potential Impact on Debt Management ..... .................. 210 Policy, Legal, and Institutional Framework for Central Government Debt Management ...... ..................... ... . 212 Strengthening the Framework for Debt Management . .217 Estimated Fiscal Impact .. 225 Annex 1. Key Elements of Vulnerability of Central Government Debt. 227 Annex 11. Analytical Framework for Fiscal Sustainability ........... ... 234 Annex III. External Liquidity and Fiscal Vulnerability Issues in Colombia- A Contrast with the Argentine Case ............. ...... 238 Annex IV. Debt Management Strategy and Portfolio Benchmarking .247 Annex V. Development of Debt Markets .... .............. ... .. 251 Box 1. Legal Framework for Debt Management in Colombia ... .... 214 Box 2 Reform of the Legal Framework for the Debt Management Office in Brazil ................... ... ........ 218 Box 3. Risk Management in Brazil .220 Box 4. Building Benchmark Issues Through Reopening and Buyback Operations. ... .... 223 Box 5. Growth and Importance of Institutional Investors . .. ...... 223 Box 6. The Role of Secondary Market Windows . . 225 Box A.V. I The Spanish Experience with Specialized Investment Funds .. 253 Figure 1. Public Sector Balances ..194 Figure 2. Interest Payments of the Public Sector . ........ ... 196 Figure 3. Gross External and Domestic Debt of the Public Sector. . 197 Figure 4. Gross Public Debt by Debtor . . 198 Figure 5. Domestic Debt of the Central Government by Creditor.. 199 Figure 6. External Debt of the Central Government by Type of Creditor .. 200 ViII COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Figure 7. Debt Structure of Sub-National Entities ..... ............... 201 Figure 8. Debt/GDP Ratio; Primary Balance Constant at 2001 Level ...... 203 Figure 9. Debt/GDP Ratio; Primary Balance Set to Keep Debt/GDP Ratio Constant .......... .......................... 204 Figure 10. Debt/GDP Ratio; Primary Balance Constant at 4 Percent of GDP Slow Growth Scenario ........................... 205 Figure 11. Domestic Real Interest Rate Volatility ... .................. 208 Figure 12. Average Coupon in Real Terms on Domestic Public Debt ....... 209 Figure 13. Public External Debt Ratios ....... ...................... 210 Figure A.I. 1. Domestic Central Government Debt by Type of Coupon, TES Bonds ...................................... 228 Figure A.I.2. Half Life and Duration of Domestic Central Government Debt . ....................................... 229 Figure A.I 3. Maturity of Domestic Central Government Debt ............ 229 Figure A.I.4. Currency Structure of External Public Debt ..... ........... 230 Figure A.I.5. Currency Structure of External Public Debt ................ 231 Figure A.I.6. External Central Government Debt by Type of Coupon .... ... 232 Figure A.1.7. Half Life and Duration of External Central Government Debt . .......................................... 232 Figure A.I.8. Maturity Structure of External Central Government Debt .......................................... 233 Figure A.III.1. International Reserves and Public External Debt Stocks ...... 239 Figure A.III.2. Total Exports and Public External Debt Stocks .... ........ 240 Figure A.III.3. Public External Debt Service and Total International Reserves ........................................ 241 Figure A.1II.4. Public External Debt Service and Total Exports .... ...... . 241 Figure A.II.5. Public Interest Payments and Tax Revenues ..... .......... 243 Figure A.III.6. Public Interest Payments and Total Revenues .............. 243 Figure A.III.7. Public Debt Stocks and Total Revenues ..... ............. 244 Figure A.III.8. Gross Public Debt and GDP ...... ................... 244 Figure A.III.9. Consolidated Public Sector Balances ..... ............... 245 Figure A.III. 10. Total Public Sector Revenues ......................... 245 Figure A.V. 1 Holders of Central Government Domestic Debt .... ...... 252 Table 1. Key Assumptions and Projections .202 Table 2. Permanent Primary Surplus: Sensitivity Analysis (percent) . 205 Table 3. Permanent Primary Surplus: Sensitivity Analysis (percent) . 206 Table 4. Short-Run Impact of Changes in Key Variables on Public Sector Debt Service .207 Table A.I. 1. Types of Domestic Treasury Bonds .......... ............. 227 8. The Financial Sector ............................... 255 I. Background ...................... ..................... 255 II. The Financial Crisis Management Strategy ............. ..... 257 CONTENTS ix III. Restructuring and Recapitalization of the State-Owned and Private Banks ..... ........................ 258 FOGAFIN's Crisis Response Instruments .259 The Development of Regulatory and Supervisory Strategies .261 IV. Fiscal and Economic Costs in Relation to Financial Sector Restructuring ......2.........6...... ... . .. 263 The Corporate Sector . ... 264 Pending Policy and Institutional Reforms-The Banking Sector . 266 V. Reform of Second-Tier Public Banking Institutions ......... .. .... 268 Characteristics of Government Banks in Relation to a Merger Proposal .. ...................................... 268 Pending Policy and Institutional Reforms-Second-Tier State Banks .270 VI. Capital Market Development and the Public Debt Market .271 VII Conclusion .277 Table 1. Financial System Return on Assets (percent) .... .. ..... . 256 Table 2. Capitalization Support Provided to State-Owned Banks, 1998-2001 . .............. ... .......... ....... 260 Table 3. Private Bank Capitalization (Credit Line) Program, 1998-2001 . .. 261 Table 4. FOGAFIN Balance Sheet as at end 2000 and end 2001 (Col$ billions) .262 Table 5 Total FOGAFIN Support to Financial Entities, 1998-2002 (C$ millions) ........... ... . 264 Table 6. Assets as Percent of GDP of Companies/Corporations in Distress (percent) ...... ...... ..................... 265 Table 7. Public Pension System Liabilities ..... 274 Table 8. Estimated Fiscal Funding of Territorial Pension Obligations (Col$ billions) . . ... . ........................... 275 9. Pension Reform ..279 1 Executive Summary ............ ........................... 279 II. Introduction ...... . ...................... ............. 281 III. Colombia's Pension System ........... ...................... 282 The 1993 Pension Reform .282 Law 100 of 1993 .............. ........................ 283 Current Structure of the Pension System ........ 286 Evolution of Coverage ............ . ............... ...... 286 Instituto de Seguros Sociales .287 Private Pension Fund Managers ..... .............. . .. . 290 Transition Regimes . .......... ... ................ 290 Exempted Regimes ... .............. ............... 292 Collective Agreements .............. ............. I ... 293 Pension Recognition Bonds ........... .............. 294 x COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE IV. Fiscal Requirements of the Current System ....... I ............... 294 The Model ............................... 294 Pension Liabilities ........ ............................. 295 V. Reform Requirements ................ 298 Overall Strategy ................ 298 Reform Options ................ 299 Specific Reforms ................ 302 Fiscal Impact of Pension Reform .................... 305 Pension Reform in Draft Law .................... 305 VI. Conclusion ........... ......... 310 Figure 1. Payroll Taxes for Social Insurance and Health in Latin America, 1995 ... ................. ................... 285 Figure 2. Pension Benefit Recipients as Percentage of Elderly Population 1980-99 ... 288 Figure 3. Contributors to Publicly Mandated Social Security as Percentage of Economically Active Population, 1980-99 . 289 Figure 4. Current Deficit of the Pension System by Component (as percent of GDP) .297 Figure 5. Reserves in Millions of 1999 Pesos; Cohort of 1,000 Women and 1,000 Men Joining ISS .301 Figure 6. Total Current Deficit of the Pension System With and Without Reform, as a Percent of GDP .306 Table 1. Colombia: Total Labor Costs (as percent of wage) .284 Table 2. Labor Force Statistics as of January 2001. 287 Table 3. Exempted Regimes .292 Table 4. Contributions to Exempted Pension Regimes (as a percentage of salary) .293 Table 5. Pension Liabilities, by Regime .296 Table 6. National Government Pension Payments (as percent of GDP) .... 296 Table 7. Estimates of Implicit Pension Debt and General Government Debt, Selected OECD Countries (percent of GDP) .298 Table 8. Implicit Pension Debt Estimates, and Pension and Public Debt Data (percent of GDP) .299 Table 9. Impact of Individual Parameter Changes on the ISS Deficit . 302 10. Housing Finance ........................................ 313 1. Executive Summary ........................................ 313 11. Genesis and Evolution of a Crisis .............................. 313 111. Government Packages of Reforms .............................. 314 IV. Main Sectoral Issues ........................................ 315 A Contaminating Deterioration of the Mortgage Portfolio .... ..... 315 A Contracting Mortgage Stock and Production of New Loans ....... 317 Legal, Regulatory, and Institutional Environments ..... .......... 319 CON1IENTS xi Figure 1. Quality of the Residential Mortgage Portfolio ........ ...... 316 Figure 2. Nonperforming Mortgage Loans, Public and Private Banks ... 318 Table 1. Recent Evolution of Mortgage Lending ................ .... 319 ll. Urban Development ... 329 1. Urbanization in Colombia .................. .. .. ..... .. 330 11. The Legal and Institutional Framework ...... ..... ............ 332 111. Urban Poverty ............... .. ....................... 334 IV. Access to Public Services ....... ........... . . 336 Land and Housing ............. .......... 338 Water Supply, Sanitation, and Solid Waste Management .341 Urban Transport .. . . . ........ 350 V. Policy Recommendations .................. .. .. ..... .. 353 Urbanization ............ ................. ..... 353 Land and Housing ............... .................. . . 356 Water Supply and Sanitation ...... ......... 357 Urban Transport ......................... .. ... ..... 357 References .......... . .. .............. ...... ......... 358 Box 1 The Medellin Metro .............. .................. . 354 Box 2. The TransMilenio System in Bogot ................. ........ 354 Figure 1. Bid Rent for Land: Poor versus Nonpoor ............... ... 344 Table I Gross Annual Rates of Internal Migration to the Main Cities and Some Conurbations, 1982-2000 ... .......... 331 Table 2. Indicators of Unsatisfied Basic Needs by Zone, 1993-2000 ...... 335 Table 3. Poverty Indicators and Income Inequality, Urban Colombia, 1978-99 ....................................... 336 Table 4 Social Indicators, Urban Colombia, 1978-99 ........... ... 337 Table 5. Housing Deficit (National, Urban, and Rural), 1997-2000 ...... 342 Table 6. Coverage of Water Supply and Sewerage Services in Colombia (percent) ............. ................... 345 Table 7 Coverage of Water Supply, Sewerage, and Solid-Waste Collection in the Main Cities (percent), 2000 . ................ 346 Table 8 Coverage of Water Supply Services in Major Cities ..... ....... 346 Table 9 Water Supply and Sewerage Utilities: Total Expenses/Operational Income, 1999 and 2000 ... .......................... 348 Table 10 Solid Waste Companies: Total Expenses/Operational Income, 1999 and 2000 ................................... 349 Table 11. Unaccounted-For Water in Major Cities, Percentage .349 Table 12. Demand for Public Transportation in Major Cities ........ ... 350 Table 13. Public Transport Vehicles and Their Productivity in the Main Cities, 2000 ............................ .... 351 Table 14. Public Transport Fares in Several Cities, 2000 .......... .... 352 Table 15. Accident Rates in Colombia's Main Cities, 1999 .. . ....... 353 xii COLOMBIA THE ECONOMIC FOUNDATION OF PEACE 12. Transport ......... 363 I. Background ............................................ 363 II. Main Sector Issues and Strategies ........ ...................... 370 Armed Conflict Negatively Affects the Functioning of the Sector .... 370 Isolation of Areas Due to Lack of Transportation Infrastructure ...... 371 The Unfinished Development of the Institutional and Regulatory Framework ..................... ......... 371 Ensuring Adequate Resources to Address the Needs of the Transport Sector ...................... 374 Private Participation in the Development of the Sector's Infrastructure................ ..... 377 Difficulties in the Mobilization of Population within Urban Areas.... 380 II. Recommendations ...... ......... ......................... 381 Contributing to the Peace Process ........................... 381 Consolidation of the Institutional Reforms of the Transport Sector . .. 382 Decentralization of the Road System ....................... 383 In Search of Financial Sustainability for the Sector ...... ......... 384 References ... .. ...... 385 Annex ..... ............................................. 386 Figure 1. Transportation and Economy, GNP, Tons per Kilometer ........ 364 Figure 2. Transport Sector Investment ............. ................ 375 Table 1. Cargo Mobilization at National Level (thousand of tons per year) ......................................... 365 Table 2. Deterioration of Paved Road System Managed by INVIAS (percent) .... . ............................ 366 Table 3. Condition of Unpaved Roads (percent) ......... ............ 367 Table 4. Private Participation in the Transport Sector ...... ........... 369 Table 5. Transport Budget 1998-2001 (millions of pesos) ..... ......... 376 Table A. 1. Road Priority Projects-Vias para la Paz Program .... ......... 386 Table A.2. Transport Entity Roles ... ............................. 387 Table A.3 Port Indicators With and Without Reform ....... ........... 388 Table A.4. Anticipated Private Participation in Transport ...... .......... 389 13. Energy Sector Strategy .................................... 391 I. Background. .............. 391 Terrorism . .....391................ 391 Oil Subsector .392 Gas Subsector .395 Power Subsector ...................................... 397 Coal Subsector .400 Energy and the Environment .401 II. Achievements and Issues ....... ........... ................. 402 Issue 1: Oil and Gas Supply and Pricing ....................... 403 CONTENTI S xiii Issue 2: Regulating Generation and Distribution .... ... ....... 406 Issue 3: Unfinished Business-Further Privatization of State-Owned Companies ... 409 Issue 4: Reforming or Strengthening the Regulatory and Supervisory Institution ......... ..................... . 411 Issue 5: Energy, Poverty, and Peace ... . ..................... 414 III Summary of Recommendations and Next Steps .415 References . .. .... ............................. . .. . 419 Figure 1 Per Capita Energy Consumption (kg oil equivalent per year) . 392 Figure 2. Comparative Reserves (Mboe). ................... .394 14. Information and Communications Technology Sector .421 I. Background .................. .. .. 421 Information and Communicationis Technology and Development .... 421 Sector Reform and Liberalization . . 422 Privatization Attempts .. 425 II Sector Issues .. . . . 425 Low Private Sector Participation ....... ..... ... . . 425 Fragmentation of Regulatory Institutions ... . 426 Subsidies and Tariffs .. 426 Low Internet Penetration .................. 428 Reduced Role of Colombia in the Information Society and the Global Economy .... .. .... .. ......... 428 III Options for Solutions .......... ........... .429 Increased Private Sector Participation ...... . .. .... ..... 429 Tariff Reform ............. ...... ...... . .... 430 Consolidation of Regulatory Institutions . . . . .. 430 Incentives for Local Telephone Companies to Compete With Each Other and Flexibility to Merge .. ... . . ........ 430 IV Colombia in the Information Society .... ..... . .. ..... . . 431 Ministry of Information and Communication Technology .431 Telecom Fund . .432 Developing and Promoting the Use of ICT Applications .... ...... 432 Figure 1. ICTs Are the Basis of the Knowledge Economy ... 423 Figure 2. Teledensity in Selected Latin American Countries and Spain 424 Figure 3. Cell Phone Penetration in Selected Latin American Countries .... 425 Figure 4. Local Subscription Rates in Selected Latin American Countries . 427 Figure 5. Internet Penetration, Selected Latin American Countries and Spain .......................... ................ 429 Table 1. Finances of Colombia's Telecommunications Companies ... .... 428 15. Urban Water and Sanitation Sector .433 I Summary ......... ......... ...... 433 xiv COLOMBIA THE ECONOMIC FOUNDATION OF PEACE II. Background. Sector Status and Issues ........ ................... 435 Sector Context .......... ............... 435 Main Sector Issues ... .................................. 436 III. Water Sector Policy Recommendations for the Incoming Government ............ ............................... 442 Improving the Water Supply and Sanitation Services in Medium-Size Cities and Small Municipalities ..... ........... 442 Improving the Water Supply and Sanitation Services in the Rural Sector ................................... .... 443 Management and Disposal of Municipal Wastewater ... ......... 444 Guidance for Major Projects in Large Cities ............. ...... 447 Improving the Regulatory Framework of the Water Sector .... ..... 447 Annex I. Improving the Water Supply and Sanitation Services in Medium-Size Cities and Small Municipalities ..... ......... 452 Annex II. Examples of Government Guidance Regarding the Water and Sewerage Projects in Large Cities ...... ................. 458 Annex III. Intersectoral and International Context of the Problems Associated with the Regulatory Reform Proposed by the Water Regulatory Commission .......... ............... 461 Table 1. Coverage of Water and Sewerage in Colombia . .437 Table 2. Effective Water Coverage in Colombia as Indicator of Level of Service ................... ................... 438 Table 3. Regional Distribution of Water and Sewerage Coverage Rates in Colombia ................... 438 Table 4. Performance Indicators in Some Large- and Medium-Size Utilities Operated by the Private Sector ................... 441 16. Natural Disaster Management: Vulnerability Reduction & Insurance .......................... 463 Part A. Hazard Mitigation and the Institutional Framework ..... ......... 463 I. Background ......... ................................... 463 Risk and Vulnerability Assessments ...................... 464 Institutions ..... ....................... 466 Part B. Catastrophe Risk Exposure of Public Assets: An Analysis of Insurance Instruments for Smoothing Fiscal Volatility . 470 I. Insurance Contractual Arrangements and the use of Reinsurance . 471 The Quantification of the Natural Hazard Threat in Colombia 472 Inventory of Public Assets and Infrastructure .475 Loss Projections for Risk-Management Planning .476 Catastrophe Insurance Pricing and Risk-Load Factors .477 Comparison of Financial Instruments for Catastrophe Protection .... 478 Financial Parameters for Establishing a Risk-Management Strategy .... ................................ 479 CONTENTS xv Risk-Linked Securities and Catastrophe Bonds ................. 479 Risk Financing with Contingent Credit . 482 Role of the Domestic Insurance Sector ....... ........ . . ... 484 11. Recommendations for the Development of a Government Risk-Transfer Policy .............. ..... .. 484 Figure 1. Quantification of Natural Hazard Threat in Colombia .. .. . 473 Figure 2. Seismic Risk-Loss Distribution Function .................. . 476 Figure 3. Loss Probability Functions for Single and Cumulative Events . . . 478 Figure 4. Catastrophe Bond Financing Structure ............... ... 482 Figure 5. Potential Role of Domestic Insurers in Government Risk Transfer. 485 Table 1. Local Industry Reinsurance Arrangements .......... . . . 471 Table 2 Catastrophic Earthquake Frequency and Loss Intensity .. 474 Table 3. Roads Vulnerability Factors as a Multiple of Standard Damage Ratio of 1.00. . ..... ... . .. ....... 474 Table 4. Capital Stock-Earthquake Damage Ratios .................. 475 Table 5. Equivalence in Pricing Between Financial Market Bonds and Cat Bonds (percent) .......... . .. . .. ...... . 480 Table 6 Example of Parametric Loss Triggers for Cat Bond "Payment" . 481 Table 7. Insurance Structure Using Risk Transfer and Risk Financing Instruments . ... . .. . .. . 483 17. Agriculture and Rural Development .487 I Background . . .. . ............. 487 II Main Issues.. . .... ................. ... 492 Missed Opportunities for Growth ........................... 492 The Poor Functioning of Factor and Goods Markets .......... ... 493 Poverty and Inequality .................. .. . . .... 499 Institutional Breakdown ......... ............. ..... 502 LII. Options for the Future ...................... .. ..... . 505 Fostering Growth in the Rural Economy ............. .. 505 Poverty .. .. ... .. .. ... . 510 Institutions ......................................... 511 Natural Resources .......... .. ....... 513 References . . . ... . . ... .. 514 Box 1. PROCAMPO .......... ............ 507 Figure 1. Growth Rates ............................... ..... 488 Figure 2. Percent Change in Area Harvested Between 1988-90 and 1999-2000 .498 Figure 3. Income Per Capita-Sen Welfare Index. .. .. . .. 500 Table 1 Poverty and Inequality (in percent) .................. 489 18. Coffee .517 I Background ........... ...... .. .. .... ..... 517 xvi COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Anatomy and Evolution of the Sector ......................... 517 Global Context and Trends ................................. 519 Production and Demand Trends ..... ........... ........... 520 Macro Trends in Established Consumer Markets ..... ........... 523 II. Main Sectoral Issues/Diagnostic ...... ............... ......... 525 Domestic Impact of Low Prices Trends and Volatility ..... ........ 526 Smallholders and Their Integration with the Process of Rural Development ..................................... 528 Policy Initiatives with Stabilization Funds and Subsidies ..... ...... 529 Institutions and Their Roles ........ ............... ....... 530 Credit ...................................... 532 Issues Across the Supply Chain ............. ................ 532 The Changing Nature of Demand and Adapting to Differentiated Markets .......................................... 536 III. Policy Recommendations . .................................... 538 Smallholders and the Rural Poor ......... ................... 538 Managing Transition: Options for Diversification ..... ........... 540 Risk Management and Safety Nets ........................... 543 Policies of Subsidy and Support .. ........................ . 545 Institutional Leadership .................... 546 Competitive Foci .................... 547 Annex I. The Nature and Characteristics of Differentiated Coffee Markets . .. 551 Annex II. Colombia's Coffee Subsector Public Supports .554 References ..555 Figure 1. Importer Demand for Exportable Production (percent) .521 Figure 2. Producer Price ($1998 = 100) and Production Volume .526 Figure 3. Key Policy Recommendations for the Coffee Sector .... ..... .. 539 Figure A.I. 1. Percent Rating Attributes as Important . .553 Table 1 The Increasing Concentration of the International Coffee Business .522 Table 2. Coffee Prices Volatility Index (Standard Deviation Percent) . 527 Table 3. Distribution of Coffee Plots (by size) .527 19. Land Policies .559 I. Executive Summary .559 II. Background and Challenges .560 The Extent of Inequality in Land Access and Reconcentration and Underlying Factors .561 The Consequence of Colombia's Skewed Land Distribution: Inappropriate Use of Natural Resources .563 Landownership Patterns and Administrative Decentralization . 564 Land Access and Landownership as Key Determinants of Household Welfare .565 CONTENTS XVii 111. Main Issues. .. . .................. ....... . .... .. 566 Security of Tenure and Mechanisms of Land Access. ........ 567 Land Titling ... . ..... ........... ......... ..... 572 Land Reform ....... . 573 IV. Policy Recommendations ........ .............. ....... ... 580 Systematic Titling .... ......... ...................... 580 Fiscal Policy and Land Taxation ............. ....... . . . 580 Ordenamiento Territorial and Sustainable Land Use .. ........... 581 Land Reform ...... ..................... . ........ 581 Annex ... ....... ....... ........................... 583 References ............ 585 Table 1. Structure of Landownership and Use in Colombia, 1984 and 1997 ......................... 562 Table 2 Actual and Potential Land Use in Colombia, 1985 and 1999 .. 564 Table 3. Means of Land Access, Colombia ................ ........ 568 Table 4. Ways of Land Acquisition in Different Periods, Colombia .. 569 Table 5. Producers' Rental and Sales Market Participation in Colombia, by Region ... .. ........................ ..... 570 Table 6 Probit Regression for Land Rental and Sales Market Participation .571 Table 7. Public Spending for Land Reform and Related Projects, Colombia Different Years (millions of current pesos) ... 576 Table 8. Cost Per Beneficiary Family for Different Modalities of Reform (millions of 1998 dollars) . ... .......... ......... . 577 Table 9. INCORA's Executed Budget, Total and by Component (millions of 1998 pesos) ...... 578 Table 10. Debt Burden of Different Farms Under Pilot and Nonpilot .... 579 Table A. 1. Determinants of Land Prices .. . . .. .583 20. Rural Finance ...... 587 1. Background .587 Overview. ....... ..... 587 Market Destruction and Distortion Due to Conflict and Violence .... .......................... ............ 589 Crisis and Reform in the Financial-Cooperatives Sector .589 The Public Sector .............. ........................ 590 Alternative Sources of Finance .................... ......... 591 11. Main Issues ........... ................................. 592 111. Policy Recommendations .... ............................. 595 The Role of the State- Banco Agrario, FINAGRO, and the Effectiveness of Public Interventions ... .............. .. 595 Modernizing Rural Risk Assessment and Risk Management . . ... 597 Enabling the Use of Nontraditional Collateral ....... .... 598 XVIII COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE Annex I. The Cooperative Sector ........ ........................ 599 Annex II Public Sector Institutions and Programs in Rural Finance ........ 603 References .... ................. .......................... 607 Figure 1. Density of Retail Financial Outlets Supervised by the SuperBancaria ........... .......................... 588 Figure 2. Total Agricultural Sector Loan Disbursements-FINAGRO Discounted Funds and Own "Substitute" Funds .... ........ 593 Table 1. Cities and Towns with at Least One Bank Branch .... ......... 589 Table A.I.1. Financial Cooperatives and Commercial Banks, Balance Sheet Aggregates and Relative Comparison, US$ Millions as of 31 December of Each Year .. .... .......... 599 Table A.I.2. Number and Balance Sheet Aggregates of Cooperatives Supervised by the SES, US$ Millions as of December of Each Year ... ................................... 600 Table A.I.3. Number and Balance Sheet Aggregates for a Subset of Cooperatives Consistently Reporting to the SES, US$ Millions as of December of Each Year ..... ........... 601 Table A.II.1. Agricultural Sector Lending 1998-2001 (disbursements, US$ millions) ..... .............................. 604 Table A.II.2. Distribution of Financial System Loan Portfolio by Sector, 1996-98 ....... 605 Sharing the Fruits of Growth with Al Colombians 21. Education .................. 611 I. The Big Picture: Colombia's Education and Challenges .... .......... 611 II. Colombia's Education System: Organization and Recent Trends. ...... 613 Preschool Enrollment: Coverage is Low ...... ................. 614 Primary Enrollment Trends: High Coverage, But Stagnating With 12 Percent Still Out of School ...... ................. 615 Secondary Enrollment Trends: Coverage is Low Especially the Poor ... 615 Higher Education Trends: Very Low Coverage and Favors the Rich ... 616 III. Policy Issues and Options in the Education Sector .................. 617 IV. Government Strategy: Addressing Education Sector Issues .... ........ 629 Programs and Policies Responding to Issues of Educational Access and Equity ..................................... 629 Programs and Policies Responding to Problems of Quality and Sector Efficiency ...................................... 631 Programs Addressing Effectiveness of Decentralization .... ........ 632 V. Conclusions and Policy Recommendations ...... ................. 634 Annex I. Possible Activities to be Implemented by the New Administration to Respond to Critical Education Sector Issues .... ......... 637 CONTl-ENTS xix References ............. ....................... ........ 650 Box 1. Hogares de Bienestar Infantzl .......... . . .. ....... 614 Box 2. Colombia's Escuela Nueva A Tested Strategy for Reaching the Rural Poor and Improving Quality ........... .. ...... 633 Box AI.1. Applying the Community College Model to Colombia . ... 639 Box AI.2. Mexico's Telesecundaria: Increasing Access to Secondary Education for the Rural Poor ....... .. .......... 641 Figure 1. Impact of the 1998-99 Crisis on Primary Enrollment ..... 616 Figure 2. Higher Education Gross Enrollment in Selected Countries, 1980 and 1994 .......... .... 617 Figure 3. Fiscal Situation and Contributions, 1994-2000... .. 634 Table 1. Average Math Achievement Test Scores of Eighth Graders, Selected Countries ........ ... .. ............... 620 Table 2 Average Years of Education by Income Decile, 1991-97 . . . 623 Table 3. An Education Action Plan. Policy and Program Option Matrix . .. 636 Table A.1 Country Scores in 1998 UNESCO/OREALC-Assessment of Student Outcomes in Latin America and the Caribbean . 644 Table A.2. Distribution of Family Spending on Education, 1997 . . 645 Table A.3. Distribution of the Primary and Secondary Education Subsidy by Income Decile, (constant 1996 pesos) ........... . . 646 Table A 4 Per Capita Transfer for Education by Department, 1994-98 ... . 648 22. Health .. . 653 I Introduction ... . .. 653 11. Background ............. .... ... ................ 657 Before the Reform .................... ... ... ..... 657 The Reform: Law 100 of 1993 .................. ........ 658 Implementation of the Reform ....... .................... 661 Financing ....... .. .. ... 664 III Sector Issues . . ....... ...... .. . . .. 666 The Social Security Institute ..... ....................... 666 Public Hospitals ....... . . ............................. 670 Financing . . . .................... ... .. ... ... 675 Health Insurance Coverage and Targeting . . .680 Public Health: The Issue of Vaccination .. . . . .... 684 IV. Government Response and Current Reforms . . . 685 V Policy Alternatives and Recommendations . . .689 The Social Security Institute . . .690 Public Hospitals ..... . .... . .. ... . .. 693 Financing . . . .. ... ....... 694 VI. Conclusion ...... ... .............................. 699 Appendix . ... . ... . ..... ........... .. ....... 702 References . . ........... . .......... . . ... ...... . 706 XX COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Box 1. The British Experience ............................ .... 674 Figure 1. Overall Health Performance and GNP Per Capita (PPP) .... 655 Figure 2. The Colombian Health System After the Reform of 1993 .... 660 Figure 3. Coverage Rates by Quintiles .................... ....... 663 Figure 4. Financing of the Health System .... 665 Figure 5. ISS Expenditures .............. .......... ........... 668 Figure 6. Balance Sheet, Consolidated EPS-IPS .. .................. 670 Figure 7. Income Statement, Consolidated EPS-IPS ........... ... 671 Figure 8. Public Hospital Financing, Millions of 1999 Pesos .673 Figure 9. Allocation of Resources-Law 60 ...... ................ .. 678 Figure 10. SISBEN Classification Document Affiliation and Health Care Copayments in the Subsidized Regime .682 Figure 11. Coverage Levels for Three Doses of DTP Among Children Younger Than 1 Year of Age, Andean Region, 1990-2000 . 685 Figure 12. Allocation of Resources-Law 715 .686 Table 1. Health Sector Indicators ....................... ..... 661 Table 2. Number of Affiliates, 1996-2000 ..... ................... 662 Table 3. Changing Patterns of Health Insurance, Percentage of Affiliates . . . 664 Table 4. Quality of Health Care, 1997 ...... .................... 664 Table 5. Waiting Time to Receive Treatment ....................... 665 Table 6. Transformation of the Fiscal Situation and the Transferred Rents, 1997-2000 ...................................... 666 Table 7. Distribution of Total Expenditure and Total Income of Public Hospitals of Levels I, II, and III, 1993 and 1998 ..... ....... 672 Table 8. Real Public Hospital Expenditures as Percentage of GDP .... .... 673 Table 9. Evasion in the Contributory Regime (millions of 2000 pesos) ..... 676 Table 10. FOSYGA: Annual Budget FY 2000 (pesos) ..... .......... .. 677 Table 11. Situado Fiscal Transformed into Demand Subsidies (millions of 1999 pesos) .............................. 679 Table 12. Number of New Affiliates in the Subsidized Regime by Type of Action ..................................... 681 Table Al. Option 1: ISS Health, Millions of US$ (in year 2000 dollars) .... 702 Table A2. Option 2: ISS Health, Millions of US$ (in year 2000 dollars, without including values of assets that could be sold) ......... 703 Table A3. Option 1: Timid Restructuring of Public Hospitals, Millions of US$ (in year 2000 dollars) . .......... ................ 704 Table A4. Option 2: Bold Restructuring of Public Hospitals, Millions of US$ (in year 2000 dollars) ............................ 705 23. The Social Safety Net .709 I. The Social Safety Net in Colombia .709 II. Role of Social Safety Nets and Recent Policy Initiatives .711 Colombia's Social Assistance Programs .711 CONTENTS Xxi Responding to the Recent Economic Crisis: The Social Support Network .. . . .. ................................ 714 III Main Sectoral Issues ......................... .. ...... ... 714 Issue 1: Sources of Vulnerability .. ....... .................. 715 Issue 2: Social Risk Management. ..... ... ... ... .... 719 Issue 3: Problems with Colombia's Social Safety Net .721 IV. Options and Recommendations for Reforming Colombia's Safety Net ... .. .. .. ............................... 723 Recommendation 1: Increase Resources for Social Assistance . 724 Recommendation 2. Focus on Streamlining the Existing System, not Creating Additional Programs .......... .. . 725 Recommendation 3: Establish a Countercyclical Element of the Safety Net ... .............................. .. .. 726 Recommendation 4 Addressing the Needs of the Internally Displaced Population .727 V Conclusions . 727 Annex . ..... ......................................... 729 References .. ........................................... 733 Table 1. Colombia's Main Social Assistance Programs . .... . ... 712 Table 2 Perceptions of Vulnerabilities by Socioeconomic Level . . 716 Table 3. Poverty Count for Different Subgroups of the Population, 1978-99 ... ................................. 718 Table 4. Social Risk Management Strategies . .720 Table A. Social Risks and Social Protection Programs in Colombia, 2000. . . 729 24. Higher Education .737 I Quality Expansion of Higher Education: Governance and Financing.... 737 II. Emerging Trends in Colombia's Higher Education .. ......... .... 737 1II A Strong But Underperforming Sector ...... ........ ........... 740 IV Policy Recommendations ........ .......... ......... 743 Ensure Clear and Progressive Governance ..... . .............. 743 Direct Public Resources to Its Core Responsibilities .. .. 744 Stimulate Demand in Public and Private Tertiary Education Through Student Aid ... ..... ... ....... ....... 745 Promote Strategic Levels of Education .... . .. ........ 746 Induce and Assure Quality . ... .... .................. 746 Figure 1. Expansion in Tertiary Enrollment Led by the Private Sector (Private and Public Enrolment, 1980-2000) ......... 739 Table 1. Urban Labor Market Indicators, by Level of Education (2000) ... . 740 Table 2. Public Spending on Education (1990-99) ... .. ......... 741 25. Science and Tchnology .749 1. Background on Science and Technology .. . 750 xxii COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE II. Strengths and Weaknesses in Colombia's Science and Technology Sector ............. ......................... 751 III. Policy Recommendations ........ ..................... ..... 753 Figure 1. Strengths and Weaknesses in the Science and Technology Sector: Technology and Innovation Indicators ..... .............. 751 26. Enhancing Employment Opportunities Through the Labor Markets .......................................... 755 I Context for Reform ............... ........................ 755 II. Issues and Policy Options .................................... 757 The System of Social Security ........ ...................... 757 Labor Unions, Wages and Industrial Relations ..... ............. 759 The Labor Market, Informal Sector, and Small Firms ..... ........ 761 Insufficient Human Capital Accumulation ...... ............... 764 III. Key Policy Conclusions ............ ......................... 766 References ................................................. 766 Table 1: Total Wedge on Payroll (as percent of wages) ....... .......... 757 Table 2. Marginal Rates of Return on an Additional Year of Education (percent for urban males age 25 to 60) ..... ............. 764 27. Gender ............................................... 767 I. Introduction ............................................. 767 II. Advances Related to Gender .................................. 768 Demographics and General Health .......................... 768 Reproductive Health ..................................... 769 Education and Literacy ................ ............... 770 Labor Force Participation ............... 770 Wage Gaps ............... ............................ 771 Rural Development ............... 772 III. Gender Issues as They Relate to Colombia's Economic and Social Development ................. 772 Conflict and Violence ................. 772 Displacement ................. 774 Domestic Violence .............. ....................... 774 Drugs, Alcohol Abuse, and Violence . ....... ............. 775 Reproductive and Sexual Health .......... .................. 775 Male Dimensions of Reproductive and Sexual Health ..... ........ 777 Family Size and Poverty ................................... 778 Educational Attainment . .................................. 778 Household Structure ..................................... 779 IV. Conclusions and Policy Recommendations ....... ................ 780 Crime and Violence ..................................... 781 Reproductive and Sexual Health ......... ................... 782 CONITENTS XXIii Support to Female Single-Parent Families .... ........... .... 782 References ..... . 783 28. Indigenous Peoples and Afro-Colombian Communities . 787 I Colombia: A Pluriethnic and Multicultural Nation . ........... . 788 The Indigenous Population ................. .............. 788 The Afro-Colombian Population ................. ........ 788 Progress in Recognition of the Rights of Ethnic Groups .789 11. The Rights of Ethnic Groups in the Context of the Political Conflict and Economic Criss .............i............ .. ....... 790 Surveying and Consolidating the Territorial Rights of Ethnic Groups ............... .......................... 790 Natural Resources ....... . ........................... . 794 Territorial Pressure on the Indigenous Homelands and Collective Territories of the Afro-Colombian Communities .796 The Vision of the Future for Ethnic Groups ....... .. ........ 803 III Policy Recommendations ................. ..... 805 Respect for New Models for Living Together and Human Rights Protection for Ethnic Groups ..... . ... . .......... . 805 Protection of Social, Economic, and Cultural Rights of Ethnic Groups ........................................ 806 Conservation and Sustainable Use of Natural Resources of the Collective Territories and Indigenous Homelands ...... . ..... 808 Annex I Legal Advances in Colombia as Related to the Rights of Ethnic Groups ..... ......... ..... ............. 810 Annex II ..... .............................................. .814 Annex III ........... ........ ................. .......... 818 Annex IV. . .. ...... ...... ........ ....... ..... .. 821 References. ...... 823 Box 1. National Development Plan for the Afro-Colombian Population: Building a Pluriethnic Nation: Poverty and Human Development Indicators, 1998-2002 ..................... ..... .... 790 Table 1. Legal Situation of Indigenous Land Tenure in Colombia, December 2001 ......................... .... ..... 792 Table 2. Comparison of the Legal Situation of Indigenous Land Tenure in Colombia Between 1980 and 2001 .............. ....... 793 Table 3. Natural Parks that Overlap with Indigenous Territories . 795 Table A.II.1. Ethnic Groups and Indigenous Population by Department, 2001 ...................... ........ 814 Table A.II.2 Ethnic Groups of Colombia, Population and Area, 2001 . 815 Table A.III. 1 Some Basic Demographic Indicators in Four Indigenous Regions, Pifieros Marion, and Ruiz Magda, 1998 .. ..... 818 xxiv COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE Table A.III.2 Colombia, Population and Indigenous Area and the Population and Total Extension by Department, December 2001 ...... 819 Table A.IV.1 ICN Resources Assigned to Homelands 1994-2001 (2001 pesos) ......... ............................ 821 Table A.IV.2 Distribution of Resources by Fiscal Sector Prioritized by Indigenous Authorities, 1994-99 ....... .. .......... 822 29. Forced Internal Displacement .............................. 825 I Summary .. .... .................................... 825 II. Background ......... 826 Causes of Displacement ............... 828 Responsible Parties .. ................................... 829 Estimation of the Magnitude of the Problem ........ ........... 829 Characteristics of the Displaced Population ......... ........... 832 Lost Property ..... .................... ...... .. .... 833 Areas Affected by Displacement .... . ....................... 833 Consequences of Displacement ......... ............ 834 The Government's Response ....... .............. 835 Other Institutions Working on the Issue .......... ............ 836 III. Main Issues ...... ...................................... 837 Issue 1: Lack of Priority to the Problem of Forced Displacement both Nationally and Internationally ... .................... 837 Issue 2: Lack of Preventive Measures to Reduce or Mitigate the Impact of the Armed Conflict, the Main Cause of Displacement . 838 Issue 3 Insufficient Understanding of the Size and Characteristics of the Internally Displaced Population ...................... 839 Issue 4: Little Success in Reestablishment and Socioeconomic Stabilization of the Population ......................... 840 IV. Recommendations ........... ............................. 841 Recommendation 1: Make the Problem of Forced Displacement a Priority and Cover the Displaced Population According to their Needs in the Governmental Programs ........ ............. 842 Recommendation 2: Prevent the Risks of Displacement by Implementing Effective Measures at the Local Level and Protecting the Assets of Displaced People ........... ........ 843 Recommendation 3: Adopt a Common Methodology Among Key Stakeholders for Estimating the Size and Characteristics of the Internally Displaced Population ....... .............. 843 Recommendation 4: Create Realistic Incentives and Opportunities for the Reestablishment-Return, Relocation, Integration-of the IDP ........................................... 844 References ..... 846 Figure 1. Urban-Rural Population ........................ 827 CONTEN1'S xxv Figure 2. Displaced Population Accumulated Between 1985-2000 ... 830 Figure 3. Plan of Action for the Prevention and Attention of the Displacement ... ... .. .. ............... 836 Table 1. Internally Displaced Population ............ .. . . 830 Table 2. Displaced Population by Country . .... ... ........... 831 30. Natural Resources and Global Commons: Capital for Sustainable Development ............................ 849 1. Background ................. .. ... 849 Contribution to Global Biodiversity . . . . 849 Contribution to Global Carbon Sinks . . . . 850 Contribution to CO2 Mitigation . . . . 850 Contribution to International Waters . . . ... 851 11. Past and Recent Policy Initiatives . . .852 Biodiversity Strategy . . .. .. 852 Regime for Land Conservation .. . . . . 854 National Strategy for Use of the Clean Development Mechanism (CDM) ........ .............. ... ... 855 III. Main Issues .. . ............ .... . ........... 856 Unsustainable Use of Biodiversity (Forced Extinction of Species and Habitats .. .......... .. . . . . ........... 857 Poor Understanding of Carbon Sequestration and Mitigation Potential .. ... . .. .... ... ... .. ... 859 Poor Linkage Between Global Issues and Local Development . .. 860 Social Conflict and Use of Natural Resources .... 861 IV. Policy Recommendations . . . . 862 Invest in Conservation and Sustainable Use of Natural Capital .... 862 References ..... 866 Figure 1. Gross Potential of Annual CER Exports from Colombia (excluding oil) ... .. ....................... 852 Building a Government of Quality 31. At the Crossroads of Decentralization, Recentralization, Federalization ..... 871 Introduction The Gap Between Objectives and Achievements ...... .. 871 I Colombia A Country Committed to Decentralization Groundbreaking Reforms, and a Precedent-Setting Decentralization Framework .... .. 872 Unsatisfactory Fiscal Results ...... . .. ............ ... 875 Insufficient Capacity Strengthening . . .. .......... .. 876 Decentralization Did Not Meet Democracy and Governance . 877 xxvi COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE II. Factors that Account for Partial Failure of Decentralization in Colombia ............................................. 877 Fiscal Decentralization Factor ....... ....................... 877 Internal Conflict ........................................ 885 Weak Governance at the Central Level ........................ 887 III. Pros and Cons of Main Policy Courses ...... .................... 888 Option 1: Partial, Explicit Reversal of Decentralization ............ 889 Option 2: Streamlining and Reinvigorating Decentralization ........ 890 Option 3: Let Policies other than Decentralization Silently Operate the Reform of the Decentralization Process ............ 891 IV. Recommendations ......................................... 892 Recommendations Regarding Fiscal Decentralization Issues .... .... 892 Proceed with Market-Oriented Regulations for Subnational Borrowing .................. ....................... 893 Recommendations Regarding Administrative Decentralization Issues .............................................. 894 Recommendations Regarding Political Decentralization Issues ....... 894 Recommendations Regarding Context Factors for Decentralization in Colombia ......................................... 895 32. The Judiciary ........................................... 897 I. Background .............................................. 897 II. Enhancing the Constitutional Framework ...... ................. 899 III. Issues Related to the New Constitutional Framework ..... .......... 902 Increased Juridical Insecurity ..... ......................... 902 The Growing Governance Quagmire ........................ 903 Managing the Inefficiencies of Success: The Protection Action ...... 903 Reducing Impunity ...................................... 905 Increasing Efficiency, Quality, and Integrity ..... ............... 913 IV. Conclusions and Recommendations ....... ..................... 926 References ................................................. 928 Figure 1. Labor Justice Demand (1994-99) ........... .............. 904 Figure 2. Protection Actions on Labor Matters (1995-99) ..... ......... 904 Figure 3. Budgetary Appropriations (1992-2000) ........ ............ 906 Figure 4. Workload of Preliminary Investigations ....... ............. 908 Figure 5. Output at the Criminal Instruction Phase ........ ........... 908 Figure 6. Criminal Justice Demand ............................... 909 Figure 7. Evolution of the Homicide Rate in Colombia, 1993-99 ........ 909 Figure 8. International Comparison on Homicide Rates, 1998- .... ....... 911 Figure 9. Recurrent Expenditure of the Judiciary, 1992-98 . .......... 914 Figure 10. Investment Expenditures of the Judiciary, 1992-98 .... ........ 914 Figure 11. Institutional Capacity, 1998 ............. ................ 915 Figure 12. Comparison of Caseloads in Andean Countries, 1992-98 ....... 916 CONTENTS xxvii Figure 13. Judiciary's Budgetary Allocations, 1998 ..... . ........ 916 Figure 14. Workload Evolution Ordinary Jurisdiction, 1993-99 .917 Figure 15. Average Process Duration Civil Procedures .918 Figure 16. Critical Factors Affecting the Quality of Justice .. . .. ..... . 919 Figure 17. Critical Factors Affecting the Quality of Justice .. . 920 Figure 18. Changes in Stock of Civil Cases (by type of proceedings), 1992-98 ... .................................. 921 Figure 19. Public Perceptions on the Independence of the Judiciary .. . ...922 Figure 20. Entrepreneur Perceptions on the Independence of the Judiciary . . . 923 Figure 21 Evolution of Judicial Branch's Expenditure .. . . 924 Table 1. Judicial Human Resources (1985-95) .907 33. Corruption, Institutional Performance, and Governance: Developing an Anticorruption Strategy for Colombia .931 I Background ........................... ... ... ....... 931 11. Main Conclusions of the Empirical Study. ....... ............... 932 State Capture and Governance .............................. 933 Moderate Incidence of Administrative Corruption ..... ....... 941 The Purchase of Public Administration Posts ..... .. ... 941 The Perception of Institutional Performance .... . . . 941 Institutional Performance .............................. 944 Presence of Corrpution at the Regional Level ......... .......... 946 Judicial Power and Corruption ... 946 Greater Impact of Administrative Corruption on the Poor ..... .... 948 III. An International Comparison of Governance ... ..... .......... 948 IV. A Strategy for Integrity .. .. .. . 950 References .......... . .. ....... 955 Figure 1. Corruption: Comparison of Colombia, Ecuador, Honduras, and Peru. ....... 934 Figure 2. Level of State Capture/Undue Ability of Elites to Influence High-Level Decisions of Authorities in Colombia, Honduras, and Peru (according to companies, 2001) .935 Figure 3. How Much Ability Do the Following Actors Have To Use Bribes To Influence the Decisions of High-Level State Authorities? . . 936 Figure 4. Meritocracy versus Presence of Corruption (Central Level) . 938 Figure 5. Level of Politicization versus Presence of Corruption . 939 Figure 6. Restrictions on Competition for Public Contracts .... . ..... 940 Figure 7 Quality and Presence of Corruption in Public Services .942 Figure 8 Quality and Presence of Corruption in Public Services . . 943 Figure 9. The Ten Most-Honest Public Institutions: How Dishonest/ Honest Is Each of the Following Institutions? .944 Figure 10. The Ten Least-Honest Public Institutions: How Dishonest/ Honest Is Each of the Following Institutions? .... ...... 945 xxv;Ii COLOMBIA THE ECONOMIC FOUNDATON OF PEACE Figure 11. Evaluation of the Judicial Sector .........................4947 Figure 12. Indicators of Governance in Colombia: 1998-2001 (based on more than 175 countries) ................ .... 949 Figure 13. Governance Indicators for 2001 (based on a study of 175 countries) .. 950 Table 1. Strengths and Weaknesses of Public Institutions, Colombia . 946 Table 2 Corruption and Level of Government .......... 947 34. Reform of Public Administration and of the State in Colombia . 957 I. Background to Current Challenges.957 II. Government Proposal for a New Public Sector Reform .............. 961 Strategic Options .................................... ..... 961 Concept of Reform .. .. ................. 965 Objectives, Strategy, and Goals of the Reform Agenda .. ......... 966 Choice of Instruments, Risks, and Costs ............ 967 Stages of Reform ............. ......................... 974 Formulation Unit and Management Unit for the Reform .975 Monitoring, Evaluation, Dissemination .976 III. Synthesis of Recommendations .......................... ... 979 Table 1. Examples of Objective and Goal Definition ... ........... ... 968 Table 2. Tools, Projects, and Actions Regarding the Transversal Reforms of the Colombian Public Administration (with multilateral financing) .973 35. Budgetary Institutions .981 I. Introduction ............................................. 981 II. Budgets and Fiscal Balance .............................. 982 The Issue ...... ................................... .. 982 Ongoing Actions ....................................... 983 Recommendations ............... ....................... 983 III. Control of the Budget . . 984 The Issue ................. 984 Ongoing Actions ................. 985 Recommendations ......... ........... ................. 985 IV The Planning and Budgetary Process ... 986 The Issue ......... .. . . .. ............ .... 986 Ongoing Actions ........................................ 987 Recommendations ............... ....................... 987 V. Budgetary Inflexibility ...... . .. ........................... 988 The Issue ............ ............................... 988 Ongoing Actions . 988 Recommendations ......................... ... ... 989 VI. Organizational Adjustment ................. ................ 990 CONTENTS XXIX The Issue ... .... .............. ........ 990 Ongoing Actions. ......... ........... ......... ... 990 Recommendations ... ........ .. .... 991 VII Information Management and Use . ........ ........ 992 The Issue ....... . ......... . . ........ .. ...... 992 Ongoing Actions ...... ........ ..... ... 992 Recommendations .. ........ ................ 993 References ................. ............ ........... . ... 993 Table 1. National Budget Arrears, 1990-2002 ....... . ....... ... 982 Table 2. Budgetary Inflexibilities ... . ......... ........... .. 989 Acknowledgments This volume is the result of a team effort and, as such, it has benefited from an array of invaluable contributions. Our thanks are therefore due to a large number of peo- ple. First, the chapters' authors, who have provided not just material of outstanding technical quality but a remarkable commitment to enriching the policy debate in Colombia. We consider ourselves fortunate to share this book with these principle authors-Jairo Arboleda, Adolfo Brizzi, Krishna Challa, Loic Chiquier, Alberto Chucca Mora, Vivian Foster, Vicente Fretes Cibils, Eleoterio Codato, Elena Correa, Maria Correia, Mauricio Cuellar, Carlos E. Cuevas, Mario Adolfo Cuevas, Shelton Davis, Klaus Deininger, Clemente Luis Del Valle, Marfa-Luisa Escobar, Eduardo Fernandez (consultant), Christina Garcia, Elsie Garfield, Jose Gil-Diaz (consultant), Daniele Glovannucci (consultant), Natalia Gomez, Lauritz Holm-Nielsen, Isabel Lavadenz, Menahem Libhaber, Matthew McMahon, Alexandra Ortiz, Panagiota Panopoulou (consultant), Vicente Paqueo, Zeinab Partow, Thakoor Persaud, John Pollner, Juanita Riafio (consultant), Laura Rawlings, Fernando Rojas, Juan Pablo Ruiz, Felipe Saez, Enrique Sanchez (consultant), Mario F. Sangines, Lisa Taber, Car- los Eduardo Velez, Eduardo Velez Bustillo, Walter Vergara, Eloy Vidal, Hermann von Gersdorff, Steven B. Webb, and Eduardo Zolezzi. All authors are affiliated with the World Bank Group unless otherwise indicated Other contributors to individual chapters are recognized in the credits of each specific chapter. While this book reflects the authors' own views (and not necessarily the views of the World Bank, its Board of Executive Directors or its member countries), its pro- duction was institutionally housed at the World Bank We thus benefited enor- mously from the general guidance of Guillermo Perry (Chief Economist for the Latin America and Caribbean Region), and from the auspices of the office of David de Ferranti (Vice-President for Latin America and the Caribbean Region). We also recognize the importance of and thank the participants at the workshop held on July 27, 2002 in Bogota. This workshop not only brought together a major- ity of the authors under one roof for a day of candid discussions but also, and more importantly included Colombian officials of both the outgoing and incoming administrations. Of the Colombian officials at the workshop, we especially wish to thank: Maria Consuelo Araijo, Carolina Barco, Jorge Humberto Botero, Carlos Gustavo Cano S., Alberto Carrasquilla, Luis Ernesto Mejia Castro, Andres Uriel Gallego Henao, Alejandro Gaviria, Gustavo Gaviria, Roberto Junguito, Fernando Londofno Hoyos, Beatriz Londofio, Juan Luis Londofio, Santiago Montenegro, Juan xxxi XXXii COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE Ricardo Ortega, Martha Pinto de De Hart, Jaime Ruiz Llano, Martha Lucia Ramirez, Cecilia Rodriguez Gonzalez-Rubio, Francisco Santos, Sandra Suarez, Miguel Urrutia, Alberto Velasquez, and Cecilia Maria Velez. Their comments, sug- gestions, and inputs along with those from the many others who attended this work- shop added greatly to this volume. Finally, we are especially grateful to Michael Geller, who so ably managed the production and support team-from initial conception and planning to final lay- out, type-setting, and dehvery. At various stages, that team included: Cristiam Rodriguez and Christopher Humphrey, who assisted in the actual preparation of various chapters; Liliana Wiesner, who organized the initial two-day country team meeting in Washington on February 25-26, 2002; Maria Teresa de Henao, who at various times before, during, and after the course of this project gave much time and energy to support its successful completion; Cristina CGfuentes, Jenner Guzman, Daniel Reyna, Julieth Rodriguez, and Adriana Salazar-all of the World Banks Bogota office-who gave many hours particularly on July 27, 2002 to create the proper environment for discussions between the Colombian officials and the Bank staff; Vinh H. Nguyen, who acted as our budget advisor; Diane Stamm, the princi- pal language editor; Santiago Pombo-Bejarano, Brenda Mejia and Thaisa Ysonde Tiglao, who spearheaded the printing efforts in English and Spanish; Richard Creighton, Brenda Waugh, Judy Gibson and the staff of the Magazine Group, who created the design and layout of the volume; and Maria Victoria de Robayo and Natalia Bonilla Maldonado of Museo Nacional de Colombia in Bogota and Ana Escallon of the OAS Museum of the Americas in Washington, D.C. who greatly assisted in contacting Diego Obreg6n to secure the art of his father, Alejandro Obreg6n, for use on the cover of this volume. Our thanks to all of them. Marcelo M. Giugale, Olivier Lafourcade, and Connie Luff Washington D.C. and Mexico City October 2002 Preface Colombia faces a future of challenge-and of opportunity. Long besieged by con- flict, it has been unable to free itself from the shackles of poverty Rich in human and natural endowments, the country has yet to become a prosperous, equitable nation. The scourge of violence has repeatedly tested the resilient effort of its peo- ple Even its impressive record of economic growth seems less remarkable when one sees the poor, the homeless, and the refugees. It need not be like that. And the time has come to leave the past behind. The resounding election victory of President Uribe on May 26, 2002 and his underlying mandate for change have opened the door for Colombians to think beyond war -to articulate a shared vision and a path toward it Economic devel- opment will be an integral part of that path. Jobs, education, and a clean environ- ment need to be as much a foundation of peace as security and political agreement. The World Bank was thus honored by the new authorities' invitation to publish this set of policy notes which, from an independent point of view, provides an account of Colombia's development agenda-the problems, many of the reasons behind those problems, and some options to address them. The analysis here does not pro- vide definite answers. Rather, the intention is to contribute to the national debate, to Colombians' own search for solutions. The required reforms would be difficult for any government. For a government trying at the same time to end a decades-old war, they are even harder. Current regional and international conditions are not facilitating matters, and are unlikely to facilitate them in the near future. The financial crisis in Argentina is restricting access to external funding for other emerging market countries, including Colom- bia. Sagging global growth is dampening Colombia's exports. Prospects of military action in the Middle East have heightened the volatility of oil prices. Terrorism and its financing have fostered the illegal drugs trade. Yet, the cost of inaction seems unbearably high- without reform, the Colombian economy risks spiraling into a cycle of recession, increasing poverty, and intensifying conflict. The government of President Uribe is determined to address, head on, the dou- ble challenge of peace and reform. In the few months since taking office, the new administration has rallied support for some major changes in policy direction-for example, in the fiscal area. We at the World Bank stand ready to support the gov- ernment's efforts. This is central to our institution's mission of poverry reduction. And, more personally, it is central to our staff's desire to see Colombia succeed. xxxiii xxxiv COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Finally, I wish to express our gratitude to the Colombian authorities for giving us the privilege of being part of their country's quest for development, and to the edi- tors, authors and producers of this book. David de Ferranti Vice President Latin-America and Caribbean Region Washington, D.C. October 2002 Editor Biographies MARCELO GIUGALE, an Argentine/Italian national, holds a PhD and a MSc in Eco- nomics from The London School of Economics, and a B. A. in Economics from Universidad Catolica Argentina. After a spell in academia, he joined the World Bank in 1989 as an economist in its financial research department. From 1990 to 1994, he was a Senior Economist in the Middle East Operations Vice-presidency, super- vising Egypt's structural adjustment program and leading the Bank's reconstruction work in post-war Lebanon. From 1994 to 1998, Mr. Giugale was a Principal Econ- omist in the Eastern Europe and Central Asia Region, responsible for the Bank's lending and analytical economic work in Lithuania and Kazakhstan. In September 1998, he became the Lead Economist for the Colombia-Mexico-Venezuela Depart- ment. He is currently the Director of the Bank's Andean countries Department. He held teaching positions at the London School of Economics and the American Uni- versity in Cairo, and has published in the areas of applied econometrics, finance, business economics and economic development. OLIVIER LAFOURCADE, a French nationial, is Director of the Colombia-Mexico- Venezuela Country Management Unit of the World Bank, located in Mexico City. Mr. Lafourcade joined the World Bank in 1973. A graduate of Ecole Nationale Superieure Agronomique of Rennes (France), and holding MSc and Ph.D degrees in Agricultural Economics from the University of Maryland (U S.A), he started a career in the Bank working largely on agriculture and rural development projects in Bolivia, Mexico, Brazil, Cameroon. From 1980 to 1982, he was Personal Assistant to two successive Bank presidents (Robert McNamara and Tom Clausen). He then was responsible for the Bank's agricultural programs for several African countries (Zaire, Rwanda, Burundi) before taking charge of the agricultural program for India. He became Director of the European Office of the World Bank in Paris in 1988. In 1992, he moved back to Washingtoni to become Director of the West Cen- tral Africa Department (Benin, Burkina Faso, Cote d'Ivoire, Ghana, Niger, Nigeria and Togo), before taking his assignment in Mexico in September 1997. Before join- ing the Bank, Mr. Lafourcade worked in Argentina for two years as a researcher in the Argentine National Agricultural Research Institute CONNIE LUFF, a United States national, is a Senior Operations Officer in the Colombia-Mexico-Venezuela Country Management Unit, located in Mexico City. xxxv xxxvi COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Since joining the World Bank in 1988, she has spent most of her time managing projects and working on operational activities such as developing country assistance strategies, enhancing project quality enhancement and managing project portfolios. She has worked extensively in Latin American and Africa. She has an MBA from ESADE in Barcelona, Spain and a BA from Michigan University. Acronyms and Abbreviations ACODAL Asociaci6n Colombiana de Ingenierfa Sanitaria y Amblental ACOLGEN Asociaci6n Colomblana de Generadores, Syndicate of Colombian Electricity Generators ACP Asociaci6n Colombiana de Petroleo, Association of Oil and Gas Companies ACPM p. 14 of Taxation Note ADBs Agricultural Development Bonds ADT Average daily traffic AFP Administradora de Fondos de Pensiones, Pension Fund Managers AGD Accountant General's Department AGO Attorney General's Office ALM Asset and Liability Management ALMACAFE Almacenes Generales de Dep6sito de Cafe S.A. APU Agricultural Production Unit ARP Administradoras de Riesgos Profesionales, Administrators of Professional Risks ARS Administradora de Regimen Subsidiado, Administrator of the Subsidized Regime ASOCODIS Asociaci6n Colombiana de Distribuidores, Colombian Distribution Companies Association AVSC Association for Voluntary Surgical Contraception BANCAFE Commercial Bank in Colombia BANCOLDEX Banco de Comercio Exterior de Colombia, S.A. (the export bank) BANRURAL Banco Nacional de Credito Rural (Mexico) BaR Budget-at-risk BCH Banco Credito Hipotecario BIBOR Bogoti Inter-Bank Operations Rate BOD Biochemical Oxygen Demand BOMT Build, operate, maintain, and transfer contracts BOT Build-Operate-Transfer BRI Bogota River Interceptor xxxvii xxxviii COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE BSE Bovine spongiform encephalopathy ("mad cow" disease) B2B Business-to-business CAA Centro de Atenci6n Ambulatoria, Centre for Ambulatory Care CAF Corporaci6n Andina de Fomento CAIPs Centros de Atencion Integral al Preescolar CAJANAL Caja Nacional de Previsi6n Social, Government Employees Pension Fund CAMEL Capital/Assets/Management/Earnings/Liquidity (an international prudential and regulatory diagnostic tool for assessing financial soundness of banks) CAN Comunidad Andina CAPRECOM Caja de Previsi6n Social de Comunicaciones CARBOCOL Carbones de Colombia CARs Corporaciones Autonomas Regionales, Autonomous Regional Corporations CAS Country Assistance Strategy CASA Corporaci6n Casa de la Mujer CASEN/93 Encuesta Nacional de Caracterizaci6n Socioeocon6mica 1993, National Survey of Socio-economic Characteristics 1993 CAV Corporaciones de Ahorro y Vivienda, Housing Savings and Loan Associations CC Cajas de Compensaci6n, Compensation Associations CCF Caja de Compensaci6n Familiar, Family Compensation Fund CCI Corporaci6n Colombia Internacional CDM Greenhouse gas mitigation CEDAW United Nations Committee on the Elimination of Discrimination Against Women CEGA Centro de Estudios Ganaderos y Agricolas CEJ Corporaci6n Excelencia en la Justicia CENCOA Central de Cooperativa Agraria, Central Agrarian Cooperative CENICAFE Colombian Coffee Research Center CEPAL Comision Econ6mica para America Latina y el Caribe CER Carbon emission reductions CESU Consejo Nacional de Educact6n Superior CGR Comptroller General CIGADE A UNDP database CISA Central de Inversiones, S.A. CMO Collateralized Mortgage Obligation ACRONYMS AND ABBRENIATIONS xxxix CNA Consejo Nacional de Acreditaci6n CNDM Comision Nacional de Doctorados y Maestrias CNSSS Consejo Nacional de Seguridad Social en Salud, National Council of Social Security in Health CODENSA The Bogoti distribution company CODHES Consultorfa para los Derechos Humanos y el Desplazamiento, Bureau on Human Rights and Displacement COLCIENCIAS Fundaci6n Colombiana Francisco Jose de Caldas para el Fomento de las Ciencias CONFECOOP Confederaci6n de Cooperativas CONFIS Consejo Superior de Politica Fiscal CONPES Comisi6n Nacional de Politica, Econ6mica y Social, National Commission for Economic and Social Policy COP Colombian pesos CORPOICA Corporaci6n Colombiana de Investigaci6n Agropecuaria CPI Consumer price index CPS Caja de Previsi6n Social, Social Health Insurance Fund CRA Comisi6n Reguladora de Agua Potable y Saneamiento Basico, Water Regulatory Commission CRE Comisi6n de Regulaci6n de Electricidad CRECE Centro Regional de Estudios Cafeteros y Empresariales CRLP Center for Reproductive Law and Policy CRT Telecommunications Regulatory Commisslon CREG Comisi6n de Regulaci6n de Energia y Gas DANE Departamento Nacional de Estadisticas, National Department of Statistics DANSOCIAL Departamento Administrativo Nacional del Sector Social DAPS Potable Water and Sanitation Directorate DAS Departamento Administrativo de Seguridad DB Defined benefit DC Defined contribution DDT Direcci6n de Desarrollo Territorial del Departamento Nacional de Planeaci6n, Ministry of Territorial Development of the National Department of Planning DESEPAZ Program Programa Desarrollo, Salud y Paz DGCP Direcci6n General de Cr6dito Publico, General Directorate for Public Credit DGPAD Direcci6n General para la Prevenci6n y Atenci6n y de Desastres DHS Demographic Health Survey DIAN Direcci6n de Impuestos y Aduanas Nacionales, National Tax and Customs Administration xI COLOMBIA THE ECONOMIC FOUNDATION OF PEACE DNP Departamento Nacional de Planeaci6n, Department of National Planning DNPAD National Department of Prevention and Attention of Disasters DRI Desarrollo Rural Integral DTF Colombia's benchmark interest rate DTP3 Diphteria-Tetanus-Pertussis Vaccine EAAB Empresa de Acueducto y Alcantarillado de Bogota EAP Economically active population EC European Community ECAT Traffic Accidents and Catastrophic and Terrorist Events ECD Early childhood development ECHO European Union Humanitarian Aid Office ECOGAS Empresa Colombiana de Gas, a public gas transmission enterprise ECOPETROL Empresa Colombiana de Petr6leos, State oil company ECOSALUD Empresa Colombiana de Recursos para la Salud ECV Encuesta de Calidad de Vida, Survey of Living Conditions EDA Enfermedad Diarreica Aguda, Acute Diarrhea Disease EDUSAT Mexico's Educational Broadcast System EEB Empresa de Energia de Bogota (Bogota electricity holding company-controls EMGESA and CODENSA) ELN National Liberation Army EMAs Environmental Management Agencies EMCALI Empresas Municipales de Cali (Cali water, electricity and telephone distribution municipal company) EMGESA Generation company under EEB EN Escuela Nueva ENCV/97 Encuesta Nacional de Calidad de Vida 1997, National Survey of Quality of Life 1997 ENH Encuesta Nacional de Hogares, National Household Survey ENH/2000 Encuesta Nacional de Hogares 2000, National Household Survey 2000 EPL Expected loss EPM Empresas Piblicas de Medellin EPSs Empresas Promotoras de Salud, Entities of Health Promotion ESE Empresas Sociales del Estado ESMAP Energy Sector Management Assistance Program ESP Autonomous public service enterprises ACRONYMS AND ABBREVIATIONS xli ESS Empresa Solidaria de Salud, Health Solidarity Entity ESTA Energy Sector Technical Assistance project ETB Empresa de Telecomuniaciones de Bogoti ETIS Entidad Territorial Indigena, Territorial Indigenous Entity EU European Union FAEP Petroleum Stabilization Fund, Fondo de Ahorro y Estabilizaci6n Petrolera FAG Fondo Agropecuario de Garantfas FAO Food and Agriculture Organization of the United Nations FAOSTAT Food and Agricultural Organization Statistics FARC Fuerzas Armadas Revolucionarias de Colombia FEC Fondo de Cr6dito Educativo FEN Financiera Energetica Nacional (the energy sector bank) FES Fundaci6n para la Educaci6n Superior, Financiera (a finance company for educational funding) FINAGRO Fondo para el Financiamiento del Sector Agropecuario FINDETER Financiera de Desarrollo Territorial FIRA Fideicomiso Instituido en Relacion con la Agricultura (en el Banco de Mexico) FNC Fondo Nacional de Cafe, National Coffee Fund FNC Fondo Nacional de Calamidades FNUAP Fondo de Poblacion de las Naciones Unid FODESEP Fondo de Desarrollo de Educaci6n Superior FOGAFIN Fondo de Garantfas de Instituciones Financieras, Financial Institutions Guarantee Fund FONADER Fondo Nacional para el Desarrollo Rural FONCOLPUERTOS Pension fund of port workers FONECAFE Costa Rica National Fund for Coffee Stabilization FONPET Fondo de Pensiones Territoriales, National Fund for Regional Pensions FONPRECON The pension fund for members of Congress FOPEP Fonde de Pensiones Publicas FOREC Fondo para la Reconstruccion y el Desarrollo Social del Eje Cafetero FOSIT Programa Para el Fortalecimiento del Sistema de Informaci6n Financiera Territorial, Territorial Financial Information System FOSYGA Fondo de Solidaridad y de Garantfa, Solidarity Fund FRECH Fondo de Estabilizacion de la Cartera Hipotecaria FSAP Financial Sector Assessment FSL Fiscal Strengthening Loan Matrix xiii COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE FSOPEN Fund for Pension Solidarity FSP Pension Solidarity Fund GDP Gross domestic product GEF Global environment facllbty GNP Gross national product GPS Global Positioning System GWh Gigawatt-hour HBI Hogares de Bienestar Infantil HCBs Hogares Comunitarios de Bienestar IAvH Instituto de Investigaciones Alexander von Humboldt ICA Impuesto de Industrias y Comercio ICA International Coffee Agreement ICBF Instituto Colombiano de Bienestar Familiar, Colombian Family Welfare Institute ICETEX Instituto Colombiano de Credito Educativo y Estudios Tecnicos en el Exterior, The National Scholarship Institute ICFES Instituto Colombiano de Fomento para la Educaci6n Superior ICO International Coffee Organization ICR Incentivo a la Capitalhzaci6n Rural ICRC International Committee of the Red Cross ICT Information and communication technologies ICVs Indices de Calidad de Vida, Quality of Life Indicators IDB Inter-American Development Bank IDEAM Instituto de Mercadeo Agropecuario, National Hydrology, Meteorology and Environmental Studies Institute IDEMA Instituto de Mercadeo Agropecuario IDP Internally displaced population IEA International Association for the Evaluation of Education IFI Instituto de Fomento Industrial (the State Industrial Development Bank) (the national development bank) IGAC Instituto Geografico Agustin Codazzi, Geographic Institute 11 Information infrastructures IICA Instituto Interamericano de Cooperaci6n Agricola IIED International Institute for Environment and Development ILO International Labour Organization IMF International Monetary Fund INAS Instituto Nacional de Salud ACRONYMS AND ABBREVIATIONS Xliii INAr Instituto Nacional de Adecuaci6n de Tierras INCORA Instituto Nacional Colombiano de la Reforma Agraria, National Colombian Institute of Agrarian Reform INDEMUN Indice de Desarrollo Municipal, Municipal Development Index INDERENA Instituto Nacional de Recursos Naturales Renovables y Medio Ambiente, National Institute of Natural Renewable Resources and Environment INGEOMINAS Instituto de Geociencias y Minas INS Instituto Nacional de Salud, National Institute of Health INURBE Instituto Nacional para la Vivienda de Interes Social y la Reforma, Urbana; National Institute of Social Interest, Housing, and Urban Reform INVIAS Instituto Nacional de Vias, National Highway Institute IOM International Organization for Migrations IPC Consumer price index IPD Implicit pension debt IPPF International Planned Parenthood Federation IPR Intellectual property rights IPS Instituciones Prestadoras de Salud IRA Infecci6n Respiratoria Aguda, Acute Respiratory Infection ISA Interconexi6n Electrica S A. ISAGEN ISA Generaci6n, a major, state-owned, power producer ISS Instituto de Seguro Sociales, Institute of Social Security IUD Intrauterinie device kbpd Thousands of barrels per day kWh kilowatt-hour LAC Latin America and the Caribbean LFR Law of Fiscal Responsibility LIBOR London Inter-Bank Operations Rate LPG Liquified petroleum gas MADR Ministry of Agriculture and Rural Development MAFP2 Proyecto de Modernicaci6n de la Administraci6n Financiera del Sector Piblica 11 Mbbl Millions of barrels MBS Mortgage-backed securities Mcfd Millioni cubic feet per day MDE Ministerio de Desarrollo Econ6mico MED Ministry of Economic Development MEN Ministerio de Educaci6n, Ministry of Education MHz Megahertz xliv COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE MISs Management Information Systems MMA Ministerio de Medso Ambiente, Environment Ministry MME Ministry of Mines and Energy MOE Ministry of the Environment MOFPC Ministry of Finance and Public Credit MOH Ministerio de Salud, Ministry of Health MT Metric tons MT Ministry of Transport MW Megawatts NBI Indice de Necesidades Basicas Insatisfechas, Index of Unsatisfied Basic Needs NBPTS U.S. National Board for Professional Teaching Standards NPR National Performance Review NCF National Coffee Fund NFCG National Federation of Coffee Growers NFPS Nonfinancial public sector NGO Nongovernmental organization NIS National Innovation System O&M Operation and maintenance OAPF Old Age Poor Fund OECD Organization for Economic Cooperation and Development OLADE Organizaci6n Latino Americana de Energia OMA Displaced Pop OMS Organizaci6n Mundial de la Salud, World Health Organization ONAD Oficina Nacional para la Prevenci6n y Atenci6n de Desastres OREALC Oficina Regional de Educaci6n para America Latina y el Caribe PAC Plan Anual de Caja PAC Plan de Atenci6n Complementaria, Complementary Health Plan PACES Programa de Ampliaci6n de la Cobertura y Mejoramiento de la Calidad de la Educaci6n Secundaria y Media, Program to Improve Access and Quality in Lower and Upper Secondary Education PAHO Pan-American Health Organization PAI Programa Ampiado de Inmunizaciones, Expanded Programme on Immunization PAYG Pay as you go PCBs Polychlorinated Biphenyls PCS Personal Communication Service ACRONYMS AND ABBREVIAT IONS xlv PCMH Por cada cien mil habitantes, for each hundred thousand inhabitants PDS Planes de Desarrollo y Expansi6n Sectoroal PDs Primary Dealers PDT Planes de Desarrollo Territorial PEIs Proyectos Educativos Institucionales, Institutional Education Projects PFMP Public Financial Management Project PICN Participacion en los Ingresos Corrientes de la Nacion, Participations of Current National Income PLANTE Plan Nacional de Desarrollo Alternativo, Alternative Development Plan PMP Plan de Medicina Prepagada, Prepaid Medicine Plan PNPAD Plan Nacional para la Prevenci6n y Atenci6n de Desastres PNR Plan Nacional de Rehabihtaci6n POAT Environmental zoning plan POI Programa de Obras e Inversiones, Works and Investment Program POS Plan Obligatorio de Salud, Mandatory Health Plan POSS Plan Obligatorio de Salud Subsidiado, Subsidized Mandatory Health Plan POTs Planes de Ordenamiento Territorial, Territorial Land Use Plans (Land Zoning Plans) PRAN Programa Nacional de Reactivaci6n Agropecuaria PRI Programa de Desarrollo Institucional PROAGRO Programa de Oferta Agropecuaria PRONATTA Programa Nacional de Transferencia de Tecnologfa Agropecuaria PSP Private sector participation R&D Research and development RAS Red de Apoyo Social, Social Support Network RED Red de Solidaridad Social R/P Reserves to Production ratio RRPSC Civil Society Reserves System RRSC Civil Society Reserves Network RSS Red de Solidaridad Social, Social Solidarity Network RUT Colombian Bishops Conference's Information System on Displaced Population SABER National Evaluation System for the Quality of Education xIvi COLOMBIA THE ECONOMIC FOUNDATION OF PEACE SAT Sistema de Aprendizaje Tutorial, Tutorial Learning System S&T Science and technology SCAA Specialty Coffee Association of America SCD SISBEN Classification Document SCJ Superior Council of the Judicature SDS Secretaria Distrital de la Salud, District Health Secretary SECAB Secretaria Ejecutiva del Convenio Andres Bello, Executive Secretariat of the Andres Bello Agreement SECF Sistema de Estimaci6n de Desplazamiento Forzado por Fuentes Contrastadas, System to Estimate Forced Displacement through Compared Sources SENA Servicio Nacional de Aprendizaje, National Training Service SF Subsidio Familiar SFAL Structural Fiscal Adjustment Loan SGP Sistema General de Participaciones SIIF Sistema Integrado de Informaci6n Financiera, Integrated System of Financial Information SINA National Environmental System SINAP Sistema Nacional de Areas Protegidas, National Protected Areas System SINERGIA Sistema Nacional de Evaluaci6n de Resultados, Evaluation System for Public Management SINTAP Sistema Nacional de Tecnologia Agropecuaria SIRAP Regional Protected Area System SISBEN Sistema de Selecci6n de Beneficiarios para Programas Sociales, Selection System of Beneficiaries for Social Programs SISDES Sistema de Identificaci6n de Desplazados SNCTA Sistema Nacional de Ciencia y Tecnologia Agroindustrial SNCyT Sistema Nacional de Ciencia y Tecnologia SNPAD National System for the Prevention and Attention of Disasters SNSSS Sistema Nacional de Seguridad Social en Salud, National System of Social Security in Health SPI Comprehensive Production Systems SPNN National Natural Parks System SS Suspended Solids SSAL Social Sector Adjustment Loan SSPD Superintendencia de Servicios Publicos Domiciliarios, Superintendency of Domiciliary Public Services ACRONYMS AND ABBREVIATIONS XIVii SUIFP Sistema Unificado de Inversi6n y Finanzas Plblicas, Integrated Public Investment System SV Superintendencia de Valores TA Technical Assistance Tcf Tera (1012) cubic feet TES Treasury bonds TIMSS Third International Math and Science Study TM TransMilenio UAESPNN National Park System Special Administrative Unit UAFs Unidad Agricola Familiars UASB Upflow Anaerobic Sludge Blanket reactor UFW Unaccounted-for water UMARCO Unidad de Analisis Macro-Economico en el Departamento Nacional de Planeaci6n UMATA Unidad Municipal de Asistencia Tecnica Agropecuaria UN United Nations UNAIDS Joint United Nations Programme on HIV/AIDS UNDP United Nations Development Programme UNESCO United Nations Educational, Scientific and Cultural Organization UNHCR United Nations High Commissioner for Refugees UNICEF United Nations Children's Fund UNSO United Nations Statistical Office UPAC Constant Purchasing Power Unit UPC Unidad de Pago por Capacitaci6n, Capitation Payment Unit UPME Unidad de Planeamiento Minero-Energ6tica URC Unified Registry of Contributors USAID United States Agency for International Development USCR United States Committee for Refugees UTC Unidad T&nica Con)unta, Joint Technical Unit UVR Unidad de Valor Real, Constant Value Unit VaR Value-at-Risk VAT Value added tax VIS Vivienda de Interes Social, Social Interest Housing WAP Working-age population WB World Bank WDI World Development Indicators WFP World Food Program WHO World Health Organization WTO World Trade Organization WS&S Water supply and sanitation services WWF World Wildlife Fund xlviii COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE XL Excess of loss ZNIs Zonas No Interconectadas (non-interconnected zones) ZRCs Zonas de Reserva Campesinas Colombia The Economic Foundation of Peace Synthesis This Chapter was written by Marcelo M Giugale. I. Rationale and Organization It is the practice and the privilege of the World Bank to provide incoming Presi- dential Administrations in its client countries with a comprehensive diagnosis of their development position-and with independent policy recommendations to advance it. That practice appears particularly pertinent for Colombia today. Presi- dent Uribe is to lead a nation of immensely talented people and with abundant nat- ural resources that, for far too long, has been trapped in a deepening circle of vio- lence, destruction, and poverty. There is now an opportunity to break that circle, and this volume is meant to help the new authorities seize that opportunity. This chapter delivers a summarized account-a synthesis-of the development challenges that the incoming President is likely to face, and of the possible policy responses to address them. It distills the four thematic chapters that make up Part I of this book. Those chapters are, in turn, based on 31 sector-specific chapters con- tained in Part II. While the findings and analysis presented here respond to the country's current realities, they have been enriched by five decades of development partnership-the first World Bank mission to visit Colombia took place in 1948, opening the way for some 150 projects and over a thousand pieces of analytical work spanning the entire development spectrum. Yet, the intentioni is not to provide definitive, foreign answers but, rather, to trigger and support policy debate in the country. Below, the central messages in Colombia's development agenda are presented, followed by the diagnoses and policy recommendations that make up that agenda The chapter closes with a suggested path for the prioritization of actions. This book was finalized in August 2002, policy actions that may have taken place after that time are not reflected 2 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE II. Central Messages There is no doubt that the long-standing armed conflict is the most Important issue facing Colombia as a nation, let alone as an economy. The cost of violence is diffi- cult to overestimate.' Since 1980, some 100,000 people have died directly as a result of the conflict, and two million desplazados have lost their jobs, homes and, increas- ingly, hope, and are now lingering at the fringe of society. Another million, perhaps the most educated million, left the country altogether. By some calculations, the conflict dampens GDP growth each year by two full percentage points-in other words, had the war stopped, say, 20 years ago, the income of the average Colombian would now be 50 percent higher, and an estimated 2.5 million more children would be above the poverty line. More fundamentally, the conflict has bred gross underal- locations and misallocatons of resources: frightened by violence, private investment has, over the last 30 years, rarely surpassed one tenth of GDP, and just the lost rev- enue due to sabotaged oil pipelines (some US$500 million per year) would be enough to double the country's annual budget for social assistance. And, behind these costs, lies a less quantifiable but even more disturbing consequence of the con- flict-the dilution of Colombia's "social capital," that is, its citizens' basic trust in societal contracts and institutions that make nations function. It is indeed a tribute to the resilience and ingenuity of the Colombian people that, in spite of so long a conflict, they managed, until 1998, to keep their econ- omy growing every single year for the previous seven decades. While other coun- tries in the region stumbled (notably in the early 1980s), Colombia achieved income, education, and health standards that made it stand out in the developing world. As recently as three years ago, the country whose image had been so tar- nished by drugs-fostered violence, still inspired investment-grade confidence among international financiers. A disturbing message is, however, embedded in that economic resilience-material prosperity alone has not and will not stop the war. But impoverishment may spin it out of control. When the economy collapsed in 1999, the country found itself unable to cater for its people, three million of whom fell into poverty and one in 10 into unemployment, and the many efforts at peacemaking faltered. It is therefore argued here that economic development will be a necessary, but not a sufficient, condition for peace-it will be one of the foundations, perhaps a cornerstone, on which an eventual settlement (and national reconciliation) will stand and from which a new nation will be built. This brings us to the cen- tral question of this book: Given its quest for peace, what is the development agenda that the new administration should follow? The answer has three main components. I Not all violence in Colombia is, however, due to the armed conflict, and much of it is driven by common criminal activities (especially drug-related) SYN I HESIS 3 * Achieve fast and sustainable growth * Share the fruits of growth with all Colombians * Build a government of quality The first priority is indeed to reach fast and sustainable growth. Growth has been Colombia's best and most effective social safety net; on average, each per- centage point increase in per capita GDP, reduces poverty by 0.6 percent (or, at present, a quarter million people). Given the historical, geographical, and military fragmentation of the country, and the related weakness of the state, the possibility of earning income within a growing economy has held the nation together. That order came to an end in 1999, wiping out within a year the previous decade of decline in poverty levels. Putting the economy back on a rapid, sustainable growth path will not be easy, but it is feasible It requires action on five fronts. First, and foremost, a more con- ducive macroeconomic framework (which Colombia once had) is necessary and, in particular, redressing the large financial imbalance in the public sector and the asso- ciated, rapid debt accumulation problem. The key will be to spend less and tax more, but without slowing down the economy. The ensuing adjustment will need to be "locked in" through a stabilizing fiscal responsibility law and a professionalized debt management function. Second, the reform of the financial sector needs to be completed, if the sector is to become an engine for growth rather than another lia- bility for the state-both the conclusion of the bank-restructuring program and the fiscally all-important pension reform are pending. Third, the institutional, regula- tory, and cross-subsidization arrangements in infrastructure provision should be adjusted for further private participation to materialize and, at the same time, for service coverage to reach the poor (the state cannot afford the cost of replenishing the country's much depreciated physical capital). Fourth, a rebirth of the rural econ- omy is difficult but possible, even despite the armed conflict, the illegal crops, and the coffee crisis, if public policy is credibly reoriented toward facilitating, rather than resisting, market signals-policies and programs that shield producers from inter- national commodity prices, distribute subsidies that are not farm-size neutral, dis- tort the land market, or perpetuate institutional arrangements that, although tradi- tional, have outlived their usefulness, should be abandoned. Finally, the intrinsic synergies between growth and environmental protection remain unexploited, both at the local and global level (notably, Colombia's potential in the carbon emissions market remains untapped). Growth will mean little to the 27 million Colombians that live in poverty if they are not given the tools to share In that growth. For most of them, better human capital formation mechanisms, and better markets in which to sell that capital, will be the answer. Both the public education and health systems need to be freed of the vested interests that now resist reform and accountability In both systems, the gov- ernment is paying too much, and getting poor and deteriorating quality. At the same time, the legal and regulatory framework of the labor market, where the poor 4 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE should be able to exploit their human capital, is out of tune with a modern econ- omy and effectively shuts out one in three workers. A comprehensive labor reform is much overdue. But flexible labor markets and enhanced human capital will not be the answer for all Colombians-certainly not for those that cannot cope with systemic shocks, have been displaced by the conflict, or still face race- and gender- based discrimination. Colombia does not yet have a proper social safety net, and badly needs one. Finally, the new government's capacity will be challenged by the breadth of the attendant policy agenda and, at the same time, handicapped by the gradual but, by now, profound loss of credibility of the state. A reform of the state itself is thus critical, but this should not be about changes in bureaucratic organigrams (much as some of these may be needed) but about improving the way the state discharges its main functions-how it budgets, procures, audits, evaluates, regulates, super- vises, and reports. Moreover, the context in which those functions are performed should be the subject of reform too, on three fronts First, the decentralization of responsibilities is on an unsustainable path, and a new incentive framework is nec- essary if departments and municipalities are to raise the quality of their public services and maintain fiscal discipline. Second, the mechanisms to fight corrup- tion need strengthening, especially because corruption in Colombia takes the per- verse form of state capture. And, third, the judicial system is caught between its pre-1991-Constitution culture and its post-1991 mandate and, thus, delivers serv- ices of poor quality and insufficient quantity. A few, but critical, reforms can cor- rect that problem (like the introduction of fee-for-services or the circumscription of the Tutela). III. The Agenda's Diagnoses and Policies a) Achieving Fast and Sustainable Growth Reaching a higher growth plateau will be a multidimensional task. Both macro and micro interventions will be necessary-adjusting the fiscal position, completing the reform of the financial sector, fostering private participation in infrastructure, diver- sifying the rural economy, and protecting the environment. Many of those inter- ventions are urgent, and none is simple. However, in no case is the "do-nothing" sce- nario really an option, let alone a preferred option. i. A Foretold Macro-Fiscal Adjustment Recovering from the crisis of 1998-99 in effect dictated the macroeconomic poli- cies of President Pastrana. Protecting that recovery will dictate President Uribe's. More specifically, during the past four years, Colombia's macroeconomic policy had to focus, and rightly so, on redressing the imbalance in the public purse. In the context of a three-year IMF Extended Fund Facility, the growth of territorial SYNTI'HFSIS 5 transfers was delinked from the central government's revenues (through the Acto Legzslatzvo),2 fiscal discipline among subnational governments was fostered (through Law 617/2000, Law 715/2001, and new regulations for municipal bor- rowing), public investment was further curtailed (regrettably, to a mere 1 percent of GDP), civil servants' real wages were reduced (a measure later reversed by the Constitutional Court), the base of the value-added tax (VAT) was widened, fuel taxes were increased, and a financial transactions tax was introduced. The Central Bank's independence was enhanced and its mandate to fight inflation was strengthened. Undoubtedly, those were all critical achievements, and perhaps all that could be done given the political space for reform. But they fell short of cor- recting the financial imbalance of the public sector-the deficit did not fall enough and still hovers at around three percent of GDP. As a consequence, the country's sovereign debt position deteriorated: total nonfinancial public sector debt as a proportion of GDP increased by half (to 45 percent) in the four years of the outgoing administration And, today, Colombia spends more on interest pay- ments than on education. The main macroeconomic task of the incoming administration will therefore be to complete the fiscal adjustment that its predecessor initiated and put the accumu- lation of public debt on a sustainable path. More specifically, even under optimistic assumptions in terms of growth, interest rates, oil prices, and Central Bank senior- age, the new authorities will have to immediately reach and maintain a primary sur- plus in the nonfinancial public sector of at least four percent of GDP per year just to keep debt accumulation on a manageable trend (that is, reaching about 52 per- cent of GDP by the end of the Presidential tenure in 2006). At present, that surplus stands at only one percent of GDP. What can be done to achieve that three-percentage-point adjustment in the pri- mary balance? A combination of three factors is in order-a major reduction in the size of the state, a growth-friendly tax reform, and better debt management. The Colombian government, at its various levels, consumes a third of what the country produces. One million people, or about five percent of the labor force, work for the state. In relative terms, these staffing figures are not particularly large; and, yet, the salaries and benefits to those employees (not to mention their pensions) amount to almost eight percent of GDP. The areas for potential contraction are many-linking subnational expenditure decisions to the political cost of having to raise local taxes to pay for them, reversing the exploding cost of servicing pension obligations, rationalizing the health branch of the Instituto del Seguro Social, eliminating over- 2 While de-linking transfers from revenues and consolidating all transfers into one budget item were its major achievements, the Acto Legislatnvo also established specific annual rates of real growth for those transfers over the next ten years, something that could increase their relative burden on the central government's budget should the economy not grow fast enough 6 COLOMBIA THE ECONOMIC FOUNDATION OF PEAcr lapping agencies, and so on.3 These structural reforms to the fiscal accounts, which are discussed in more detail later, will mean that the Colombian state will spend less. A parallel effort should also be put in place for the state to spend better. This implies upgrading the budgeting, procurement, financial management and, critically, evalu- ation and accountability mechanisms. Lower expenditure will, however, not suffice-it will need to be accompanied by higher tax collection. The six partial reforms implemented over the last decade, all triggered by immediate revenue needs, left Colombia's tax code riddled with loop- holes, inequities, and distortions, and cumbersome to administer. This facilitates the rampant evasion of the two most important taxes-the income tax and the VAT, which together represent about 80 percent of the total tax collected. On average, about a quarter of every peso due on taxes is evaded. The current problems with the tax system are, however, also an opportunity-there is room to both increase rev- enue and make taxation fairer by removing the array of exemptions that plague it and without overrelying on hikes in tax rates. Specifically, in the corporate income tax, industry-, firm-, and location-based exemptions should all be eliminated (the list is long: public service enterprises, livestock funds, funds organized by financial institutions, firms with operations in the Paez River, publishing houses, lotteries, liquor producers, and cooperatives); presumptive income taxation should be based on gross assets (not net worth); and inventory valuations should be adjusted for inflation. Similar rationalizations should also be applied to the personal income tax-the flat exemption of 30 percent of the wage should be eliminated, the general exemp- tion should be set at three times the per-capita income (from its current three min- imum wages), and the three rates should be scaled down to 15, 25, and 35 percent (from 20, 29, and 35). Finally, the VAT should be recovered as the broad-based tax it was supposed to be-a single rate of 16 percent should apply to all goods and serv- ices, instead of multiple rates and multiple exemptions. This will extend the 16 per- cent VAT to among others, food and medicines (currently at zero rate), capital goods (currently at 10 percent), and all previously exempted transactions (like those involving paper, printing, beer and tobacco, personal services, transportation and construction); it will also discontinue the "implicit" VAT, exempt exports, and make the VAT creditable against the VAT itself (not other taxes). At the same time, a flat- rate, national excise tax on luxury goods should be introduced. Those reforms to the tax code, taken together, would not only make the system fairer and simpler to administer, but would also generate a much needed three to four percent of GDP in additional revenue. This will allow for compensating the poor for the impact of extending the VAT to food and medicines-even doubling the size of the total social assistance budget to make cash transfers to the poor will 3 It should be noted that some of those "reforms of the state," while necessary, would reduce fiscal costs only in the medium- and long-term (and may increase them in the short-run, for example, because of related severance payments) SYNTHESIS 7 only cost some 0 7 percent of GDP.4 But a word of caution is called for if too much of a "tax shock" is applied, the risk that the fragile growth recovery will evaporate (or even turn into recession) is high. This could put the economy on a downward spi- ral-tax pressure reduces growth, which increases the need for a larger primary sur- plus, which calls for further tax pressure, and so on. The spiral would soon make the debt burden unbearable, especially in terms of access to refinancing (a regrettable example of this phenomenon is Argentina 1999-2002) That is why Colombia's tax reform should be more about tax rationalization and simplification than about higher or new taxes. The adjustment in public finances will be politically costly, and pressLire to undo it will not diminish. Thus, mechanisms to "lock in" the adjustment will need to be institutionalized. Two such mechanisms stand out. First, the currently proposed "fis- cal responsibility law," which sets up a primary balance goal and regulates contin- gencies, arrears, tax expenditures, forward allocations (vzgenczasfiuturas), and other accounting tools that might weaken fiscal discipline, should be expanded to include mechanisms for aggregate revenue stabilization (for both budgeting and reserve accu- mulation purposes) and automatic expenditure reduction rules in case of revenue shortfalls. Second, the public debt management function needs to be professionalized and made directly accountable to the taxpayers for its performance; this will strengthen the incenitive against the unwarranted assumption by the central govern- ment of liabilities from the parts of the public sector that refuse to reform (the con- tingent liability of the central government for the debts of subnatiolals governments, banks, pensions, and toll roads already amounts to several times the size of GDP) ii. Completing the Reform of the Financial Sector Much as the macroeconomic policy of the new Presidential Administration will be dictated by the need to complete what its predecessor initiated, policy in the finan- cial sector, the second key element in achieving fast and sustainable growth, will also be about completing and consolidating reforms With decisive action and sheer ingenuity, the outgoing government managed to avert a systemic banking collapse in 1999. That action was, however, not without costs-the fiscal liability arising from rescuing, to varying degrees, borrowers, depositors, and bankers is so far esti- mated at about 8 percent of GDP. And the cleanup is far from over. While the let- ter of the legal and regulatory environment for banking operations is now up to interniationally accepted standards, the government's deposit insurance and resolu- tion agency (FOGAFIN) is still in the midst of supporting (through credit lines to their shareholders) a number of private banks and housing finance institutionis that 4 Still, the extension of VAT to food and medicines may be politically and, in some cases (like primary agricultural products), administratively infeasible In this case, a second-best reform package could maintain the exemption for food and medicine, reduce the thresh- old of minimum non-taxable personal income to twice (rather than thrice) the per capita income, and generate a total additional revenue of 1 5 to two percent of GDP 8 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE have not yet been fully capitalized or restructured; disposing of two large, failed, and rapidly deteriorating banks (Bancafe, Granahorrar); articulating a suitable strategy for an array of second-tier, state-owned lending institutions (IFI, FINDETER, FEN, FINAGRO, BANCOLDEX) that have a quarter of the industry's assets, an outdated mandate and a weak balance sheet; and instilling needed expediency in its asset disposition subsidiary (CISA). In other words, Colombia's banking system will over the next four years need to undergo a second, and perhaps larger, round of reform. The key will be to facilitate that reform without incurring further fiscal costs (which the government can Ill afford anyway) or interrupting, and optimally foster- ing, the flow of bank lending to the private sector (which, as a percentage of GDP, has been continuously falling and is still only two thirds of what it was in 1999) Two sections of the banking industry merit a closer look, because of both the fragile financial position of their main actors and the social role they are supposed to play-housing and rural finance. The mortgage industry has not yet been restored to soundness, following its virtual collapsed in 1999 (when one in every five mortgaged properties had to be repossessed). A quarter of its loans are still nonper- forming (and the rate is rising); various rulings by the Constitutional Court (cap- ping interest rates, forbidding out-of-court foreclosures, and so forth) have fostered a spreading culture of nonpayment; the legally mandated inflation indexing mech- anism (the WVR) makes payments highly volatile; and the mortgage-backed securi- ties and mortgage bond market has been slow to take-off (the passing of the enabling legal framework-Housing Law 546/1999-has not yet been fully matched by investor interest). Not surprisingly, the financial sector's portfolio of housing loans has shrunk every single month since 1999.5 This hints at a difficult reality: apart from technical improvements that can be achieved by amending various regulations (for instance, in basing the UVR on a moving average of, rather than on a previous month's, inflation), a rapid revival of housing finance in Colombia may, for the time being, be beyond the reach of policymaking. This is a particularly distressing reality, because Colombia is estimated to have a deficit of one million new housing units, and two million in need of repair. A similar situation exists in rural finance. Some 2,300 financial cooperatives, with primary focus in rural markets, spawned unregulated in the two decades to 1999; at their peak, they controlled one tenth of the financial system's deposits. When the crisis took place, most of them went bankrupt, saddling the government (through FOGAFIN) with assuming responsibility for the largest among them. The government's own rural banks, which as a group provided one quarter of all agri- cultural credit, fared no better (notably, Caja Agraria had to be reorganized into Banco Agrarto). While arguments about the efficiency (that is, cost) and effectiveness (that is, impact on poverty) of the government's continuing supply of rural funding (through Banco Agraro, FINAGRO, FONADER, and the like) are debatable, the 5. Although the stock of housing credit remains flat, the flow of new housing loans has recently restarted SYNTI HESIS 9 likelihood that private banks may be attracted back into rural finance, especially for small and medium borrowers, is marginal as long as the armed conflict rages While there may be room for divesting from Banco Agrarto and FINAGRO and using the released resources to subsidize private mobile rural banking, part-time branches, and risk-assessment tools that lessen the reliance on collateralization, the fact remains that, until peace returns to the rural areas, private banks will not venture in any sig- nificant way to provide additional services there-let alone services that cater for the poorer segments of the population. The banking industry is not the only component of the financial sector the reform of which remains incomplete and poses a major contingent liability to the fiscal accounts-a more complex reform and larger liability are pending in thepen- sion system. The consolidated pension system for public sector workers is technically insolvent, the total actuarial value of its liability (the "implicit pension debt") is twice the size of GDP, and the annual fiscal transfers required to cover its cash deficit is now some 2.5 percent of GDP-that is, equivalent to the total health budget of the nation. This is a worrisome position for a country that is still demographically young and whose pension system only covers a third of the labor force Moreover, partly because of the rich political debate that it has generated, pension reform has become an icon of the Colombian governmenit's resolve to address deep-rooted structural imbalances in its fiscal accounts-and international financial markets are pricing Colombia's country risk accordingly How did Colombia reach this point, and what can it do now to correct the situation? Until 1993, Colombia had a large series of pension institutions for the public sec- tor (the largest of which-CAJANAL-serving central government workers) and one for private workers (administered by the Instituto de Seguro Social, ISS). All of them were in practice unfunded, pay-as-you-go schemes with unsustainably gener- ous benefits and high-levels of evasion. That year, a major reform was implemented, with five main components. First, the central government assumed explicit respon- sibility for all accrued pension tights. Second, the array of pension regimes was con- solidated into one (with the exception of the staff of ECOPETROL, teachers, armed forces, and Congress) and its benefits package was curtailed by an increase in retire- ment ages and a tightening of eligibility conditions This reduction in the benefit package would, however, be introduced only gradually through 2014, that is, a 20- year "transition" period was built in. Third, all public and private workers were given the choice to contribute to the pay-as-you-go regime (still administered by ISS) or to a defined-contribution system of individual accounts administered by new private fund managers but with a minimum pension guarantee. Workers were, however, given the option to switch back and forth between the two systems every three years Fourth, the contribution rates were raised for all workers, in whichever regime. Fifth, workers above a certain salary level were forced to contribute toward a new nonicontributory pension scheme for the poor (the so-called Solidarity Tax). In retrospect, while the 1993 reform was important in introducing new concepts (lower benefits, defined contributLion, private fund managers, and the like), it had 10 COLOMBIA THE ECONOMIC FOUNDAnON OF PEACE too long a transition period and too many exceptions. While the increase in contri- butions and the gradual reduction in the benefit package reduced the implicit pen- sion debt of the government (by some 37 percent of GDP), it did not correct the structural imbalance in the system-between effective benefits that are still too gen- erous and the contributions made by the workers. In its current form, the system is not sustainable, because the necessary fiscal transfers to cover its annual deficits will continuously grow, up to six percent of GDP in 2020. There is therefore no question that a new round of pension reform is urgently needed. At the same time, contribution rates (13.5 percent for all concepts) are already too high and discourage participation in the formal labor market. Further reforms should thus primarily focus on a reduction of the benefits package and, crit- ically, in how fast that reduction is introduced (to avoid yet another long-lingering "transition"). The guiding principle should be that, at the very least, the marginal worker entering the system after the reform, should not increase the implicit pen- sion debt of the government. This can be achieved with many combinations of changes in the system's parameters (retirement ages, replacement rate, and so on).6 However, this type of reform will not bring fiscal relief soon. For that, it is also imperative to accelerate the "transition" of public sector workers out of the pre-1993 regime-that is, to actually implement the 1993 reform.7 Taken together, the acceleration of the 1993 transition and a new reform to the current regime could reduce the implicit pension debt by some 40 to 50 percent of GDP. It should be mentioned that complementary amendments will also be needed to armor the new system thereafter-like the reform of the currently exempted regimes, a predictable and sustainable mechanism to index the minimum pension guarantee, a provisioning mechanisms for regimes arising from collective bargaining agreements of public enterprises, and the discontinuation of the "switching" option between defined-contribution and defined-benefit regimes. Finally, banking and pension reforms would be much facilitated if Colombia could count on a deeper, more liquid capital market. Today, financial intermediation is vastly dominated by banks, since not even creditworthy corporations venture much into the private bond markets. Similarly, administrators of the newly created private pension funds do not find many private debt securities into which they could diversify, especially out of the dominant Treasury bills. The capitalization of the country's stock market (equivalent to about 13 percent of GDP) is one of the low- est in the region. Fortunately, much activity is planned to develop Colombia's capi- 6 As an exarmple, one possible option is to introduce retirement ages of 60 for women and 65 for men, a replacement rate of 50 percent after 1,300 weeks of contribution, and an increase in contribution rates of one percentage point 7 Again as an example, that acceleration could take the form of leaving untouched the regime for men and women that in 1994 were 50 and 45, respectively, or older; acceler- ating the transition for those that at that time were between 42 and 49, and 36 and 44, respectively; and bringing all other workers into the full effect of the 1993 reform SYNTHESIS 11 tal markets. a proposed draft securities law currently in Congress would reduce obstacles for private issuance and wotld standardize mortgage instruments; a new securitization firm is now operating; market-making has been institutionalized for government bonds; and, critically, an almost complete range of Treasury bill matu- rities are now available as risk-free benchmarks. The new government should see these various reforms through, and prepare the relevant supervisory institutions to implement them (specifically, the Superintendency of Securities). iii. Infrastructure: Making Private Participation Work for All An adequate provision of physical infrastructure will be the third key element in returning Colombia to a path of fast and sustainable growth This is particularly rel- evant for a country where armed groups systematically target and destroy bridges, transmission pylons, and pipelines-and kidnap those that could help rebuild them. It is therefore not surprising that private investment has been very low But even at those low levels, Colombia has been unusually successful in attracting private par- ticipation into the provision of infrastructure services-with the exception of telecommunications, the private sector has a major role in all subsectors (oil, gas, coal, power, water, and the like). The problem is that those infrastructure services are both insufficient to support a higher growth plateau, seldom reach the poor, and there is no fiscal space to complement them with additional public investment. Thus, the policy dilemma for the next administration is how to make private infra- structure investment more appealing in the context of the conflict, increase the cov- erage for those that cannot pay for infrastructure services, and protect the fiscal accounts, all at the same time. The answer lies in removing the remaining institu- tional and regulatory constraints that prevent further private participation and in adjusting the mechanisms for subsidization across users. More specifically, in energy, while the government has been extremely successful in adjusting the rewards of oil exploration to the heightened security risks (in the so- called contratos de asociacidn), the process of environmental licensing has become unduly cumbersome-and needs institutionial resources to upgrade it Similarly, wellhead gas prices are regulated rather than market determined, precluding the nec- essary investments to tap even larger fields (especially Cusiana) or explore for new ones. Price regulation is also an impediment to entrants in the downstream oil industry-the ex-refinery prices for gasoline and diesel are still set by the Ministry of Mines and Energy (and then taxed at the pump) rather than by the market. Reg- ular disputes over the "capacity charge" for thermal power plants (a surcharge on the regulated price meant to ensure a minimum income in a primarily hydro system) and over the allowed return-on-assets for transmission and distribution services have cooled private appetite for further investment in electricity after the initial round of privatizations. Most important, the government still holds ownership of one major power generator (ISAGEN) and some 20 distribution companies (electrificadoras). The privatization of the former was delayed in light of anti-trust concerns, whereas the privatization of the latter is proving elusive because of their already-fragile finan- 12 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE cial health and governance problems-an uncomfortable position, because they serve about half of the country's electricity consumers. Beyond regulation, pricing, and ownership problems, the institutional frame- work of the energy sector needs reform, as well. On one hand, the Superintendencia de Servicios Pi;blzcos Domiciliarios (SSDS) is an ombudsman for consumers, a work- out unit for distressed companies, and an asset manager for the corporations that the state decides to keep. Naturally, it struggles to handle such diverse roles and to fence off political interference in its management. On the other hand, the Comzsidn de Regulacidn de Energiay Gas, supposedly an independent regulatory commission, sees many of its functions overtaken by the line ministry (for example, in oil price set- ting) and, partly as a result, its professional capacity diminished. Similar impediments to private participation exist in water and sanitation, although, in this area, the investment gap is even bigger. While coverage is high by Latin American standards, only half of all drinking water and less than eight percent of municipal wastewater receive any treatment. The Comisidn Reguladora del Agua (CRA), who deals with an atomized group of 1,700 operators, is not isolated from local political pressure and, thus, tends to set tariffs that do not cover costs-even if they did, the tariff-setting process is based on reported costs rather than perform- ance benchmarks and quality standards, allowing for the cost of inefficient manage- ment to be transferred to consumers. Almost all providers in small and medium-size cities exhibit operational deficits that call for fiscal support. In many of those cities, and especially in rural areas, the consumer base is not affluent enough to pay for the investment and operation of the service, and public funding is indeed required. CRA's relative role vis-a-vis the policy-setting line ministry (the Ministry of Eco- nomic Development) and the consumer-protecting SSDS is not well defined, gen- erating further uncertainty for private participants. And environmental standards for wastewater, as currently specified in the regulations, are so high as to be self- defeating; to meet nationwide the absolute level of minimum removal of biochem- ical oxygen and suspended solids (80 percent) would require an additional invest- ment of about five percent of GDP-so the standard is ignored. A standard more endogenous to the use of the receiving water body may be more suitable. The private sector has also participated in a major way in transport (especially toll roads, railroads, maritime terminals, and main airports). Given its dominant role in cargo, its absorption of 70 percent of public investment in transport, and its rapidly deteriorating physical status, the road subsector is perhaps the one in most urgent need of policy attention. The institutional setup is similar to other sectors: a line ministry establishes policy (Ministry of Transport), an independent commission reg- ulates (Comisidn Reguladora de Transporte), national and local institutions actually manage the roads (national INVIAS being the largest), and a supervisory authority controls and enforces (Superintendencia de Transporte). In practice, the division of responsibilities is less clear-cut: those institutions overlap in their activities, are not mutually independent, and engender uncertainty. This institutional weakness is par- ticularly unfortunate in the context of the armed conflict-the government SYNTHESIS 13 (national or local) can neither reach certain areas to perform road maintenance nor protect cargo from hijacking This all but smothers interest in further road conces- sions; the first generation of those concessions have demanded, and have called on, a minimum revenue guarantee or complementary public funding absorbing about one third of the central government's budget for roads. Like other infrastructure sectors, and also following the mandate of the 1991 Constitution, the telecommunications service was opened to private participation- but with less success. Less than four percent of telephone lines are in the hands of private companies, compared to 85 percent for Latin America as a whole. There are onily two cellular providers in Colombia, with high prices and low quality of serv- ice. Private providers did enter BOT arrangements with public enterprises to pro- vide local services, and the long-distance and international business was opened to competition. But this put financial pressure on the national carrier (Telecom, which, because of labor opposition, was not privatized) and forced the government to default on the terms of the BOTs. The ensuing legal battles have stiffened further private interest (notably in the otherwise profitable cellular service), already weak- ened by the imposition of a hefty 16 percent tax on long-distance service and a pro- posed 5 percent tax on the operators' gross revenues. To make matters worse, the overlapping of regulatory functions seen in other infrastructure sectors is present here as well-a tariff-setting Comisidn Reguladora de Telecomunicaciones, a policy- setting Ministry of Comunicaciones, and an enforcing SSPD. The above suggests that the uncertainties associated with the armed conflict and with the entanglement of institutional functions hampers all infrastructure sectors alike. There is, however, a third, equally binding impediment for all sectors-cross- subsidization. The Public Services Law classifies residential users of natural gas, elec- tricity, water, sanitation, and telephones in six strata and forces the fifth, the sixth, and commercial to pay a surcharge (up to 20 percent of service cost) in order to pro- vide subsidized service primarily to the first and second (that is, the poorest). While the system is ingenious and has politically facilitated the opening of the various infrastructure sectors to private entrance, its application has been complicated by the fact that not all companies have a balance between rich and poor consumers (call- ing for support from the government) and the strata certification process is less than accurate (it is based primarily on residence characteristics-neighborhood's overall quality and house size-that are in practice misleading predictors of well-being). In some municipalities, the entire population has been classified as poor, pushing, for example, telephone operators into bankruptcy. This hints at two policy alternatives: either the stratification system is improved to include better means-detection and means-testing tools (not just housing quality); or a government-funded connection and bare-minimum level of service is provided to all consumers, above which marginal tariffs grow steeply to recover the initial investment. The problems with the various infrastructure sectors also have a common geo- graphical point of convergence-urban centers Two-thirds of Colombians live in urban areas, of which more than half live in poverty The armed conflict feeds the 14 COLOMBIA THE ECONOMic FOUNDATION OF PEACE migration flow to cities, giving birth to peripheral towns burgeoning around already-congested metropolises (for example, Soacha on the outskirts of Bogoti). As will be explained later, those cities now operate within more binding budget con- straints (a consequence of various laws and policies putting order into the decen- tralization process). This limits their capacity to support the arriving poor with cash, while suboptimal regulations (like minimum floor-to-area ratios and maximum con- struction densities) limit the support they can give through land allocations. Not surprisingly, it is estimated that one fifth of all urban land developed each year is developed in informal settlements, where the subsequent provision of basic services is not only more expensive but also, under current legislation, technically illegal. In principle, local governments are compelled by law to produce Planes de Orde- namlento Territorial, nine-year land-development plans that are supposed to direct the new settlements toward efficient locations, but because of lack of resources, few municipalities do. iv. Facilitating the Rebirth of the Rural Economy The opening of the Colombian economy in 1991 brought about major reforms to the country's rural economy and, in particular, its agriculture. Protection against imports, subsidized credit, price supports, marketing monopolies, and the like were all to be abandoned as part of a change in development strategy domestically called the apertura. This was expected to foster productivity increases, a diversification away from traditional products, and eventually an export boom. It did not happen. The reason is simple: the apertura never really occurred. Marketing boards returned in 1992 (until 1997). Price bands, stabilization funds, and import barriers were rein- stated on and off for certain crops, for certain periods of time. Although many of those interventions were relaxed in recent years, the ensuing uncertainty about pol- icy direction left the sector's performance essentially unchanged-its productivity has not increased, its flagship product (coffee) is on the brink of collapse, and the one product that proves profitable and skews market incentives is illegal (coca leaf). More profoundly, 80 percent of the rural population remains poor and subject to the trauma of an armed conflict whose tactical epicenter Is the control of land In that inauspicious context, what, if anything, can pohcymakers do now to unlock the otherwise rich potential of Colombia's rural economy? The answer lies in a combi- nation of market-based diversification and direct social assistance; the former is addressed below, while the latter is addressed in the next section. The diversification, and through it, the revival of the rural economy, calls for four main policy reforms. First, the new authorities need to make an early, credible, and consistent statement about the direction of their sector policy-signaling their intention to put the sector permanently on a path of open markets and private- sector-led efficiency. One powerful signal will be a reduction in the dispersion and average level of agricultural import tariffs (which vary widely across crops) and a relaxation of nontariff barriers, both of which are perceived as volatile and captured by special interest groups SYN1i IESIS 1 5 Second, all public interventions in the sector should be made farm-size neutral. At present, only two percent of small farmers have ever managed to access the govern- ment's matching grant facility for investmiienit financing (Incentwvo a la Capitalhzacidn Rural). Even the state-owned institutions that remain engaged in rural finance (Banco Agrarzo, Fondo Emprender, anid FINAGRO) require collateralization levels that, in practice, exclude small farmers and crops not linked to commodity markets. Those farmers increasingly find the municipality-based technology extension services (so- called UMATAs) not as relevant as they could be for the needs of small farming units. Much of this is a natural market response-between the uncertainty of ever-shifting sectoral policies and the insecurity of the armed conflict, a few standardized agricul- tural products turned out by large commercial establishments with export reach hold the best chance to survive and, thus, attract the bulk of the available support. The key is then to link those larger establishments to the rest of the sector without causing fur- ther distortions-this is the objective of the existing Acuerdos de Competitrvziad, whereby the government provides demand-driven support (financial, informational, trade representation) to local, private associations of producers that risk their own capital to exploit synergies among local participants of all sizes. Peso for peso, these Acuerdos are perhaps the most efficient use of public money in the sector Third, distortions to the land market should be removed. Colombia has one of the most concentrated land ownership structures in the World (with a land Gini of 0.86), and is getting still more concentrated. More than half of the land is consid- ered "large farms," and only two thirds of plots-mostly the largest two thirds- have titles Money laundering by the illegal drugs industry fuels the concentration of land Limited cadastre coverage, market-insensitive valuation mechanisms, and weak collection capacities keep municipalities from exploiting the land tax (thepre- dia), exacerbating the regressivity of land ownership. Limited cost recovery for irri- gation favors low-margin crops and pasture, activities more suitable for large plots. And, most of all, the insecurity associated with the armed conflict pushes those that have no alternative means of living (typically small farmers) to cash out, and those that can retain their land (typically commercial ventures and richer absentee owners) to refrain from difficult-to-reverse investmenit. All these factors combine to produce a worsening mismatch between the natural vocation of the soil and its actual use: livestock uses double the land that would be technically appropriate, while agricul- ture uses a third of what is suitable for it Policymakers in Colombia have been trying since 1961 to correct the concentra- tion problem through a land reform program (last adjusted by Law 160 of 1994). On the whole, the program has failed to redistribute land, has grossly distorted the land markets, and should IIow be shut down or, at a minimum, overhauled. In its current form, the program provides funding for beneficiaries to buy farming units; 80 percent of the funding is a grant, while the rest is a loan from a state-owned bank (Banco Agrarzo). TIhis absorbs about one fifth of the central government's budget for agriculture. However, lacking working capital, the prototypical beneficiaries cannot set up business, defaults on its loan (virtually all loans have been defaulted on), and 16 COLOMBIA THIE ECONOMIC FOUNDATION OF PEACE abandons the land. Moreover, the public institution that administers the program (INCORA) spends two thirds of the program's budget on its own administrative costs and, predictably, has not been free of political influence and rent-seeking. Clos- ing down the land reform program would free resources that could be better employed in direct social assistance, a faster titling mechanism, the provision of title safekeeping services, or defraying the cost to municipalities of improving the collec- tion of the predial tax. If closure were politically infeasible in the short-term, a series of program modifications, currently being piloted, hold some promise for enhanced effectiveness-decentralization of decisions to local participants, delinking the grant from the purchase of land itself, and permitting the purchase of smaller than full- farm units. Finally, the diversification of Colombia's rural economy will be spearheaded by the structural transformation of the coffee industry. This transformation is already underway, and the best policy response is to facilitate it-rather than resist it. In Colombia, the income of one in every five rural households, and one in every four agricultural jobs, depend on coffee. Coffee accounts for just under 5 percent of GDP and 6 percent of exports. The country controls one tenth of the world's cof- fee market, and a highly valuable brand name. But formidable external forces are changing the business of coffee-the international prices are at their lowest point in almost 200 years (and are unlikely to recover, given the increase in tree planting worldwide), the concentration of international roasters, traders, and retailers has squeezed the share of earnings going to producers; consumer preferences have shifted toward specialty coffees, and an international regulatory structure has effec- tively emerged (through the WTO's phytosanitary standards, global environmental agreements, and so on). Those forces have hit Colombia's coffee hard. Low earnings are trapping the domestic growers in a vicious circle-low profitability pushes them to cut back on production costs, which reduces quality, leading to lower earnings, and to low profitability. By now, the domestic price stabilization fund has run out of money, forcing the government, for the first time in history, to launch a rescue package in 2001. That package, which is supposed to be a temporary three-year arrangement, is meant to support a certain price level, the renewal of coffee trees, technical assistance, and debt refinancing. It costs some US$100 million per year. While the rescue package is supposed to end in 2003, political realities will likely prevail and force a continuation. The key will then be to convert further support into direct income transfers to poor households in the coffee-producing areas. This will give them the choice to stay in the industry under the new market conditions (and use the cash to invest accordingly) or diversify away from it. Part of the funds could also be used to provide technical assistance for diversification into specialty coffees, rather than more production of the same type. Most of all, using the lever- age of its financial support, the government will want to elicit a fundamental reform of the industry's institutional framework-of the semipublic, policy-setting National Coffee Committee and of its regulatory arm, the National Federation of Coffee Growers. In the new reality, the Federation will need better governance, SYNTHESIS 17 enhanced public accountability, a separation of its role as regulator and market par- ticipant, more partnership with new local associations, and an end to its exclusive right to broker coffee exports. Will this transformation mean the demise of Colombian coffe& It need not. There may well be lower production volume of the traditional single type-but there will be better earnings thani at present. The new industry will likely be better serving the domestic market (as Brazil has done); exploiting Geographic Indications of Origin (as Jamaica and Antigua have done), integrating vertically to sell its own brand directly in the retail markets of consumer countries (at least in part); and developing niche coffees (organic, fair-trade, and so forth) Some growers, especially medium-size ones, will phase out and move onto other parts of the rural economy. The transformation, and the government's support for it, will be crucial for Colom- bia in one other respect-if the authorities signal their commitment to efficiency- enhancing transformation in coffee, the leading and most traditional agricultural product in the country, there will be no doubt about, and less resistance to, its sim- ilar intentions in other parts of the rural economy. v. Exploiting the Synergies Between Growth and Natural Resource Protection Colombia is the third-most biodiverse country in the world. It has 65 different types of ecosystems and 18 ecoregions. The diversity of its birds, amphibians, and vascu- lar plants is unparalleled on the planet. With close to a thousand permanent rivers, its water supply is the fourth largest; the country is also home to headwaters of large tributaries to the Amazon and Orinoco Basins. Much of this natural wealth is being rapidly destroyed, by three main factors. First, and foremost, the armed conflict pro- tects and fosters illicit crops. These take place in primary forest in fragile areas with no consideration for long-term soil preservatiol, discharge chemical residues from coca laboratories, and attract chemical fumigation-which pushes them further into new virgin areas to clear. Second, much of Colombia's growth (and badly needed fis- cal resources) is increasingly dependent on extractive industries (like oil, gas, and coal). Extractive industries, even if well regulated, can have a substantial negative impact on the surrounding environment where they operate, especially amid the sensitive ecosystems of the western Amazon. Third, as described earlier, perverse incentives in agriculture have caused a major gap between the natural vocation of the land and its actual use. Successive governments have over the last decade tried to address the degradation of the country's natural resources (for example, through Law 99/1993 which established the ministry of the environment; the National Strategy for the Conservation of Biodiversity of 1997; the creation of various protected areas; the ratification of the Kyoto Protocol; the signing of the Stockholm convention; and the membership in the Montreal Protocol). These initial steps have, however, brought little success. This evokes the question: Given the country's war, poverty, and fiscal realities, can growth be made compatible with environmental protection? This book argues that, in Colombia, growth and environmental protection not only can be compatible but also are mutually reinforcing-that is, they are intrinsi- 18 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE cally synergistic. Two types of policies-local and global-can help exploit those synergies. On one hand, the central government should use its (shrinking) environ- mental budget (about half a percent of the central government's total expenditures) to support, technically and financially, the municipalities' effort to conclude their legally mandated Planes de Ordenamiento Territorial (Territorial Development Plans), and in particular the environmental components of the plans. These plans will highlight concrete areas where protector-producer reserves are most urgently needed (in effect, the reserves are individual contracts with landowners in environ- mentally fragile areas whereby the owner agrees to a set of protective local activities in exchange for some form of compensation). At the same time, the central and municipal governments should further foment and recruit local associations to help enforce existing regulations and reserves. On the other hand, untapped opportunities remain at the global level. First, Colombia is not yet fully profiting from the carbon emission trade.8 A first, pio- neering deal for carbon mitigation was recently completed, but vast amounts of car- bon credits are still possible in the "cleaning" of the power sector and industry, and in switching to renewable sources of energy. While still globally less developed, an eventual market in carbon sequestration (both through reforestation and forest con- servation) would also carry large potential for Colombia, a country with some 300,000 square kilometers of largely pristine rainforest and, thus, an important car- bon "sink."9 If well developed, total exports of carbon emission credits could gener- ate an estimated 0.5 percent of GDP per year-that is, about three times the size of the current environment-related expenditure in the national budget. [Channeling part of those resources back to the local communities that host the sequestration and, more generally, the environmental protection efforts would enhance the sus- tainability of the initiative.] The government, with active private participation, should hence take the lead in organizing this market domestically-strengthening the institutional capacity of the recently established Office for Optimal Use of Clean Development Mechanisms, funding methodologies for project preparation and cer- tification, disseminating information on external opportunities, and assembling a portfolio of carbon trade projects. Most critically, standardized risk-management mechanisms need to be articulated for carbon emission credits of Colombian origin to be globally valuable. Second, Colombia is not exploiting its enormous potential for biocommerce (oils, medicine plants, ornamental plants and flowers, exotic tropic fruits, natural dyes, and the like). Very small public investments in bringing foreign market information and technological know-how to local communities could help spur this industry. Third, 8 Under the Kyoto Protocol, an international market for carbon credits was created to assist overall global reductions in carbon emissions. The carbon trade enables parties (including industrial firms) in developed nations to meet their greenhouse gas reduction targets through purchases of carbon credits in developing nations. 9 Carbon "sinks" are areas that can absorb the C02 that would otherwise feed global warming. SYN I'HESIS 19 the country has not taken full advantage of nature-for-debt swaps, whereby foreign sovereign debt is written off in exchange for the preservation of natural reserves, indigenous areas, or commitments to set aside areas for future generations. Finally, much as the environment needs to be protected from growth, growth also needs to be "protected" from the environment. Colombia is a country of prevalent natural disasters. In the last quarter century, it has experienced six major earthquakes, three volcanic eruptions, three landslides, and three avalanches, all at a tremendous cost to the human and physical capital of the country These disasters have taught important lessons and, through legal, regulatory, and institutional action over the last decade, the country is now better prepared to deal with them. It essentially oper- ates a decentralized system (the Sistema Nacional para la Prevencuin y Atencion de Desastres Naturales) where local commissions are responsible for mitigating risk and dealing with the ex-post consequences of disasters (for which they can access a National Calamity Fund). The system is, however, weak on two critical fronts, both of which call for policy action by the central government With very few exceptions like the city of Bogota, local governments lack financial and technical capacity to invest in future risk identification, mitigation, and prevention (as menitioned earlier, most of them have not yet completed their Planes de Ordenamiento Territorial, which are supposed to take into account the probability of natural disasters). And insur- ance instruments (be they the pooling and reinsurance of risks, the issuance of catas- trophe bonds, or other financial tools) that could smooth the fiscal volatility that natural disasters entail have not yet been developed; this is of concern, as the asset value in the balance sheet of the state that is at risk from natural disasters is estimated at around 20 percent of GDP. b) Sharing the Fruits of Growth Colombian society is highly unequal. While in the past three decades, sustained growth brought about marked reductions in poverty, inequality continued to increase This not only affects the distribution of income but also of assets and of access to infrastructure-the two top deciles control 60 percent of income, while the bottom two accrue less than five percent; the measure of land concentration is one of the largest in the world (Gmi 0.86); and the coverage of electricity, water, sewerage, and other public services mostly stop at the door of fast-growing informal settlements where most of the urban poor live. Moreover, inequality seems resilient to public pol- icy-it has not been significantly dented by 40 years of land redistribution efforts, low-income housing programs, and schemes for cross-subsidization among users of infrastructure. In fact, many of those policy efforts were not only ineffective in reduc- ing inequality but also hurtful to growth (as the ensuing market distortions damp- ened investment). This is an inauspicious position to be in because, in Colombia, inequality has been shown to be an important determinant of violence The new government is therefore faced with this question What can policymak- ing do to spread the fruits of the growth process without )eopardizing it? This book 20 COLOMBLA THE ECONOMIC FOUNDATION OF PEACE suggests two strategic lines of response: break the lock that interest groups have on the public systems of human capital formation (education, health, labor markets), and attend directly to the needs of those for whom markets will not cater (those that need a "safety net" to cope with systemic shocks, the displaced, and those discrimi- nated against on the basis of race or gender). i. Recovering the Mechanisms for Human Capital Formation Colombia's public education can show remarkable achievements. Those achieve- ments, however, mask serious inequities. The richest 10 percent of the population have more than double the education of the poorest 10 percent. At the current rates of relative educational attainment, the rural sector is 30 years behind the urban areas. Some two million, primarily poor, school-age children and youth are out of school, a worrisome fact since reaching nine years of education in Colombia has been found to reduce the risk of poverty by one third. The large majority of higher education students come from the upper two quintiles of the income distribution- and 40 percent of those go to free public institutions. Those inequities are exacerbated by problems with the quality of public educa- tion, in turn brought about by outdated teacher training (still mostly based on frontal, teacher-centered models), weak teacher supervision (both technical-peda- gogical and administrative), inflexible curriculums (that put emphasis on memoriz- ing information) and, most important, weak accountability mechanisms (standard- ized tests results are not published and do not cover all departments; parent participation is limited). Interestingly, Colombia has experimented in poor rural areas with a different education model, the so-called Escuela Nueva, which puts emphasis on student-centered learning of critical thinking and communication skills in an accountable, community-based environment. By all standards, the model has produced excellent results; however, it has not spread, least of all to the urban areas. The sharp inequity and the stagnant quality of the public education system is not due to lack of resources-over the last three decades public education expenditures quadrupled in real terms, while the school-age population grew by less than one third. The problem is rather that the resources are captured by the vested interest of teachers unions which, on one hand, are reluctant to reform the teaching status quo (especially if it means recruiting fewer teachers, or putting them under heightened accountability) and, on the other hand, find it relatively easy to press politically the subnational entities (departamentos and large cities) to which the responsibility for public education was decentralized in the early 1 990s The subnationals themselves, fearing that their future education transfers could be curtailed if they cut costs and reallocate resources away from teachers, have had no incentive to negotiate reforms with the unions (a few progressive mayors-like Bogota's-did, with good results). It is thus clear that, in the current setup, additional resources will not improve edu- cation coverage or quality; if anything, Colombians are overpaying for the public education they now get. This points to the policy priorities for the new government in regards to education: to link subnational transfers to the number of students SYN IlHESIS 21 rather than teachers (the failed intention of Law 60/1994 and, when actually applied, the main objective of the recent Law 715/2001), and to elicit public accountability for the results obtained with those transfers (for example, linking part of the transfers to the evolution of published, standardized test results). A similar problem of capture by special interests affects the performance of the health sector as a tool for human capital formation-and is bringing major parts of it to the brink of financial collapse. In 1993, Colombia implemented a sweeping, pionecring reform of its health system (though Law 100/1993). The new arrange- ment would ensure efficient, effective, and universal coverage. It was based on sep- arating the funding of health services from the provision-health "promoters" would compete to Insure clients who would then use their insurance benefits to con- tract health services from competing public and private suppliers. People would pay for that insuranice according to their means, with the poor being subsidized by the government and by a portion of the premium for the nonpoor. Public funding would thus go to "subsidize the demand" by the poor, rather than the "supply" by public health providers. The initial results of the reform were impressive-total cov- erage more than doubled, rural coverage increased sevenfold, and coverage among the poorest decile rose tenfold. The World Health Organization now ranks the over- all performance of Colombia's health system as the best in Latin America, and 22nd in the world. Those achievements, however, are currently jeopardized by two main factors First, the state-owned InstitUto de Seguro Social-Salud (ISS-Health), the largest insurer and provider in the market, has not adapted to the new market rules simply because it has been unable to reduce the bloated cost of its 30,000-strong labor force and, thus, to compete with other providers This has called for successive financial rescues from the government (some financed by FOSYGA, the fund that is sup- posed to administer the cross-subsidies for the poor) that, in turn, have perpetuated inefficienit management practices, disregard for service quality, and ballooning arrears wilth other parts of the system Second, similar labor cost rigidities have pre- vented public hlospitals from competing for business and, as a consequence, have all but stopped the transformation of public funding for "supply" subsidies to "demand" subsidies-only some 60 percent have so far been transformed, putting the whole concept of the reform on hold. In fact, public resources for public hospi- tals more than doubled in real terms since the reform began, but two thirds of the increase went to pay for a larger wage bill. What is the appropriate policy response to these problems? As with education, in health there is a need for additional reform, not for additional money The outgoing government attempted, with partial but positive results, to negotiate the reform of ISS-Health and to restructure 27 of some 170 public hospitals. Both initiatives need to be deepened by the new authorities. In particular, two options should now be con- sidered for ISS-Health (both its insurer and provider arms): a complete closedown, or a closedown and reopening on commercial terms (splitting the insurer and the provider into separate companies, and rehiring workers according to competence and 22 COLOMBLA THE ECONOMIC FOUNDATION OF PEACE need). The estimated total net saving to the fiscal accounts from each option would be 1.2 and 2.4 percent of GDP, respectively, compared to the do-nothing-further sce- nario. On the public hospitals front, all of them should now be put under restruc- turing; this will call for an up-front cash injection of about US$500 million but, depending on how deep costs can be cut (the initial restructurings achieved only a five percent reduction) that investment can be recovered in four to seven years. Improving the two main mechanisms for human capital formation-education and health-will mean little in terms of inequality reduction if the poor cannot count on functioning labor markets in which to convert that capital into employ- ment. However, Colombia's formal labor markets are excessively rigid. High and binding minimum wages, a perceived weak link between hefty social security con- tributions and benefits, and powerful unions that are primarily concerned with maintaining real wages result in a so-called "natural" rate of unemployment that is extremely high (about 11 percent), while actual unemployment is even higher (17 percent in mid-2002). The "wedge" between the cost of labor to the employer and the compensation received by the employee is worth about 50 percent of the wage bill (compared to 39, 31, 19 percent in France, Mexico, and the United States, respectively). Not surprisingly, 60 percent of the workforce operate in the informal sector; in contrast, only one in 10 Colombian workers belongs to a union. This calls for a major overhaul of the legal and regulatory framework in which the labor markets function. In particular, contributions for which the individual worker sees no direct benefit (parafiscales) should be eliminated and the correspon- ding programs (like training and family welfare) should be funded out of general tax revenues."° The minimum wage should be nonbinding and should not apply to apprenticeships, temporary workers, or public workfare programs. Collective bar- gaining should be done at the firm level, not at the industry level. Legal provisions extending equal benefits to all workers in an "economic entity" (principal firm, sub- sidiaries, and affiliates); those imparting seniority-based promotions, compensation and training; and those mandating indefinite rollovers of fixed-term contracts, should be abolished ii. Helping Those for Whom Markets Will Not Work While strong human capital and functioning labor markets will be an effective means for many Colombians to share in the fruits of growth, there will be others for whom market-based solutions will simply not work and for whom direct govern- ment support will be necessary-those that cannot cope with systemic crisis, those that have been displaced, and those that are discriminated against on the basis of race or gender. I0 The elimination of all parafiscales may not be politically feasible, especially those associ- ated with family welfare. A second-best solution would be to keep those that fund the most sensitive, best-targeted social programs and, for these, to separate financing from provision (opening the provision to competition by public and private suppliers). SYNTIHESIS 23 Colombia never really had a formal social safety net, something that became painfully clear during the unprecedented macroeconomic crisis of 1999. Existing social programs (like the Servicio Nacional de Aprendizaje, SENA; the Caja de Com- pensaczdn; and the Instztuto Colombiano de Bienestar Familiar, ICBF) are paid out of payroll taxes and only cater to formal workers. The sweeping reforms that accom- panied the enactment of the 1991 Constitution brought about neither a strategy for social assistance nor additional funding for it Thus, in response to that crisis, a tem- porary five-year Social Support Network (Red de Apoyo Social) was created to fund temporary employment (Empleo en Accu5n), conditional cash transfers to poor fam- ilies (Familias en Accid5n), and firm-based apprenticeship grants for youth (Jovenes en Accidn). The network, which took two years to unfold fully, has several loopholes, but one is particularly worrisome-it does not sufficiently protect children and ado- lescents at risk. About 70 percent of urban children under the age of seven are poor, a proportion that reaches almost 90 percent in rural areas. Roughly half of all chil- dren have not been vaccinated against basic communicable diseases. And the armed conflict limits the effective reach of the government's safety net in certain rural areas, precisely where the youth are being recruited (mostly by force) into combat. All this is both a challenge and an opportunity for the new administration. Based on the lessons that the completion of the RAS will bring in 2004, it should proceed to establish a permanent and comprehensive Social Risk Management System. This need not mean new budget appropriations; rather, it should absorb the funding cur- rently going to the SENA, the Caja de Compensaczo'n, and the ICBF This should be accompanied by an overhaul of the means-testing mechanisms (the SISBEN), some- thing that could also improve the targeting of subsidies in infrastructuie services (discussed earlier). Overall, the incominig government will have to decide whether Colombia, which at the moment spends a low 0.7 percent of GDP on social assis- tance, will have a standing safety net arrangement and, if so, what would be its prod- ucts and its countercyclical features. Even a properly functioning social safety net would, however, need to be adapted to cater to the urgent needs of Colombia's displaced population. Unofficial estimates put the number of desplazados at some two million people (about five percent of the population) They have been traumatized out of their rural homes by the armed conflict; have lost jobs and assets (and, in most cases, even the titles to their assets); can no longer access formal education or health services; and are lingering in slums at the outsklrts of large cities. The ma)ority are women, about half are under 18 years old, and one fifth is either Afro-Colombian or Indigenous. Virtually all are poor. Most distLrbinigly, they are stigmatized and isolated in the receiving localities, something that, until very recently, made them invisible as a political and policy pri- ority. It is difficult to see how Colombia will attain a functioning level of social cohe- sion if the plight of the desplazados is not attended to, a plight that will not go away and will get worse if unresolved. It was not, however, until 1997 that a legal framework (Law 387/1997) was enacted providing for "the integral care" of the displaced, and a single institution 24 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE (the Red de Solidaridad Social) was designated to implement it. The RSS, in coop- eration with the Interniational commUiuity, issued a corresponding action plan in late 1999. So far, some US$50 million have been invested in that plan, and only an esti- mated 17 percent of the desplazados have been reached. The new government needs to bring the aid program for the desplazados to a much higher plateau. To this end, there are five priorities. First is to give them means of political representation, including mobile voter registration services-this would ensure sustained policy attention. Second, support mechanisms should move from mere humanitarian aid to schemes for return, relocation, or integration, as the case may be, and adequate budget resources should be made available (a strong, well-articulated commitment by the Colombian government is likely to attract substantial donor funding). Third, the operating rules of all existing social assistance programs, including those that are based on means tests through the SISBEN, should be modified to allow for priority consideration of displaced cases. Fourth, local governments should be co-opted to take the lead in preventive action in high-risk regions, for which additional resource transfers from the central government will be needed. Fifth, a uniform methodology to track, classify and, ultimately, understand the displaced populations should be pursued. Finally, race- and gender-based discrimination still distorts the way many Colombians can profit from the growth process. Colombia is pluriethnic nation with some 800,000 indigenous peoples (belonging to 82 different groups) and over one million Afro-Colombians. The country has a strong tradition of recognizing ethnic rights (cultural institutions, communal properties, political participation) and enshrined them in the 1991 Constitution. However, several factors have now combined to weaken those populations' ownership of one of their defining assets- their land-and, thus, their capacity for market-based development. Some of those factors are technical in nature and within reach of public policy (like more precise surveying, adequate collective titles, better rules for assessing public services and pro- grams); others are more difficult to address (the colonization process, illicit crops, and, critically, the internal conflict). Contrary to common observations elsewhere, in Colombia the ascription of gender-based roles is more harmful for men than women. Girls have higher school enrollment levels than boys, women's fertility and maternal mortality have decreased, women's labor market participation rates have increased, and gender gaps in wages have narrowed. Not only do boys perform worse in education, but also males are more affected by HIV/AIDS-, alcohol-, and drug-consumption-related diseases. More critically, males are the disproportional victims of violent death because of both the armed conflict and crime. Available research explains this pat- tern as low-income, unemployed males seeking to fulfill through violence a socially expected identity as family providers, a role they have failed to play through pro- ductive work. School curriculums and learning materials that refrain from gender stereotypes and teach conflict resolution skills; media-based campaigns that provide nonviolent role models; and community-based programs that seek to identify and SYNTHESIS 25 prevent local sources of violence (like the government's Convzvenczay Seguridad Cmu- dadana) are the best ways to address gender-driven social issues for males c) Building a Government of Quality Reaching a path of fast and sustainable growth, and giving people the tools to share in the fruits of that growth, is a difficult policy agenda for any government. In today's Colombia, it will be a formidable challenge, for the armed conflict, the ille- gal drug industry, and the capture of policymaking by interest groups have all but paralyzed the effective authority of the state. The new administration will thus need to rebuild early on the government's management credentials in front of its citi- zens-in other words, they will have to enhance the quality of the government func- tion. There are, of course, very many ways in which the quality of government can be improved, and prioritization will be crucial. This section argues for five areas of immediate focus-reform of the state's functions, decentralizationi, budgetary insti- tutiols, corruption, and justice. Colombian governments have for decades exhibited a dichotomic performance in discharging their functions. On one hand, a technically competent, domestically independent, and internationally accountable set of institutions (among others, the Central Bank, the Ministry of Finance, the Departemento Nacional de Planeaczon) consistently watched for and delivered solid macroeconomic policy. On the other hand, the array of sector institutions administering microeconomic policies and pro- grams have had less success, and have been regularly perceived as captured by their own bureaucracies and by the surrounding sectoral interest groups. This perception covers a large array of such institutions operating in many fields-from the Cajas de Compensacidn Familiar and the ICBF (operating family welfare programs); to INCORA, INVIAS, the SSPD (in land, roads, and residential public services); to labor unions (stifling decentralization efforts in education and health). This background of macro-competence and micro-capture hints at three basic principles that should guide a meaningful reform of the state in Colombia. First, while some organigramic reorganization will be necessary, and redtindant and duplicative agencies will have to be merged or closed (notably in agriculture), the reform should be about fixing critical "functions" within the state, not about reordering reporting lines and shifting personnel among ministries and institutions. Those ftinctions include much-enihanced accountabiliry in decentralization transfers (see below); an initial shift toward results-based investment budgeting in the central budget (also discussed below); a major overhaul of procurement systems (with trans- parency as the guiding principle); much stronger internal and external auditing of the central government (especially, the Contraloria General de la Repablica, which is the external auditor of the Executive); currently-missing tools for evaluating public expenditure (in terms of their impact and beneficiaries, not Just their procedural propriery); more efficient and more independent regulation in infrastructure (espe- cially in roads, water, electricity, and gas), and better mechanisms to detect, report, 26 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE and punish abuse of power and rent-seeklng (notably, through capacity upgrading at the Procuraduria General). Second, given the entrenchment of special interests, early reform successes in a few focal functions (and in their related institutions) will be preferable over sweeping, across-the-board undertakings that are likely to run out of political steam very quickly (Colombia has tried time and again to pass and implement comprehensive public sec- tor reforms-most recently in 1992-93 and 1998-99-but was soon derailed by the political resistance of vested interests). Third, some of the functional reforms will, at times, trigger changes in major agencies; support at the highest level of government will then be necessary to proceed and, thus, set Important precedents for lesser insti- tutions For example, better evaluation of public expenditures and better procurement systems may have implications for, respectively, the ISS and ECOPETROL; reform of these, would send a clear signal to the rest of the administration. Focusing on state functions will naturally lead to the second key priority in upgrading the quality of government in Colombia-decentralization. Since the mid- 1980s, the country led Latin America in embracing the principle of bringing poli- cymaking closer to its beneficiaries (partly as a tool to address the causes of the armed conflict), enshrined it in the Constitution of 1991, and implemented it almost at once. The results have fallen well short of expectations, and have threat- ened not only the efficiency of sectoral expenditures but also the fiscal pillar of macroeconomic stability. This presents a strategic question to the new administra- tion: Should Colombia further decentralize, should it recentralize, or should it pro- ceed with an all-out federalization? It is argued here that decentralization remains the best option, but major adjustments to the process are needed to deal with its main weaknesses-lack of incentives to rely on local taxation, to enhance efficiency in service delivery, and to maintain fiscal discipline. First, while the constitutional amendment Acto Legislativo of 2001 brings all transfers into one single fund and fixes its growth, there is still no mechanism to evaluate, and hold subnationals accountable for, the results they achieve with those transfers. Nor has Law 715/2001, which clearly shifted responsibility for core social services to the subna- tionals, established delivery standards (this is naturally worrisome, because subna- tionals now handle almost half of all public sector revenues-about 15 percent of GDP). Further legislative action to that end is necessary, especially in education, to make the very transfer of responsibility conditional on performance."1 Second, the volume of transfers (like those meant to cover teacher payrolls) should not be based on historical supply cost, but rather, on need-and-effort (like cost per student or "capitation"); the above-mentioned Law 715/2001 calls for that. However, for the transformation to be at all possible in practice, teacher salaries (the largest decentralized expenditure) and employment policies can no longer be deter- mined by national negotiations with the unions, but at the subnational level. Third, 11. Law 751/2001 provides enough flexibility to instrument performance agreements (acuer- dos de desempeno) with the sub-nationals SYNTHESIS 27 subnational governments, especially large municipalities like Bogota and Medellin, should be allowed to surcharge and fiscalize the income tax and the VAT (the draft law on Taxation Territorial currently in Congress is a good step in that general direc- tion). Fourth, subnational borrowing regulation, currently based on certification by the central government-semaforos-and clauses in past bailout agreements-under Law 617/2000-need to become market based. For this, the current, essentially vol- untary banking regulations linking capitalization requirements for loans to subna- tional borrowers to published credit ratings, must become compulsory Fifth, while in theory a rearrangement of political divisions and levels of government could bring improvements to Colombia's decentralization process (that is, a so-called Reorde- namiento Territorial, regrouping current departments into regions or provinces), the country's political and fiscal reality advises against it at the moment. Finally, central government resources should be invested in developing a common budgeting, accounting, and reporting framework at all levels of the state, as well as common procurement standards. Both the reform of the state's functions and the adjustment of the decentraliza- tion process will be mutually reinforcing with the third key element in improving the overall quality of government-better budgetary institutions. In recent years, Colombia has made significant progress in this area, notably with the establishment of a single treasury account for its central government (Cuenta Onica del Tesoro). This technical improvement has, however, not been matched by better policies in designing and operating the budget itself Several areas for policy action stand out. The National Development Plan should become a strategic, indicative roadmap, rather than a bloated, legally binding list of wished-for projects the execution of which is then controlled discretionally through cash management. International accounting standards (like the IMF's Government Finance Statistics Manual) should replace the rather loose definitions currently applied in Colombia's budget (for example, in regards to investment outlays). The array of legal entitlements and earmarks accumulated over the years should be reviewed and their costs and benefits put out for public consideration Congress and the public should be given reading access to the government's financial infor- mation system (Sistema Integrado de Informaci4n Financiera) and to the currently- under-construction results assessment system (SINERGIA). The split responsibility for the "operational budget" (current expenditure) under the ministry of finance, and the "investment budget" under the Department of Planning should be unified, and only one of those institutions made responsible for the whole budget, some- thing that will avoid funding investment projects without adequate maintenance allocations in subsequent years. More broadly, public sector budget information should be "regulated" and clear responsibilities assigned; at present, several institu- tions collect and publish (sometimes inconsistent) data on the same matter, an obvi- ously inefficient and confusing use of resources. The fourth pillar in raising the quality of government in Colombia is the fight against corruption. As with other aspects of the country's social life, the 1991 Consti- 28 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE tution set out a framework to reduce corrupt practices in public office (by, for exam- ple, establishing the officc of the Attorney Gencral), but had little impact on the per- ceived integrity of the government function. Surveys indicate that Colombians see corruption as a rampant problem in their society. The traditional indicators of integrity (in licensing, tax administration, customs, and the like) are, however, much better than the average, certainly the average in Latin America. The reason is that, in Colombia, corruption takes a particular form-"state capture"-that is, high-level government officials affecting, for their own financial, social, or reputational benefit, the rules that govern the relationship between interest groups and the state This is an especially perverse form of corruption, because it goes well beyond specific transac- tions and puts a cloud of illegitimacy over the continuum of government dictums. As such, the solution lies not as much in improving administrative processes (although some of these need reform), but in the reforms of the party and electoral systems. Many of those reforms, like changes to the campaign finance laws, fall outside the purview of this book. But others are a matter of development policy: a profes- sionalized civil service, top-level management appointments that are done on the basis of public competitions run by outsourced human resource agencies, and com- pensation and job stability that are based on performance; a procurement system that is online and open to public scrutiny; enhanced independence for key regula- tory institutions (not surprisingly, the above-mentioned surveys show the constitu- tionally independent Central Bank to be among the least corrupt and the most com- petent institutions in the nation); the establishment of a single institution in charge of fighting corruption that reports directly to Congress; and, as discussed below, a much more effective judicial system. A better-functioningjudicialsystem is, indeed, critical to abate the problem of cor- ruption and, more generally, is a core priority to improve government quality in Colombia. The system was one of the main targets of the 1991 Constitution-a new constitutional control jurisdiction was created in the form of a Constitutional Court (over and above the existing Supreme Court and State Council); a new arrangement for criminal justice was set up, headed by the new office of the Attorney General; a new legal figure (the Tutela) was instituted giving citizens the right to stop and dis- pute at any judicial level any action that may hurt their constitutional rights (which, in turn, were widely defined, like the "right to life"); the administration of the judi- cial apparatus was given to a new Superior Council of the Judicature; and a new juris- diction of community-based justice was opened (the so-called Justices for Peace). The new justice function was then given a major increase in its resources (the judicial sec- tor's budget has tripled in real terms since 1991). The overall objective was to raise the judicial system's performance, especially in terms of congestion, quality, and chronic corruption. The results have been disappointing; if anything, access and res- olution rates deteriorated (some 4 million cases, one for every 10 Colombians, await disposition). Why, and what can now be done to fix it? The new Constitutional Court has pro-actively, and distortingly, intervened in matters of economic policy (like reversing a freeze in public sector wages and cap- SYN rHFSIS 29 ping interest rates for mortgages); the roles of the Constitutional Court and the Supreme Court are in practice overlapping, causing a great deal of judicial uncer- tainty, and the Tutela has become a much-abused way to dispute contracts (notably, labor contracts) and has clogged the courts. Optimally, such structures should be changed. Political support for reversing some of the mandates of the 1991 Consti- tution is, however, highly unlikely to materialize. Adapting the old judicial appara- tus to the new reality is, therefore, the best reform strategy in the sector This can be achieved through a combinationi of six main actions. First, justice services can no longer be completely free in all branches and for all cases, if not to generate revenue at least to restrain superfluous use of the system. Dur- ing the last decade, 80 percent of the cases in the civil jurisdictioni were related to commercial payments collection; these should pay for all or part of their adjudica- tion, even if it is on a means-tested basis. Second, procedural rules and their associ- ated logistical arrangements should be adapted to deal with Tutelas through swift, oral proceedings. Third, the governance of the administrative Superior Council of the Judicature should be made more accountable to the system's users, its board should be dominated by stakeholders outside the judiciary itself Fourth, additional resources within the overall judicial budget should be reallocated to the criminal justice branch (where the demand for )ustice services far outstrips the installed capaciry) but, in par- allel, the Attorney General's Office should be made publicly accountable for an agreed set of socially relevant performance indicators (like serious-crime resolution rates). Fifth, the union-driveni rigidities in retraining and reassigning (let alone dis- missing) judicial personnel to functions more compatible with the new reality should be overcome (a telling example- there is no need for a lawyer to play the role of court secretary). Finally, a sizable investment in improving the physical infrastructure through a computerized case-management system seems long overdue. IV. A Path to a New Colombia The development policy agenda described above calls on the new governmenit to reach a higher and more sustainable growth plateau, to give all Colombians a chance to share in that growth, and to make the state itself an icon of quality. Even in nor- mal circumstances, this would be a formidable challenge. For a country dealing with an armed conflict and pent-up social demands, the agenda is nothing but over- whelming. But it is also feasible. Colombia can still marshal the talent and resilience of its people, the wealth of natural riches on its land, and the deserved support of the international community to create a new economic pillar on which to build peace. The presidential elections of 2002 brought about a clear mandate and a unique opportunity to construct that pillar. The tasks ahead are many, and the ability of the government to implement reforms is necessarily limited by the capacity of the state's institutions and the polit- ical realities of the country. Prioritization will thus be crucial, and a clear imple- 30 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE mentation map will be necessary. Table 1 presents an example-and only an exam- ple-of such a map. Determining what is possible when remains, however, the sole prerogative of the new authorities. It will be up to them to assess their preferred sequence of policy actions according to their own strategic objectives and the avail- able consensus. Not all policy initiatives will succeed. In some cases, reaching agree- ment among stakeholders will prove elusive, as is rightly expected in a vibrant democracy like Colombia. In other cases, first-best solutions will simply have to give way to more practical arrangements. Most of all, not everything that needs to be done can be done within a single presidential term. So, what kind of country would Colombia be, four years from now, if a policy agenda along the lines of Table 1 were implemented? It may not be rid of the inter- nal conflict. It would certainly not be free of all its economic problems. There would be a lot more jobs, but many people would still be jobless. The painful emigration flow and its associated brain drain would have slowed down, but not reversed. Colombia would not even be a "rich" country. But it would be a country on a fast path of construction, where development would give hope and opportunity to all. Table 1. Colombia: A Possible Prioritization of Policies for the New Presidential Administration From the very beginning... Protect the macroeconomic framework Send early, confidence-building signals Make human capital development an * Launch a growth-friendly tax reform (with * Pass a comprehensive pension reform explicit priority compensatory programs for the poor) * Put specific privatization efforts on the * Commit to quantifiable improvements in * Announce reform of the state agenda (for example, telecommunications) the quality of education * Pass a strengthened "Fiscal Responsibility * Fill senior positions in the civil service * Commit to freeing the health system from Law" through open competition vested interests * Reach out to the Desplazados After thefirst 100 days... Take first steps toward higher growth Gather consensus for reforms in human Bring governance to the forefront of the * Launch the reform of the coffee sector capital formation agenda * Press on with the commercial banking . Begin publishing the disaggregated results * Announce a detailed plan to combat sector's consolidation and reform of existing standardized tests corruption * Enhance regulatory independence in Launch participatory evaluation (cost and * Put the government's procurement and infrastructure beneficiaries) of all social programs financial management systems on-line Eliminate all parafiscales * Launch reforms in the judicial system After thefirstyear... Focus on enhancing long-term growth Begin reforms in human capital formation Begin to exploit the synergies between * Overhaul the SSDS and make all sector * Link all education transfers to cost per growth and environmental protection regulatory commissions truly independent student and performance agreements . Bring Colombia to the international car- * Reform cross-subsidization mechanisms in * Close or reengineer ISS bon trade market infrastructure * Put all public hospitals under restructur- * Fund the completion of all Planes de * Close the land redistribution program, ing plans Ordenamiento Territorial and covert it into cash subsidies for the * Multiply local producer-protector reserves rural poor (continues on next page) Table 1. (continued) By mid-administration... Consolidate the higher growth plateau Complete and Consolidate Reforms in Bring government quality to the local level o Articulate an exit strategy for all second- Human Capital Formation . Give municipalities larger taxation powers tier state-owned banks * Convert all health subsidies into Link all transfers to open performance o Complete privatization in electricity and "demand" subsidies standards in service provision telecommunications o Launch labor reform o Put market-discipline in sub-national * Open agriculture to international o Create a permanent "Social Risk borrowing competition (a true apertura) Management System" And leave behind at the end of the presidential term... An economy that is growing A stronger human capital A "greener" Colombia And a state that is respected for fast and is sustainable * The poverty headcount has * Where deforestation has been its improved efficiency * Real long-term growth is at fallen by at least ten percent- stabilized * With institutional capacity at least five percent per year, and age points since 2000, and a . Extractive industries do not the central, department and unemployment is below 10 permanent social safety net is systematically damage the municipal levels percent in place environment * A judicial system of quality, * The fiscal accounts are in The education system begins Blodiversity is preserved, speed, and access surplus and inflation is to turn out skillful, critical celebrated, and profitable . And internationally acclaimed rn comparable to that of the thinkers transparency 0 G7 countries Basic health coverage is * External and public debt efficient, universal and positions are comfortable financially sustainable '11 0 z 0 Part I Thematic Notes 1 Violence, Sustainable Peace, and Development This Chapter was written by Elsie Garfield andJairo Arboleda. I. Introduction There is a clear consensus among Colombian citizens of all walks of life that the current state of insecurity and violence that prevails in most of the country is unacceptable. It has resulted in a poor quality of life, and has created a poor investment climate, a serious obstacle to economic growth and employment gen- eration, and, many would argue, a threat to the social fabric and very survival of Colombia as a nation. In a poll taken in February 2002 following an upsurge of violent acts related to the end of negotiations with the Fuerzas Armadas Revolu- cionarias de Colombia (FARC), 92 percent of Colombians ranked violence as the problem with the most perverse effect on them and their families, far above any other social or economic issue. As was the case in 1996-97 when the World Bank was developing with Colombia its Country Assistance Strategy (CAS), there still seems to be widespread agreement that a top priority in Colombia's development agenda is peace. II. Violence Trends and Characteristics Colombia is a country with a long history of violence going back to the war of One Thousand Days in 1899, which left 100,000 dead, to La Violencia in the late 1940s and early ]950s. Since the 1980s, the multidimensional problem of vio- lence has become more widespread and is exacting an increasing economic and social toll This thematic chapter draws heavily on Alberto Chueca Mora's research on the sub)ect 35 36 COLOMBLA THE ECONOMIC FOUNDATION OF PEACE TRENDS IN VIOLENCE HAVE WORSENED AND ARE SOME OF THE HIGHEST IN THE WORLD. Using the homicide rate as a proxy for violence levels, official figures increased from 15 to 92 per 100,000 inhabitants between 1974 and 1995, with levels growing dramatically in the post-1985 period. The homicide rate is one of the highest in the world: three times higher than in particularly violent countries such as Jamaica or Russia, seven times higher than in the United States, and 50 times higher than in a typical European country (see Figure 1). In some cities murder took on almost epidemic proportions in the early 1990s Medellin Figure 1. Murder Rate per 100,000 Population, 1993-94 C olom bia - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Jamaica Russia Estonia Ecuador Scotland Nicaragua Bulgaria Israel I Costa Rica Uruguay Kuwait Northern Ireland Italy India Jordan 3 Sudan Ungria I Finland - Greece `I France- - , Chile---| Slovenia -! Canada-I Sweden -|i Malaysia - 1 Singapore - Australia -- Hong Kong -- Denmark Korea Belgium Austria 3 Cyprus -i Indonesia I Spain i Japan Egypt _I I II I 0 20 40 60 80 Source Global Report on Crime and justice. United Nations VIOLENCE, SUSTAINABLE P'EACE, AND DEVELOPMENT 37 recorded a rate of over 400 per 100,000 population. The rate of homicides attrib- utable to internal armed conflict has also risen in the last 20 years and reached lev- els experienced during the period of La Vtolencia (see Figure 2). In terms of dura- tion and loss of life, Colombia's armed conflict (1948-62 and 1984 to the present) is among the five longest and most intense in the world, comparable only to coun- tries like Afghanistan, Angola, Rwanda, and the Sudan (Echeverry, Salazar, and Navas 2001). All types of crime-including extortion, kidnapping, car theft, and armed rob- bery-flourished in the 1980s and continued to rise significantly during the 1990s. Further, the crime boom was geographically concentrated, such that the offenses grew tenfold in some places, while in others it remained almost unchanged Between 1985 and 1998, the number of reported kidnappings rose from about 9 per million persons to about 80 per million persons, with the majority of victims being civilians (Mejia 2000); this was an important source of revenue for some armed groups. An upsurge in crime in urban areas is linked to the upsurge in drug trafficking, partic- ularly in Cali and Medellin, and to unemployment and poverty, and to precipitators such as availability of arms and consumption of alcohol. INSURGENT ARMED CONFLICT Is MORE ENTRENCHED AND WIDESPREAD THAN AT ANY POINT IN THE PAST THREE DECADES AND AREAS WITH AN ACTIVE STATE PRES- ENCE HAVE DECLINED. During the past five decades, the number of actors involved in the armed conflict has expanded from the guerrilla and armed forces to include drug cartels and right-wing armed paramilitary groups. This has been accompanied by increasing density-both spatially and in terms of interdependent reciprocal net- works between different actors. The internal armed conflict has become a struggle to control territory with many areas of the country, particularly rural, under active dispute between guerrilla and right-wing armed groups. Municipalities with some type of guerrilla presence increased from 17 percent in 1985 to 58 percent in 1995. If areas experiencing paramilitary, drug, and armed forces activity are added, approx- imately 75 percent of the country is experiencing some level of armed conflict. The Colombian military and police have been unable to ensure the security of its citi- zens, and impunity is widespread in the face of increasing human rights violations of all kinds. As a result there has been a loss of sovereignty of municipal adminis- trations and a growing number of displaced persons. VIOLENCE ESCALAT'ED WITH TIHE UPSURGE IN THE DRUG INDUSTRY IN THE 1980S. Numerous studies have shown that Colombia's involvement in the illegal drug trade is a key factor explaining the dramatic increase in violence, particularly homicides, over the last two decades. The drug trade has exacerbated levels of violence related to organized crime around drug production and processing, the militarization of the fight against drugs, and disorganized violence around distribution and consump- tion It has also fueled violence by providing a steady source of funding for several of the key illegal armed groups. Figure 2. Murder Rate, 1946-98 130- 120 Deaths attributable to Deaths today _ "La Violencia" 100 __________ 90 C) 80- 70 70 F- __ ___ ____ 40~~~~~~~~~~~~~~~~~~~~~~~~~~~~ v~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ C. - ~~~~~~~~~~~~~~~~~~~~~~~~~~~0 1 0 - - - - - - - - - . .. . . . . 0 z~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I. I ~~~~~~~~~~~~~~~~0 2 0Source National P e-I Source Nauional Police VIOLENCE, SUSTAINABLE PEACE, AND DEVELOPMENT 39 VIOLENCE INCREASINGLY AFFECTS SOME GROUPS DISPROPORTIONATELY. All Colom- bians have been terribly affected by violence. However, the rural population in gen- eral, youth, and ethnic minorities have been most affected, particularly by the armed conflict. Younger, lower-income, and less educated men are more likely to be both perpetrators and victims of homicide. The homicide rate for men thus multiplied by 13.5 between 1980 and 1995, with shifts throughout this period toward younger men. The number of internally displaced people has increased over the last decade and is still growing. According to a 1995 study, 600,000 people were forced to leave rural areas during the previous decade (Meertens and Segura-Escobar 1996). The Presidential Council for Human Rights estimates that the current number of dis- placed people has reached 1 million, about 2.5 percent of Colombia's population. Women and their dependent children living in rural areas are the group most affected by displacement due to the armed conflict' 58 percent of the displaced pop- ulation are female, and 75 percent are under 25 years old (Conferencia... 1994). Indigenous communities have been disproportionately affected by armed conflict throughout the country. III. Sources of Violence and Conflict While there is broad agreement that the current state of insecurity and violence is unacceptable, there are significanit differences within Colombian society, among academic and other analysts of Colombia, and within the international community about the causes of the situation, and the appropriate strategy for progressing on a path to peace and development The many excellent studies and analyses of violence in Colombia, particularly of the last two decades of the 20th century, reflect these differences. It is our contention that no single analysis or perspective has sufficient explanatory power, but that taken together they provide important insights that are useful (Moser 2000). This is due to the multidimensional and interrelated nature of the sources of violence in Colombia, the importance of looking at its evolution over time, and the rapidly changing dynamics particularly at the local level. This section discusses the sources of violence and conflict,' and Section IV discusses the costs as a basis for drawing some preliminary conclusions about a path to sustainable peace and development. A somewhat stylized, simplistic presentation of the principal explanations of the current situation are that (a) the driving force for the escalation of violence in the 1980s is the illegal drug trade which has fueled criminal activity and financed the armed groups, which are motivated primarily by greed, tending to lead to a conclu- sion that punitive actions involving force are the solution; and (b) the latest phase I This discussion does not touch on interpersonal and domestic violence, which is a serious problem in Colombia These are discussed in more depth in Moser (2000) and Moser and Mcllwaine (2000) 40 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE of the armed conflict is rooted in long-standing social conflicts related to an exclu- sionary political, economic, and social system, and has been aggravated and perpet- uated by funding obtained by the armed groups through the drug trade, tending to lead to a conclusion that achieving sustainable peace requires structural changes involving a redistribution of power and assets within Colombian society. As indicated above, we consider each of these stylized explanations to have valid- ity, particularly if one enriches them by looking at the situation of the last two decades from the perspective of Colombia's long history of violence. We will briefly examine the sources of the complex phenomenon of violence in Colombia by mak- ing a distinction among: • The armed conflict (often referred to by many Colombian analysts as "politi- cal" violence) of the more recent period involving the guerrilla forces (FARC, National Liberation Army (ELN), and so forth), peasant self-defense move- ments, paramilitary groups, and the Colombian military and police * The long-standing, underlying social conflicts at the local, regional, and national levels, which have also manifest themselves in violent ways and which are claimed by various armed groups as a motive and justification for their actions o The illegal drug trade and related criminal activities (arms trade, money laun- dering, kidnapping, and so forth), which have interacted with and con- tributed to the worsening of both the armed conflict (with the drug cartels becoming an armed actor) and social conflict in complex ways that are diffi- cult to disentangle. a) Armed Conflict Colombia has a long history of violence: confrontations between liberals and con- servatives in the 19th century, the period of La Violencia in the mid-20th century, and the escalation of armed conflict and crime since the 1980s. In the past decade, the scale and intensity of the internal armed conflict has changed from one confined to certain rural areas to a generalized violence that dominates the daily lives of most citizens. As remote guerrilla activity has turned into a conflict that includes new armed actors including paramilitary groups and drug cartels, and a steady source of funding, the nature of the conflict has been transformed and its negative impact on Colombian life has become enormous. A distinguished analyst of Colombia's situation over the last two decades describes the current armed conflict as a "war against society" where a generalized state of violence prevails and the majority of citizens are in effect held hostage by the various armed groups. Domination of people and territory appears to be the core military strategy of each of the principal illegal armed groups; the original social jus- tice concerns that motivated the founding of the guerrilla groups seem to be of less importance (Pecaut 2001). Many areas of the country are either dominated by one VIOLENCE, SUSTAINABLE PEACE, AND DEVELOPMENT 41 group or are under active dispute The inability of the State, through the military and police, to exercise a monopoly of force and protect people and property has per- mitted the illegal armed actors to move into these areas and exercise authority The municipality has become a key arena for the armed struggle, making it nearly impos- sible in many parts of Colombia for citizens to exercise their rights and for local gov- ernments to carry out their responsibilities. The pervasive impact of this situation is discussed further in Section IV. There have been many peace initiatives in Colombia since 1982, with a small measure of success. The faltering of the peace process initiated by the Pastrana gov- ernment in August 1998 and the roller-coaster nature of the negotiations process with the principal guerrilla group, FARC, coupled with the many brutal acts com- mitted by each armed group against people and property, have led many Colom- bians to lose confidence and hope for a negotiated settlement of the armed conflict. Yet, as noted above, it would be too simplistic to reduce the discussion of violence and peace in Colombia to the armed conflict. b) Social Conflict A variety of studies indicate that Colombia's violence and armed conflict stem from a complex interaction of economic, social, historical, and political factors at the local, regional, and national levels (L6pez and Garcia 2000). International evidence also supports the importance of these factors. POVERrY. While Colombia made significanit progress in reducing the poverty rate (measured in terms of per capital income) between 1978 and 1988, a serious rever- sal in the second half of the 1990s brought these rates back to their 1988 levels. Growth has nor reduced the large gap between the incomes and well-being of rural and urban residents: the level of extreme poverty in rural areas is three times that of urban areas. Although it may not be possible to establish a direct causal relationship between poverty and the armed conflict, the relationship appears to be dual: poverty favors the appearance of violent conflict, and the conflict itself is a major cause of poverty as evidenced by the large number of persons and families from humble rural origins who have been forced to flee to poor outlying areas of cities leaving behind their land and other possessions which they have little hope of recovering. There is no doubt that the poor are the main victims of violence, whether in rural or urban settings Both in the regular armed forces and in the illegal armed groups, young men from poor families are on the frontline of the armed conflict. Likewise, lack of access to jobs, good education, and health services, the basic amenities of life, and opportunities offered by the drug trade are undoubtedly linked to an upsurge in crime in urban areas INEQUALITY The increase in violence over the last 20 years has occurred simultane- ously with an increase in economic disparities. The growth of the guerrilla move- 42 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE ment has run parallel to the deterioration in all income distribution indicators. The number of guerrilla fronts has risen from 14 in 1980 to 102 in the mid-1990s; simultaneously, between 1982 and 1999 the share of the poorest 20 percent of the population in income fell 30 percent (from 4.9 percent to 3.4 percent of total income, according to the National Household Survey). At the same time, the top 10 percent earners have increased their share by 5 percent (up from 37.1 percent to 43.6 percent). In the last 20 years, it is not only the gap between rich and poor that has grown, but also the gap between the cities and rural areas. In 1975, an urban family earned 1.5 times more than a rural family, but 20 years later it earns 4.5 times more. A study comparing municipalities during 1985-96 found that the group of municipalities with an increasing rate of violence had a higher level of inequality measured in terms of the Quality of Life Index, and lower rates of political partici- pation and level of education (Sarmiento 1998). A recent study of 39 countries found a positive relationship between high levels of violence (measured by homicide and robbery rates) and income inequality (Fajnzylber, Lederman, and Loayza 2002). Colombia shares with other Latin American countries a turbulent and often vio- lent history of land relations. Access to land is a major source of political and social conflict, in addition to being a key determinant of the productivity of the rural economy. Land ownership is highly concentrated, particularly in certain regions. From 1984 to 1997, there was a clear tendency of increasing concentration of land ownership, which is explained in part by drug monies being invested in land; the physical area of large farms showed a marked increase-from 32 percent to 45 per- cent of the total farming area. Colombian experts have shown that regions with a history of a high concentration of land ownership and conflict over land are those with a high presence of illegal armed groups and purchases of large farms with drug monies resulting in increased inequality in land distribution (Reyes Posada 1998). Many experts and journalists have also discussed the financing of paramilitaries by larger landholders to provide a measure of protection for life and assets. POLITICAL EXCLUSION, POOR GOVERNANCE, AND CORRUPTION. Many analysts point to the concentration of political power in the hands of an elite and the exclu- sion of new actors-particularly during the power-sharing arrangement between the two traditional parties during the National Front period (1958-74)-as a source of conflict and discontent (L6pez and Garcfa 2000) A recent survey about corruption in Colombia revealed that the perception and experience of public servants and business people is that legislation and special privileges are negotiated at the top among business concerns, politicians, and the executive branch. The survey revealed a generalized lack of credibility and legitimacy of public institutions, which are not thought to serve the public interest. HIGH LEVELS OF IMPUNITY WITHIN THE JUSTICE SYSTEM. The Colombian State has proved unable to effectively serve as the arbiter of conflict among citizens, though either formal or informal institutions. The perception is that the more powerful pre- VIOLENCE, SUSIAINABLE PEACE, AND DEVELOPNIENT 43 vail, especially by use of influence or force. This, in turn, sends the signal that con- flicts need to be resolved by private means because the State is unable to do so, which creates lawlessness and a climate conducive to more use of force by various actors. This situation is examined in more detail in the Chapters on The Judiciary and Corruption c) Drug Trade It is widely agreed that the emergence of drug trafficking has been one of the most important developments in Colombia's recent history. Cocaine production has grown from less than 100 tons in 1980 to more than 500 million tons in 1999. The 500 percent growth in production and the 700 percent increase in growing areas have made Colombia the world's largest cocaine supplier. This has had serious neg- ative consequences: a sharp rise in crime and violence, corruption, weakening of institutions, health and social impacts, displacement of legal crops, and environ- mental degradation. Recent analysis suggests that the long-term sustainability of guerilla activity since the early 1980s and the rise of the paramilitary are closely asso- ciated with increasing involvement with the drug industry. These relationships are shown in Figure 3. An empirical analysis of the determinants of violence, measured by the homicide rate, found that in the seven main cities the main explanation for the increase in vio- lence in the 1980s was drug trafficking and, to a lesser extent, the collapse of the judicial system. Evidence for 700 municipalities showed that high homicide rates in the 1990s were primarily explained by the presence of armed actors (paramilitary, guerrillas), intensity of drug trafficking, inefficiency of the judicial system, and interaction between armed actors and drug trafficking; socioeconomic variables such Figure 3. Cocaine Production, Homicide Rate, and Guerrilla Front 120 - -800 Guerrilla Fronts v C, I600 uc 80 - _ _ 5 ° 80- Cocaine Production in Tons 60 - ---------- --.-- -----,,-- - -400 o o 40- Homicide Rate 200 E 20- - - 0 -1 I I I I I -0 1958 1964 1970 1976 1982 1988 1994 2000 Source Cardenas (2001) 44 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE as poverty and Inequality also contributed, but to a much smaller degree (Sanchez and Nunez 2000). There are two very different views expressed by Colombian analysts concerning the current relationship between the drug trade and the armed conflict: (a) some argue that the existence of the armed conflict and its increasing intensity (that is, need for funds to finance the war) is the reason for the expansion of illegal crops and drug traf- ficking, implying that finding a settlement to the conflict is key to stopping this process; and (b) others argue that illegal crops and drug trafficking are financing the armed conflict, and have become ends in themselves for the armed actors, and there- fore should be directly attacked in order to end the conflict. Irrespective of which view one adopts, it is evident that there are no easy solutions to either of these problems. As indicated earlier, the drug trade also relates in complex ways with underlying social conflicts. For example, much evidence indicates that money laundering and profits from the drug trade went into extensive land acquisition in rural areas, with negative impacts on employment, equity, and security (Reyes Posada 1998). Another example is the lack of viable economic alternatives to offer small farmers growing coca, who are concerned about the high social costs to them and their families of being involved in this activity. These are just a few examples of the interrelated issues of armed conflict, social conflict, and the drug trade as sources of violence. IV. Costs of Violence and Conflict The impact of violence and conflict on the well-being of the Colombian people is enormous, resulting in a pervasive sense of uncertainty and vulnerability. For pur- poses of this discussion, these will be divided into economic, social, political, and environmental costs, though these are, of course, interrelated. Although much important analysis has been done on this subject, these costs remain difficult to measure in quantitative terms.2 a) Economic Costs of Violence Discussion of the economic costs of violence and conflict must be seen in the con- text of the prevailing interpretation that between 1950 and 1980 the armed conflict and violence did not significantly affect Colombia's economic growth and develop- ment. Throughout that period, Colombia avoided significant macroeconomic imbalances, maintained a respectable record of economic growth, and made signif- icant improvements in key social indicators, though progress in rural areas contin- ued to lag behind urban areas. This situation began to change in the 1990s; Colom- 2 See Moser (2000) for a more in-depth discussion of these costs, which synthesizes the work of Colombian experts and instirutions like the Departamento Nacional de Planeactdn on this subject. Also see DNP (1998), and Ca-macho Guizado and Leal Buitrago (2000). VIOLENCE, SUSTAINABLE PEACE, AND DEVELOPMENT 45 bia's macroeconomic imbalances, particularly fiscal deficits, started to build up, and internal and external debt steadily increased. At the same time, the armed conflict expanded and intensified, and an increasing area of Colombia's territory was under active dispute or controlled by the armed groups. In the second half of the 1990s, as the armed conflict and violence were dramatically increasing, economic growth slowed down, and toward the end of the decade there was a serious recession. ECONOMIC COSTS MEASURED IN TERMS OF GROSS DOMESTIC PRODUCT. While there are various interpretations as to the relationship of these phenomena, it is clear that the economic costs of violence and conflict are high. DNP's comprehensive study on violence, La paz el desafio para el desarrollo, brings together and builds on the growing number of studies on the economic costs of violence in Colombia. It concludes that during 1991-96, the cost of violence, that is, the cost to the victims, referred to as "gross cost") was COP17.2 billon (1995), equivalent to 25.3 percent of GDP, or an annual average of 4.2 percent.3 The cost to the economy as a whole (that is, the macroeconomic cost, referred to as the "net cost," which does not include transfers) was estimated to be COP12.5 billon (1995), equivalent to 18.5 percent of GDP, or an annual average of 3.1 percent (Rubio 1998) Loss of life accounted for 43 percent of the amount, "excess" military expenditures 30 percent, other security expenditures 23 percent, terrorist acts such as blowing up the oil pipeline 3 percent, and health costs 1 percent (DNP 1998). Another way of look- ing at this is the effect on the Human Development Index (HDI)4 a broader meas- ure of development that includes, in addition to per capita income, variables such as life expectancy, and educational level. In 1998 it was estimated that the fall in life expectancy in Colombia results in an economic loss equivalent to between 2.7 per- cent and 3.6 percent of GDP (Sarmiento 1998b). DECLINE IN INVESTMENT. Common sense dictates that the current situation in Colombia is unfavorable to investment: risks are high due to insecurity, including targeting of violent acts such as kidnapping, and destruction of costly assets; pro- duction costs are high due to security costs, extortion, and disruption of trans- portation and electricity supply; and legal recourses are limited. The situation is par- 3 The "gross cost" includes the cost of robbery, extortion, and kidnapping, which are not included in the "net cost," because for the econonmy as a whole these are transfers. Thus, the distribution of costs looks different for the two measures In the case of cost to the vic- tims, loss of life accounted for 31 percent of the amount; "excess" military expenditures 22 percent, other security expenditures 17 percent; crimes against patrimony (delitos pat- rzmoniales) 15 percent, kidnapping, extortion, and robbery 12 percent, and terrorist acts such as blowing up the oil pipeline, 2 percent. 4 HDI is calculated based on life expectancy, individuals' educational level, and income level. This Index is a reply to the general criticism made of measuring the development of a country based only on the GDP because it reflects the income distributlon. 46 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE ticularly dramatic in rural areas. A quantitative analysis shows that when indicators for human capital and homicide rates are factored into the traditional equations for calculating inflation and the capital costs of investment, violence had a negative and significant impact on investment in Colombia. One study found that each percent- age point of increase in the murder rate reduces the rate of private investment by 0.66 percent (Parra 1998). The decline in the rate of domestic and foreign invest- ment, from 19 percent of GDP in 1995 to 5 percent in 2001, is now attributed by many to the increase in violence: In Colombia today, in addition to the constant changes in regulations, the main factor discouraging investment is the armed conflict and its conse- quences in terms of loss of life, forced displacement and indiscriminate attacks on production infrastructure. From this point of view, the armed conflict is undoubtedly the main obstacle to the distribution of the benefits of growth in our society as a whole (Segundo parlamento de paz 2000). DESTRUCTION OF PHYSICAL CAPITAL AND Loss OF PRODUCTWVE ASSETS. Violence has eroded physical capital (also known as man-made or produced capital) by reduc- ing the stock of plant, equipment, infrastructure, and other productive resources owned by individuals, the business sectors, and the government. Costs associated with violent attacks on infrastructure-especially petroleum and electrical installa- tions, roads, and airports-are mounting dramatically in Colombia. The petroleum, coal, and electrical power industry have been particularly hard hit. The costs of pro- tecting and repairing this infrastructure, not to mention the foregone sales and other indirect impacts, are high. The impact of the armed conflict on rural areas has been dramatic. Many landowners do not have physical access to their land or are unable to make full use of its productive potential. Others, mainly poor, peasant farmers have been forcibly evicted by the competing interests of guerrillas, paramilitaries, and drug traffickers, or have voluntarily fled to escape death threats. According to a recent study in Cali, for instance, 53 percent of the rural displaced had land ownership rights that they relinquished upon migrating, and 83 percent of them were landowners (CODHES 1997; Angarita Canas and Osorio Moreno 1998). Since many peasants do not hold legal title to their land, if they abandon it, they lose their legal rights to ownership. Even those who have legal title to land do not necessarily have recourse to judicial processes to regain their land or claim compensation. IMI'ACT ON GROWTH. Colombia's annual GDP growth fell from an average of 5 per- cent between 1950 and 1980 to 3 percent between 1980 and 2000. A recent study found that the decline in the growth rate is fully explained by a fall in productivity. It concludes that "the implosion of productivity is related to the four-fold increase in criminality. The existing literature has already shown that the explosion of crime was the result of the rapid expansion of drug-trafficking activities and the intensifi- VIOLENCE, SUSTAINABLE PFACE, AN) DEVELOPMENT 47 cation of the internal armed conflict (fueled by the rents from the drug trade)." The study concludes by indicating that "the period between 1980 and 2000 can be char- acterized by a vicious circle of high crime, negative productivity gains, low growth, and increasing concentration" of income, reversing the positive pattern of the earlier period (Cardenas 2001). Analysis of the rural sector indicates that the armed con- flict and drug monies are associated with the increased concentration of some of Colombia's highest potential land for crop production and its use for livestock pro- duction; this is not economically efficient and represents a significanit social and environmental problem. International evidence supports the idea that high crime rates and high concentrations of income are linked, and that countries with those characteristics suffer from low productivity and low growth Conversely, GDP growth has a significant impact on the reduction of violence when it is accompanied by a distribution of income, which reduces poverty (Fajnzylber, Lederman, and Loayza 2002). b) Social Costs of Violence Colombia's social fabric has been seriously weakened by violence and conflict, basic human rights are not being met, and the quality of life for all has declined in the cli- mate of intimidation and uncertainty that prevails, particularly in rural areas and poor urban neighborhoods. The cost to Indigenous and Afro-Colombian peoples has been particularly high. Violence has eroded human capital by limiting access to education and health care by both users and providers. For examnple, death threats and assassinations of rural teachers and health workers by armed groups have led to the abandonment of many rural facilities across the country (Parra 1997). It has also created an added burden for the health care sector, with trauma care consuming an important pro- portion of health resources This includes not only physical injuries and disability caused by violence, but also the psychological consequences of victimization or wit- nessing violence among adults The social costs of forced displacement, both in terms of those displaced and the recipient areas, is also high. Those displaced usually flee to the periphery of urban centers where housing and work are difficult to obtain, and often lose access to health and social services. A 1995 study of pre- and post-displacement unemploy- ment rates showed male rates increasing from 6.2 percent (when most worked in rural agriculture) to 34 percent (when they had relocated on the urban periphery) In Medellin, where the situation is particularly acute, the municipal government has struggled to cope: displaced persons have been evicted from the city on the grounds that their presence was a risk to public order and could cause a natural disaster due to their settlements being in geologically unstable areas (Angarita Canas and Osorio Moreno 1998). Violence has also eroded household relations as an asset by reducing the capacity of many households to function effectively as a unit. In rural conflict zones, where 48 COLOMBIA. THE ECONOMIC FOUNDAnON OF PEACE many men have joined illegal armed groups, family life is seriously disrupted by high stress levels. In poor urban communities, many women have identified a direct link between male unemployment, alcohol abuse, and increased domestic violence. In the case of internally displaced populations, research shows that women are more vulnerable than men at the moment of eviction, when they are exposed to unex- pected widowhood, threats, clandestine action, flight, and separation from their homes. Men, in contrast, seem better equipped to cope at such times, but the reverse is true when displaced households restructure their lives. Then the impact is greater on men, who become unemployed and experience a loss of status as breadwinners and a rupture of their sense of masculine identity. Gang involvement is an example of the negative social cost: young people, bereft of strong family and community support, form mutually reinforcing groups. Violence has eroded social capital5 by reducing trust and cooperation within for- mal and informal social organizations and among their members. The importance of social capital relates to its recognized contribution to sustainable development and the evidence that the size and density of social networks and institutions, and the nature of interpersonal interactions, significantly affect the efficiency and sus- tainability of development processes (Putnam 1993). A 2001 cross-country study (Fajnzylber, Lederman, and Loayza 2002) provides empirical evidence showing a strong negative relationship between violent crime and social capital. Evidence for Colombia is consistent with this, showing higher levels of participation in commu- nity action groups in less violent areas and lower levels in more violent areas (Cuel- lar de Martinez 1997) This is not surprising as the capacity for community-level organizations to function depends on their cohesiveness and personal safety, and the ability to meet locally. Sustained conflict, arbitrary killings by armed groups of sus- pected sympathizers of their opponents, and widespread death threats have system- atically reduced trust between neighbors and communities across the country. c) Political Costs of Violence Important progress was made toward opening up the political system in the late 1980s and early 1990s, most notably with the election of local and regional author- ities, and the adoption of the 1991 Constitution, which recognized citizens' basic rights, provided mechanisms for citizen participation and oversight of public affairs, and recognized Colombia as a multiethnic nation. However, this progress has been seriously eroded by the inability of the Colombian State to guarantee its citizens basic rights such as security, and to prevent the domination of many areas of the country by armed groups. The decentralization process and exercise of democracy under such conditions is precarious. 5 Social capital is the rules, norms, obligations, reciprocity, and trust embedded in social relations, social structures, and societies' institutional arrangements, which enable its members to achieve their individual and community objectives. VIOLENCE, SUST-AINABLE PEACE, AND DEVELOPMENT 49 Violence has severely eroded faith in the relevance and effectiveness of many social institutions as a consequence of both human rights violations and a reported 98 per- cent impunity rate. The growing concern that Judicial, educational, health, media, and security institutions are no longer viable is testing the institution of democracy. The effect of drug trafficking on the judicial system is a primary example of the impact of violent crime on institutions. Furthermore, as Colombia's police and judi- cial institutions weaken, the privatization of security is a growing phenomenon in both rural and urban areas: in 1980, Colombia had 2.5 policemen for every private security agent, by 1995, this ratio had dropped to 1 to I (Ospina 1997). Systematic threats and attacks against the communications media have been meant to alter media behavior and information content, thus affecting a key channel of information necessary to an informed, democratic civil society. More generally, continued violence has created a climate of fear, anxiety, and mutual distrust, which tends to suppress the "voice" of civil society institutions to participate effectively and peacefully in political decisions at community and national levels (Quintero and Jimeno 1993) d) Environmental Costs of Violence Colombia is rich in natural resources and is recognized as being one of the most bio- diverse countries in the world, with nearly all the world's ecosystems represented. Environmental degradation has only recently been recognized as a significant cost of violence, particularly the armed conflict and drug-related activities. One of the most dramatic examples is damage due to oil spills resulting from regular attacks on the pipeline. Environmental degradation in and around urban areas has been exacer- bated by the influx of displaced populations, worsening the problems of solid waste disposal and water contamination. In rural areas, land degradation and inappropri- ate land use have been exacerbated by the armed conflict. The presence of armed groups in and around many of the national parks and other protected areas (many of which overlap with Indigenous reserves and Afro-Colombian community prop- erty) has been a major obstacle to effective conservation and protection of these pre- cious ecosystems. Finally, drug-related activities result in contamination of soil and groundwater due to aerial spraying of illicit crops and chemicals used to process coca, and increased deforestation rates and destruction of fragile ecosystems for coca and heroin poppy cultivation. V. Toward Building Peace and Sustainable and Equitable Development It is apparent that Colombia is enduring a crisis of terrible proportions, with enor- mous economic, social, political, and environmental costs. Building conditions for peace and tackling the sources of violence and conflict are goals that will address the basic development problems of the country. International experience in the post- 50 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE cold-war era with peace processes that mark the transition from violent conflict to development, make it clear that building peace requires a process of reinforcing mutual confidence and financial support to lay the foundation for a more open and inclusive State (Colleta and others 2000). There is also growing recognition that development must continue even when all or part of a country is in conflict. Posi- tive incentives for peace, security, and development must prevail over perverse incentives for violence, fear, and destruction (World Bank 1999b). The goals of achieving peace and securing equitable economic and social development comple- ment and reinforce each other. In Colombia, an agenda to build peace and tackle the sources of violence in order to create conditions for equitable development must address simultaneously (a) the armed conflict, (b) the drug trade, and (c) the underlying factors of social conflict. The first two aspects need military and political measures and a hemispheric approach to deal with them; the third one demands an integrated social and eco- nomic policy based on broad consensus. The key issues to address are (a) land, (b) access to justice, (c) security of life and property, and (d) fighting corruption. The best way to deal with these issues is through a region-based approach, for example, peace and development programs in critical regions; and a population-based approach, for example, priority on youth. Finally, the formulation and implementa- tion of policy and programs must follow a partnership approach between govern- ment, private business, and civil society organizations. a) Armed Conflict and Drug Trade The first two elements of the agenda will encounter formidable opponents in the armed groups (both guerrilla and paramilitary) and in the drug cartels, which fuel the conflict by financing the armed groups in exchange for the protection of their business. The reasons for this opposition are based on the following facts. First, these extralegal armed groups pose a direct challenge to the sovereignty of the State, both in a territorial sense and in an administrative and legal one. They, in essence, sub- stitute for the State and provide basic services, including administering justice, edu- cation, and social services. Lately, they have even forced the exile of legitimate local authorities. Second, they have an efficient economic base based on drug money and other illegal and inhumane methods such as extortion, kidnapping, and forced con- tribution. Third, they have declared a de facto war on the civilian population and the State through acts of terrorism and the violation of fundamental human rights. The illegal drug industry and armed groups have created a lawless environment that threatens not only Colombia, but the neighboring countries. To the extent that the drug trade, the violation of human rights, and the acts of terrorism are matters of international concern and consequence, dealing with these elements of the agenda demands a hemispheric approach. The aim of a plan of action is to regain control and sovereignty of the State over the entire territory of the nation, and increase the security of citizens. VIOLENCE, SUSTAINABLE PEACE, AND DEVELOPMENT 5 1 b) Underlying Social Conflicts Even If the armed conflict and the drug trade are handled through a variety of mil- itary and political measures, including a negotiation process with the armed groups, several structural elements associated with violence and the violent conflict will remain unresolved, unless there is a plan with the objective of attacking the struc- tural problems of violence and social conflict. Addressing this element of the agenda should not be dominated by the negotia- tion themes proposed by the various armed groups, even if there is a new dialogue for peace. This aspect of the agenda needs to be constructed, independently of their willingness to start a peace process, through an integrated economic and social pol- icy which is formulated on the basis of a broad social consensus on content and pri- ority, and carried out in a partnership between civil society, the private sector busi- ness community, and government at various levels. Evidence from other countries indicates that while the specific context of violent conflict varies, the exclusion of some groups and the alienation of citizens have always been major factors. Most important, dealing with such structural factors is necessary, regardless of the exis- tence of an armed conflict, because of their implication for the reduction of poverty and inequality in Colombia Such integrated economic and social policy must deal with at least the following unresolved issues: access to land and productive assets, reduction of impunity, pro- tection of life and property, and fighting corruption. i. Land and Productive Assets Access to land has been a major source of political and social conflict in addition to being a key determinant factor of the productivity of the rural economy. Displace- ment of population from rural areas has been associated with the conflict over land, and the demand for access to land and other productive assets a permanent cause of peasant movements. Measures toward the removal of land-related conflict factors may include: * Facilitating the integration of land purchase with productive investments and financial services within a broader context of local development. This will remove impediments for the integration of smallholders into a more equitable development process and thus a source of wealth, employment, and growth. The emphasis of this measure is on gaining access by smallholders to existing mechanisms for public support to the agricultural sector (Incentivo a la Cap- ztalrzaczSn Rural [ICR], credit, Fondo de Garantias) to improve livelihoods and avoid their capture by the largest producers. * The availability of working capital is an option that might have considerable positive impact on improved livelihood as well as preventing the involvement of small producers in the production of coca or finding alternative crops for those already involved. In effect, a major element that sets the production 52 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE process apart from other crops grown by small farmers is the easy availability of working capital to produce coca. It is common that downstream agents in the coca "value-chain" prefinance working capital needs of small coca pro- ducers for the entire season. This contrasts sharply with the widespread lack of access to credit by small farmers in Colombia for the cultivation of other crops. Increased efforts to ensure access to rural credit by small farmers could well be one of the elements of a more active policy to simultaneously provide economic opportunities to the rural poor and combat drug production. ii. Reduction of Impunity Security and justice are essential elements for renewed trust in the capacity of the State to protect citizens. However, the Colombian State has proved unable to effec- tively serve as the arbiter of conflict among citizens, either though formal or infor- mal institutions or to exercise control over criminal behavior. The recent survey on Corrnpczdn, Desempefio Insitutucional y Gobernabilidad (March 2002) shows that most of those interviewed think that justice in Colombia is unfair, favors those with power and influence, and that the judicial system is not independent from the exec- utive or from politicians. Consequently, gradual reduction of impunity and establishing the rule of law on an equal footing for all citizens in all regions of the country must be a priority of the new administration in order to increase the credibility of the State and to increase the security of citizens. The State must regain the monopoly of justice and its coer- cive capacity on individuals. In particular, the judicial system must increase its capacity to apprehend those involved in serious criminal activity. In the past, the reform efforts were geared toward governance issues within the judiciary, the change of codes, or the increase in penalties. The challenge now is to break the cycle of impunity, which reinforces the idea that crime pays because the probability of being apprehended is small. This cycle, some argue, may also generate a demonstration effect particularly among youth who will give social value to the attitudes and possessions of delinquents. Toward the reduction of impunity, the criminal system must become highly selective and strate- gic to be socially relevant in the middle of an acute structural imbalance between criminal activity and investigative resources. This would imply that the criminal sys- tem should aim at dealing effectively with those criminal problems that are most dis- turbing to Colombian society, such as a drastic reduction in the number of homi- cides by organized crime. iii. Security of Life and Property Colombia's social fabric has been seriously weakened by violence and conflict, basic human rights are not being respected, and the quality of life for all has declined in the climate of intimidation and uncertainty that prevails particularly in rural areas and poor urban neighborhoods. The cost to indigenous and Afro-Colombian peo- ples has been particularly high. VIOLENCE, SUSTAINABLE PEACE, AND DEVELOPMENT 53 In this context, the protection of the lives and property of citizens becomes a pri- ority for the government, State agencies, and all other social actors. This is bolstered by international experience that indicates that the degree of security enjoyed by cit- izens affects the amount of time peace will last. Measures toward this end may include (a) increasing the presence of the State, particularly in the territories with low or no presence, (b) promoting a culture of respect for human rights, and (c) securing the rights of land and assets. * The increased presence of the State should have three dimensions (a) the pro- vision of basic services and the special protection of vulnerable groups, (b) the presence of institutions to intermediate or serve as arbiter to resolve conflicts among citizens; and (c) the presence of an agent that can exercise the monop- oly of force as representative of the State. * Bolster the voices of ordinary citizens that have decided to stand up before the armed actors to claim their fundamental human rights. In particular, reinforce with economic and security resources the organized Indigenous communities and small municipalities of the emerging movement of civilian resistance to counterbalance the domination exercised through terror by armed actors, both guerrilla and paramilitary. Recognizing the weakness of the State, pro- mote among private business, civil society organizations, and the international community their contribution to this local effort from the strength of their own specialty. This action is very likely to be successful because there is a growing number of citizens from all walks of life tired of being actual or potential victims of the same violations of their human rights and willing to make an effective contribution. The burgeoning culture of respect for human rights and citizenship and a growing conviction of the urgent need for political solutions-reached through a negotiated settlement, rather than military solutions-means that today a social movement around peace is developing. * Securing rights to land and assets in rural and urban settings, as a factor of production, as a source of livelihood, and as asset endowment for the market. A massive land titling and securing of assets program would be of particular importance for rural populations at high risk of displacement (as a preventive measure) and for urban dwellers that could secure their property and acquire a market value. iv. Fighting Corruption The capture of the State-understood as the capacity of interest groups to influ- ence through corrupt practices the high levels of decisionmaking of the State- appears to be one of the most predominant forms of corruption in Colombia. About 70 percent of respondents among business people and civil servants indi- cate that this is a significant pattern in the high levels of the business environment in Colombia. 54 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE This high level of agreement around the incidence of the capture of the State poses a significant challenge regarding the framework of governance in Colombia. Based on the analysis of the results of this large survey, the following measures to reduce this level of corruption are proposed: * Increase political competitiveness by reforming the financing rules of political campaigns * Reduce clientelistic practices inside public administration, particularly between the executive branch and Congress * Introduce meritocracy in public institutions to discourage corrupt practices in government and state agencies * Establish social accountability mechanisms to oversee budget and investments of public resources at all levels of public administration, and municipal, departmental, and national agencies. c) Critical Elements of Strategic Approach i. Community-Based Regional Development Given that roots of unresolved social conflicts are within regions and local contexts, the main issues of access to land and productive assets, reduction of impunity, secu- rity of life and property, and fighting corruption must be addressed at those levels. This emphasis, however, should be complemented with outside political and eco- nomic support by parties not embroiled in the conflict themselves. Therefore, the intervention of civil society groups, the private sector, commu- nities, and producer organizations should be emphasized so that they can better participate in local planning, definition of local priorities, and implementation. The local and regional development approach will integrate rural-urban popula- tions in terms of on-farm and off-farm opportunities. This approach promotes better integration of communities and rural producers, integration with the sup- ply chain, development of labor and financial markets, provision of basic infra- structure and services, and sustainable management of natural resources. The regional space becomes the main scenario where social cohesion can be rebuilt and participation in the decisionmaking process becomes the basis for consensus building and ownership of the local development process. The Program for Development and Peace in the Magdalena Medio region shows the enormous potential of this approach. Community-driven development will increase the stake of the population in peace and their willingness to resist the pressure from armed actors. The economic and social opportunities will facilitate the productive occupation of the territory. This occupation supported by the State and government agencies will also con- tribute to regaining the control and sovereignty of a more legitimate State that sup- ports and receives the support of its citizens. VIOLENCE, SUSTAINABLE PEACE, AND DEVELOPMENT 55 ii. Targeting Regions and Youth Poverty, inequality, and conflict vary gready throughout Colombia. This is not surpris- ing given Colombia's incredible physical, cultural, social, and economic diversity, history of strong regional identities, and important rural-urban differences. These differences need to be kept in mind in the formulation of policies and programs to deal with the underlying factors of violence and conflict. Being poor or discriminated against in the Pacific region or in the southern departments of Colombia is quite different from being poor or discriminated against in Bogota or the Central region. The Pacific region, for example, has social indicators similar to those that Colombia as a whole had 20 years ago. To make matters worse, much of the Pacific region, along with much of Colombia outside of Bogota, are territories under active dispute among contending armed parties for the control of resources and people. As a result, the poor-the majority of people in such territories-are subject to the devastating, combined effects of unemployment, lack of basic services, poor infrastructure, and the threat of forced displacement. Forced dis- placement, mainly as a result of the social and armed conflict, is the most serious humnan- itarian and social problem in Colombia today. It causes uprooting of people and disrup- tion of family life; loss of human, social, and physical capital; generalized impoverishment, and high costs for the families and for the State. Therefore, a special effort must be made to identify the crintcal regions most affected by poverty, mequality, and social and armed conflict, and target them for government programns of the kind described above. In these regions, access to safety nets, education and health services, infrastructure, and new economic opportunities are essential, as are measures to reduce or mitigate the armed conflict, the main cause of displacement. Youth, on the other hand, is the population group most affected by violence, unem- ployment, and lack of educational opportunities. They are also subject to voluntary or forced recruitment by armed groups or drug lords and to the contamination effect of ille- gal and criminal activities that go unpunished. Young men and women should receive priority attention in the design of local or regional peace and development programs, and special opportunities in urban programs in intermediate and large cities. Particularly important are opportunities for postprimary education, job training, and employment. In countnes affected by conflict, job creation is a central instrument to pro- vide alternatives to youth and adults, and thus to prevent armed fighting This is an area where the involvement of the private sector and specialized nongovernmental organiza- tions is promising. In effect, the use of special trust funds to finance job training, offer microcredit, and increase livelihood chances have proved effective. The creation of new jobs is good not only for economic development but also for security. VI. Conclusion Colombia is a country with an enormous base of human and natural resources, a good geographic location for international trade and foreign investment, and a 56 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE valuable democratic and cultural tradition. The Colombian people have demon- strated enormous courage and creativity in coping with and trying to address the long-standing issues that are a threat to their well-being and that of their nation. There are an enormous number of positive experiences and achievements through- out Colombia to build on, from the National Rehabilitation Program, to the Pro- grama Desarrollo, Saludy Paz (DESEPAZ Program) in Cali, to the pioneering Pro- gram for Development and Peace in the Magdalena Medio region. While the obstacles are enormous and sacrifices will need to be made by all, there is great hope that Colombia's many assets and strengths, particularly the courage and determi- nation of its people, will be sufficient to build a sustainable peace and achieve equi- table development. References Angarita Canas, P. E., and C. A. Osorio Moreno. 1998. "Analysis of Displaced Pop- ulations in Colombia." Washington, D.C.: World Bank. Camacho Guizado, A., and E Leal Buitrago. (Compiladores). 2000. Armar la Paz es Desarmar la Guerra. Bogota, Colombia: Ciro Editores. Cardenas, M. 2001. Economic Growth in Colombia- A Reversal of Fortune? Center for International Development, Harvard University, Working Paper No. 83, December. CODHES (Consultoria para los Derechos Humanos y el Desplazamiento). 1997. Desplazamiento rural: violenctaypobres. Bogota: CODHES. Colleta, N., M. Kostner, P. Cleves, and J. Mendelson. 2000. "Toward an Architec- ture for Sustainable Peace and Development: Lessons from the World Bank's Experience." In A. Solimano, ed., COLOMBIA Essays on Conflict, Peace and Development. Washington, D.C.: World Bank, pp. 159-93. Conferencia Episcopal Colombiana. 1994. "Secci6n de movilidad humana." Mimeo. Conferencia Episcopal Colombiano, Bogota. Cuellar de Martinez, M. 1997. "Valores y capital social en Colombia " Coporaci6n Porvenir and Universidad Externado de Colombia, Santa Fe de Bogoti. Mimeo. DNP (Departamento Nacional de Planeaci6n). 1998. La paz: El desaflo para el desarrollo Bogota: Tercer Mundo Editores. Echeverry, J. C., N. Salazar, y V. Navas. 2001. "Si nos parecemos al resto del mundo? El Conflicro colombiano en el marco internacional." Seminario de Economia de la Catedra Colombiana, Bogoti. Fajnzylber, P., D. Lederman, and N. Loayza. 2002. "Inequality and Violent Crime." Journal of Law and Economics, Vol. XLV, April: 1-40. L6pez, C., and A. Garcia. 2000. "The Hidden Costs of Peace in Colombia." In A. Solimano, ed., COLOMBIA: Essays on Conflict, Peace and Development. Wash- ington, D.C.: World Bank, pp. 78-158. Meertens, D., and N. Segura-Escobar. 1996. "Gender, Violence and Displacement in Colombia." Singapore Journal of Tropical Geography, 17 (2):165-78. VIOLENCE, SUSTAINABLE PEACE, AND DEVELOPMIENT 57 Me)ia, D. 2000. "El secuestro en Colombia: una aproximaci6n econ6mica, en un marco de teorfa de juegos." Coyuntura Econ6mzca, Vol. XXX, No. 2, Junio. Moser, C. 2000. "Violence in Colombia: Building Sustainable Peace and Social Capital." In A. Solimano, ed., COLOMBIA Essays on Conflict, Peace and Devel- opment. Washington, D.C.. World Bank, pp. 9-77. Moser, C., and C. Mcllwaine. 2000. Urban Poor Perceptions of Violence and Exclu- sion in Colombia. Washington, D.C.: World Bank and the Swedish International Development Agency. Ospina, J. 1997. "Hacia una nueva estrategia de manejo del orden publico." In Par- tido Conservador Colombiano, ed., Inseguridad e impunidad en Colombia. Bogota Partido Conservador Colombiano. Parra, C. 1998. "Determinante de la Inversi6n en Colombia: evidencia sobre el cap- ital humano y la violencia." DNP, Archivos de Macroeconomia, Doc. 84, June. Pecaut, D. 2001. Guerra Contra la Sociedad. Bogota: Editorial Planeta Colomblana. Putnam, R. D 1993. Making Democracy Work. Civic Traditions in Modern Italy. Princeton, N.J.: Princeton University Press. Quintero, M., and R. Jimeno. 1993 "Los medios de comunicaci6n y la violencia." CINEP-APEP, Bogata. Mimeo Reyes Posada, A 1997 "Compra de Tierras por Narcotraficantes." In Franciso E. Thoumi, ed., Drogas ilicitas en Colombia: Su impacto econcimico, politico y social Bogota Ariel Ciencia Polinca. Reyes Posada, A. 1998. "La Cuesti6n Agraria en la Guerra y la Paz" In Departa- mento Nacional de Planeaci6n, La paz- El desaffo para el desarrollo. Bogota: Ter- cer Mundo Editores, pp 205-26. Rubio, M. 1998 "Los Costos de la violencia en Colombia: El estado del debate." Washington, D.C. World Bank. Mimeo. Sanchez, F., and J. Niunez. 2000. "Determinantes del crimen violento en un pais altamente violento : el caso de Colombia." Mimeo. Universidad de los Andes, Bogota. Sarmienro, A. 1998a. "Violencia y Equldad." In Departamento Nacional de Planeaci6n, La paz. El desaflo para el desarrollo. Bogoti: Tercer Mundo Editores, pp. 227-61. . 1998b. Informe de Desarrollo Humano para Colombia, 1998. DNP-UNDP, Tercer Mundo Editores, pp. 122-25. Segundo Parlamento de Paz. 2000. "Desarrollo Econ6mico y Social." Camino de Conciliaci6n Nacional. Instituto de Derechos Humanos y Relaciones Interna- cionales. Universidad Javeriana. Bogota, March Solimano, A., con F. Saez, C. Moser, and C. L6pez, Editores. 1999. Ensayos sobre la Paz y Desarrollo. El caso de Colombia y la experiencia internacional World Bank, Bogota: Tercer Mundo Editores. Solimano, A., ed 2000. COLOMBIA Essays on Conflict, Peace and Development. Washington, D.C.: World Bank. English version of Ensayos sobre la Paz y Des- arrollo, 2000. 58 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE World Bank and the Carter Center. 1997. "From Civil War to Civil Society: The Transition from War to Peace in Guatemala and Liberia." Workshop Report, July. Washington, D.C. and Atlanta, GA. . 1999a. Violence in Colombia: Building Sustainable Peace and Social CapitaL Washington, D.C . 1999b. Security, Poverty Reduction & Sustainable Development: Challengesfor the New Millennium. Washington, D.C., September, p. 9. . 2001. World Development Report 2000/2001 Attacking Poverty. New York: Oxford University Press. 2 Enabling Sustainable Growth This Chapter was written by Krishna Challa. I. Introduction Despite the major problems Colombia is facing in the areas of security and violence, it continues to have considerable growth potential. The most important reason for this is the country's impressive human capital, which has kept the country on a growth path for all but one year over the past five decades. Through a tradition of prudent macroeconomic management, Colombia has so far managed to avoid the major cycles of macroeconomic instability that have plagued a majority of countries in the Latin America region in recent decades. There is a good probability that Colombia will foster a sustainable growth path in the coming years, provided that the security situation does not worsen significantly, prudent macroeconomic man- agement continues, and that certain enabling conditions are maintained for a good business environment and responsible management of natural resources. Overall, the economic record of Colombia in the last three or four decades is one of quite impressive progress, and can be characterized as very good performance under duress and against significant odds However, there remains a substantial unfinished agenda. Requirements for a sustainable growth path in Colombia, apart from a security situation that does not worsen significantly and development of well-qualified human capital via improved human development policies (areas addressed in the other accompanying notes), include: * Prudent macroeconomic management, including sustainable fiscal policies and management. * A well-functioning finanicial system that can provide adequate services to all seg- ments of the productive sector and the population at large, including a healthy and sustainable banking, housing finance, pension, and insurance systems. 59 60 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE * A business environment that would release the potential of the private sector, including adequate regulatory and competition frameworks, and administra- tive procedures governing the establishment and operation of firms that are not excessively cumbersome. o Efficient, effective, and well-managed infrastructure sectors that facilitate competitiveness of the Colombian productive sector vis-a-vis international markets (for example, through reductions in transport, communications, and logistics costs of firms and workers) and adequate access to good-quality basic infrastructure services for all parts of the society. o Responsible management of natural resources and the physical environment (land, water and wastewater, forest and agricultural resources, urban air qual- ity, waste and hazardous waste management) to ensure their sustainability and positive contribution towards the long term economic growth objectives. II. Providing a Sustainable Macroeconomic and Fiscal Framework a) Background Colombia's recent path could lead to an unsustainable increase in deficits and debt, along with poor prospects for GDP growth and high unemployment. The rapid expansion of the public sector over the previous decade, from spending one-fourth of GDP in 1990 to over one third of GDP by the end of the decade, combined with growing liabilities, particularly on the pensions front, have led to significant and per- sistent structural deficits. The private sector, on the other hand, languishes: while government consumption grew by two-thirds between 1994 and 2001, that of house- hold remained basically unchanged, and private investment fell by over 37 percent. The weakness of the private sector is at the center of Colombia's current economic woes. Private GDP has grown little since the mid-1990s, and private investment has deteriorated from its peak of near 12 percent of GDP in 1994 to under eight percent today. Total investment now stands at about 15 percent of GDP, incompatible with the resumption of significant growth. Public sector consumption has in effect dis- placed the private sector as the major contributor to domestic demand. As public spending rapidly outstripped the sector's revenue generating capacity, much of the fiscal deficit, particularly since the mid-1990s, has been financed with debt, both domestic and external. Today, net public sector debt in Colombia stands at about 46 percent of GDP. In 1997, prior to the onset of the recession and to the rapidly deteriorating fiscal balances of 1998 and 1999, public debt stood at 27 per- cent of GDP. Spending on interest payments to service public debt now stands at near five percent of GDP, constituting approximately a quarter of the central gov- ernment's expenditures and more than twice its investment budget. Colombia now faces the combined challenges of reducing its debt levels to ensure sustainabihty and achieving growth rates that are sufficient to reduce poverty levels ENABLING SUSTAINABLE GROWFH 61 and increase employment. Continued fiscal adjustment is unavoidable if debt sus- tainability is to be attained; just as important, the quality of the adjustment is key for a return to growth rates in the four to five percent range. b) Fiscal and Debt Sustainability To avert an explosion of the debt-to-GDP ratio, the combined public sector has to generate a permanent primary surplus of about four percent of GDP.' This calcula- tion incorporates the present value of all net public sector liabilities, including pub- lic pension payments (of the implicit liabilities) under a no-reform scenario, the sce- nario also assumes long-term GDP growth rates of 3.5 percent. In 2001, the public sector primary balance stood at 1.4 percent of GDP. In other words, to achieve fis- cal sustainability, the Colombian public sector will have to generate 2.6 percentage points of GDP in additional savings on a permanent basis. If the pension system were fully reformed along the lines recommended in this report, the required pri- mary surplus would be reduced to 1.5 percent by 2010, a much more feasible tar- get. This estimate does not take into account the incoming administration's program from significantly increasing military and social expenditures over the coming years. Even under the scenario of maintaining a primary surplus of four percent of GDP for the coming decade, it is estimated that the total net debt of the public sec- tor would nevertheless continue increasing, reaching about 57 percent of GDP by 2010. To further illustrate the vulnerability of Colombia's fiscal path in the absence of comprehensive reforms, one can project the path of public sector debt under the assumption that the primary balance is kept constant at its 2001 ratio with respect to GDP (1.4 percent). Under this scenario, debt would rise from about 46 percent of GDP in 2001 to about 80 percent in 2010 (assuming an annual real GDP growth rate of 3.5 percent). On the other hand, to maintain the debt-to-GDP ratio at today's level of 46 percent of GDP, primary surpluses between five and six percent of GDP would have to be maintained over the next decade. The magnitude of the required adjustment means that significant measures on both the expenditure and revenue fronts are needed. Before addressing the options for fiscal adjustment, however, a brief review debt management is presented below. c) Debt Management Colombia's public debt management is hampered by the absence of a comprehen- sive and coherent conceptual and legal framework for debt management, and the existence of overlapping responsibilities among various institutions. Colombia also needs to strengthen its debt management capacity to increase the efficiency and effectiveness of managing a vulnerable and complex portfolio. To date, debt policy has been largely driven by the need to finance the fiscal deficit. Consequently, debt 1. See the Public Debt Susrainabiliiy and Management Chapter for details. 62 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE decisions have been taken with limited systematic analysis of risks, and without a strategic framework for debt management. The rapid growth of the debt stock over recent years has also led to a deterioration in the structure of the portfolio, whose management has been constrained due to shallow and incomplete domestic markets, limited access to international markets, and weak debt management capacity More- over, Colombia's debt portfolio is vulnerable to unexpected changes in both real interest and exchange rates. Strengthening the framework for debt management would involve attention to a number of areas. First is the creation of a legal and governance framework for debt management operations, consolidating all legal and regulatory instruments of debt management. Specifically, the new framework should provide for a separation of policy formulation, implementation, and evaluation functions, the clear definition of responsibilities and accountabilities of the various government entities, as well as clear objectives including risk tradeoff issues as a central dimension of debt man- agement. Second, specific steps can be taken to improve debt management strategy and portfolio structure. This should include improved monitoring of the govern- ment's contingent liabilities, the systematic use of active operations (such as swaps, exchanges, buybacks, etc.) to modify existing debt stocks, as well as the integration of long-run fiscal and debt sustainability analysis into portfolio and funding strate- gies. Third, capacity may be strengthened by centralizing operations in a single debt office with the necessary resources and skills, and sufficient administrative flexibility to respond to changes and opportunities in financial markets. Finally, public debt management would be facilitated by deeper domestic debt markets. In this regard, important steps would include the simplification of the benchmarks for the mini- mum rate of return of pension funds, the identification and removal of main con- straints to increasing the efficiency of collective investment vehicles and mutual funds, and an assessment of the suitability of the Interest Rate Index (DTF) as a market reference rate for secondary market activities. d) FiscalAdjustment Options The magnitude of the primary surplus required to avoid unsustainable debt expansion means that continued fiscal adjustment is unavoidable, and that it will need to include measure to both reduce spending and increase revenues. Nevertheless, international evidence has shown that adjustments based on expenditure reduction are more favor- able to growth than those based largely on increased tax revenues. As such, expendi- ture reduction and substitution will need to play a central role in government policy. i. Rationalization of Expenditures Expenditure adjustment has often been inefficient and inequitable, as it has relied on short-term measures with relatively low political costs, such as de facto invest- ment cuts, payroll freezes, and wage increases below inflation. This approach is unsustainable in the longer term, and creates an illusory fiscal adjustment as assets ENABLING S(JSTAINABLE GRO\TrH 63 of the state deteriorate, undermining future growth. Implementing significant spending cuts is a difficult challenge in Colombia, given large, inflexible expendi- tures such as debt service payments, territorial transfers, pensions and public sector salaries Together these categories constituted nearly 75 percent of the central gov- ernment's expenditures, and 112 percent of its tax revenues in 2001. This points to the importance of carrying out the necessary reform in order to inject greater flexi- bility into public sector spending. A ntimber of policy options have been under consideration in Colombia to address these challenges. Primary among these is addressing pension transfers through a cut in the transition period between regimes. It is difficult to expect rationalizing public sector employment and salaries without a review of expendi- tures, and their efficiency and effectiveness, in the justice and defense sectors, which currently account for fully 70 percent of the central government's personnel and goods and services outlays. Territorial transfers constitute the other main area of expenditure inflexibility for the central government, while at the same time formlig the bulk of the country's spending on social sectors. The recently approved constitutional amendment on territorial transfers guarantees that unless growth performance improves substantially, territorial transfers will continue to pose a bur- den on expenditures. In effect the amendment created a contingent liability in the event that real growth of the economy falls short of 2-2.5 percent by guaranteeing a lower threshold below which transfers will not decline even if growth, and rev- enties, fall short. Given recent evidence regarding the ineffectiveness of continued increases in expenditures on health and education as a means of delivering improved coverage and quality, the question arises of the appropriateness of real increases on the order of 2-2.5 percent in the level of territorial transfers during a period of fiscal austerity and low growth. A temporary transfer freeze, in real terms, would generate substantial fis- cal savings. If local governments are simultaneously empowered to manage central transfers and to raise own revenues, this should not result in any weakening of Colombia's drive towards decentralization. Currently, local governments have little incentive to responsibly manage resources whose assignation and management in effect remain under the control of the central government. The enhancement of serv- ice delivery, one of the main justifications for decentralization, is not merely a func- tion of more resources, but of their appropriate assignment and distribution. ii. Raising Revenues New ways to increase revenues should focus on comprehensive reforms that make the tax structure less complex and more efficient. However, the margin for increas- ing revenues significantly in the short term, without affecting growth prospects, is likely to be small. Moreover, passage of a tax bill that relies mostly on rate increases but does not correct the deficiencies of the current regime will only necessitate, as it has in the past, further rate increase within a year or two as "tax fatigue" sets in fol- lowing a temporary rise in revenues. 64 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Changes introduced over the last decade in the tax structure have boosted collec- tion, but they also introduced distortions into the system. Five major tax reforms were approved in the 1990s. The later reforms in particular were largely aimed at collecting more revenues in order to close widening fiscal gaps, rather than improving the archi- tecture of the system. The incremental approach has resulted in an excessively complex tax structure characterized by comparatively high rates and narrow base. Tax reform should be based on the principle of enhancing the quality of adjust- ment and incentives for growth. For example, caution should be exercised in con- sidering increased levies on the use of capital, since they will raise the cost of under- taking private investment projects and thereby reduce investment. Emphasis in tax reform must be placed on expanding the tax base and removing exemptions and exceptional regimes (which tend to promote evasion and inequalities), rather than increasing tax rates per se beyond a threshold. Specifically, the two most important taxes in Colombia, the VAT and the income tax, both suffer from significant distortions: their bases have been successively eroded by the reforms of the 1990s, and evasion rates are high, estimated at 20 per- cent for the income tax and 28 percent for the VAT. Regarding the income tax, major issues are the distortions of the corporate income tax and exemptions granted for personal income tax. For the corporate income tax, the categories of exempted firms and exempted sources of income largely explain the difference between actual and potential collection. It is therefore recommended that exemptions, such as those for firms operating in the Paez river area, for publishing firms, lotteries, coopera- tives, and itcoreras, be eliminated. The tax base should be widened to include cur- rently exempted public services enterprises, livestock funds, and funds established by financial institutions. Moreover, the corporate income tax creates a number of dis- torting incentives. In this regard, the adjustment for inflation, particularly with respect to inventories, should be reintroduced, as should the presumptive tax on gross assets. A presumptive tax on net wealth creates a bias in favor of contracting large debts, a problem which should be addressed when reintroducing the tax. For the personal income tax, the large number of potential taxpayers who fall outside the scope of the tax is an important constraint to expanding collection: more than 90 percent of wage earners are exempt from paying income taxes. Individuals with an income below three times the average per capita income are currently exempt; other wage earners enjoy an exemption of 30 percent, paying income tax on only 70 percent of their income. Suggested changes include the reduction of the general exemption for individuals to twice the per capita income, elimination of the 30 percent exemption for wage earners, and the introduction of new rates of 15, 25, and 35 percent (versus current rates of 20, 29, and 35 percent.) Similar problems have caused a low yield from VAT, with the ratio of collection to GDP below a third of the VAT rate. Exemptions are high, numerous rates make the tax complex, and further intractability results from the implicit VAT (aimed at protecting domestic activities) and complex procedure for VAT on capital goods. A reform with the objective of widening the base and eliminating preferential rates in ENABLING SUSTAINABLE GROWrH 65 the VAT should include a unification of rates, increasing the preferential 10 percent rate to the general rate of 16 percent; a widening of the base to include paper prod- ucts and printing, beer and tobacco, personal services, transportation and construc- tion; an elimination of the implicit VAT; and an introduction of a VAT on capital goods as a credit against the VAT. Multiple exemptions have also created difficulties for VAT administration, and the VAT on capital goods has added further complexities. In recent years, the Colombian tax administration (DIAN) has undertaken a number of programs to improve compliance and reduce evasion. DIAN's strategy incorporates measures to improve performance in the areas of collection, control, cashing, and efficiency. Clear indicators for tax administration, such as the recovery of overdue taxes, dead- lines for administrative actions and efficiency ratios, should be established and mon- itored. There is also scope to enhance controls on tax returns and for improvements in cross-checking of VAT-related data. III. Ensuring a Healthy Financial System Colombia managed to prevent the severe economic downturn in 1999-2000 from triggering a systemic financial sector crisis. The unprecedented economic contrac- tion of this period came on top of a combination of many other major challenges affecting the health of the financial system. Among the most prominent ones are: (a) the steadily deteriorating security conditions in the country; (b) the adverse effects of the large drug-related financial transactions on the financial system, such as money laundering and large financial transactions taking place outside the for- mal financial system in the shadow economy, which nevertheless had important effects on the real economy; (c) the resulting major political uncertainties and weak political leadership; (d) a prior history of sustained high real interest rates, which together with the recession had substantially weakened the finances of the corporate sector and consequently the portfolios of banking institutions; (e) the important setbacks suffered by housing finance institutions as a result of unex- pected Constitutional Court rulings declaring null and void certain features of the Constant Purchasing Power Unit (UPAC) indexing system for housing finance and requiring the establishment of a new indexing system (which had both a direct and indirect negative effect on the quality of loan portfolios of most hous- ing institutions, with a perverse effect on the willingness to pay of even borrowers with good payment records); (f) the social problems associated with the failure of the nonbank financial institutions in the cooperatives sector and homeowners who no longer had the economic capacity to service their loans; (g) the need to restructure, close and/or privatize several major public or "officialized"2 banks 2 "Officialized" banks are those that were originaly private institutions, but were intervened and taken over by the government 66 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE facing financial problems; and (h) the negative effects of all these developments on the overall fiscal situation of the economy. During various stages of the crisis, the percentage of non-performing loans (NPL) in the overall banking system reached 12-15 percent, with certain acutely affected segments such as the public and officialized banks, housing finance institu- tions and cooperatives showing much worse ratios of NPL. The situation was also aggravated by a non-payment culture in some parts of the system (stemming from the uncertainties created by the Constitutional Court rulings regarding the legality of some of the prevailing rules), expectations of debt relief, and absence of clear supervisory and enforcement rules in some areas. By putting in place a balanced set of carefully chosen corrective actions, the gov- ernment managed to avert the onset of a financial sector crisis of major proportions. Over the last three years, the government has had to delicately balance the need to provide relief to some of the most severely affected borrowers in the system (for example, low-income homeowners, depositors into the failed cooperatives system), the legal requirement to abide by the Constitutional Court rulings, and the need to put in place a resolution process for failed financial institutions. It was important that the financial sector resolution process minimize the perverse incentives and neg- ative consequences that could stem from debt or debt service relief to borrowers and/or financial institutions facing problems. To respond to the crisis, the government secured passage of major financial sec- tor reform legislation (in 1999) and a series of associated regulatory and adminis- trative measures Among the most notable measures were to: (a) increase the mini- mum capital rules for financial institutions, strengthen loan classification, provisioning policy, and procedures for prompt correction actions by the Superin- tendency of Banks when problems arise, and strengthening the powers and financial capacity of the Fondo de Garantza de las Instituciones Financieras (Financial Institu- tions Guarantee Fund, FOGAFIN, the deposit insurance and bank resolution agency) to expedite mergers and other financial restructuring solutions as needed; (b) close, restructure, and/or sell to the private sector several publicly owned finan- cial institutions; and (c) overhauling the housing finance sector in terms satisfactory to the Constitutional Court, involving the establishment of a new housing finance indexing system (Unidad de Valor Real, Constant Value Unit, UVR) that corrects past problems while avoiding excessive interest capitalization in housing loans, and including a carefully circumscribed debt or debt service relief for housing loan recip- ients acting in good faith but with the least capacity to pay. Additional actions included: (a) substantially revamping bank supervisory author- ity and putting in place clear, streamlined rules to deal with loan classification to address different categories of risks facing Colombian financial institutions and con- trol illicit financial transactions; (b) providing long-term loans, subject to strict guide- lines and limits via special recapitalization programs of FOGAFIN, to banks and housing finance institutions that have clearly evidenced their commitment to a full recapitalization of their enuties using sufficient quantities of their own funds and risk ENABLING SUSTAINABLE GROWTH 67 capital; (c) strengthening guidelines to control money laundering; (d) overhauling the financial cooperatives segment, accompanied by a newly established supervisory and deposit insurance system; and (e) upgrading the Superintendency's comprehensive risk-management system. Overall, the government spent more than US$4 billion during 1999-2001 on financial sector emergency measures and to close, downsize, restructure, or otherwise resolve problems associated with financial institutions Due to a number of financial sector reforms, the institutional landscape for pub- lic and private financial institutions operating in Colombia's rural sector has shifted dramatically in recent years. With the closure of Caja Agrarta in 1999, the govern- ment has taken significant steps toward eliminating distortions in rural financial markets and rationalizing government support to agricultural and other rural enter- prises. In addition, changes in the legal and regulatory framework have led to the massive restructuring of bank and nonbank financial intermediaries in Colombia, with mergers and liquidations drastically reducing the number of banks, coopera- tives, and other financing companies operating in the country. Banco Agrario, Fondo Emprender, and several guarantee and titling programs administered mostly through the Fondo para el Financiamiento del SectorAgropecuarzo (FINAGRO), are still oper- ating in the rural sector. While the actions noted above did help avert a systemwide crisis, a substantial further agenda of actions remains if a healthy financial system providing the requi- site services to the population and productive sector is to be assured. The agenda of further actions to put the financial system on a sound footing for the future should include: * Wind up, restructuring, and/or privatizing the public and officialized finan- cial institutions (most important, BANCAFE and Granahorrar) as early as possible to avoid further fiscal drain and erosion of the value of these entities. * Review and overhaul policies, procedures, and action plans being followed by CISA, the subsidiary of FOGAFIN responsible for the management and dis- position of assets transferred from the bad loan portfolios as part of the process of the resolution of problem banks, in order to achieve more rapid and efficient disposition of the corresponding assets. * Review the operations of the various second-tier financial institutions cur- rently under state control (Instituto de Fomento Industrial, State Industrial Development Bank, IFI; Financiera de Desarrollo Territorial, FINDETER; Financiera Energe'tica Nacional, FEN; FINAGRO, and BANCOLDEX), with a view to rationalizing and consolidating their operations and thereby increas- ing the efficiency and effectiveness of second-tier bankung, ideally via the operation of only one or two second-tier agencies with clear rules that would be applied consistently across different sectors. * Reform the pension and social securiry system (see the Pension Reform Chap- ter). The main elements of a reformed system should include (a) establishing a fully capitalized, actuarially sound, defined contribution pension system at 68 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE the national level for public and private employees with parameters related to eligibility rules, benefit accumulation rates, retirement age, and maximum benefit rules, adjusted to ensure its long-term financial viability; and (b) har- monizing the system across the different employee populations by moving Instituto de Seguro Social (Institute of Social Security, ISS) and the various spe- cial and "excepted" public employee pension regimes toward such a system as early as possible and by clearly specifying a transition phase to the extent that an immediate switch to such a new system is not politically or legally feasible; and (c) reforming the various sub-national level public employee pension schemes by implementing essentially the same principles for pension reform at the decentralized levels of government. * Review and reform judicial system procedures as they affect credit and finan- cial system operations. Most important, reduce the extremely long delays imposed in exercising guarantees and foreclosure of properties provided as col- lateral for credit transactions, which places a major burden on creditors and imposes, in effect, a high risk premium in all credit transactions in the econ- omy. Specific attention should be given to continued strengthening of effec- tive commercial arbitration and extrajudicial methods to reduce the economic burden of these legal and judicial processes. * Develop a comprehensive strategy to foster capital markets development. This strategy should include completion of the work started under the proposed new capital markets and securities legislation that established clear disclosure and transparency rules and strengthened the legal framework supervisory guidelines for entities and instruments to facilitate housing mortgage securiti- zation. Streamline regulations for efficient securities issuance procedures, to increase the capacity of the Superintendency of Securities (for example, to detect and investigate cases of market fraud, manipulation, and insider trad- ing), and to deepen the primary and secondary markets for fixed-income secu- rities issued by the Treasury By establishing the appropriate anchor for interest rates and a yield curve in the market, a government securities market could provide the necessary platform to further develop capital markets operations in other segments. The new securities legislation should be passed without delay. * Overhaul the insurance sector and the regulatory framework guiding it, given the relatively weak information systems and database in this area and an underdeveloped regulatory system that could threaten the heath of the sector, as important products such as those covering natural disaster insurance and crop- and weather-related insurance are developed in the future (see section on Disaster Management and Disaster Insurance Mechanisms). • Review problems of access to finance reportedly experienced by medium-size, small, and microenterprises throughout the economy, and rural enterprises. The progressive withdrawal of public banks and the consolidation of private sector institutions may have set the stage for the growth of more competitive and stable financial intermediaries. However, their impact on the availability ENABLING SUSTAINABLE GROWIH 69 and terms of services for different strata of rural households has not yet been compensated by a growth in credit flows from the financial system to small/ micro enterprises and the rural sector In the case of agriculture sector, the cur- rent agricultural credit scheme, which is anchored in the Banco Agrario and FINAGRO, needs to be reviewed with the objective of shifting the role of the state from retail lending to an institutional development role with more par- ticipation of private sector and nonbank intermediaries. Incentives and tech- nical assistance for intermediary capacity building should be provided to pro- mote expanded services in rural areas by sustainable bank and nonblank institutions. Instruments adapted to cater to the special needs of the rural pro- ducer in terms of savings mobilization, insurance schemes, contract farming, investment funds, and facilitation for the use of collaterals will therefore need to be developed. Innovations that reduce transaction costs and spread risks more effectively, such as index-based insurance and commodity price hedging, also ought to be considered for wider adoption IV. Creating a Business Environment to Foster Private Sector Development Although Colombia has produced notable private initiatives and penetration of new markets (for example, development of export markets for flowers and educational material; and private participation in water, electricity, and road concessions), the overall trends in small and medium-size enterprise development, export growth, investment rates, direct foreign investment, and overall factor productivity have not been promising. A recent reform of the corporate governance and bankruptcy legis- lation has brought some essential reforms to prevent creditors from pushing trou- bled companies into liquidation prematurely. However, it may need some further adjustments to streamline and modernize the rules for extrajudicial debt restructur- ing agreements (Acuerdos de Reestructuracuin under Law 550) and to ensure efficient closing or bankruptcy procedures where needed. Regulatory and competition frame- works have been improved in several infrastructure areas (see below). The performance of small and medium-size businesses and microenterprises, a segment critical to employment generation in the economy, has been poor. Colom- bia has had a strong historical tradition of innovative private sector and philan- thropy-based schemes to support small business and micro-enterprise development, and this needs to be revived, with government support, if necessary, in the areas of technical assistance, business advisory and information technology services, and cat- alytic financial assistance where appropriate. A major obstacle to fluid private sector operation is the large administrative bur- den (tramitologia) imposed on enterprises by federal, departmental, and municipal government authorities in the form of a myriad of administrative and license requirements, slow processing of required authorizations, and special fees or mone- 70 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE tary compensation that officials may demand. The impact of the administrative bur- dens imposed by these procedures, fees, and required authorizations is dispropor- tionately high on the smaller businesses and microenterprises because they represent fixed costs applied to a very small scale of business, and because in many cases the entrepreneur, head of the enterprise, or the senior-most official often have to devote substantial amounts of their own time to these chores, which is time away from managing the firm's productive operations, looking for new areas of business, and developing a customer base The government should commission a special study of the current requirements and their adverse impact on productive operations of dif- ferent sizes and in different sectors, and develop an agenda of reforms to ensure more streamlined and less costly procedures. Effort to develop nontraditional exports should be given high priority. With the spread of free trade arrangements and the gradual coming into force of World Trade Organization provisions, export development becomes an important ingredient of Colombia's economic growth strategy. A combination of assistance to budding exporters in the areas of market identification, product development, information and business advisory services, improved access to financing, and other technical assistance sources would help accelerate the development of export potential by Colombian entrepreneurs. Given the critical importance of private sector led growth and employment gener- anon, developing a consensus view across the different areas of the government (e.g., planning, finance and trade ministries, and departmental/municipal authorities) on a specific integrated set of actions that should be taken in these areas is a matter of urgent priority. This is best done via a quick comprehensive review that would help develop more detailed options and actionable recommendations in each of these areas for early consideration by the Consejo Nacional de Politicas Econ6micas y Sociales (National Council on Economic and Social Policies, COMPES) and rhe President. V. Improvement in Infrastructure and Public Services to Foster Competitiveness Adequate physical infrastructure can play a major role in enabling economic growth and competitiveness. Econometric studies using data from a large number of devel- oping and developed countries show that as much as one third of economic growth can be directly linked to the availability of adequate infrastructure, and Colombia is no exception. Poor logistics alone have been estimated in some economies to lead to a 15 to 20 percent increase in the total cost of production in some sectors. Indeed the role of infrastructure in enabling growth may actually be even greater now in Colombia, given the significant erosion of physical infrastructure facilities that took place in recent years as a result of both inadequate maintenance budgets, and vio- lence that includes periodic sabotage of highways, bridges, power transfer stations, distribution towers, and other facilities. ENABIING SUSTAINABLE GROWrH 71 Despite very challenging circumstances, Colombia has made major efforts over the last decade to improve the organization and efficiency of infrastructure services, and has implemented several thoughtful initiatives. Although the nature of the efforts and their relative success varied greatly across different infrastructure sectors, there has been progress in most sectors. In general, these efforts paid off in terms of serv- ice improvement and greater amounts of foreign and domestic investment in many of the infrastructure sectors, even while hindered in specific sectors by the problems created directly or indirectly by the security situation (such as problems in road trans- port concession projects and privatization proposals in electricity generation and dis- tribution companies). In the last three years, foreign and domestic investment rates have turned downward sharply, reflecting the difficult macroeconomic and security conditions, the need for fiscal austerity, and the volatility and uncertainty that has characterized the financial markets and Colombia's credit rating in international mar- kets. Given the limited public resources available for investment in these sectors, unless private sector investment picks up substantially, Colombia will fall short of the minimum infrastructure investments needed to ensure adequate maintenance, replacement, and expansion of the services to keep up with the demands of the grow- ing population and to fuel the requisite economic growth. a) Energy Sector Over the last decade Colombia has been a pioneer in implementing far-reaching structural and regulatory reforms in the electricity sector. The current structure pro- vides for an unbundled electricity sector structure (between generation, transmis- sion, distribution, and marketing) that is conducive to competition and private sec- tor participation in electricity generation and distribution. A well-functioning wholesale electricity market has been established and has already demonstrated sig- nificant gains in overall efficiency, cost reduction, and quality of service (number of service interruptionis, loss and access indicators) and maintenance of relatively stable electricity prices to consumers since the inception of the wholesale market. The state-owned generation and transmission company (Interconexidn ElMctrica S.A., ISA) continues to be one of the most efficiently run public sector companies in Colombia, and has even shown an institutional capacity to cope with the traumatic events related to guerilla violence without major systemic breakdowns. Guerilla-related violence and threats have, however, caused major damage to power and oil transmission and distribution facilities, typically causing major inter- ruptions in service through the blowing up of towers and other facilities. The con- templated privatization of ISA (which was to be unbundled and sold) had to be sig- nificantly delayed as a result, but a proceeding more gradually via public share offerings in the market. Privatization of ISAGEN and several regional distribution companies was also delayed as a result of an anti-trust dispute (ISAGEN) and cer- tain financial and governance issues (distribution companies). Efforts are under way to complete the process with some innovative shareholder structures to offset the 72 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE uncertainties posed by guerilla activities, including the very successful offering of a significant block of ISA shares to the public at large via rhe stock market. The sec- tor regulatory agency, the Comiszdn de Regulaczdn de Electricidad (CRE), developed good institutional capacity and had been instrumental in the healthy development of the electricity wholesale market. However, over the last two years, some impor- tant questions have been raised by private sector participants as to whether CRE's interpretation of the regulations in imposing certain price restraints in the operation of the wholesale market was sound and whether CREG is becoming excessively short term oriented. Key concerns relate to the way the recovery of capacity charge is dealt with under the rules of operation of the wholesale market. There may well be some legal challenges to the jurisdictional authority of CRE or the manner in which it is being implemented. If not resolved in a manner amicable to all parties, doubts about such interpretations may have a negative effect on the attraction of fur- ther private sector investment into the sector. Colombia is a country with major coal reserves. The government has been largely successful in privatizing the coal industry and thereby developing an export-ori- ented, private-sector-led industry. The logistics of the transport of coal and arrange- ments for sharing transport and railway facilities among firms under clear rules are important issues in this sector. The government has generally been successful in arranging for reasonable sharing, although this may need some additional work. The government, through the Empresa Colombiana de Petrdleos (ECOPETROL), has been fairly successful in maintaining the interest of oil companies in exploration, primarily by adjusting the parameters of the association contracts and pricing to be in line with the evolving international scenario and the local reservoir characteris- tics. Oil exploration and production are conducted mainly under "association con- tracts" between ECOPETROL and private companies. Work is also under way to strengthen the regulatory structure governing natural gas transmission and distribu- tion throughout Colombia. The strategy is to strengthen the sector institutions in implementing the regulations, and to promote more effective use of natural gas through an expansion of the gas pipeline network and distribution stations through- out the country. In part to make up for the adverse developments in the investments stemming from the deteriorating security situation, the terms of the standard joint venture contract among ECOPETROL and oil and gas companies were modified during 1994-2000. The 1999 and 2000 reforms included royalty relief and a reduc- tion in ECOPETROLs participation requirement from 50 percent to 30 percent. There are, however, potential opportunities to improve incentives and move towards further deregulation in the natural gas and oil sectors. A special effort is needed to improve access to electricity and other energy sources in the rural and isolated areas. A focus in developing off-grid electricity supply based on renewable energy sources such as wind, geothermal energy or mini-hydro plants would seem a promising avenue to correct the deficiencies in access. In this connec- tion, Colombia should aim to take advantage of the incentives for renewable energy production offered under the Kyoto Protocol and the Prototype Carbon Fund. ENABLING SUSIAINABLE GROWTH 73 ENERGY/ENVRONMENT ISSUES. Oil exports have decreased in recent years, from 810,000 bbl/d in 1999 to 630,000 bbl/d in 2001. This decline is partially due to more than 180 bombings of the Cano Limon-Covenas, Transandino, and Ocensa pipelines in 2000 and 2001. Oil spills due to these bombings have had a significant negative environmental impact. Given the regulatory uncertainty caused by the current envi- ronmental licensing system, has led to regulatory uncertainty. There is an urgent need to reform the regulations on environmental licensing for oil and gas projects (Decree 1753/1994). The government has strengthened the capacity to monitor the environ- mental impact of energy-related policies and projects e.g., to monitor the environ- mental impact of the use of fossil fuels in the transport sector. Work is also under way to strengthen both the regulatory structure governing natural gas transmission and dis- tribution throughout Colombia and the sector institutions implementing the regula- tions, and to promote more effective use of natural gas through an expansion of the gas pipeline network and distribution stations throughout the country. Among the key strategic issues to be addressed by the new Administrationi in the energy sector are: * Adequacy of the incentives for petroleum exploration and development (including tax policies, royalties, sharing agreements, etc.) in light of the cur- rent security situation and how these may need to be adapted as security con- ditions in Colombia improve or deteriorate. * Appropriate division of roles between the public and private sector; including in particular ECOPETROL's dual role as a producer and as the entity with which all other producing companies have to enter into association contracts. * Possible further deregulation of the natural gas, in respect of well head prices, downstream operations and prevailing restrictions and exports. * Best strategies and options for proceeding with the privatization of ISA, ISAGEN and the distribution companies which are still in public sector hands. * A full review of the performance of CREG, required institutional improve- ments in CREG, and improving the clarity of the nature of its regulatory role and scope of functions, while preserving its autonomy from the line ministries. * Improving the capacity, operations and performance of SSPD. * Putting in place a holistic strategy to improve access to electricity/energy to rural and isolated areas via public-private solutions to energy services. * Taking advantage of opportunities offered by the Kyoto Protocol and the Pro- totype Carbon Fund (PCF) to finance off-grid solutions (e.g., wind, small hydro plant). * Reexamining the efficacy of the cross-subsidization scheme on electricity prices currently based on classification of customer by income strata. Consid- eration of options that may be available to make the system simpler and easier to operate-for example one in which the basic minimum level of services is offered at a subsidized price for all customers, with higher (and possibly esca- lating) prices for additional blocks of service. 74 COLOMBiA THE ECONOMIC FOUNDATION OF PEACE Further enhancement of institutional capacities to monitor and minimize adverse environmental impact of energy operations. b) Telecommunications Over the last five to 10 years Colombia has taken steps to open up the cellular tele- phone market, conduct spectrum auctions and, more recently, redesign the ongoing Personal Communication Service (PCS) concessions to make them viable. The fixed-wire telephone services are, however, lagging behind in terms of reforms, and are still dominated by government-owned local and long-distance telephone capac- ity (the Bogota telephone company and TELECOM, respectively). The telecommunications sector in Colombia faces several issues: * Low Private Sector Participation. Only 3.4 percent of Colombia's telephone lines are in the hands of private companies today, compared to 85 percent for Latin America as a whole. In general, telecommunications is considered to be a private sector business throughout the region, with the exception of Colom- bia and only a few other countries. * Fragmentation of Regulatory Institutions. Colombia's regulatory institutions are fragmented: (a) the Commission for Television Services oversees the television subsector; (b) the Ministry of Communications manages the Radio Frequency Spectrum, assigns frequencies for different use, and controls the radio subsector; (c) the Telecommunications Regulatory Commission dictates the fundamental technical plans and manages other scarce resources for the telecommunication operators; (d) the Superintendency of Industry and Commerce regulates the operators' anticompetitive behavior; and (e) the Superintendency of Public Domiciliary Services supervises the relations between customers and operators. * Subsidies and Tariffi. Compared with other countries and the rest of Latin America, Colombian tariffs are below the average, and in some cases, below the cost of providing the service, causing financial problems for electricity companies and contributing to government fiscal imbalance. o Low Internet Penetration. Colombia had only 0.54 Internet hosts per 1,000 inhabitants, well below countries of similar GDP per capita, and below Argentina, Brazil, and Mexico. * Reduced Role of Colombia in the Information Society and the Global Economy. Even though the Pastrana administration's "Connectivity Agenda"3 has made an impressive contribution to bring Colombia into the Information Society, much remains to be done. There are still too few computers in Colombia, fewer con- nections to the Internet than in other Latin American countries, and lower use of information and communication technologies (ICTs) for development. 3. "Agenda de Conectividad," Departamento Nacional de Planeaci6n, Bogota, 9 February 2000, Documento COMPES, 3072. ENABLING SUSTAINABLE GROWTH 75 In addressing these issues, the government should give serious attention to the following specific actions: * Increased Private Sector Participation The current scenario of public sector controlled companies playing a dominant role in providing telecommunica- tion services should yield way to significantly increased private sector partici- pation via a proactive restructuring and privatization program. To facilitate that, several other measures listed below need to be taken simultaneously to guarantee a vibrant and competitive sector. * Tariff Reform Local telephone tariffs need to be liberalized within the framework of a price cap-based regulatory system to stimulate investments in local service. Such a system would allow companies freedom to set their own tariffs as long as they comply with the price cap, which would be set at a level high enough to guarantee a reasonable rate of return to companies that invest in networks to provide new telephone lines. An international benchmarking system could be used to set and periodically adjust the price cap. To ensure access to minimum basic service to the poorest and least priv- ileged segments of the population, a "lifeline service" operated via public phones or a prepaid access cards system should be considered. This would replace the current system of stratification of subscribers for tariff purposes which has not been effective or economically efficient. In order to harmo- nize better with international rates and avoid adverse incentives for compa- nies and consumers, the government should also consider replacing the taxes imposed on long-distance calls with a general tax on telecommunica- tions company profits and setting the new proposed levy for universal serv- ice at a more moderate level.4 * Consolidation of Regulatory Institutions. The current system requires telecommunications enterprises to submit informationi and a multitude of reports to satisfy the requirements of diverse regulatory institutionls. More- over, given the overlapping jurisdictions, it is difficult to have clarity on with regulatory entity should have the ultimate say on a specific issue The creation of a single regulatory entity for the telecom sector covering tariffs, service licensing as well as spectrum control would be desirable. The gov- ernment, through the Ministry of Communications, should in any case retain the responsibility for formulating and implementing overall sector strategy and policies. * Incentivesfor Local Telephone Companies to CompeteAgainst Each Other and the Flexibility to Merge To provide incentives for the private sector to invest in local telephone companies, they should be allowed to also offer mobile services (cel- lular and PCS services) which is the segment of the industry that has higher growth potential for the future and needs large investments in line with the 4 In Latin America the fee is usually one percent of gross revenues 76 COLOMBLA THE ECONOMIC FOUNDATION OF PEACE oncoming third generation of mobile services.5 In addition to the 1900 MHz band (the North American Standard) already open, the government should also make available the 1800 MHz band (the European Standard) to operators. Telecommunication assets, concessions and frequency bands for mobile serv- ices should be auctioned via open public offerings to facilitate structured com- petition in the sector, efficiency enhancing mergers where appropriate. Colombia and the Information Society. The government should increase efforts to insert Colombia into the new Information Society. A key measure would be to assign consolidated responsibility for developing and implementing an integrated information and communication technology (ICT) strategy. Work- ing closely other ministries, it would help develop and implement new ICT applications to improve the efficiency and effectiveness in these other eco- nomic and social areas, and insert Colombia firmly into the rapidly evolving world knowledge economy. The government should also consider all or part of the revenues mobilized from the telecommunications companies to fund ICT applications for development purposes, outsourcing the actual provision of ICT-based services to the private sector as much as possible. c) Transport Colombia had been a pioneer in the private sector-driven development of ports, which handle 98 percent of exports flowing out of Colombia, with a large portion of it through specialized private terminals (coffee, bananas, and oil). The opening up of these sectors for private sector participation has on the whole been very success- ful and has yielded positive results in terms of increases in productivity and effi- ciency of port operations, reduction in total costs and shipping times, and corre- sponding reduction in rates. In the civil aviation sector, three hub airports and the second Bogota airport terminal have been given under concession contracts, and the independent regulatory authority (Aerocivdl) established earlier continues to func- tion satisfactorily. In railroads, two networks (Pacifico and Atl4ntico) and the rolling stock were recently devolved to the private sector under two concessions. Although these initiatives are also expected to yield productivity, efficiency, and quality-of- service gains, the ultimate effects cannot yet be assessed. In the area of highway development, Colombia had been innovative as well, lay- ing out a strategy that explicitly considered ways to maximize private sector partici- pation via different types and "generations" of toll road concessions, together with methodologies for prioritizing budget allocations among federal and departmentally funded highways and secondary and tertiary roads. The Instituto Nacional de Vias (National Highway Institute, INVIAS), the government agency responsible for managing the national highway program, was actively pursuing innovative methods 5 The third generation usually refers to mobile telephone systems that provide fast Internet access for data, voice, and video ENABLING SUSTAINABLE GRO\vTH 77 to achieve this objective, such as providing certain types of assurances regarding traf- fic demand, and "negativc concessions" where concessionaires bid for the minimum present value of the toll road revenue they will retain. Colombia has a long experi- ence of contracting maintenance through microenterprises, achieving substantial improvement in the condition of the road network by the end of the 1990s The government also took action to reduce accidents and hijackings, and introduced monitoring of vehicles in transit via Global Positioning Systems (GPS). Recently, however, all parts of implementation of the highways program have experienced severe setbacks. Security concerns have made it extremely difficult to reach agreements with the private sector on new concession arrangements, especially where the size of the transaction or technical capacity requires the participation of foreign investors and concessionaires. Severely constrained budget resources com- bined with an increasing incidence of sabotage have limited new work on even nationally or departmentally sponsored roads. Local communities that depend on access roads for education and health facilities and trade links to the rest of the Colombian economy have suffered severely. Management of road concessions has also suffered from weaknesses in the institutional capacity to follow up, control, and supervise the concession contracts, and from the lack of an overarching framework linking the concessions program to the functioning of the transportation system as a whole. Given that the availability of good quality roads is typically a key factor in the ability of the population (especially the poorest families) to access basic education and health service facilities and to participate in economic activity and the productive labor force, it is critical to quickly correct these deficiencies. One bright spot has been the development of the urban transport system in Bogota, with a clear and consistent policy anchored in the development of the highly successful transmilento and improved attention to road safety, service improvement, and reduction in adverse environmental effects of vehicular emissions (for example, through the introduction of the pico yplaca system which requires all cars to be off the roads one day a week). Given the uncertainties and lack of a sector strategy that can be effectively imple- mented, INVIAS and the Ministry of Transport have also suffered the loss of key staff and decreased motivation among those who remain in the agencies. The con- sequent erosion of institutional capacity in these agencies is an important problem that needs to be addressed. It is also critical to establish and build the capacity of a transport regulatory agency to provide clear guidelines for sector activities, and to build the capacity of agencies responsible for road building and maintenance at the departmental and local levels. Looking ahead, it is critical to * Rebuild and strengthen the institutional capacity of (a) the Ministry of Trans- port, to enable it to better perform its role as policymaker and ensure the devel- opment of a good master plan for transport; (b) INVIAS in its role as the entity with responsibility for detailed road planning and design and execution of road concessions; and (c) a well-functioning transport regulatory commission. 78 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE * Clarify the division of responsibilities between the federal and decentralized levels of the government, and build appropriate institutional capacities at the decentralized (department and municipal) levels of government. * Overhaul the highway concessions program to adapt it to the reality of the Colombian security situation, if necessary incorporating appropriate transi- tion and contingency management strategies, ensure longer-term financial and environmental sustainability of the system. Should the security condi- tions improve, be prepared to revert to a viable longer-term strategy. * Develop innovative methods for building and maintaining rural and regional roads that fully incorporate the participation of the local communities and promote a concept of integrated regional development; features could include contributions (in klnd or in cash) of the local communities in road construc- tion and/or maintenance, links to tertiary roads and nonmotorized transport methods where appropriate, and promoting participation of local microenter- prises in the work and in housing and community facilities development. d) Urban Development and Low-Income Housing In Colombia, 31 million people reside in urban areas, compared to about 12 mil- lion in rural areas. Indications are that there will be an even greater shift of popula- tion toward urban areas in the future. Among the problems faced by Colombia's growing urban population are (a) lack of systematic urban land management; (b) poorly developed land-titling systems and land markets; (c) congestion and trans- portation bottlenecks in the large urban centers; (d) large slum areas surrounding major cities that suffer from lack of access to basic infrastructure services such as piped water, electricity, community roads, and telephones; and (e) inadequate access roads to link the poorer neighborhoods to the main city centers. The problems tend to accumulate and worsen over time because of inadequate institutional capacity of the municipalities and departments to develop strategies to correct these problems. Given the limitation of three-year, nonrenewable terms for mayors of municipallties, there is also typically very little incentive for city administrations to take corrective actions that are strategically thought through. Also, notwithstanding specific provi- sions under Law 134 of 1994 to strengthen citizen participation in local decision- making, there is often less-than-sufficient consultation with and participation of local communities in designing and implementing urban development solutions. Despite some overall improvement in aggregate urban poverty indicators, there are large localities in metropolitan areas that are clearly suffering from acute poverty, overcrowding, unhealthy environmental and social conditions, lack of basic services such as water supply and sanitation, and extremely poor-quality housing with haz- ardous conditions. The history of inadequate shelter options and the lack of an effective targeted subsidy system for the lowest-income stratum has led to disorderly growth of slum areas. These areas are characterized by informal settlements typically initiated through illegal but organized "invasions" of public lands and accelerated by ENABLING SUSTAINABLE GROWTH 79 pirate developers. Since the settlers in these areas almost never receive legal property titles to the land they are occupying, there are few incentives for them to improve the land and the services thereof unless the government takes specific action to facil- itate an orderly transformation of these areas. To address these problems Colombia must focus on: * Progressive upgrading via community-based development, together with practical solutions for problems of land tenure and property titling, and effi- cient mechanisms to help the poor help themselves. * Other effective measures to expand the shelter supply to the poor that, among other things, could stem the tide of informal settlements and "invasions" of public lands. * More effective ways of addressing the problems of access to basic services such as water, sanitation, electricity, community roads, and other amenities, within the framework of actions advocated in the individual sectors as described in the preceding sections * Improved functioning and capacity of the environmental management agen- cies (EMAs) or the equivalent units operating through the autonomous regional agencies to ensure environmentally responsible development of these areas, removing the current overlapping responsibilities among the local, regional, and central level agencies. Also of high priority are measures to improve the operation of the land markets by removing or relaxing the excessively cumbersome rules and regulations that govern land acquisition, titling, and transfer. Programs to upgrade land, roads, and housing in slum areas surrounding the main cities of Colombia and to improve access of basic services to those areas could go far in improving the quality of life in urban slums. The experi- ence of other countries in Latin America and elsewhere (Brazil, Thailand, Venezuela) indicates that a systematic approach to upgrading slums will help integrate them into the cities and ensure access to basic services for the population living in those areas, and will instill in them a greater sense of community, ownership, and social capital. A well- designed subsidy system, with clear targeting of the subsidies to the lowest-income seg- ments of the population and with incentives built in for development of "progressive housing" (that is, gradual improvements in the housing units maklng use of household savings) could play an important role in improving the lot of these less-privileged com- munities and integrating them into the main urban areas. It would be important to combine and closely link these measures to initiatives to increase the supply of shelter, specifically to accommodate the needs of the poorest of the poor e) Superintendency of Domestic Public Services The Colombian institutional structure includes a unique mechanism for ensuring that public services provided to households by municipal or other public or private 80 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE service providers abide by certain basic rules and regulations. These supervision functions are preformed by the Superintendency of Domestic Public Services (SSPD). The SSPD complements the functions performed by the individual sectoral regulatory agencies such as the CRE, the Comisidn Reguladora de Agua Potable y Saneamiento Bdszco (Water Regulatory Commission, CRA), and others. Among its important functions are to ensure that the entities providing public services to citi- zens are being run effectively and in a financially responsible and sustainable way, that service quality is being respected, and in the case of specific concessions, that the rules agreed upon under the concession agreement are being respected. The SSPD has the authority to take over the management and control of a public utility under its jurisdiction if it finds that the utility is in violation of the rules and that its operation is deemed unsustainable financially or otherwise. In general, this means that if the SSPD intervenes in the operation of a utility, it will have to take over the operation of that utility and maintain satisfactory service provision during an interim period, until an appropriate restructuring and transfer of the utility to a new ownership and/or management can be achieved. While this rather unique structure offers the advantage of a clear separation between the roles of sector policy and tariff regulation and the supervision and con- trol of the operations of the specific entities providing services, in practice, this statute has given rise to a number of problems. Most important, it has been difficult for the SSPD to acquire the requisite institutional skills to interpret and enforce the regulations applying to the broad array of domestic services over which it has juris- diction (water, sanitation, solid waste disposal, electricity and natural gas distribu- tion, telecommunications, and so forth). There have also been gray areas as to its jurisdiction vis-a-vis the individual sector regulatory bodies (CRA, CRE, and so forth). Finally, while the SSPD is entrusted with the authority to intervene in the operation of a utility that is performing less than satisfactorily and poses the risk of defaulting financially or in the discharge of its responsibilities, it generally lacks the technical, administrative, or other capacity (or the incentives structure) to operate these public utilities effectively even during an interim period, let alone to shepherd a rational restructuring of the entity. The functions of intervention, restructuring, and spinning off the operations of public utilities require very advanced technical, administrative, and financial engineering skills that the SSPD lacks. This has become quite clear from the way the SSPD tried to manage the interventions of a number of electricity distribution companies and water supply and sanitation com- panies over the past two years. For the SSPD to be able to discharge its responsibil- ities effectively will require, at a minimum, highly focused efforts to precisely define the rules of operation governing SSPD interventions, the scope of authority and responsibilities of the SSPD during the period of intervention, and creation within the SSPD of the requisite advanced level of institutional capacity and sophistication. If this is considered untenable in the long term, the government may need to search for other more appropriate institutional mechanisms, including a possible reorder- ing of the entire supervisory apparatus for basic public services. ENABLING SUSTAINABLE GROwrH 81 VI. Management of Natural Resources and Physical Environment In pursuing growth, productivity, and employment objectives, it is critical that the government preserves and protects the rich natural resource endowment of Colom- bia. Clearly, growth achieved at the expense of damage to physical environment or poor management of natural resources will not be sustainable and cannot assure long term improvement of the quality of life for Colombians. Worldwide experience has shown that correcting environmental degradation will almost always prove to be much more expensive than prevention through prudent natural resource manage- ment policies based on appropriate incentives. The following sections examine five critical areas related to natural resources management and environmental protec- tion: (a) agriculture and land resources management; (b) water resources protection and management; (c) potable water supply, sanitation and wastewater treatment; (d) solid waste management; and (e) management of natural disasters. a) Rural and Agricultural Development Skewed incentives and lack of appropriate technologies, compounded by the expan- sion of illicit crops, are taking a toll on Colombia's natural resources, and its remark- able environmental capital is being rapidly degraded. Colombia has amongst the richest biodiversity resources but also faces major soil degradation and water pollu- tion problems which are having a detrimental effect on many watersheds and on the contamination of the hydrological systems.6 In the Orinioquia and Amazonia regions, the impact comes from recent colonization, but it is already showing in the Guanfa, Guaviare, Putumayo, Caqueta, and Meta departments, where more than half of the area has already been deforested. The gap between the natural vocation of the land and its actual use points to a distorted incentive structure and inappro- priate use of resources. It is estimated that the agriculture/livestock frontier advances at a rate of 200,000 hectares per year.7 AGRICULTURE The agriculture sector is of major importance in the Colombian economy accounting for 14 percent of GDP, 23 percent of the labor force, and 28 percent of total exports. In the aggregate, it shows relatively high labor productivity and a remarkable development potential thanks to the extraordinarily rich natural resource endowment and the quality of its human capital. However, recent produc- tion trends suggest that the natural resource base is not being used efficiently. There has been a sharp decrease in the area sown with annual crops during the nineties, 6 The volume of sediment transported by the Magdalena River to the sea (130 million tons) is equivalent to 40 percent of the total sediment carried by the entire fluvial network of the country (IDEAM 1998, "El Medio Ambiente en Colombia") 7 "Estudio sobre cobertura y uso actual de la tierra en Colombia," (IGAC-CORPOICA 2001). 82 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE resulting in an estimated decrease of one million hectares to a level of 1.6 million hectares. This decrease in annual crops has been only partially offset by an increase in permanent crops. Most of the land taken out of crops has been converted into extensive livestock farming of low productivity. Almost all of the livestock farming is based on the latfijndios with about 38 percent of the production based in 3.5 per- cent of the farms and much of the expansion in permanent crops is also in the large- holder sector. The gap between the natural vocation of the soil and its productive use suggests an inefficient allocation of resources stemming from biased incentives in agriculture, land, rural finance, etc. This is evident even in the irrigation sector where the absence of rigorous cost recovery in public irrigation schemes and skewed incentives helps explain why most of the land is under low-margin crops and pasture. More- over, the insecurity generated by violence reduces the incentive to invest in agricul- ture and biases the pattern of investment toward non-intensive farm activities. At the prevailing low coffee prices, marginal coffee producing areas (especially small- holder farms), which have so far managed to remain relatively insulated, are now likely to be tempted to move towards illicit crops. Given the critical role of the coffee sector in the rural agricultural areas, avoiding a rural poverty crisis will require strong action to help improve productivity in the coffee sector. One strategy should be to help Colombian coffee producers move toward the country's comparative advantage, e g niche markets for high quality cof- fee. This requires targeted promotion and marketing effort and strengthening of public and private institutions to help implement such a strategy. A diversification strategy will be key to increasing productivity. Effective diversi- fication will require improvements in rural infrastructure, contract farming, mar- keting aliances, access to land and credit, and innovation strategies for small farm- ers. Some coffee producing farms and areas may have to get out of coffee. This will have large social implications and the issue of social safety nets and social compen- sation will have to be confronted while taking into account the fiscal costs of such strategies. The government, which has played a minor role in the coffee industry in Colombia up to now, should probably become more proactive in helping to imple- ment a multi-pronged approach along the above lines. FOREST LAND MANAGEMENT. Growing crops to produce illicit drugs contributes to deforestation. At the same time, spraying drug crops to destroy them contributes to environmental contamination and to the deterioration of natural resources, and is also affecting the health of the population living in these areas. In addition, coca cultivation is extremely destructive to forests in fragile areas. The recent expansion of the fumigation of coca fields has accelerated the itinerant characteristic of illicit crop growers in the search for new land to be cleared. In addition, chemical residues from coca laboratories are contaminating water flows and aquifers. The conflict will continue to pose a serious limitation on any attempt to manage natural resources in a sustainable way because violence is particularly acute in the richly endowed areas. ENABLING SUSIAINABLE GROwrWH 83 HAPHAZARD INFRASTRUCTURE DEVELOPMENT. Rural infrastructure work, espe- cially building roads, has also accelerated the natural ecosystem transformation process. Despite progress on the regulatory and institutional front, there has been little implementation of the Planes de Ordenamiento Territorial (Territor- ial Land Use Plans, POTs) contemplated in Law 388 of 1997, nor diligent application of environmental assessments to infrastructure development. ENHANCING COMMUNITY PARTICII'ATION. Efforts to restore natural resource man- agement practices need to be built from the local level up. Communities need to reestablish a certain level of social cohesion to enable them to make appropriate trade-offs among agriculture, livestock, and the sustainable use of natural resources at the local level. Unfortunately, the indigenous reserves and the communal land rights granted to Afro-Colombians have also been affected by decreasing social cohe- sion, as well as the disorder caused by the armed conflict b) Water Resources Management WATER QUALITY PROTECTION The ambiguities in the regulatory framework regarding pollution standards, environmental licensing, and pollution charges related to water resource management pose potential risks to the protection and conservation of this crucial resource as well as potential for private sector mivest- ment in this area The case of pollution charges illustrates the instability of the regulatory environment In 1993 Colombia became a regional leader in the development and implementation of economic instruments to mitigate and control environmental pollution. Under Law 99/93, Congress established pol- lution charges (tasas retributivas), and Decree 901 of 1997 established the actual implementation mechanism for wastewater charges. However, the ambiguity and other weaknesses of the system have allowed only few regional environ- mental authorities (Corporaciones AutSnomas Regionales) to fully implement the wastewater pollution charges. Even in those cases, the rules of the game are not completely clear. The regional environmental authorities determine the locally appropriate level of organic load and suspended solids in the water and set the pollution charges in a discretionary way. The wastewater charges could increase several times, in short periods of time, until the water quality goal is achieved. In light of this, and certain technical issues of water and sanitation regulation, it has become apparent that the system's regulatory methodology needs improve- ment. For example, it would be necessary to: (a) move away from the organic and total suspended solids loads for which the dischargers have to pay now, toward the actual water quality parameters that have a more significant effect on human health, such as heavy metals and toxic organic compounds; and (b) revise the actual charges such that they are less discretionary and more consis- tent across the country, ensuring clear "rules of the game" for industry and also for PSP in water utilities. 84 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE DECENTRALIZED WATER BASIN MANAGEMENT. The system of Corporaciones Autdno- mas Regionales (Autonomous Regional Corporations, CARs) has a mandate to oversee allocation of from individual water basins, act as regional environmental regulatory agencies, and to develop large infrastructure projects, water resources projects, basic sanitation, and other public works. There is a conflict of interest in the responsibilities of these authorities, since they act as regulators and at the same time implement irri- gation, drainage, water reservoirs, wastewater treatment plans, landfills, forestry, and other projects. The mandates of these authorities to implement large infrastructure development projects also overlaps with the mandates of national agencies, depart- ments, and municipalities. A reform of these 34 CARs that exist in Colombia is nec- essary to strengthen the environmental regulatory responsibilities and to overcome the conflict of interest and the overlapping of responsibilities mentioned above. c) Potable Water Supply, Sanitation, and Wastewater Treatment Given that water supply, sanitation, and wastewater treatment responsibilities fall mostly under local jurisdiction, the role of the central government is primarily to guide the overall policies and set the regulatory and tariff principles. As part of the provisions of the general law governing the economic management of all public serv- ices (Law 142 of 1994), water supply and sewerage charges levied on consumers (like most other public services) are required to be set at levels that ensure at least full recovery of operation and maintenance costs and the long-term financial sustain- ability of the utilities concerned. The law allows a scheme of cross-subsidization under which the commercial and industrial sectors and the two highest strata of res- idents could be required to pay up to 20 percent more than the value of their con- sumption to help finance the basic consumption of lower strata. Working within this framework, local governments can set the detailed tariff structures and decide on any explicit subsidies or cross-subsidies they are willing to provide from their own budget resources for the poorer segments of the population. This system of checks and balances is beginning to function, although there is still much to be learned as full implementation of the system proceeds. Most water and sewerage systems providing service to the major urban centers of Colombia are operated directly by the respective municipalities or enterprises fully owned by the municipal government. Although the major urban centers of Colom- bia have had operating water and sewerage systems for a long time, most of them suffer from problems of low operational efficiency, high water wastage, poorly main- tained pipelines, limited (if any) access of piped water to the poorer neighborhoods, and water and sewerage charges that are inadequate to allow a financially sustainable operation by the utility. Wastewater discharges into bodies of water are also required to fully abide by the applicable environmental regulations and charges. However, in reality, few municipalities in Colombia (including major cities) have satisfactory wastewater treatment facilities to treat the water prior to discharging it into bodies of water. Thus few of them manage to comply with the applicable environmental ENABLINCG SUSTAINABLE GROWTH 85 regulations on such discharges. This is a problem that needs urgent attention and appropriate corrective measures, because some of the major bodies of water in Colombia are already severely contaminated and pose significant health and envi- ronmental risks. The government recently launched a nationwide program to help small and medium-size water supply and sewerage systems improve the efficiency and quality of their services and ensure adequate water access to the poorer neighborhoods. The local communities themselves would have a major role in developing the available options to improve the quality and access of water service and choosing one that best fits their needs. It 1s anticipated that a majority of the communities will choose a pri- vate sector concessionaire, manager, or operator to run the services. The government initiative aims to attract the largest possible amount of private investment resources into the sector and to catalyze improved efficiency, reduced wastage, and good cost recovery The program is aimed at learning from the initial experience with water supply concessions and management and operating agreements, and then applying what is learned to the operation of a large number of water and sewerage systems throughout Colombia. This program of gradually extending private sector partici- pation and adapting the specific mode of private participation to the characteristics and size of the local water and sanitation system is highly commendable. To tackle the major issues facing the sector, Colombia must: * Focus on improving water supply and sanitation services in small and medium-size municipalities while improving the efficiency of operation, qual- ity of service, and access provided by them through the progressive applica- tion of economic efficiency principles and innovative methods to increase engagement of private sector participation in investments and operations. * Improve the operation, efficiency, quality, and access to water and sanitation systems in the large metropolitan areas, where there are even greater opportu- nities to move toward and increase the private sector role in all aspects of investment and operations. Pilot schemes to test alternative solutions should be tried in the near term in this area. * Develop and implement a policy for increasing the coverage and level of serv- ice in the rural sector, which has steadily deteriorated over the last 15 years. * Improve management and disposal of municipal wastewater throughout Colombia by (a) harmonizing environmental legislation and guidelines related to water and sanitation to ensure public health and to prevent further environmental deterioration, in general, and of the quality of drinking water sources, in particular; and (b) identifying municipalities of high priority in terms of the need for wastewater treatment to prevent public health risks and degradation of water resources. * Implement a simple but effective water sector regulatory framework under the authority of the Water Regulatory Commission (CRA) to address the prob- lems identified above. 86 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE * Develop innovative options to finance the above activities by creating frame- works to attract increasing investment and financial resources from the private sector. * Reform the CARs to strengthen their institutional capacity to discharge their responsibilities and overcome the conflict of interest and overlapping respon- sibilities mentioned above. d) Solid Waste Management More than 21,000 tons per day of municipal solid waste are improperly disposed of in urban centers in Colombia. There is an urgent need to address this problem. The national government has developed policies and a methodology for the man- agement of solid waste in the country. However, the implementation of these poli- cies falls almost entirely in the hands of the municipal governments, and their record is very mixed. Although a number of municipalities have well organized col- lection systems, and with support from the Fondo de Regalias, have been investing in building landfills for the adequate final disposal of municipal waste, much remains to be done in this sector. For example, the regulatory framework needs to be revised, the technical assistance to municipal governments must be strength- ened, and additional resources for investments for solid waste management would have to be made available. e) Disaster Management and Disaster Insurance Mechanisms Colombia also confronts a variety of challenges related to natural disasters such as floods, droughts, and earthquakes. A separate section that follows identifies the key issues that bear consideration in achieving rational management of the impact of the potential natural disasters. Specific attention is paid to strategies to prevent or min- imize the disaster related damages and methods to anticipate the contingencies and financial costs and plan for them in a way that does not disrupt Colombia's long term economic development. Among other things, rational use of disaster insurance mechanisms and other contingent financial instruments should help mitigate and smooth the fiscal and financial costs of natural disasters. Colombia has historically been affected by a variety of natural disasters including earthquakes, hurricanes, floods, and droughts. Shifts in weather patterns make agri- cultural crop production particularly vulnerable. A systematic approach to antici- pating and putting in place appropriate safeguards could go a long way in reducing the human and economic costs of natural disasters. Specification and enforcement of better building codes at the national and local levels (tailored to the natural phe- nomena that characterize a particular geographical area), promotion of earthquake- proof construction methods, strengthening of information systems, a systematic approach to risk and vulnerability assessments, closer coordination amongst the dif- ferent institutions involved, and putting in place tighter standards for public infra- ENABLING SUSrAINABLE GROWTH 87 structure in disaster-prone areas could play an important role in minimizing dam- age and losses caused by natural disasters. Good contingency planning and innovative disaster insurance techniques should be used to help smooth the costs over time of coping with disasters, and minimize the trauma caused to the society by major catastrophes. Statistical and actuarial models have been applied successfully in other countries to derive probabilistic estimates of the types and frequency of the occurrence of disasters in a particular region. This information can in turn be used to design insurance schemes that achieve risk diver- sification across regions and risk smoothing over time. This will require develop- ment of appropriate statistical models, cost estimates, and design of progressive risk coverage, with different layers of risk being covered by different sources of funding. However, a government insurance scheme for public infrastructure could benefit from a substantial portfolio size to achieve pricing efficiencies and negotiating lever- age in terms of external reinsurance funding. Sources of funding to cover the risks could also include, progressively or in par- allel, insurance premiums and deductibles paid by the property owners, interna- tionally available insurance or reinsurance schemes, issuance of catastrophe insur- ance bonds to institutional investors or reinsurers, public funding of certain basic minimum costs in cases of the poorest segments of affected population and, possi- bly, contingency public funds that are especially set aside to cover catastrophic losses exceeding certain limits. The government has made a welcome initiative in strengthening contingency planning in order to increase its state of readiness to deal with disasters physically, logistically, and financially (for example, through the creation of a pooled contingency fund). Much more work is needed, however, to build a fully developed contingency planning and appropriate disaster management and insurance framework. The additional effort is well worth supporting in the coming years Parametric weather insurance instruments can also be very effective in helping farmers hedge against the risks of crop losses because of adverse shifts in climate. Other countries have initiated interesting pilot programs in this area that could be very relevant to Colombia, given its diverse agricultural climatic variations. Although commodity price-hedging markets already exist to manage world price fluctuations, even in stable pricing environments, unexpected weather developments can spell losses for sectors impacted by flood, drought, or freezes. The compilation of the historical record of such contracts and the development of financial hedging instruments (for example, options) to provide compensation against such events can be viable, particularly when counter parties might be available to take the "other side" of the risk, such as energy companies that actually benefit from weather extremes by providing more heating during colder weather. The benefit of using weather-based contracts, however, rests with the mechanism for contract triggering, that is, the use of degree measurements as the contract driver, without the need for on-site assessment of crop damage by area, which would otherwise significantly raise the costs and moral hazard risks of such contracts. 88 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE VII. Conclusions Sustainable growth is obviously key to the realization of the dreams of any society of achieving better standards of living and quality of life for all segments of the popu- lation. A thoughtful and responsible approach towards achieving sustainable growth will be one that provides a platform not only for achieving positive economic results in the short and medium term, but for a long term growth trajectory that avoids undue economic and financial volatility and its associated social and economic development disruptions, and unleashes the dynamism, energy, and innovation of the country's private sector. Moreover, it does this while fomenting the talents of Colombia's human resources and allowing the nation's considerable natural resources to be conserved and if possible bettered. A well conceived sustainable growth strategy should in the process also help bring down the high incidence of poverty and unemployment currently witnessed in Colombia. Achieving true finan- cial, fiscal, environmental, and social sustainability is a complex challenge that can be dealt with successfully only if the policy and institutional frameworks in the var- ious areas discussed above are designed to be mutually reinforcing, and the govern- ment is able to apply a strong determination to pursue and sustain the needed reforms, and good )udgement in striking a balance between competing considera- tions where needed. 3 Shared Growth, Poverty, and Inequality This Chapter was written by Carlos Eduardo Velez, Laura Rawlings, Vic Paqueo and Juanita Riafio. I. Introduction The equitable distribution of the fruits of economic growth is the most powerful tool available for fostering economic development. Until 1996, Colombia enjoyed high, sustained, and very stable growth that allowed for substantial achievements in poverty reduction and social progress. Between 1978 and 1995 the share of Colom- bians living in extreme poverty fell from 45 percent to 21 percent, while impressive gains were achieved in primary and secondary school completion, health insurance coverage, access to basic infrastructure, child labor, infant mortality, and life expectancy. In the second half of the 1990s, the most severe economic crisis to hit Colombia since the 1930s erased over a decade of progress in poverty reduction, and under- mined Colombia's ambitious social agenda. Macroeconomic performance deterio- rated significantly, growth rates plummeted, and unemployment escalated to almost 20 percent. Simultaneously, risk indicators worsened, mostly due to increasing eco- nomic volatility and financial sector fragility, and persistently high levels of crime and violence. Moreover, de facto authoritarian regimes enforced by paramilitaries and local guerrilla warlords took hold in some isolated rural areas, fueled by profits from the drug trade Today, Colombia faces the triple challenge of securing peace, restoring robust economic growth, and ensuring that the benefits of growth are shared within Colombian society. Under present circumstances, economic growth of 4 percent over the next decade is required in order to reduce poverty to its 1995 prerecession level. Yet greater gains can and should be achieved by focusing on enhancing equity, efficiency, and security to ensure that the poor, in particular, are able to benefit from restored growth This Thematic Chapter focuses on how to improve equity and effi- 89 90 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE ciency in the use of resources and application of public policy to ensure that the ben- efits of restored growth are optimally applied to reduce poverty and inequality. This Chapter has two objectives. First, it provides a brief diagnosis of the chal- lenges of poverty and inequality facing Colombia. After summarizing Colombia's historical progress and recent shocks, the Note considers three areas critical to improving the welfare of poor: (a) enhancing their levels of income and consump- tion, (b) ensuring their equitable access to basic social services, and (c) reducing their exposure to risk. In each of these areas, the Note identifies key issues and policy options that should be addressed in order to establish an equitable and efficient foundation for shared growth. Second, this Chapter synthesizes the key themes presented in the eight papers on education, the social safety net, health, labor, pensions, housing, the internally dis- placed population, gender, and Indigenous peoples and Afro-Colombians. Those Chapters provide an in-depth diagnosis and explicit policy recommendations in each of those areas. This Chapter concludes by summarizing a short-term and long-term agenda for reform, drawing on the ample agenda for reform in each of these areas, recognizing that priorities must be established in the face of a constrained budget. The Appendix at the end of this Chapter provides a summary in table form. II. Colombia's Tradition of Economic and Social Progress During the 1980s and early 1990s, when much of Latin America was suffering for its "lost decade" of growth, Colombia's strong economic performance allowed it to achieve remarkable success in poverty reduction and social progress. a) Economic Growth Was Instrumental to Poverty Reduction From 1978 to 1995, a period of rapid poverty reduction and social progress in Colombia, economic growth rates averaged over 4 percent, mean income per capita almost doubled, and unemployment remained below 10 percent.' From 1978 to 1995, the percentage of Colombians living below the poverty line fell by 20 per- centage points, from 80 percent to 60 percent. Extreme poverty declined even more rapidly, falling by more than half, from 45 percent to 21 percent. Although rural poverty is much worse than urban poverty, rural and urban areas made similar sub- stantial progress in reducing poverty from 1978 to 1988. However, between 1988 and 1995 rural rates saw a decline of only 1 percentage point in poverty, while poverty in urban areas dropped by almost 4 percentage points (see Table 1). Stable economic growth was instrumental to poverty reduction during the 1980s and early 1990s. Recent economic analysis conducted for the "Colombian Poverty Report" indicates that that economic growth explained, almost totally, the 22-percentage- I Unemployment rates exceeded 10 percent briefly from 1983 to 1985 SHARED GROWTH, POVERTY, AND INEQUALITY 91 Table 1. Poverty and Inequality Indicators, Colombia 1978-99 1978 1988 1995 1999 Poverty indicators National Poverty rate' 80% 65% 60% 64% Poverty Gap2 46% 32% 29% 34% Extreme poverty rate 45% 29% 21% 23% Mean income per capita3 112 183 216 210 Urban Poverty rate 70% 55% 48% 55% Extreme poverty rate 27% 17% 10% 14% Rural Poverty rate 94% 80% 79% 79% Extreme poverty rate 68% 48% 37% 37% Inequality indicators National Gmi Coefficient 0.53 0.54 0 56 0 57 Quintiles (Top/Bottom) 17.2 17.6 17 2 20 2 Inequality Urban-Rural Decomposition (Theil) Total 0.54 0.60 0 70 0.68 Within 0 42 0.50 0 59 0.58 Between 0.12 0 10 0.11 0 09 I Poverty rate Percentage of populationi under poverty line 2 The poverty gap describes the average distance to the poverty line 3 Thousand 1999 pesos per month 4 See other definitions in the appendix Source Colombia Poverty Report (2002) point reduction in urban poverty during this time frame (Velez and others 2002)." Much of this progress was made possible through the steady accumulation of human and physical capital, resulting in high productivity and employment. Moreover, thanks to this solid foundation of economic growth, today Colombia has a great deal more resources to tackle poverty b) Persistent Social Progress Colombia has also achieved clear social development gains during the last two decades. Table 2 shows that in education, completion rates for primary and second- ary schooling showed substantial improvement, reaching 90 percent and 59 percent, respectively. However, school enrollment rates progressed somewhat slowly and suf- fered a small reversal during the recent economic downturn. Illiteracy rates had a similar evolution. In health, life expectancy increased 20 years during the last four decades, with more than proportional gains for females. Equally important, child 92 COLOMBIA. THE ECONOMIC FOUNDATON OF PEACE Table 2. Social Indicators, Urban Colombia' 1978-99 1978 1988 1995 1999 Average education > 18 years 6.2 7.7 8.4 8.9 Illiteracy rate2 5.3% 3.3% 2 8% 2.6% School enrollment Ages 7 to 11 91.8% 94.8% 96.5% 95.3% Ages 12 to 17 76 9% 80.5% 84.4% 82.2% Ages 18 to 22 31 2% 35.8% 41.0% 363% Complete primary school (ages 12 to 17) 67 0% 78.7% 77.7% 89.8% Complete high school (ages 18 to 22) 21.6% 35 3% 48.7% 59 2% Child labor Ages 12 to 16 12.0% 11 5% 9.9% 9.5% Ages 12 to 14 5.8% 5.0% 5 2% 3 7% Child Malnutrition3 Stunting, low height for age 16 8% 10.1% 8.4% 6 7% Wasting, low weight for height 22.4% 16 6% 15 0% 13.5% Low weight for age 4.9% 2.9% 1.4% 0.8% Crime4 Homicides (Per 100,000 pop.) 26 62 65 59* Access to public utilities Electricity NA 99 3% 99 6% 99.4% Water NA 97.4% 97 7% 99 0% Telephone NA 62 2% 71.0% 84.2% Sewerage NA 94 8% 96 0% 97.3% 1 Urban Colombia represents 67 percent of Colombia urban area Barranquilla, Bucaramanga, Bogota, Cali, Manizales, Medellin anid Pasto 2 For population 12 years old & older 3 For population under 5 years old; represents national data for 1977, 1986, 1995, and 2000 4 Levitt and Rubio, 2000 * 1998 figure Sources Authors' calculations based on Encuesta Nacional de Hogares; Profamilia, and Encuesta Nacionial de Demograffa y Salud. malnutrition and infant mortality were significantly reduced. Despite being pro- cyclical-as in other Latin American countries-child labor has had a decreasing trend. Finally, basic infrastructure showed progressive gains. Most of the growth in coverage of water, sewerage, electricity, and telephone results from catch up by cities least covered in 1978 and the progressive extension of coverage to the poorer deciles. Sewerage and telephone connections still lag behind. The missing percentages mask thousands of individuals, concentrated in regional pockets of poverty, where these basic needs are not yet met. c) Generally Propoor Expansion of Social Expenditures in the 1990s Much of the social progress achieved in Colombia can be attributed to progressive social policies applied during the last three decades. Colombia's strong economic SHARED GROwTH, POVERTY, AND INEQUALITY 93 performance allowed for a dramatic expansion of social expenditures during the 1990s, many of which were of substantial benefit to the poor. Driven by mandates in the 1991 Constitution and a series of related legislation, the level of public social expenditure (defined as expenditure on health, education, housing, public services, and social security) doubled in Colombia between 1992 and 1997, raising its share of gross domestic product (GDP) from 8 percent to 15 percent, and its share of overall public expenditure from 30 percent to 35 percent. The new Constitution and accompanying laws also translated into major changes in the composition and administration the health, education, and public service sectors, notably because of the decentralization of power to departmental and municipal levels of government. The result of this expansion of expenditure and introduction of decentralization reforms was a rapid expansion of social service coverage, particularly among the poor. With the exception of childcare and primary education, all services showed much faster rates of expansion among the lower income quintiles (see Table 3). This is particularly striking in the case of health insurance and treatment, and public util- ities. Owing to increases in the underlying target population, the growth in cover- age rates understates the true expansion of services in all cases. A better indication can be obtained by looking at the growth in the total number of people served. While health insurance succeeded in doubling the number of beneficiaries during the five-year period, in the utility sectors, the modest growth in coverage entailed a substantial increase of around 35 percent in the number of household connections. Finally, in the case of childcare, we can appreciate that the drop in the coverage rate was larger than the drop in the number of beneficiaries. Against this generally positive scenario, pension obligations raise a note of con- cern. Social security accounts for the largest share of pubic social expenditures (40 percent), with negligible benefits for the poor. Due to the generosity of its design, the pension system is technically insolvent and government pension subsidies end up being clearly regressive-highly paid individuals receive larger subsidies. Public Table 3. Change in Social Service Coverage Rates, 1992-97 Education ChoLdcarm Health Utzihes Income quntile Primary Secondary Ternary Insurance Treatment Eletr. Water Sewerage Change in Coverage (percentage points) Quintile 1 9 13 3 -5 35 14 9 14 15 Quintile 2 7 7 2 -7 31 0 1 1 3 Quintile 3 9 14 5 -9 28 0 0 -2 -2 Quintle 4 5 12 6 -9 21 0 0 -1 I Quintile 5 12 13 17 -17 15 -3 1 1 -2 Total 8 12 6 -7 27 3 2 3 3 Growth rate of the number of beneficiaries (percent) Total 4 15 99 -4 116 19 35 36 38 94 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE pension payments have been growing at an explosive annual rate of 7 percent during the last decade. According to government estimates, the present value of the opera- tional deficit of pension payments is worth two times the GDP (Acosta 2000). This rate of growth raises serious concerns about the fiscal sustainability of these obliga- tions, placing pension reform as a key policy issue on the agenda for reform. These concerns are compounded by the distributional characteristics of the pension regime for public workers: in addition to subsidizing the nonpoor with taxpayer funds-vio- lating vertical equity principle-the multiple pension regimes for government work- ers discriminate among Colombians-violating horizontal equity principle.2 In addition to these problems of vertical and horizontal inequity, the payments under the current pension system impose a high opportunity cost on the use of pub- lic funds, a high efficiency cost on taxation, and labor market inefficiencies. In terms of the opportunity cost, the public funds demanded for pensions could be used for alternative public investment in human assets with higher economic return, notably education. The taxation and labor market problems are related in that the additional taxes being levied to finance the pension bill raise even further the tax burden on the Colombian economy, and induce labor market distortions. III. Recent Setbacks During the last decade Colombia has faced simultaneous setbacks in three areas: a severe economic recession that erased the poverty gains of the previous decade, ris- ing macroeconomic instability that raised the exposure of households and businesses to economic risk, and increasing violence. The impact of these shocks was exacer- bated by two structural issues: increasing economic inequality and inefficiency in public social expenditures. a) Economic Recession Since 1997, after two decades of positive and sustained growth, economic activity in Colombia plummeted to the point of reaching negative figures in 1999. Until the early 1990s, prudent management of the Colombian economy allowed for low gov- ernment debt levels that, together with low inflation rates by Latin American stan- dards, led to steady-although moderate-growth rates. However, public spending rose from 24 to 36 percent of GDP between 1990 and 1998. Fiscal imbalances helped pushed Colombia toward a slowdown after 1996, and into recession in 1998-99. The fiscal position continued to deteriorate as the economic slowdown adversely affected tax revenues. 2 According to Gomez-Buendia (2002), teacher pensions cost 1.7 times the average cost for an ISS affiliate, police pensions cost 1.9 times, Caja Agraria 2.3 times, ISS employees 3.9 times, and for members of Congress and court magistrates 22 times. SHiARED GROWrH, POVERTY, AND INEQUALITY 95 The recession of the late 1990s reversed the long-term poverty trend previously en)oyed by the Colombian economy, and raised the national poverty rate to 64 per- cent in 1999, a return to its 1988 level, reversing a decade of progress. However, the impact on extreme poverty was less severe, with rates rising to 23 percent, still well below 1988 levels. On the social front, the historical rate of unemployment doubled, and the reces- sion contributed to the deterioration of several key human development indicators that has left some groups in a highly vulnerable situation, notably young children, adolescents, and the internally displaced population. Finally, the recession placed severe fiscal constraints on the government's ability to address those vulnerabilities, and on households' ability to successfully manage the economic impact of the recession. b) Increasing Macroeconomic Instability and Judicial Uncertainty After being considered one of the most stable economies in the region, recent per- formance has shown greater volatility.3 The Colombian business cycle has suffered dramatic change. Cycle duration is now much shorter and much more volatile. Before 1996 the Colombian economy experienced relatively long business cycles of six to eight years in duration. However, the last cycle's duration was less than five years; not only the duration changed, but the amplitude of the last cycle was 50 per- cent larger than the amplitude of cycles experienced between the late 1970s and early 1990s 4 Moreover, according to Partow (2002), volatility of the Colombian economy during the 1990s increased relative to the two previous decades. Volatility nearly doubled for a number of macroeconomic and fiscal variables, with the exception of the terms of trade (Table 4). This went against the Latin American trend. A num- ber of authors-among them Gavin and others (1996), Herrera, Perry, and Quin- tero (1999), and IDB (1997)-have pointed out that domestic volatility in many countries of the region declined during the 1990s. Judicial uncertainty is leading to unpredictability in key policy areas, thus rein- forcing market uncertainty. A number of recent Constitutional Court rulings cre- ated increasing uncertainty about the "rules of the game" in several key economic markets, such as labor, mortgage credit, and the private provision of education. In 3. Rodrik (1999) estimates the probabilities of entering episodes of high volatility for vari- ous countries, and shows that Colombia faced a probability very close to zero for the last 30 years 4 See Echeverry, Santamarfa, and Escobar (2002) According to this article the main cause behind the change suffered by the Colombian business cycle is the Central Bank reform of the early 1990s Countries with monetary policy set by an independent body are sup- posed to experience greater volatility but lower resulting inflation, which is supposed to bring greater economic growth. 96 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Table 4. Colombia!s Volatility 1970s 1980s 1990s 1990s: 1980s* GDP Volatility 1.8 1.5 2 7 81% Fiscal Volatility 1.3 1.4 2 7 91% Exchange Rate Volatility 1.1 1.5 2.4 57% Monetary Volatility 6.8 3.5 3.7 56% Terms of Trade Volatility 13 0 10.4 5 8 -44% '1990s 1980s indicates the change in volatility in the 1990s relative to the 1980s Note Data represent the standard deviation, in percentage terms, of changes in GDP, in the fiscal bal- ance/GDP, in the real exchange rate index, In MI growth, and in the terms of trade index Source Partow (2002). addition, constitutional rulings on pensions rights have cast doubts about the con- tingent liabilities of the public sector.5 As a result, the ability of political mecha- nisms to solve the fiscal crisis has been weakened. For example, it is uncertain whether the pension reforms eventually passed by the elected legislative branch will become a public mandate, since the reforms may be overturned by the Constitu- tional Court. These combined factors led to the recent downgrading of the risk-based ratings given to the Colombian economy, increasing the cost of external borrowing. Finan- cial sector performance has deteriorated and has become more fragile, with signifi- cant bailout costs (Partow 2002; Acosta 2000). c) The Growing Burden of Violence and Crime Whereas most of the social indicators in education, health, and infrastructure show substantial and persistent long-term improvements over the last two decades, the simultaneous escalation of violence over this period has caused a deterioration in liv- ing conditions, imposed a considerable economic burden on the State and house- holds, undermined investor confidence, and presented itself as one of the preemi- nent challenges facing the incoming presidential administration. Empirical evidence indicates that the drug trade and the presence of illegal armed groups linked to this trade are the main factors fueling this rise in violence and crime. Colombia has the highest homicide rate in the world and, except for El Salvador, the highest crime rate in Latin America. Homicide rates almost tripled between the early 1970s and late 1980s, reaching their peak in 1991 when close 1 in every 1,000 Colombians was murdered. Although homicide rates fell in the 1990s, the incidence of extortion, car theft, armed robbery, and other crimes grew dramatically during this period, bringing the rate of victimization to more than 35 percent of households in 1997. 5. See, for example, "Colombia Poverry Report" (2002), Box 4, about mortgage credit. SHARED GROWTH, POVERTY, AND INEQUALITY 97 The poor are more likely to be victims of homicide, and the nonpoor are more likely to be victims of property crime, extortion, and kidnapping (Gaviria and Velez 2001). Domestic violence is also concentrated among the poor, with uneducated women and spouses of uneducated men bearing a disproportionate share of domes- tic violence. In Colombia, young men face an extremely high risk of being mur- dered. According to Velez and others (2002), men age 15 to 35 are 15 times more likely to be homicide victims than women the same age, and rwice as likely as men over age 45. Due to this phenomenon, gender differences in life expectancy have nearly doubled during the last five decades.6 This situation imposes a psychological and economic burden on victims and their relatives, and increases victims' chances of suffering from enduring pathological and dysfunctional behavior. The concentra- tion of these problems among children and youngsters ultimately erodes socioeco- nomic mobility and contributes to the perpetuation of poverty. Due to the escalating civil war an increasing proportion of Colombians have been forced to leave their homes. This internally displaced population (IDP) constitutes one of the most serious social problems in Colombia. The growing numbers of IDP and the limited absorptive capacity of the urban sector, especially in the context of the recession, are placing a growing number of vulnerable groups at increasing risks. The IDP faces major economic losses, tremendous psychological hardship, family disintegration, a loss of assets, and barriers to access to social services. These com- bined vulnerabilities put them at risk of falling into a vicious circle of extreme poverty. The well-being of children of IDP households, present and future, is par- ticularly jeopardized by the problem IDP households have with food security, and by shrinking opportunities for school enrollment in the overburdened receptor municipalities. Colombia's judicial system is stretched to its limits, and in remote areas the rule of law is tenuous at best. Almost every type of crime and violence-political, drug related, domestic, and normal street crime-remains rampant.7 The widespread practice of extortion and kidnapping, which pervades all strata of society, is increas- ingly weakening property rights over physical assets and thereby undermining the market economy. Moreover, de facto paramilitary or guerrilla authoritarian regimes enforced by local warlords effectively rule some rural areas and are making incur- sions into urban areas, as well. d) Rising Income Inequality: Induced by Higher Wage Skill Premia and Eroding Potential Welfare Gains Colombia's high levels of inequality need to be addressed to ensure that the benefits of restored growth are shared equitably. Colombia's income inequality is extreme in the 6 The gender gap in life expectancy reached 8 3 years in 1995 and tends to be larger in more prosperous regions of the country (see "Colombia Poverty Report" 2002, p 32) 7 Homicide rates are roughly 59 per 100,000. 98 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE international context but relatively moderate in Latin America, a region of relatively high income inequality, where it ranks just above the median.' Rising inequality reduced welfare gains made during Colombia's period of strong economic growth up to 1995, and exacerbated the impact of the recent recession on the poor. Average income increased by 88 percent between 1978 and 1995, then leveled off and decreased during the second half of the 1990s. But the Gini Coefficient jumped 7 percentage points between 1978 and 1999, reaching 0.57 in 1999. Had inequality remained constant during the last two decades, the poor would have benefited more from growth. The Sen welfare index estimates income correcting for inequalities by randomly selecting two individuals and assessing the expected welfare of the worse off of the two individuals. Had inequality remained constant, the income of the "second best" Colombian would have been 18 percent higher in 1995, and 23 percent higher in 1999. The two main factors behind changes in income inequality-skill-wage differen- tials and differences in educational attainment-operate in opposite directions. While growing skill-wage differentials between wealthier and poorer Colombians raised income inequality in the 1990s, the increasing equalization of schooling had the opposite effect. Between 1978 and 1999, skill-wage differentials explain nearly 150 percent of the 6.8-percentage-point increase in the urban Gini Coefficient. This negative effect of rising skill-wage differentials dominates the progressive impact of education endowment equalization that accounted for a 6.6-percentage-point reduction in the Gini Coefficient.9 Colombia's high wage-skill premia are illustrated in Figure 1 which shows that, on average, the labor earning of individuals with postsecondary education are 4.3 times those of individuals with incomplete primary education. The Colombian dif- ferential is nearly twice the differential in the United States (2.5), and well above Mexico (3.3) and Brazil (3O7).iO Why are wage skill premia high in Colombia, and contributing factors to income inequality) Is it because the supply of education is insufficient or regressive, or is it because the demand for high-skilled workers has grown above available supply? Edu- cational attainment shows a growing, progressive trend in the last 25 years: Colom- bian school-age cohorts have persistently reached higher educational attainment with less inequity.'1 But this trend has been present mainly in basic education, and 8. Out of 17 countries listed in the World Bank data for 1999, Colombia ranks seventh after Nicaragua, Brazil, Honduras, Bolivia, Paraguay, and Chile (see "Colombia Poverty Report" 2002). 9 See Table A.4 in the Appendix 10. Those differences are even higher when compared to high school graduates. In this case, Colombia leads with 2.4, followed by Brazll (2.0), Mexico (1.9), and the United States (1 6) 11. On average, the number of years of education that the 1975 cohort attained by 1995 was approximately 10 years, 4 years greater than achieved by the cohort born 40 years earlier. Furthermore, the inequality of education within each cohort (measured by the coefficient of variation) fell by more than half during the same period, from 0.75 to 0.33. SHARED GROxrrH, POVERTY, AND INEQUALI VY 99 Figure 1. Labor Earnings by Skill Level, Urban Colombia, 1999 4.5 4.0 - Brazil Mexico - C.Z I* _ 6 25 _-_____ ____ 10 _--- Us a Colombia ~0 5 partlall 1 atbto 4 5 toh 6l s 7upl tof hig9-toiend wokes oitpotsecondary education. The public policy problem underlying income inequality iS Colombia's bottleneck In postsecondary education. Colombian data show that, despite significant improve- ments in equity and effiacincy in finishing all levels of basic education-particularly in finishing secondary school-there is still a bottleneck at the entrance to postsec- ondary education, with increasing frustration for the lower-middle class and the poor. Similarly, while improvements in equity are substantial for high school graduates, they are nonexistent for students beginning their postsecondary education. From this perspective, the substantial Improvements in equity and efficiency in basic and secondary education in the 1990s are a mixed blessing on one hand they have provided an abundant pool of candidates for the potential expansion of post- secondary education, but on the other, these high school graduates face an inequitable bottleneck in accessing postsecondaty education. Those same high school graduates could move readily into postsecondary education if they were offered credit or some demand subsidy. In sum, despite propoor gains in secondary school attainment, access to postsec- ondary education has become a bottleneck and a source of inequality. Public policy should search for the most cost-effircient way of increasing the supply of postsec- 100 COLOMBiA: THE ECONOMIC FOUNDATION OF PEACE ondary education graduates in order to reduce the excessive level of skill wage dif- ferentials. This could be addressed through the provision of education credits for the poor, and demand subsidies for the middle class. Unless this trend is reversed, one would expect an increasing deterioration of inequality in the medium term. During the current period of internal conflict, another emerging important inequality is the socioeconomic bias of the military draft. Estimates from Velez and Riafio (2002) show that Colombian boys and girls do not enjoy equality of oppor- tunities with respect to military service. In Colombia, eligibility for full military service-including combat participation-is reserved for the Colombian young men-not women-without a high school degree. As a result, young men belong- ing to the poorest 20 percent of the population face 11 times more chances of being eligible for the military draft than those belonging to the richest 20 percent. IV. Resulting Vulnerability The combined result of the economic recession, increased macroeconomic instabil- ity, the rise in violence, and growing inequality has seen a marked rise in vulnera- bility. Those most vulnerablc to poverty today include children; young, low- and mid-skilled household heads; recent migrants; and non-homeowners. These groups are clearly worse off than pensioners, the college educated, the elderly, and non- recent migrants. As illustrated in Table 5: o Children under age 18 consistently present higher poverty rates than the gen- eral population, with an increasing proportion of infants and preschoolers in poverty. o Homeownership clearly provides protection against poverty. o Migrants used to fare better or similar to the overall population until 1995, but recent migrants are more likely to fall into poverty in the recession years. This must be partially associated with a shift from voluntary to forced migra- tion, and underscores the high vulnerability of the internally displaced popu- lation due to the armed conflict. o The disabled have remained consistently poorer than the rest of the urban population since 1988, although incidence rates are decreasing for this group. o Contrary to what could be expected, pensioners and the elderly do far better than the rest of the population, and their relative standing seems to be improving over time. Today, as throughout the 1990s, the likelihood of escaping poverty is increasingly dependent upon having fewer children, more working-age family members with postsecondary education, and access to employment for members other than the household head. SHARED GROWTH, POVERTY, AND INEQUALITY 101 Table 5: Poverty Count for Different Subgroups of the Population, Urban Colombia 1978-99 (percent) Share of population 1978 1988 1995 1999 in 1999 Urban Colombia 70 55 48 55 100 Children under 2 yrs 80 71 63 72 3 8 From 2 to 6 yrs 81 70 63 69 10 0 From 7 to 13 yrs 80 70 62 69 13.3 From 14 to 17 yrs 73 61 55 64 7 7 Migrants/Just moved' NA 50 50 64 1 1 Non home owners 77 64 57 63 51 0 Disabled 69 68 60 60 0.6 Migrants < 52 NA 51 46 60 1.2 Women 69 55 48 55 53.4 Migrants < 103 NA 49 43 54 1.7 Home owners 62 46 40 47 49.0 Migrants < 254 NA 50 42 44 3.8 Over 65 years old 52 42 35 37 4.7 Pensioners 37 32 20 24 1.9 1 Migrants/)ustmoved refers to people who have lived less than I percent of their lives at the current city 2 Migrants < s percent refers to people who have lived less than 5 percent of their lives at the current city 3 Migrants < 10 percent. refers to people who have lived less than 10 percent of their lives at the current city 4 Migrants < 25 percent refers to people who have lived less than 25 percent of their lives at the current city. Source Columbia Poverty Report (2002) Three specific groups call for special attention in the crafting of public social pol- icy because of their high levels of vulnerability: children, adolescents, and the inter- nally displaced population. * First, since children are the demographic group most exposed to poverty, pro- grams are needed to prevent irreversible losses of human capital investments that carry with them perverse effects on social mobility. Increasing the cover- age of nutrition, childcare, and preschool education will benefit not only the children at risk but will give many poor families the opportunity to increase the labor force participation of female spouses and improve the odds of escap- ing poverty. The dramatic decline in vaccination coverage rates among chil- dren must also be addressed. * Second, the vulnerability of adolescents needs attention. Young high school graduates face a disproportionate share of unemployment. Labor skill devel- 102 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE opment through technical training and special exemptions from payroll taxa- tion could help improve the employment opportunities and improve their odds of escaping poverty. Moreover, the enormous exposure of young males to homicide, with documented post-traumatic effects on the survivors and witnesses, )ustifies specific preventive action for this population through the application of both judiciary and family welfare policies. Finally, the increas- ing risk of unwanted pregnancy among poor, young females, with negative effects on social mobility, justifies the expansion of reproductive health and counseling programs to this demographic group. Third, the internally displaced population, a result of Colombia's internal con- flict, constitutes another critically vulnerable group. The drug-related economic and military strength of the guerrillas and paramilitaries has led to a dramatic increase in violence, deaths, kidnappings, extortion, and displacement, especially among the rural civilian population. This population has been evicted from areas where they have been engaged in productive economic activities and relocated in urban slums where employment prospects are limited, access to social services is constrained, and violence and crime are rampant. Available data suggest that most of the displaced are women (56 percent) and children (55 percent are under 18), with limited skills and education. In their migration from rural to urban areas, they have abandoned their main asset-their land-and face barriers to accessing jobs and social services. Their concentrations in a few cities in Colom- bia has placed an added strain on limited municipal budgets, a situation com- pounded by a severe lack of financing for Colombia's main national strategy for the internally displaced population during the height of the recession. Finally, looking beyond the vulnerabilities of specific groups to focus on households underscores the critical role of the labor market in determining vulnerability. Those households experiencing the largest increase in poverty during the recent recession were those with only self-employed workers or only nonlabor income, whereas house- holds composed entirely of wage earners experienced the smallest increase in poverty. Of particular note, unemployment of the household head is catastrophic for poverty risks; in urban areas poverty is 25 percentage points higher if the head used to work and is presently unemployed (Velez and others 2002). Paralleling the findings from the survey data, participants in both rural and urban areas in focus groups conducted for the recently completed social safety net study ranked economic insecurity as their prin- cipal source of risk, followed closely by violence (Rawlings and others 2002). V. Addressing the Emerging Challenges: Employment, the Efficient and Equitable Provision of Public Services, and Security To achieve the benefits of shared growth and address present vulnerabilities, Colom- bian public policy should focus on three dimensions of welfare improvement for the SHARED GRowrH, POVERTY, AND INEQUALITY 103 poor: enhancing their levels of income and consumption, ensuring equitable access and efficient provision of basic social services, and reducing their exposure to risk. Public policy instruments can tackle poverty by attacking its source, by attacking its most detrimental consequences, or both: that is, attacking the lack of income opportunities or attacking the lack of access to basic social services and social protection. First, sound macroeconomic management is critical to raising the income of the poor, since short-term income opportunities for the poor are strongly correlated with business cycles and unemployment levels. This makes the welfare of the poor quite sensitive to macroeconomic challenges, including management of the fiscal balance, interest rates, exchange rates, and inflation, through their impact on both the level and the variability of aggregate economic activity. Second, barriers to access to basic social services must be removed for the poor. These barriers usually bring irreversible human capital losses to vulnerable cohorts, especially children. Since children are among the most vulnerable to poverty in Colombia, subsidizing human capital accumulation helps to break the vicious cycle of poverty by putting assets in the hands of the children of the poor, thereby improv- ing social mobility and reducing inequality in the long term. At the same time, human capital investment constitutes a strategic component of long-term growth, reinforcing long-term poverty reduction. Finally, since economic insecurity has shown an increasing trend during the last decade, effective social risk management is a critical component of addressing vul- nerability. Safety assistance in the form of direct transfers in cash, kind, or physi- cal assets are becoming more relevant remedies for transitory lack of income among the poor In addition, social insurance mechanisms that rely on the pool- ing of risk, such as health insurance and pensions, can play an important role in risk management. These three challenges of enhancing the levels of income of the poor through employment, improving equity and efficiency in the provision of public social serv- ices, and reducing exposure of Colombians to risk are discussed below. a) Restore Economic Growth to Accelerate Job Creation and Reverse the Rise in Poverty Recovering high economic growth is a necessary condition for returning to the poverty-reduction path that Colombia followed until the recession of the late 1990s. In order to reduce poverty to its prerecession level of 1995, average annual GDP growth would have to be at least 4 percent during the first decade of this century. Figure 2 shows how from 1978 to 1995 poverty fell at an equal rate, with positive income per capita growth, and reached 48 percent in 1995. It also shows how the recession of the late 1990s brought negative income per capita growth and increased poverty to 55 percent. If the Colombian economy grows at 2 percent-barely above population growth-income per capita growth will be positive but negligible and, 104 COLOMBIA THE ECONoMic FOUNDATION OF PEACE Figure 2. The Poverty Rate Under High and Low Growth Scenarios, 2001-10, Urban Colombia Y 6%- -75% - 5%_ _- - _ - 70% o \<1978-1988 °°4%- A6 _ 5% 1988-1995 u 3% -_-1 - 60% _ 0 o~~~~~~~~~~~~~992 G 2 0%- 1995-1999 l -45% ~~~~~~~~5 ~~ ~ ~ ~ ~ 0 io-1%- ___ ______-____ - 40% < -2%- 35% 1978 1988 1995 1999 2010 Bars represent average annual income per capita growth rate Line represents poverty rate consequendy, poverty will fall by less than I percent during the next decade.i2 How- ever, if economic growth jumps to 4 percent, income per capita will grow at nearly 2.5 percent and, after 10 years, poverty will fall back to its 1995 level. In addition, the expected impact on extreme poverty will be even larger, considering that in the past it has been more sensitive to growth than the "normal" poverty rate. Recovering propoor growth will require addressing both macroeconomic and microeconomic reforms. Poverty-reducing growth will depend crucially on macro- economic reforms to stimulate the recovery of the private sector, the job provider for Colombia's poor households. Achieving this outcome will require fiscal reform to ensure sustainable public expenditures and to recover a favorable investment cli- mate, along with labor market reforms to ensure propoor access to employment. On the fiscal front, pension reform is critical to generating substantial public resources in the medium term without hurting the poor. In addition, real rates of interest must be set at efficient levels. On the labor market front, the elasticity of employ- ment to economic growth must be recovered, especially for low-skilled workers, via recovery of the building sector. This recovery will depend not only on reestablishing a vigorous flow of mortgage credit, but also on labor market policies that avoid inef- ficient payroll taxation and an excessive minimum wage. Attention also needs to be paid to the long-term micro determinants of income per capita growth and poverty reduction, particularly education and fertility, 12 Under the assumption that inequality remains constant SHARED GROWrFT, POVERTY, AND INEQUALITY 105 because these have important effects on labor force participation and productivity The rise in education levels of the labor force and the reduction in the dependency ratios via smaller family size explain most of income per capita growth during the last two decades. Moreover, in Colombia, more educated spouses tend to have fewer children, and in turn those two factors raise their chances of being employed and getting their household out of poverty. Income per capita gains among households with low-skilled heads are also explained by higher real wages.13 b) Improve the Equity and Efficiency of Public Social Services Section II discussed the growth in public social expenditures in the 1990s and the accompanying reforms in pensions, health insurance, and decentralization The evi- dence presented below suggests that Colombia devotes enough resources to public social expenditure relative to the magnitude of the poverty problem, but that these resources are allocated neither efficiently nor equitably. Public policy must address three main problems in the area of social expenditure: (a) pension reform to restore vertical and horizontal inequities and fiscal sustainability, (b) improvement in the cost inefficiencies in the provision of social services, and (c) attention to the equi- table coverage of social public services. i. Colombia's Ability to Tackle Poverty Available evidence demonstrates that thanks to its robust economic performance until 1995, Colombia is a great deal more capable of tackling poverty. Eliminating the extreme poverty gap in Colombia would require less than 3 percent of total household income in 1999 (Figure 3). Moreover, closing the poverty gap-that is, getting 55 percent of Colombians out of poverty-would require 25 percent of total household mcome.`4 Comparing the resources required to close the poverty and extreme poverty gap to the amount of resources devoted to public social policy leads to the unequivocal conclusion that today the Colombian public sector has more than enough resources to tackle the poverty problem, without abandoning other core mandates. In 1999 13 In fact, for the representative urban household, three quarters of the predicted rise in income between 1978 and 1999 is explained by growth of school endowments-32 per- cent-and one quarter by redLictions in the adult-to-family-size ratio-9 3 percent However, until 1995-before the recession-positive changes in employment ratios and wages helped to increase income per capita and made a contributioll to poverty reduc- tion, as well (see "Colombia Poverty Report" 2002, chapter III) 14 From 1978 until 1995, both economic growth and the reduction of the poverty gap- from 46 percent to 29 percent-reduced the share of total household income necessary to close the poverty gap from 58 percent to 20 percent, just a quarter of its original value. Similarly, the share of household income necessary to close the extreme poverty gap fol- lowed a similar trend 106 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE Figure 3. Share of Total Household Income Required to Close Poverty and Extreme Poverty Gaps 60%- 12% Poverty 50% \- \ _10% D 40%-- -- __ _ _ _ __ 8% & C o 30%- 6__ N _ _ ___6%_E u 20%-- ___ _ _4% U H Extreme Povert 0u 10% _- ________ 2% H 0%_ I I I _0% 1978 1988 1995 1999 Share 1978 1988 1995 1999 To close the extreme poverty gap 10.7% 4.3% 2.1% 2.5% To close the poverty gap 58% 25% 20% 25% public social expenditure was 15 percent of GDP, while total public expenditure was approximately three times larger. In summary, this evidence shows that the realloca- tion of 10 percent of public expenditure-a moderate amount-would have a major impact on the welfare of the poor. ii. Addressing Inefficiency in the Provision of Public Social Services Gains in the coverage of public social services achieved in the 1990s were not com- mensurate with the increase in expenditures. Table 6 shows changes in public expenditures, coverage rates, the total number of beneficiaries, and the public sec- tor share of coverage. By subtracting the growth rate in the numbers of people served from the growth rate in real expenditure, it is possible to make some indi- rect inferences about the growth in unit costs. This calculation reveals that in most cases the unit cost of services broadly doubled during 1992-97. The most severe problem seems to be in the area of health insurance and treatment, for which fund- ing increased by over 300 percent, but individuals insured or treated increased in much smaller proportions. Unless quality improvement was colossal, this seems to imply that despite the gains in the number of individuals treated (19 percent), much of the growth of expenditures registered over this period was absorbed by cost-inefficiency. SHARED GROWrTH, POVERTY, AND INEQUALITY 107 Table 6. Coverage and Expenditure for Social Services, Colombia 1992-97 (growth) Public Numbers served Real expenditure Coverage rate sector share Childcare -4% - -7% 2% Education Primary 4% 114% 8% 0% Secondary 15% 108% 12% 8% Tertnary 99% 164% 6% -8% Healthcare Insurance 116% 326%* 27% -25% Treatment 19% 6% - Utinties Electncity 35% - 2% - Water 36% 194% 3% - Sewerage 38% 194% 3% - 'Note Real expenditure in healthcare calculations are based on data from the National Deparrament of Planning that do not provide information about how total healthcare expenditure is allocated between insurance and treatment Table 7. Coverage Shortfalls and Targeting in Selected Social Programs Education Healthcare' Utilities Users Prnnary Secondary Tertiary C/ildcar Insurance Treratnent Water Sewerage Electr. Numbers (mill) Potential users 6.2 4 9 1.8 4 4 39.6 6.1 39 8 39.8 39 8 Covered 5.3 3.7 0.8 1 2 23.3 4.5 33 4 27.0 37.2 Not covered 0.9 1.2 1 1 3 2 16.3 1.6 6.4 12 8 2.7 Concentration Coefficient Potential users -0 19 -0.05 0.28 -0 19 -0.05 -0.02 0.01 0.01 0.01 Covered -0 16 -0.04 0.45 -0.07 0.05 0.02 0.05 0 11 0 03 Not covered -0 34 -0.07 0 14 -0.23 -0.21 -0.12 -0.26 -0 25 -0 3 Note A more negative concentration coefficient means more that more thani proportional benefits go to the poor * Figures for 1997 Source Columbia Poverty Report (2002) 108 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE iii. Addressing Equity in the Provision of Public Social Services There are still substantial shortfalls in the coverage of social programs concentrated among the poor (Table 7). Overall, 66 percent of the corresponding age group lacks access to institutional childcare and 15 percent to primary school. Of those who complete primary school, 25 percent do not go on to secondary education, while of those completing secondary school, 57 percent do not go on to university. Further- more, 41 percent of the population is still not covered by health insurance, and 57 percent lacks connection to the sewerage network. The absolute size of the unserved population varies substantially across programs. Those who are not covered by health insurance represent the largest unserved group, amounting to 16.3 million people. The next-largest group is composed of those living in households without a sewerage connection, which accounts for 12.8 million people, while 6.4 million people live in households without access to water. It is no surprise to find that the population without access to public services is disproportionately concentrated among the lower-income deciles. For example, in the case of primary education, 19 percent of the unserved population fall into the first decile. An interesting exception is found in tertiary education, where the largest shortfalls in coverage are in the middle classes. Consequently, the most neg- ative concentration coefficients correspond to primary education, public utilities, and childcare. Today, the highest priorities for social service expansion among the poor are childcare, sewerage, and healthcare and health insurance. Among all basic social services, those four sectors have the largest relative access gaps between the poor and the better off.i5 iv. Instituting Pension Reform to Restore Equity and Fiscal Sustainability The public pension system is insolvent and pension subsidies (payments beyond contributions) disproportionately benefit the nonpoor. Reform is urgently needed to recover fiscal sustainability and reverse the offensive inequities imbedded in the system In relation to equity, the main principles of this reform should be to (a) unify the pension regimes to established horizontal equity-any two Colombians with the same productivity and age should receive equal treatment; (b) homogenize the conditions of accumulation of future benefits, accelerating the transition peri- ods; (c) adopt measures to limit the subsidy of retirement account of lower-income workers; and (d) impose contributions to persistent extraordinary benefits. 15 Relative to households in the sLxth income decile (that is, for the poorest 50 percent of households), the gaps in coverage for childcare, sewerage, and healthcare and health insur- ance are 20, 19, 17, and 10 percent, respectively Those rankings are relatively robust. Two sectors show similar ranking independent of the comparator income group used to calcu- late the coverage gap of the poor. childcare is almost always the highest priority service for expansion, while sewerage ranks as the second or third priority for expansion. SHARED GROWTH, POVERTY, AND INEQUALITY 109 c) Establish a Functional Social Safety Net A third major challenge facing Colombia is the establishment of a functional social safety net allowing Colombians to better manage their exposure to risk. This social safety net should take the form of a social risk-management system functioning in both normal and crisis periods, calling on both social assistance and social insurance components. This reformed safety net should aim to provide benefits to the chronic poor in extreme poverty, special vulnerable groups such as the internally displaced popula- tion, and those affected by idiosyncratic shocks such as poor health, unemployment, or disability. The safety net should also contemplate a countercyclical mechanism to ensure that assistance is available to those harmed by large covariate shocks (shocks that affect many people simultaneously) such as economic crises and natural disas- ters. A focus on an effective countercyclical mechanism that could be activated dur- ing times of economic crisis is critical to reversing historically procyclical patterns of social assistance spending.16 i. Colombia's Social Safety Net Historically, economic growth and the parallel expansion of basic social services served as a substitute for implementing a social safety net. Notably, social assistance was not included in the dramatic social sector reforms of the 1990s that resulted in increased spending and decentralization for health and education. Consequently, social assistance programs remain not only ill-suited to respond to large covariate shocks, but also underfunded, centralized, and lacking a strategic focus. In regard to the financing of social assistance, for example, whereas funding for health and edu- cation rose dramatically during 1991-96, central-government-budgeted expendi- tures on social assistance fluctuated around 1 percent of GDP during the first part of the 1990s, and fell to less than 0.7 percent by 2000.17 This is a very low level of social assistance spending compared to countries at a similar stage of development,18 16. A retrospective public social expenditure review reveals that for each peso of reduced GDP, social assistance spending fell by nine pesos, making social assistance the most pro- cyclical component of social sector spending (CRECE 2001). 17. Social assistance is defined as spending on the three main social assistance programs reported in aggregated budgetary data-the Colombian Institute for Family Welfare (ICBF), the Social Solidarity Network (RSS), and the National Institute of Social Inter- est, Housing, and Urban Reform (INURBE). The expenditure estimate does not include spending on the new temporary Red de Apoyo Social (RAS) conditional cash transfer, workfare, and youth training programs, nor does it take into account the substantial sub- national spending on social assistance, including utilities subsidies and the Cajas de Com- pensact6n programs 18 Each of the nine Latin America and Caribbean region countries included in a recent social protection expenditure review devotes a higher share of GDP to social assistance, including Argentina (0 9 percent), Mexico (1 1 percent), Peru (1.4 percent), Uruguay (3.4 percent), and Venezuela (I percent) (Dulitsky, Gragnaloti, and Lindert 2001). 110 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE and compared to the needs of specific vulnerable groups that lack access to key social assistance programs In sum, although substantial progress was made in health insur- ance for the poor, Colombia remains without a strategic social risk-management sys- tem, and social assistance programs remain fragmented, diffuse, and largely outside of the reach of those in the informal sector. ii. Priorities for Reform As in the other sectors, the program for strengthening social protection must address the issue of efficiency because of the country's tight financial constraints. But it is imperative for the government to find additional funding for social assis- tance because of its serious underfunding and because of the special attention required today by these vulnerable groups: children, adolescents, internally dis- placed people, and poor workers facing protracted unemployment. Drawing from the Social Safety Net and Pension Policy Chapters, the following measures should be considered in establishing a strongly propoor and efficient social safety net in Colombia: * Establish a countercyclical element of the safety net with clear saving and spend- ing rules that would allow Colombia to provide assistance to those most affected by the next large shock. Part of this process involves evaluating the new Red de Apoyo Social programs and integrating them into a coherent, reformed social safety net. If these programs are not assessed as part of a broader social safety net reform, they risk becoming another institutional layer within the diffuse collection of social assistance programs and a drain on financing for public social expenditures. * Expand the budgetfor social assistance to at least 1 percent of GDP through the redirection of funds to well-targeted social assistance programs and the increased use of poverty targeting. * Focus on streamlining the existing safety net system, not creating additional pro- grams, perhaps drawing on principles of social insurance. At the very least, reforms should focus on working within the existing system to improve poverty targeting, eliminate excessive program overlap and fragmentation, and restructure existing programs to improve coverage and quality. A more radical approach would develop a wholesale alternative to the existing system by replacing existing programs with a new system drawing on social insurance principles, as was done in the health system. What should clearly be avoided is the creation of new programs as add-ons to the existing system. * Address the needs of the Internally Displaced Population (IDP). Colombia needs to establish a strategy that would effectively address the IDP needs, starting with the dismantling of barriers to access preventing the IDP from benefiting from critical social assistance, health, and education programs. Such a strategy should also address security and stigmatization concerns regarding the use of an IDP registry. SHARED GROWTH, POVERTY, AND INEQUALITY 111 VI. Conclusion This Chapter has shown that for almost two decades Colombia made remarkable progress in reducing poverty and advancing the living standards of millions of citi- zens. It also underscored the serious setbacks experienced during the past decade the recession of the late 1990s reversed a decade of progress in poverty reduction, increased levels of violence affected the poor and nonpoor alike, and growing inse- curity left Colombians exposed to new types of risk. In addition, increasing income inequality in the 1990s exacerbated the negative impact of recession on poverty, and inefficiency and inequity in the provision of basic social services diminished the potential impact of these investments, particularly on the poor. These shocks and structural deficiencies ushered in new problems and exacerbated older ones, with the combined impact still apparent in the rise in unemployment, the protracted increase in poverty, the erosion of the social fabric, and the rise in vulnerability. To address those emerging challenges, public policy must concentrate on improv- ing three dimensions of welfare of the poor: raising their levels of income, ensuring equitable and efficlent provision of basic social services, and reducing their exposure to risk a) Raising the Income of the Poor Poverty-the lack of income opportunities-is derived from lack of assets and/or the inability to find markets for those assets. Hence, there are two types of poverty- reducing policies: policies that try to attack the source of the problem-lack of income-and policies that try to eliminate the most undesirable and intractable consequences of poverty-lack of access to basic social services and lack of social protection. Developing a strategy to attack those three dimensions of welfare for the poor involves selecting a range of policy options, with attendanit tradeoffs for gen- eratmig benefits at different points in time. In the short term, the easiest way to raise income of the poor is by achieving higher rates of economic growth and lower unemployment, which in turn increase the market demand for their human assets. Recovering higher sustainable rates of growth requires consistent macroeconomic policy and restoring security. An alter- native way of raising income in the short term is through direct cash transfers or sub- sidized access to nonhuman assets, such as housing. Increasing the wages and the demand for less-skilled workers can also be achieved though stimulating growth in low-skilled, labor-intensive sectors such as construction. Over the medium term, poverty can be reduced by raising the labor force partic- ipation of working-age adults in poor households, typically by incorporating female spouses into the workforce. This can be accomplished through different policy mechanisms, including increasing access to childcare and preschool services and lowering fertility rates through access to reproductive services. A more roundabout way to raise the income of the poor is by lowering the dependency ratio (that is, rais- 112 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE ing the adult-to-family-size ratio), which is typically achieved through better female education and easier access to reproductive health. Finally, the most effective long-term strategy for reducing poverty is to raise human capital endowments. This is generally achieved through providing educa- tion, which generates positive intergenerational effects of higher social mobility. b) Ensuring the Equitable and Efficient Provision of Basic Social Servzces Equitable access to basic social services is a crucial step in breaking the vicious cycle of poverty, and putting assets into the hands of the children of the poor. Improving the efficiency and equity of social public expenditures is critical to elevating the human capital accumulation for the poor. Therefore, the public provision of social services and subsidization of human capital accumulation can lead to improvements in social mobility and reductions in inequality and poverty in the long term. c) Reducing Exposure to Risk The persistent features of vulnerability provide an opportunity to target both social sector and social assistance programs to specific demographic groups. The "faces" of the poor are children of all ages, young low-to-middle-skilled household heads, recent migrants, and non-homeowners. These groups are clearly worse off than pensioners, the college educated, the elderly, and non-recent migrants related to displaced population. Moreover, certain low-frequency events may throw households into poverty, such as the presence of a disabled family member in the household, the loss of employment for the head, or being part of the internally displaced population. These phenomena require special attention through the mobilization of social insurance mechanisms that rely on the pooling of risk, or through programs to address groups like the internally displaced that encounter particular vulnerabilities. The presence of these vulnerabilities reflects Colombia's need for a functional social safety net to allow for better management of the increasing levels of risk. This social safety net should incorporate three components: (a) social assistance for the chronic poor and the internally displaced population, (b) a countercyclical compo- nent to alleviate covariate risk, and (c) social insurance to attend idiosyncratic risk. Priorities for action include expanding the budget for social assistance to the extent possible given the tight financial constraints; reforming the pension system to make it sustainable, and reducing the subsidization of well-off workers; and streamlining the existing safety net, and existing social assistance programs, perhaps drawing on principles of social insurance, while avoiding the creation of add-on programs. In summary, these propoor public policy options call for finding markets for assets of the poor and providing social assistance to the most vulnerable in the short term, and putting assets in the hands of the poor while permanently protecting those assets through social protection in the long term. SHARED GROWTH, POVERTY, AND INEQUALITY 113 Appendix Table A. 1. Summary of Issues, Objectives, and Recommendations by Sectors Eduncation Issues Policy Objectives Recommendations * About 12 percent of * Eliminate the educa- I Establish a stronger linkage between children age sLx to 11 tion inequities in the achievement of desired outputs and are still out of school, long term financing through with nearly all of * Good basic education * Sustained, if not accelerated, imple- these excluded chil- of allyoung children, mentation of the Law 715 formula dren belonging to the in particular the allocating education transfers to ter- lowest income achievement of uni- ritorial entities according to the num- groups versal completion of ber of students, population to be * The education gap primary education in attended, and poverty indicators. between the richest the next 4 years. * Encourage the territorial entities to income decale and the * Continuous improve- adopt policies that support schemes poorest decile is ments in the learning that link resources with desired out- roughly 7 years, and achievement of poor puts, e.g, PACES, Bogoti's con- increases over the children in basic edu- cession system (although, adapted to years cation, including en- local circumstances). * The gap against the hancement of their 2. Ensure. poor is wider at the ability to learn actively. * An explicit system of identifying secondary and terti- * Selective expansion of the poor and disadvantaged chil- ary levels. pre-and postbasic dren who merit special assistance * The education gap is education, focusing in education. exacerbated by differ- on the enrollment of * A credible program that makes sure ences m the learning poor children in that disadvantaged children get effec- achievement of poor effective schools. tive and efficient assistance, includ- and better-off chil- ing both supply- and demand- based dren in school. assistance; e.g., Famulias en accidn. 3. Improve the accountability and oversight of the performance of the Ministry of Education, De- partments, Municipalities, and schools in regard to the universal- ization of basic education comple- tion and the learming achievement of disadvantaged students through- Regular production and public dis- semination of a Report Card on the performance of these educational agencies, with special attention to the education of the poor (continues on next page) 114 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Table A.1. (continued) Education Issues Policy Objeetives Recommendations • Establishment of an independent body that would implement the Report Card system, consisting of eminent people from the business, teaching, and research communi- ties, and NGOs the work of which is focused on poverty and human development o Mobilization of poor parents to enable their effective participation in the development and implementa- tion of a propoor educational agenda and enlist the support of civil soci- ety and teachers sympathetic to the cause of poverty reduction and edu- cational equity 4. Strengthen the Ministry of Educa- tion's capacity to support the terri- torial entities to acquire the techni- cal capacity and information needed to increase their ability to reach the out-of-school children and improve the learning achievement of poor and disadvantaged chlldren through strategies such as Telesecundaria, Escuela Nueva, Accelerac7dn deApren- dazaje, and SAT (Tutorial Learning System). Health issues Policy Objectives Recommendations Passage of Law 100 Improve the health 1. Continued expansion of health in December 1993 status of the poor and insurance coverage of nonaffiliated made the system reduce their financial population including the poor and more equitable and risks from illness through: the self-employed through accessible to the poor o Achievement of uni- Full implementation of the health and vulnerable, com- versal health insur- reform law (Law 715) mandating pared to the begin- ance coverage transformation of supply to demand ning of the 1990s. subsidies for healthcare SHARED GROwTH, POVERTY, AND INEQUALITY 115 Table A.1. (continued) Health Issues Policy Objectives Recommendations However, implemen- * Strengthening of the * Reduction in the evasion and tation of the reform control of commu- underpayment of health insurance remains incomplete, nicable diseases contributions. (about 32 percent of * Exclusive use of the FOSYGA Sol- the poor are not cov- idarity Account resources for sub- ered by the health sidizing health insurance coverage. insurance system) * An improved communications For various reasons program informing the public in including inadequate general and the poor in particular financing and lack of about the affiliation process, eligi- clarity in the assign- bility rights, and allocation of ment of responsibility resources to territorial entities for for immunization, demand and supply subsidies. control of communi- 2. Full implementation of Law 715 cable diseases has mandate on quality assurance, deteriorated with de- promotion of public accountabil- centralization and iry of insurers and healthcare health reform providers with the support of a strengthened Health Superinten- dency, and development of citizen oversight (Veeduria Csudadana) of the use of public health resources 3. Optimization of the financial flow of health sector resources, includ- ing prevention of illegal use, reten- tion, or use for other purposes. 4 Restructuring the network of pub- lic hospitals in order to operate as efficient providers of healthcare services with limited direct govern- ment subsidies and adoption of a new healthcare model focusing on primary healthcare. 5. Full implementation of the meas- ures to strengthen the Expanded Program of Immunization based on the recommendations of the above- mentioned National Evaluation. (continues on next page) 116 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Table A. 1. (continued) Safety Net and Pensions Issues Poliey Objectives Recommendations * The social sector pol- Establishment of a * Expand the Budget for Social Assis- icy reforms intro- reformed Social Risk tance to at least I percent of GDP duced in the 1990s Management System * Reform the pension system guar- by the 1991 Con- to (a) improve the anteeing that (a) new generations stitution, Law 60, state of readiness of do not create new liabilities, (b) lia- and Law 100 did Colombia to assist the bilities from intermediate genera- not include social vulnerable poor dur- tions are cut, (c) unification ofpen- assistance. ing times of crisis; sion regimes (entry only into ISS * The public pension and (b) strengthen and private managed funds), and system is technically social assistance and (d) homogenization of conditions insolvent (net value insurance to chroni- for accumulation offuture benefits. of pensions in 2000 cally vulnerable poor * To develop the social risk-manage- amounted 200 per- during normal times; ment system, focus on streamlin- cent of GDP) and Improvement of the ing the existing system, not creating constitutes a ma)or quality and coverage additional programs risk to fiscal sustain- of early childhood * Establish a countercyclical element ability over the meds- development services, of the safery net. um and long term. and * Address the needs of the IDP * Causes of insolvency: - More effective assis- generous pension tance to the internally benefit guarantees, displaced population perverse incentives for (IDP). reserve management, * Special attention to and recent Constitu- the following groups tional Court rulings needs to be given: expanding pension children, adolescents, benefits. the elderly, the IDP, * Colombia's pension and the poor workers system constitutes a who have been serious drag on the unemployed due to ability of the govern- lack of employment ment to finance pro- opportunities poor social assistance * Reform the pension and other programs. system to make it sus- tamable and to reduce public subsidy of well-off workers SHARED GROwwH, POVLRTY, AND INEQUALITY 117 Table A. 1. (continued) Labor Market Issues Policy Objectives Recommendations • The labor market in * Labor market reform * Rationalization of payroll taxes (sub- Colombia is charac- to stimulate employ- stitutnng some of them with more terized by being rigid: ment growth, provide efficient taxes) real wage rigidities a more favorable envi- * Reduction of transaction costs for (minimum wages and ronment for acceler- "being formal." unions) The impact ating investments in * Elimination of the rigidities imposed of these rigidities is human capital, facil- by the minimum wage including, disproportionately itate movement of among others, allowing emergency negative on the poor, workers from low to public workfare programs for the rhe young, and high productivity poor to pay less-than-minimum- women activities, and pro- wage salary and exempting them * High unemployment mote equity from payment of obligatory fringe rates. average annual * Labor reform should benefits (prestaczones) so that they unemployment rate have in mind that can become self-targeting and reach during the last 10 payroll taxes (a) more poor, unemployed workers. years has been about finance major social Development of self-financing 14 percent assistance programs workers' income insurance scheme * The labor market in that benefit the poor to protect workers against spells of Colombia is highly (e.g, ICBF and social unemployment to replace costly job segmented-infor- health insurance), (b) security regulations mality close to 60% that they are a source * Acceleration of investment in the of the workforce. of funding for work- human capital of poor workers, * The labor market ers' training, and (c) including a reform of the Servzc:o presents low degree that job security reg- Nacional de Aprendizaje (SENA), of turnover between ulations protect (al- separating the finance and provi- the formal and infor- though, at high cost) sion of training programs based on mal sector There are currently employed workers' choice and competition low opportunity costs workers against un- involving botlI public and private of self-employment employment and the providers (recent data indicate that because of the low consequent loss of the impact on workers' salaries is productivity of the income. considerably greater for private than formal sector. public training) Improving the efficiency of pro- grams (e g, ICBF, SENA, and Social Health Insurance) funded by payroll taxes so that the actual value of the benefits from those programs is commensurate with the cost of the obligatory contributions (continues on next page) 118 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Table A. 1. (continued) Housing Issues Policy Objectives Recommendations Since 1990, commer- o Restore mortgage The government should urgently take cial banks were al- portfolio, expand measures to protect mortgage holders lowed to compete in mortgage stock as well against predatory lending behaviors the construction and as new productive and restore a sounder balance among mortgage markets, loans mainly through judges, lenders, and borrowers Actions which were domi- increasing housing to be considered include the following: nated by Corpora- demand. 1. Revise the system of a capped inter- ciones deAhorro y o Implement a legal and est rate for mortgage loans. The revi- V¼vuenda (CAV). regulatory framework sion should be consistent with the * The liberalization in- that restores housing Constitutional provisions set up to creased the cost of as collateral for the avoid predatory practices. CAV funds because lending institunons 2. Limit debt restructuring to cases of UPAC-savings be- Restore housing own- an unpredictable loss of borrower came less attractive. ership as a protection income followed by some credit The UPAC formula against poverty and solvency has changed several as a worthy form of 3. Fully inform mortgage consumers times to reflect that collateral. about new regulations regarding reality, introducing their rights and duties. instability into the 4. Adjust the Unidad de Valor Real housing market. (UVR) to reduce its role as a source • There was a boom in of volatility. poorly designed credit 5 Improve the overall structure of products between housing credits: 1995 and 1997, - Housing appraisal rules could be when the level of better adjusted to comparative mar- nonperforming loans ket trends was rising and hous- - The credit reporting system should ing prices started to not be limited to current records decline for upper seg- - Reduce delays in property ments. The deterio- registration. ration of the portfolio 6. FOGAFIN should consider was evident before strengthening the collateral require- the interest rate shock ments under the recapitalization of 1998. line by requiring collateral with lower-risk assets. 7. The Superintendency of Securities needs to expand its staffing and training by oversight inspections, detailing its regulations, and imple- menting prompt corrective actions. SHARED GROWrH, POVERTY, AND INEQUALITY 119 Table A.1. (continued) Indigenous Peoples and Afro-Colombian Communities Issues Policy Objectives Recommendations • The Afro-Colombian * Increase public invest- * Provision of adequate basic services, population has faced ment for indigenous particularly health, education, and historical deficit of peoples. services related to production public services such * Improve healthcare * Consolidation and demarcation of as health, education, services for these indigenous homelands and the com- and support services communities munal lands of Afro-Colombians for production * Improve quality of * Design of strategies for the protec- * The African-Colom- educational services, tion and conservation of special bian population is especially supply of ecosystems that contribute to the showing an accelera- well-trained teachers strengthening of territories and tra- tion of the migration in these communities. ditional systems of knowledge, con- process explained by servation, and sustainable use of the internal conflict biodiversity (Uraba, low and medium Atrato) and the violent process associated with the expansion of illegal crops * There is pressure on the indigenous home- lands and collective territories of Afro- Colombian commu- nities because of the colonization of farm- ers-especially those dedicated to illicit crops-and the inter- nal armed conflict. (continues on nextpage) 120 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Table A. 1. (continued) Forced Internally Displaced Popuiation (IDP) Issues Policy Objectives Recommendations Forced displacement o Make the problem of * Remove access barriers for the par- in Colombia is main- forced displacement ticipation of the IDP in the poverty- ly the result of the a priority targeted social programs including social and armed con- e Reduce displacement application of the Sistema de Selec- flict. It is the most by implementing ef- cion de Benefictanos (SISBEN) ques- serious humanitarian fective preventive tionnaire and issuance of the and social problem in measures at the local SISBEN cards through a plan Colombia today. level. designed jointly with the municipal - Adopt a common governments methodology among * Empower the IDP at the national key stakeholders for and local level. estimating the size Focus on regions, municipalities, and characteristics of and localties at high risk, and within the IDP those on the groups at high risk of - Create realistic incen- being threatened. tives and opportuni- * Increase incentives to resist dis- ties for the reestab- placement by ensuring protection of lishment-return, property rights on assets relocation, and inte- * Promote direct dialogue among local gration-of the IDP. authorities, organized citizens, and * Ensure access of IDP armed groups to demand respect for children to public human rights of citizens and to education. exclude from armed conflict areas such as hospitals, cultural centers, and homes. * Conduct a national survey based on a probability sample of the entire population displaced. * Establish regionally focused social funds as instruments that promote self-reliance in reestablishment proj- ects oriented to community action to secure or recover assets, commu- nity-driven reintegration and devel- opment activities, and social and economic stabilization of particu- larly vulnerable groups, such as female heads of household and youth (male and female) in those focused regions, municipalities, and localities. * Promote regional agricultural ini- tiatives that offer opportunities for the IDP to undertake productive activities. SHARED GRowrTT, PovERTY, AND INEQUALITY 121 Table A.1. (continued) Gender Issues Policy Objectives Recommendations * Colombia has made * Gender indicators I Crime and Violence significant progress in should continue to be * Retrain teachers to ensure they do reducing gender in- monitored to ensure not promote violent behavior among equalites-particu- that gains for women boys and submission among girls larly in improving the in education, literacy, Eliminate gender stereotypes in well-being of women health, and employ- textbooks and other pedagogical * While the status of ment are not eroded materials females has improved, by the continuing cn- - Develop innovative programs to male human capital sis in Colombia. teach children nonviolent resolu- has eroded on the * The approach to tion skills and promote civic values. health and education gender needs to be * Address gender issues in programs fronts. broadened from a aimed at reducing homicide rates, * Important interlink- conventional focus on such as Convivencia Ciudadana. ages exist between women to address the 2. Reproductive and Sexual Health gender, poverty, and negative dimensions * Develop integrated programs that inequality in Colom- of gender for men. address employment, violence, sex- bia Female-headed * Given the immediate Lality, and sexually transiiltted dis- households are more needs resulting from eases (STDs) and AIDS likely to be poor and the crisis, it is impor- * Use youth workers to reach other vulnerable than those rant to consider gen- youth, and men to reach men. headed by males. der in social safety net * Use participatory learning in the programs for the form of role-playing and assertive- poorest and most vul- ness exercises rather than traditional nerable-in particu- didactic training methods lar single-parent fe- * Address causal factors related to ado- male households. lescent sexual activity, including self- esteem, decisionmaking, gender roles, and valies. * Target youth and men where they normally gather 3 Support to Female Single-Parent Families * Increase poor women's access to childcare and family planning, con- tinue improvements in education, and reduce barriers to poor women's participation in the labor market Table A.2. Summary of Policy Handles that Impact Welfare in Three Dimensions: Income, Access to Social Service, and Risk Management Objective Short Term Medium and Long Term 1. Raise employment 1.1 Recover high positive economic growth. 1.5 Increase female participation by facilitating 1 2 Increase female participation by providing access to birth control mechanisms to reduce childcare and preschool services, and reducing fertility rates and by improving educational urban transport costs. endowments. 1.3 Reduce labor market rigidities such as minimum wages and costs of labor contracts. Labor 1.4 Reform temporary public employment Income programs and privilege household heads. 2. Raise productivity by g Increase 2 3 Workfare and youth training programs such as 2.1 Increase productivity by reducing education ° Income Jovenes en Acc:dn for youth and Manos a la inequities (access) and improving quality of Obra. education. 2.4 Provide services related to production for 2.2 Promote technical postsecondary education, I Indigenous and Afro-Colombian communities private sector and vocational training, and t internship and apprenticeship programs. 0 z 0 0 z 0 Table A.2. (continued) Objective Short Term Medium and Long Term 1. Promoting owsnership of assets I I Prevent predatory lending on mortgage credit. |1.2 Housing acquisition and housing subsidies. 0 2. In-kind or cash transfers such as: Nonlabor 2 1 Targeted subsidies via block tariff pricing of Income utilities. 2 2 Programs such as the PACES vouchers to allow poor students to pursue secondary education. 1. Provide basic education in poor neighborhoods 6. Improve water supply and sanitation in medium 2. Develop demand-side financing mechanisms in cities, small municipalities, and in the rural areas the form of conditional cash transfers to attract 7. Promote initiatives for adequate monitoring, More Equitable Access to and retain children in school. management, and disposal of wastewater Social Services 3 Provide healthcare for the poor. 8. Recover cost from nonpoor group beneficiaries of 4. Provide adequate basic services, particularly public social programs health and education, for Indigenous and Afro- Colombian communities. 5. Provide scholarships and educational credit for postsecondary education (continues on next page) Table A.2. (continued) Objective Short Term Medium and Long Term 1. Self-insuranec 1I.1 Promote housing ownership and real asset accumulation 1 2 Protect human capital assets (nutrition programs for the poor). 1.3 Reduce unwanted pregnancy rates, especially among teenagers. 2. Market insurance: 2.1 Implement policies that seek to expand () employment and increase productivity in the informal sector. o Economic Risk 2.2 Reduce costs of "being" formal (registration process and labor regulations) 3. Safety-net socialprotection m 3 1 Create alternatives for unemployment support 3.5 Establish a countercyclical element of safety net. ° Management systems (insurance scheme avoiding perverse 3 6 Expand the budget for social assistance to at 0 of Risk incentives). least I percent of GDP by redirecting funds to n 3.2 Create health insurance for the unemployed. well-targeted social assistance programs. o 3.3 Develop protection mechanisms for the 3.7 Expand minimum pension schemes. c informal sector. 3.8 Create assistance prograrms for the elderly 3.4 Subsidize healthcare and health insurance. (1% of payroll contribution that today subsidize o contributions for the poor) z > Table A.2. (continued) Objective Short Term Medium and Long Term 1. Vulnerability to violence and crime 1.1 Protect households, especially poor ones with 1 4 Protect the rights of common citizens to life low educational endowments, from intrafamilial and against insecurity, homicide, rape, fights, violence gangs, loitering, threats, and guerrilla and Other Risk 1 2 Reduce exposure to homicide risk, especially of paramilitary presence the poor, and the young male population. 1.5 Protect property rights. 0 1.3 Improve securty of employment generators z such as small businesses owners in poor neighborhoods 2. Internally DispLaced Population (IDP): 2.1 Facilitate access of IDP to public subsidies and Reduce rate of displacement by in-kind services by application of the SISBEN 2 5 Focusing on regions, municipalities, and questionnaire and issuance of the SISBEN cards localities at high risk and within high-risk 2.2 Empower IDP at the national and local groups governments 2.6 Increasing incentives to resist displacement by 2 3 Adopt a common methodology among ensuring protection of property rights on assets. stakeholders for estimating the size and 2 7 Promoting dialogue among local authorities, characteristics of the IDP. armed groups, and citizens to guarantee respect 2 4 Create incentives and economic opportunities for human rights of citizens for the return, relocation, and integration of the IDP. 126 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE Table A.3. Definitions Associated with Table 1 Indicator Definition Poverty line Value of a very basic basket of food products that provides mini- mum caloric intake of individuals of average age and sex. Extreme Poverty An Individual will be considered extremely poor if his or her household per capita income falls below the extreme poverty line defined by DANE, based on the 1985 Survey of Expenses and Income. Poverty A person is considered to be poor if his or her household per capita income is below the poverty line, a multiple (between 2 and 2.5) of the extreme poverty line, to take into account other basic necessities. Mean income Average household per capita income. Poverty gap Describes the average distance to the poverty line. Gini Coefficient Measures the expected difference between two Colombians' income picked at random. Quintiles (Top/Bottom) The ratio of the share of income held by the richest quintile to that held by the poorest. Inequality Urban-Rural Decomposition (Theil) Within Amount of income inequality explained by differences within rural and urban groups weighted by its share in the whole population. Betwveen Amount of income inequality explained by differences between rural and urban groups of population SHARED GROWTH, POVEWIY, AND INEQUALITY 127 Table A4. Inequality Decomposition, Urban Colombia Change in Gini Coefficient Income per Capita (% points) Decomposition Returns to Education Education Direct effect Indirect effect via Endowment Observed via price mean endowments Total Effect 1978-88 00 -19 33 1 4 -1.9 1988-95 4 2 -0 1 2 2 2.1 -1 7 1995-99 2.6 1.3 3 5 4 8 -3.0 1978-99 6.8 -07 9.0 8 3 -6 6 Source Velez et al (2002) 128 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE References Arboleda, J., and E. Correa. 2002. "Internally Displaced People Policy Note." Wash- ington, D.C.: World Bank. Acosta, 0. L. 2000. "Gasto Puiblico Soclal y Arquitectura Financiera: C6mo las Condiciones Fiscales Existentes y Esperadas Afectan la Provisi6n de Servicios Sociales en Colombia." Washington, D.C.: World Bank. Bourguignon, E Ferreira, and P. G. Leite. 2002. "Beyond Oaxaca-Blinder: Account- ing for Differences in Household Income Distributions Across Countries." Washington, D.C.: World Bank Chiquier, Loic. 2002. "Housing Finance Policy Note." Washington, D.C.: World Bank. Correia, Maria. 2002. "Gender Policy Note." Washington, D.C.: World Bank. Centro de Esrudios Regionales Cafeteros y Empresariales (CRECE). 2001. "Disefio del Marco Normativo e Institucional y del Esquema de Toma de Decisiones de la Red de Protecci6n Social." Manizales, Colombia. Davis, S., and E. Sanchez. 2002. "Indigenous Peoples and Afro-Colombian Com- munities." Washington, D.C.. World Bank. Dulitsky, D., M. Gragnaloti, and K. Lindert. 2001. "Social Protection Expenditure Review." Washington, D.C.: World Bank, processed. Echeverry, J C., M. Santamaria, and A. Escobar. 2002. "Tendencias, ciclos y dis- tribuci6n del ingreso en Colombia: Una critica al concepto de 'modelo de desar- rollo "' Archzvos de Economia. Departamento Nacional de Planeaci6n, Colombia. Fretes Cibils, Vicente, and Vicente Paqueo. 2002. "Labor Markets Policy Note." Washington, D.C.: World Bank. Gavin, M., R. Hausmann, R. Perotti, and E. Talvi. 1996. "Managing Fiscal Policy in Latin America and the Caribbean: Volatility, Procyclicality and Limited Cred- itworthiness." Office of the Chief Economist, Working Paper #326. Washington, D.C.: Inter-American Development Bank. Gaviria, A. 2001. "Who Bears the Burden of Crime and Violence in Colombia?" Washington, D.C.: World Bank. Gaviria, A., and C. E Velez. 2001. "Who Bears the Burden of Crime and Violence in Colombia?" Washington, D.C.: World Bank. Gomez-Buendia, Hernando. 2002. "La Reforma Pensional." Revista Semana, July 1. Herrera, S., G. Perry, and N. Quintero. 1999. "Output Fluctuations in Latin Amer- ica: What Explains the Recent Slowdown?" Washington, D.C.: World Bank. Inter-American Development Bank (IDB). 1997. Llbhaber, Menahem, and Vivien Foster. 2002. "Urban Water and Sanitation Sector Policy Note." Washington, D.C.: World Bank. Panopoulou, Panagiota, and Maria-Luisa Escobar. 2002. "Health Policy Note." Washington, D.C.: World Bank. Panopoulus, G. 2001. "Subsidized Health Insurance, SISBEN and Demand for Health Care Among the Poor in Colombia." Washington, D.C.: World Bank. SHARED GROWTH, POVERTY, AND INEQUALITY 129 Partow, Z. 2002. "Colombia: Public Expenditure and Fiscal Adjustment" Wash- ington, D.C.: World Bank. Rawlings, Laura B. 2002. "Social Safety Net Policy Note." Washington, D.C.: World Bank. Rodrik, D. 1999. "Why is there so much insecurity in Latin America?" World Bank, October 1999. Wlez, C. E., and V. Foster. 2001. "Public Social Expenditure in Colombia: Inci- dence and Sector Priorities in the 1990's." Washington, D.C.: World Bank Vdlez, C. E., B. de la Briere, and N. Millan. 2001. "Poverty and Welfare in Rural Colombia During the Last Two Decades." Washington, D.C.. World Bank. Velez, C. E., J Leibovich, A. Kuigler, C. Bouill6n, and J. Nnfiez. 2001. "The Rever- sal of Inequality Trends in Colombia, 1978-1995: A Combination of Persistent and Fluctuating Forces." Washington, D.C.: World Bank. Velez, Carlos Eduardo, Vivian Foster, Mauricio Santamarfa, Natalia MillUn, and B1nMdicte de la Briere. 2001. "Colombia Poverty Report." Volume I. Washing- ton, D.C.- World Bank. Processed. Vlez, Carlos Eduardo, and Juanita Rlafio. 2002. "Colombia Policy Note on Inequality and Poverty Facing Increasing Poverty and Inequality in Colombia" Background Material. Washington, D.C.: World Bank. von Gersdorff, Hermann. 2002. "Pension Reform Policy Note." Washington, D.C.: World Bank. World Bank. 2002. "Colombia: Social Safety Net Assessment." Report #22255-CO. Washington, D.C.: World Bank. 4 The Demand for Governance and Quality of Government This Chapter was written by Fernando Rojas. I. Summary Since the 1 960s, Colombia has adopted two different governance standards.' One is government capacity to formulate and implement macroeconomic policies or reach aggregate economic targets. Another, much lower one, is the capacity of the national level to propose, enact, implement, and monitor sector-by-sector policies. The country was able to live with this separation while economic growth made sector inefficiencies relatively invisible. That changed recently with globalization, the intensification of the internal conflict, and-most obviously-when the country experienced years of stagnation or negative economic growth for the first time in nearly 50 years. It then became necessary to dig into the deeper causes of, and the linkages between, fiscal and monetary imbalances, on one hand, and sector ineffi- ciencies, low investment, political and social instability, and an inadequate business environment, on the other. This Chapter discusses (a) how and why Colombia came to need a comprehensive-rather than today's fragmented-approach to enhanced quality of government, (b) the main areas where higher quality of government appears to be most needed in the next few years, and (c) how the country could strategize about strengthening the quality of government. 1. This Chapter focuses on the quality of government As proposed in this Chapter, there is quality of government when governments not only have the necessary capacity to make and implement policy, but also the capacity and commitment to select policies that are socially, politically, and economically sustainable As such, the quality of governance includes checks and balances, legitimacy, and accountability Since quality of government includes repre- sentation, participation, social control, and anticorrtption capacity, it covers the Executive, the Legislative, the Judiciary, and all independent sanctioning and control organisms 131 132 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE II. Background: A Dual Quality of Government How DID COLOMBIA COME TO Two STANDARDS OF PUBLIC SECTOR? It could be argued that the separation between economic policymaking and other public poli- cies began with the creation of the Junta Monetaria in 1963. Although the Minister of Finance was the President of the Junta, its creation was a significant step toward separation of economic policymaking from political interference. That separation was later ratified by the pathbreaking Estatuto Cambiario, or Decreto-Ley 444 of 1967. Those two reforms, together with the constitutional and administrative reforms of 1968, gave the government the initiative on (in some cases full control of) monetary, trade, exchange, and fiscal policy. Subsequent reforms, including strengthening of the independence of the Central Bank and elimination of budget subsidies in the Constitution of 1991, have further consolidated the independence of economic policymaking. Presidents have generally supplemented constitutional and legal provisions by appointing technically sound, politically independent ministers of finance and Central Bank presidents. As a mat- ter of fact, Colombian ministers of finance are usually selected out of a self-referred invisible college of distinguished economists that look to the international commu- nity of prominent economists and to international monetary and development insti- tutions as their mamn reference points. When the government's economic policy is restricted to regulation and supervi- sion (as in monetary, financial, and exchange matters), policy application generally has been satisfactory. When the government is also in charge of the administration (as in the case of internal taxes, customs, and the budget) (see Box 1), policy appli- cation is usually less satisfactory. Yet, substantial progress has been achieved over the years. Strengthening tax administration is a never-ending task in a country that has an average of one substantial tax reform every other year-and where the revenue impact of tax reform is generally short-lived. Likewise, budget transparency in terms of key budget classifications and reporting to the Congress and civil society is still far from contemporary good practices. Fiscal data are often late or unreliable (par- ticularly in the case of subnational governments) and the country still has a long way to go in terms of performance and results evaluation and in terms of e-government for purposes of real-time data collection and reporting. There is still room to either reinforce on the margin the independence of the Central Bank2 or protect it against politically pressured exchange or debt measures. As a result of the privileged institutional development of the economic policy sec- tor over the last 40 years, this sector is currently much better equipped than any other sector to make and implement policy. Likewise, the whole area of economic policy has been less contaminated by accusations of government corruption, and the area of macroeconomic policy has been largely isolated from corporate-based 2. As discussed, for instance, in the recent Alessina report regarding the composition of the Board of the Central Bank. THE DEMAND FOR GOVERNANCE AND QUALI IY OF GOVERNMENT 133 Box 1. Budget Implementation and Control The area of budget implementation and control reflects the ambivalent devel- opment of Colombia's governance and quality of government The country recently adopted the single account (Cuenta Unica del Tesoro) and integrated financial management. Both tools enhance the government's capacity to moni- tor the flow of public expenditures and enforce hard budget constraints and overall fiscal targets. However, Colombia still has a long way to go with respect to disclosure and transparency in budget reporting to the Congress and civil society.a The country is also far behind other Latin American countries such as Brazil, Chile, and Mexico with regard to transparency and control in the pro- curement process, anticorruption tools, and strategies for electronic govern- ment. Again, the primacy of macroeconomic goals has diverted institutional development from efficiency, accountability, and protection from anticorrup- tion in day-to-day government operations. a See Chapter on Budgetary InstitLtions. demands, stakeholder capture, or clientele-based political parties In sum, high qual- ity of government has ensured rather sound, sustainable economic policy and imple- mentation tools throughout the last 40 years. Other sectors have not followed the pace of macroeconomic policy in terms of governance strengthening, efficiency, or political and fiscal sustainability. Until recently, the general approach to governance strengthening was to create autonomous or semiautonomous institutions that were somehow protected from clientele capture and other forms of political interference and corruption.3 This approach generally failed, not only because most autonomous institutions proved to be vulnerable to political patronage and corporate control, but also because most of those institutions entailed some degree of bureaucratic duplication or parallelism, additional fiscal cost, and diffusion of monitoring and control. In addition, autonomous institutions often escaped from official controls and civil society super- vision. When one of those autonomous or semiautonomous institutions played a critical role in economic policymaking, as in the case of Empresa Colombiana de Petrdleos (ECOPETROL), the national oil company, economic policymakers pre- ferred the regulatory way to ensure transfer of the oil surplus for fiscal stabilization purposes, rather than the more personally painful and politically expensive way of 3 Depending on their nature and the fashion at the time, those autonomous or semi- autonomous institutions may have been called such things as empresasplbWcas, establec- imientos pdbhcos, or special funds 134 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE reforming the institution from within.4 When the country began to reconsider the role of the public sector, some of those autonomous agencies were either suppressed or transferred to the private sector. As reported in several other Chapters, most key Colombian government sectors have-to different degrees-partially fallen prey to their own corporate stakehold- ers or have been exposed to general patronage bonds. In both cases, rent-seeking practices feed political and individual interests on the supply side of government services. Transparency is restricted and control of the sector nurtures more-than- ordinary opportunities for corruption. Above all, protective barriers prevent either policymaking or effective implementation, thereby weakening governance. Education is perhaps the most visible case of corporate (teachers' trade union) control and prevention of reform. Health reform has also been either controlled or diverted by service providers. The same can be said about infant, youth, and family services under the umbrella of the Instituto Colombiano de Bienestar Famlizar (ICBF). Although the Calas de Compensacidn Famzliar appear to be more receptive to user demands than other social services, access to the Cajas still follows a closed- corporative structure-which is in turn due to the difficulty of proposing, agreeing to, enacting, and implementing overdue labor reform, including pension reform.5 As in other sectors, the country has not been able to appoint technically independ- ent managers for the labor sector. Or, when temporarily appointed, as in the recent case of the Instituto del Seguro Social (ISS), the interests of trade unions and con- tractors tend to prevail over the interest of the country as a whole. Restricting providers of labor, goods, or other services is also common practice in transportation or infrastructure investment. Since sector expenditure policy is geared toward guaranteeing supply or maintaining a certain production function- rather than satisfying demand or otherwise ensuring efficiency in results-stake- holder capture of a given chunk of sector expenditure generally goes unchecked. Trade unions often prevent expenditure rationalization and productive efficiency in electricity or public transportation.6 Although the procurement regulatory frame- work has been reformed time and again, the country is still far from the most advanced procurement reform, as exemplified by Brazil, Chile, or Mexico in the 4. One exception worth noting is the public utiliry company of Medellin, Empresas Ptiblicas de Medellin (EPM). A full account of this rather isolated best practice case remains to be added to the literature on Colombian public administration. 5. According to a recent World Bank survey, the Caja Nacional de Prevision Social (Govern- ment Employees Pension Fund, CAJANAL), the Caja de Previsidn Social de Comunica- ciones (CAPRECOM), and other pension and social security institutions are perceived to be among the most politicized and most corrupt in the country. The survey consulted employees, users of public services, and entrepreneurs. See Chapter on Corruption. 6. According to the World Bank survey cited in footnote 6, the Instituto Nacional de Vias (National Highway Institute, INVIAS) is perceived as one of the most corrupt and most politicized government institutions. See Chapter on Corruption. THE DEMAND FOR GOVERNANCE AND QUALITY OF GOVEFRNMENT 135 Latin American context. It is no surprise to see potentially strong suppliers shy away from bidding competitions because the odds of their winning have been largely reduced by predetermined government inclination in favor of another potential sup- plier. Accusations of overpricing and underreporting are also frequent. Unequal development in supervision, auditing, and sanctioning has also meant stronger governance capacity in the macroeconomic areas than in other areas of public sector activity. The Central Bank closely monitors monetary and exchange policies, while the Superintendencia Bancaria has kept a vigilant eye on savings and credit insti- tutions. More important, the quest for sustainable macroeconomic policies and a sound financial sector is matched by the cooperative ear of both the Ministry of Finance and the Planning Department. Taken altogether, the monitoring and supervision capacity of the macroeconomic and financial sector has prevented-with only a few, notable exceptions7-ma)or scandals in either macroeconomic policymaking or the day-to-day management of public debt, the country's reserves, or banking institutions. Supervision, auditing, and sanctioning of the public sector is a different story. Indeed, the country's capacity to monitor, audit, and sanction irregularities in the public sector is still far from the needs of market formation, globalization demands, and an enabling business environment. Different from the Superintendencia Ban- caria, most superintendencias were quickly captured by their own stakeholders (as in the case of the Superintendencia de Cooperativas for the solidarity sector) or by polit- ical parties and their clientele (as seems to be the case with the recently created Superintendencia de Salud). The institutional development of the Contraloria Gen- eral de la Repilblhca is another case in point. Although inspired by the same Kem- merer mission that recommended creation of the Central Bank and the Superinten- dencia Bancaria in 1923, the Contraloria was soon tied by the bonds of political patronage. Still, today the Contraloria General de la Reptiblca has been restricted by a mammoth organization where political appointees still manage to defend their interests at the cost of modernization and reform-minded staff. Other auditing and sanctioning institutions have roughly followed the fate of the Contraloria. Attempts to liberate the human resource policies of the Procuraduria General de la Reptiblica from political patronage ties have largely failed. Indeed, internal reform of the Procuraduria does not appear to have advanced as much as the reform of the Contraloria. The constitutional reform of 1991 followed a regionwide path toward the creation of technically competent judicial management. However, similar to other sectors, the top level of the Conseqo Superior de la Judicatura remains in the hands of former Judges and magistrates that tend to represent primarily their own corporate, monopolistic interests. There are even symptoms of magistrates' cap- ture of the Constitutional Court for the pursuit of their own political careers. The country is still far from timely and efficient administration of )ustice Likewise, Colombia has achieved only limited progress (compared to, say, Canada and the 7. Like the infamous case of the vanishing of US$13 million from the reserves managed by the Central Bank in the mid-l 980s, or the always debatable rescue of banks in times of crisis 136 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE United States) in developing means of alternative dispute resolution; unequal access to justice still reinforces other sources of inequality of opportunity and reduces spillover of growth benefits. Since violence and the internal conflict have further debilitated property rights and safety and other basic human rights, large sections of the population are not protected in their lives or assets. As a result, large segments of the population cannot provide either personal or real guarantees to access capital or consumption markets. Because Colombia is one of the most decentralized countries in Latin America, strengthening governance there necessarily means strengthening the capacity of the public sector to deliver service and monitor compliance at subnational levels. How- ever, the way Colombia conceived and/or implemented decentralization has con- fused the division of responsibilities and weakened accountability. As discussed in the Decentralization Chapter, sector capture of deconcentrated or decentralized social services has now been compounded by subnational stakeholder protection of the current level of fiscal transfers. Mayors' and governors' associations have equated democracy with maintaining or increasing fiscal transfers regardless of the coverage and quality of services. The fiscal demands of subnational governments have been reinforced by trade unions and patronage-based political leaders that claim the need to equate transfers with the cost of clientele-controlled rigid production functions. As demonstrated by all reforms passed since the Constitution of 1991 and Law 60 of 1993, Colombia decentralization is supply targeted rather than demand driven. Not surprisingly, the most recent government evaluation of one decade of Colombia's decentralization is primarily a study on social expenditure growth-not an assessment of how decentralization per se enhanced quality, raised coverage, or otherwise stimulated higher efficiency in allocation and production.' In spite of Colombia's widely proclaimed regional leadership in fiscal decentralization, the country has yet to develop the essential monitoring and evaluation capacity to demonstrate how and to what extent decentralization has contributed to more effi- cient service provision than the previously centralized model of service delivery. III. The Problem with the Coexistence of Strong and Weak Governance WHY IS STRENGTHENED GOVERNANCE REQUIRED IN AREAS OTHER THAN ECONOMIC POLICY? For some years Colombia was able to grow and maintain good fiscal and financial performance while keeping the dualism in governance between economic policy and other sectors' policy. So-called first-generation economic reforms appear to demand only strong policymaking capacity in the areas of trade, monetary, exchange, and overall fiscal targets. Poliucally difficult as those reforms might have been, they penalized only a limited number of rather silent and shameful stakeholders that used 8. See Decentralization Chapter THE DEMAND FOR GOVERNANCE AND QUALITY OF GOVERNMENT 137 to benefit from protectionism or profit from inflation or erratic changes in the exchange rate. The damaging experience of hyperinflation or abrupt fluctuations of the exchange rate in Argentina or Brazil moved the country to remain faithful to the solid principles of the Estatuto Cambiario, first established in 1967. In addition, trade liberalization was never as swift in Colombia as it was in other Latin American coun- tries. The attentive eye of the international community has also helped the country maintain its traditional priority for sound, effective macroeconomic policy Several factors combined to put pressure on the country to adopt second- and third-generation reforms. The country urgently needs to offset the effects of low or negative economic growth in recent years on poverry expansion and reduced invest- ment. Coexistence between strong macroeconomic governance and weak govern- ment capacity in other sector policies was sustainable only while economic growth provided a cushion for sector inefficiencies The context of relatively open economies in an increasingly globalized world is finally pushing Colombia to explore all potential avenues for enhanced efficiency in resource allocation, higher productivity, a better savings and investment climate, and strengthened country competitiveness Sector rigidities that were long tolerated are increasingly perceived as adding unnecessary transaction costs that limit the country's competitiveness. As illustrated by the recent World Bank survey on corruption and politicizatioln, both employees and managers are now fully aware that built-in inefficiencies in social services prevent labor productivity enhancement, and restricted government con- tracting protects low capital productivity. It is nIow more clear-than, for instance, at the time of the 1991 Constitution-that labor rigidities and capital overpricing reduce government efficiency to the point where the only solution is additional gov- ernment indebtedness and crowding out the financial needs of the private sector. Besides the urgent need to reestablish a sustainable pace of growth and counter- act the frightening poverty effects of the ongoing economic slowdown, the fiscal and financial demands added by the Colombian internal conflict have made it evident that the country can no longer postpone long-overdue, sector-specific and across- sector reforms. Besides adding pressure on internal security expenditures, the inten- sification of Colombia's internal conflict has multiplied both poverty and the need for social safery nets. Enhanced quality of government is required not only for effi- ciency-seeking, sector-specific reforms, but to reestablish protection of the most basic personal and economic guarantees and behavioral incentives. Across-sector reforms are needed to extend land titling and registration and to guarantee conti- nuity in and equal access to justice, social services, and public utilities in conflict- ridden areas. It is not only that governance and high-quality government are needed to bring the internal conflict to an end; it is that governance is challenged and debil- itated by the mere continuation of the conflict. Bringing the conflict to an end is a prerequisite for savings and investment, and mobilization of resources that have been marginalized by the internal war. More so than sector-specific reforms, across- sector reforms for strengthening overall governance are required to develop a credi- ble enabling business environment. 138 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Today's economic policymaking and implementation goes beyond first-genera- tion reforms. Although all reforms are interlocked, making significant progress in transversal or across-sector areas such as monitoring, auditing, sanctioning, and pro- tecting basic rights is necessary to supplement sound macroeconomic management and enhance the investment climate and service sustainability. Colombians have known for some years what are the broad policy guidelines for second- and third-generation sector policy reform. In some cases, the specifics of those reforms-as proposed by certain other sector Chapters-have also been iden- tified. Subsequent to the initial wave of reforms in the early 1990s, the country is already aware of the implementation risks and the potential resistance of stakehold- ers that daunted the reform spirit and jeopardized sustainability of reforms in, for instance, health, labor, the judiciary, education, and decentralization. As demon- strated by recent surveys and electoral results, Colombians appear to be persuaded that those reforms-either sector or across sectors-would come to supplement tra- ditionally sound macroeconomic policy with the necessary microeconomic frame- work and incentives for economic agents to develop the full potential of labor, capi- tal, land, and technology. The problem is that those second- and third-generation reforms require strengthened governance and enhanced quality of government-and this is precisely what the country has been short of for the last 20 or 30 years. There appears to be a vicious circle that goes from the incapacity of government to formu- late and implement most needed reforms, to the fact that those reforms are precisely needed to strengthen governance in Colombia. Strategizing is needed to develop a virtuous cycle that connects incremental reform with the gradual building of high- quality government, then using new, strengthened government capacity for policy- making and implementing those reforms that are more difficult to pass and effec- tively apply. IV. Strategizing to Strengthen Governance: Selectivity and Gradualism in Building High-Quality Government As mentioned, Colombia is entering a new phase of its history-one that appears to have a propitious climate for strengthening governance. The new administration seems to give high priority to improving the quality of government, and citizens are demanding government reform for efficiency and sustainability of those other reforms analyzed in the accompanying Chapters. The challenge lies in translating growing awareness and demand into effective government strengthening for making and implementing sustainable policies. It should be noted at the outset that strengthening governance and enhancing quality of government is more an art than a science. The best public sector leaders (including the Legislature and the Judiciary) and analysts can do is to follow guid- ing principles, assess how those principles have been applied elsewhere and with what results, and adjust to government culture and local idiosyncrasies. THE DEMAND FOR GOVERNANCE AND QUALITY OF GOVERNMENT 139 THE SEARCH FOR HIGHER QUALITY OF GOVERNMENT AS A MEDIuM-To-LONG- TERM GOAL IN ITSELF. Quahty of government is needed for sustainable policies, and sustainable policies may in turn reinforce quality of government. One of the most frequently encountered problems when trying to build government capacity is to confuse strengthening governance and quality of government with sector reforms. Strengthening governance and enhancing quality of government is an end in itself. As such, it requires identification of problems, strategizing, implementation, moni- toring, and evaluation. As with any other program and project, improving the qual- ity of government requires-above all-a program structure and clearly identified responsible agencies for program formulation, financing, implementation, monitor- ing, dissemination, evaluation, and control. (See Box 2 for a description of how it was done in the Clinton-Gore Administration in the United States and the Fox Administration in Mexico.) Box 2. Efforts to Improve Quality of Government in the United States and Mexico The Clinton-Gore Administration in the United States created the National Performance Review (NPR), a White House Policy Council designed to rein- vent government. The NPR orchestrated the largest peacetime downsizing in U.S. history. In addition, the NPR prepared two major procurement reform bills, conducted two major performance audits of the federal government, and oversaw a major reform initiative aimed at improving the government's regula- tory agencies. The NPR also formed the National Partnership Council, which reshaped labor-management relations in the federal government. The Fox Administration in Mexico established the National Plan for Trans- parency and Anti-Corruption, primarily focused on the federal level. The federal government appointed an interministerial commission to monitor, evaluate, and revise the program. The Commission relies on a technical secretariat that moni- tors the program on a day-to-day basis and recommends program reforms. The recommendation for Colombia is to appoint either a presidential coun- cil-a la the United States-or an interminisrerial commission backed by the President or the Vice-President-i la Mexico. Implementation and dissemina- tion of the government's plan should go together. It is necessary to make citizens aware of the government challenge-and the need to support it if sector reforrms are to be effectively implemented. Good practices from other countries may also be widely publicized since the country needs a mirror in which to see itself. Monitoring and evaluating governance strengthening should be shared by nongovernment bodies and widely disseminated. Civil society, the Congress, and even subnational governments may all play a role in assessing progress in the government's plan to enhance quality of government. 140 COLOMBIA- THE ECONOMIC FOUNDATION OF PEACE SELECTING A MANAGEABLE PROBLEM AND PROPOSING STRATEGIES TO STRENGTHEN GOVERNANCE AND ENHANCE THE QUALITY OF GOVERNMENT. When it comes to strengthening governance in key sector areas, it is necessary to know what are the loopholes through which certain stakeholder interests restrict access-either legally or illegally-to government contracts, or otherwise restrict policymaking and imple- mentation to the point of their prevailing over the public interest. Targeting reforms on specific loopholes or corruption opportunities tends to be more effective than broad guidelines and corresponding uniform monitoring indicators (see Box 3). SELECTING THE RIGHT GOALS AND TooLs. Colombia's strategy to enhance good governance needs to include specific public sector reforms (goals and tools). What is needed in order to improve quality of government is to start with those gover- nance-enhancing reforms that are least likely to arouse stakeholder opposition-or at least opposition so strong that the government is not prepared to cope with it. Multisector reforms that strengthen government capacity-such as those targeted at the procurement system or integrated financial management-usually arouse less frontal opposition than sector-specific reforms that affect long-entrenched interests of sector-organized groups-such as trade unions in the areas of health and educa- tion, and corporate associations for road construction and water provision. (See Box 4 for a discussion of what Colombia must do to achieve good governance.) Similarly, focalized reforms-such as concentrating on a small number of gov- ernment agencies at a time or mapping and simplifying sector procedures-are more doable than all-encompassing, final reform proposals such as changing government culture or eliminating party patronage (see Box 5). Strengthening governance should follow a sequential strategy. Initial reforms should be selected in such a way as to ensure production of demonstration effects and some capacity building. From then on, the government may pursue more chal- Box 3. Targeting Reforms to Plug Loopholes The Colombian public sector usually has been characterized as patronage ori- ented and stakeholder or corporate dominated. However, that broad character- ization needs to be refined for governance policy reform. It is also widely accepted that the clientele-dominated rwo traditional political parties have been weakened by urbanization, party fragmentation, and electoral support of inde- pendent candidates. At this point the country needs to know how to formulate, discuss, and implement reforms that open up markets for public services, be it the labor market for public employment or the procurement market for outside provision of goods and services. In addition, the government needs to make sure that closing the loopholes that create rent-seeking opportunities also close the door to corruption-rent opportunities. THE DEMAND FOR GOVERNANCE AND QUALITY OF GOVERNMENT 141 Box 4. Steps to Ensure Good Governance in Colombia Reform of administrative processes is usually less vulnerable to strong political opposition (from political parties or other rent-seeking stakeholders) than elec- toral reform, congressional reform, or labor law reform. Colombia needs to clearly identify why and how corporate control infiltrates the procurement process so that fair competition is reestablished. It has been claimed, for instance, that respect for the specially designed procurement process at the Empresas Puiblicas de Medellin (EPM) largely accounts for this success story within the Colombian public administration. By analyzing key processes the Colombian government may close the loop- holes that prevent good government in policymaklng and implementation. Other multisector processes that impinge on Colombia's governance and quality of government are: * Results indicators and transparency in budget reporting and disclosure. As exemplified by the Australian, British, and New Zealand examples, budget classifications and reporting can be organized in such a way that it facilitates discussion and participation by the Congress and civil society regarding pol- icy objectives, result indicators, and budget execution.a * Open-ended contracts for intergovernmenr results agreements and corre- sponding earmarked transfers. Central government capacity to monitor, evaluate, and sanction noncompliance in case of "incomplete contracts" is essential to strengthen government capacity under decentralization. b a See Chapter on Budgetary Institutions b See Chapter on Decentralization. lenging-and politically riskler-reforms. It must be borne in mind that institution building for strengthening governance and enhanced quality of government is nec- essarily a phased-in, medium- to long-term process. (See Box 6 for a discussion of the experience of Singapore with phased-in government strengthening.) Therefore, identification of intermediate goals is key to assess progress and to gain political sup- port. In today's Colombia, it is necessary to avoid the risk of postulating-even accepting that-the end of the internal conflict is the true test of strengthening gov- ernance. Perhaps the best way to avoid such a risk is to elaborate on and demonstrate the linkage between intermediate achievements in building higher-quality govern- ment and the likely scenarios and phases for finalizing the conflict. Publicly com- mitting government to achieving certain intermediate goals, combined with disclo- sure and dissemination of progress evaluation, is perhaps the most powerful marketing tool for strengthening governance, and announcing broad goals and not being able to report on partial progress is usually a recipe for failure. 142 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Box 5. Focalizing Reforms Colombia has tried time and again to pass and implement comprehensive pub- lic sector reforms-most recently in 1992-93 and 1998-99. According to les- sons learned from best practice experience, analyzing an agency and coming up with a tailor-made plan for that agency-and it is better if the plan matches and enhances the feasibility of ongoing sector reform-is usually more doable and more productive than across-the-board reforms for the entire public sector. Reform of a handful of individual agencies that play a key role in preventing and sanctioning corruption, such as the Contralorfa General, the Procuraduria General, or some of the Supenntendencias, is likely to strengthen governance in most sectors of State administration. Likewise, reform of hard-to-transform government agencies such as the ISS and ECOPETROL could be precedent setting for the rest of the administration.2 a. Reform of key government agencies could be supplemented with case studies and dis- semination of best practice agencies such as the EPM. Box 6. Singapore's Experience with Phased-In Governance Strengthening Colombia might follow the best-practice case regarding phased-in governance strengthening in Singapore which, in a period of just a few years, transformed itself from being one of the most corrupt countries in Southeast Asia to being one of the Jewels of the world in the areas of transparency, anticorruption, and enhanced governance. The story of Singapore is one of giving the government the right tools to implement and evaluate reforms in any government sector. E-government, process simplification, improved regulatory frameworks, reduc- tion of public employee direct contact with private citizens, institutionalized citizen participation, strengthened disclosure and accountability, enactment of new guidelines for human resource management, and agency-by-agency publi- cation of customer service goals are some of the tools that empowered the gov- ernment and strengthened governance in Singapore. Enhancing the quality of government appears to be a prerequisite for policymaking and implementation of sector-by-sector reforms. In this regard, strengthening governance should come before-or at least be simultaneous with-sector-specific reforms pro- posed in other Chapters. Part II Achieving Fast and Sustainable Growth 5 Macroeconomic and Fiscal Frameworks This Chapter was written by Zeinab Partow. The crucial requirement for the Colombian economy at the new millennium is to return to a path of sustained and substantial economic growth that characterlzed the country until the mid-1990s. Analysis has shown that economic growth is the key fac- tor in poverty reduction in countries (Dollar and Kraay 2000). For Colombia, in par- ticular, growth has been the key to poverty reduction: each additional point of growth in gross domestic product (GDP) per capita reduced poverty by approximately 0.6 percentage points. At the same time, it has been shown that the substantial progress in poverty reduction achieved during the 1980s and 1990s was largely erased by a few years of low growth, and by the 1999 recession. In fact, the recession in the second half of the 1990s brought poverty levels down to 1988 levels (Velez 2002). Colombia's recent path threatens an unsustainable increase in deficits and debt, along with poor prospects for GDP growth and high unemployment. New realities have emerged in fiscal domain, including significant and persistent deficits and large future fiscal liabilities. The rapid expansion of the public sector over the previous decade, from spending one fourth of GDP in 1990 to over one third of GDP by the end of the decade, combined with growing liabilities, particularly on the pensions front, have led to significant and persistent deficits of a structural nature The pri- vate sector, on the other hand, languishes: while government consumption grew by two thirds between 1994 and 2001, that of household remained basically unchanged, and private investment fell by over 37 percent. The weakness of the pri- vate sector is at the center of Colombia's current economic woes. Private GDP has grown little since the mid-1990s, and private investment has deteriorated from its peak of near 12 percent of GDP in 1994 to under eight percent today. Total invest- ment now stands at a level of about 15 percent of GDP, incompatible with the resumption of significant growth Public sector consumption has in effect displaced the private sector as the major contributor to domestic demand. While the current 145 146 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE administration has recognized the need to strengthen fiscal accounts and has begun to set a solid foundation for further reforms, these will need to be expanded and accelerated in the coming years. As public spending rapidly outstripped the sector's revenue generating capacity, much of the fiscal deficit, particularly since the mid-1990s, has been financed with debt, both domestic and external. Today, net public sector debt in Colombia stands at about 46 percent of GDP. In 1997, prior to the onset of the recession and to the rapidly deteriorating fiscal balances of 1998 and 1999, public debt stood at 27 per- cent of GDP. Spending on interest payments to service public debt now stands at near five percent of GDP, constituting approximately a quarter of the central gov- ernment's expenditures, more than twice its investment budget. Colombia now faces the combined challenges of reducing its debt levels to ensure sustainability and achieving growth rates that are sufficient to reduce poverty levels and increase employment. Continued fiscal adjustment is unavoidable if debt sustainability is to be attained; just as important, the quality of the adjustment is key for a return to growth rates in the four-to-five percent range. Macroeconomic management over the past decade has thus failed to deliver the habitual steady growth that characterized the country. The mix of reforms at the beginning of the decade, and more importantly, their implementation, generated inconsistencies that made the reforms incompatible as a package. The rapid fiscal expansion, and the resulting appreciation of the peso, was inconsistent with the goals of the apertura, or opening up of the economy (Echavarria, 2000).i The per- ceived "failure" of the apertura generated opposition to deeper reforms,2 and the combination of fiscal expansion and incomplete reforms resulted in Colombia's much-vaunted stability giving way to increasing volatility. Table I summarizes the volatility of a number of macroeconomic and fiscal vari- ables and shows that volatility increased for all of these with the exception of the terms of trade. This did not generally occur in other Latin American countries; a number of authors (among them Gavin et al. (1996), Herrera et al. (1999), IDB (1997)) have pointed out that domestic volatility in many countries of the region declined over the 1990s. As mentioned above, one hypothesis is that in Colombia, 1. Fiscal expansion was a major culprit in the revaluation of the peso during the 1990s, along with capital inflows, expectations of greater revenues from oil, the use of the exchange rate as nominal anchor, and productivity growth Analysts (e g Edwards, 1998; Cardenas, 1996; Zuleta aind Arias, 1997) disagree as to the relative importance of each of these fac- tors in the peso's devaluation. 2 Echavarrfa (2000) argues that Colombia's trade reforms were partial and relatively super- ficial, and that the country did not enjoy a significantly more open economy at the end of the 1990s as compared to the beginning of the decade. He notes that trade in goods and services as a portion of GDP has not changed over the past thirty years, and while there was an "opening" in trade of goods in particular, this was less marked that in other Latin American countries. MACROECONOMIC AND FiscAL FRAMEWORKS 147 Table 1. Increased Volatility in Colombia in the 1990s 1990s:1980s 1970s 1980s 1990s (percent) GDP volatillry 1 8 1 5 2 7 81 Fiscal volatility 1 3 1.4 2.7 91 Exchange rate volatility 1.1 1.5 2.4 57 Monetary volatility 6 8 3.5 3 7 56 Terms of trade volatility 13.0 10.4 5 8 -44 Note Data represent the standard deviation, in percentage terms, of changes in GDP, in the fiscal bilance/GDP, in the real exclaisge rate iidex, in Ml, an1d in the terms of trade as a result of the forces at play during the decade, namely a new constitution, large oil discoveries, an acceleration of illegal activities and an intensification of the inter- nal conflict, there was a loss of momentum in completing the reform agenda, and this exacerbated volatility. Along these lines, the IDB (1997) finds that on the whole, major reformers in the region enjoyed greater economic stability in the 1990s. Gavin et al. (1997) perform formal statistical tests of the empirical relation- ship found by Lora and finds that the relationship is signsficant and robust.3 It is important to note that while the country's long-standing internal conflict has undoubtedly played a role in heightening the degree of uncertainty facing economic agents, and is a constraint on Colombia's achievement of a higher level of potential GDP growth, it is not the sole constraint to growth. In this spirit, this Chapter reviews some of the factors that underlay the severe recession of 1998-99, and out- lines policy priorities to address the current challenges facing Colombia. These chal- lenges can be summarized as the need to attain primary balances that are sufficient to ensure sustainable debt-to-GDP ratios, while attending to the quality of adjust- ment such that growth is not compromised. I. The 1990s: From Rapid Growth to Recession Prior to the 1990s, Colombia was among the better performers in Latin America, with the enviable record of unbroken annual GDP growth for over 50 years, even during the "lost decade" of the 1980s.4 In the early 1990s, this trend seemed set to continue, or even accelerate, as important reforms were undertaken in the areas of trade, the financial sector, private sector participation, and decentralization. Sup- 3 The studies are not conclisive as to the direction of causality which may, of course, be in either direction from reforms to stability or vice versa 4 Garcia and jayasuriya (1997) argue, however, that government reaction to economic crises over the past 30 years and the policies chosen, while enabling the country to enjoy a rare stability, also prevented Colombia from enjoying higher growth and greater prosperity 148 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE ported by the introduction of these structural reforms, output grew by 5 percent a year during 1992-95. By the middle of the decade, however, as many of its neigh- bors aggressively pursued wide-ranging reforms and saw improved growth, Colom- bia's performance began to weaken. The country's gradualist approach to policy- making and reform did not keep pace with that of other countries in the region.5 Throughout the 1990s, Colombia's fiscal accounts experienced a continual and accelerating deterioration. The Non-Financial Public Sector (NFPS) deficit wors- ened from a level of 0.2 percent of GDP in 1991 to 6.4 percent of GDP in 1999, largely as a consequence of the rapid growth in Central Government deficits which had increased from 0.4 percent of GDP in 1991 to 7.1 percent in 1999. Growing deficits and declining growth culminated, in 1998-99, in a recession of a magnitude unprecedented in the country's history. One of the major changes that took place in the fiscal arena during this period was the public sector's unprecedented expansion. Colombia began the decade with a rela- tively small state, but ended it with a comparatively large one. Public sector (NFPS) consumption grew from an average of 10 percent of GDP between 1970 and 1993, to an average of 18.5 percent during the 1994-2001 period. The most rapid expansion took place between 1996 and 1999. As a proportion of total consumption, the public sector accounted for 12.5 percent prior to 1993, but grew to 21 percent by the 1994-2001 period. In South America, only the Brazilian public sector consumes more. The confluence of two major events that occurred at the beginning of the decade of the 1990s helps explain the origins of Colombia's fiscal problems: a new consti- tution and expectations of large oil discoveries. A new constitution came to replace one that had been in place for nearly 100 years. The new constitution reflected a new political and development agenda whose major goal was to enable Colombia to take full advantage of changing global realities. The new agenda aimed to rational- ize the role and scope of government by favoring the concentration of public sector activities in the supply of public goods with high externalities, and to establish the necessary incentives and institutions for the promotion of resource allocation according to market-based principles. These objectives were to be achieved by liber- alizing external trade6 and the financial sector, and by increasing the role of the pri- vate sector through privatizations, concessions and a greater participation in the pro- vision and administration of physical infrastructure and public services. 5. Colombia, along with Uruguay, is characterized by the IDB (part 11, 1997) as one of the continent's slower reformers. On an index of structural reforms, Colombia's rating declined from above the Latin American average in the 1980s, to below it in the 1990s 6 Between 1990 and 1999, the average nominal tariffcame down from 44 6 percent to 11.6 percent, and effective protection decreased from 59 percent 20.5 percent during the same time span. Tariff dispersion was reduced from 23 rates to four, and quantitative restric- tions were eliminated. At end-1998, Colombia scored a rwo on the IMF's 10-point scale for classifying the restrictiveness of trade regimes (a higher score indicates a more restric- tive regime.). (IMF Selected Issues paper, mimeo, 2001) MACROECONOMIC AND FISCAL FRAMEWORKS 149 At the same time, the manner in which a number of other reforms were Imple- mented tended to expand the size and role of the State. The Constitution, and sub- sequent laws, included provisions to accelerate the devolution of power to commu- nities through accelerated decentralization, and the enhancement of the development of human capital and the delivery of social services to favor income redistribution and fulfillment of basic needs. The parallel contraction of the Central Government that should have logically accompanied the decentralization of social services, did not, however, materialize. The expansion of the judicial sector and of defense spending followed the argument that the absence of the State from rural and remote areas was associated with growing violence and guerrilla conflict. Thus, while the intent may have been increases in citizen participation, in the devolution of power through decentralization, and in political accountability and rationalization of government activities, the outcome, paradoxically, was the gener- ation of major new obligations for the public sector and for the central government in particular Half of the total increase in central government spending during the decade can be attributed to the new obligations enshrined in the Constitutioll and in other laws passed during the initial years of the 1990s (Table 2) 7 Substantial new spending was targeted to education and health through departmental and local gov- ernments. Transfers by the central government to the social security sector repre- sented further significant resource requirements, as did the restructuring of the judi- cial sector. It is important to note, however, that in addition to greater spending due to new commitments, central government expenditures on wages, salaries, and goods and services also grew, by almost 50 percent between 1990 and 1998. This Table 2. Growth in Central Government Spending Due to Legal and Constitutional Changes, 1990-98 (percent of GDP) 1990-98 Territorial Transfers Municipal Participations I 0 Situado Fiscal 0 5 FEC 0 6 Social Security Transfers Law 100/1993 1 2 Electricity Subsidies 01 Modernization of the State 0 1 Judicial Sector Restructuring 0 4 Total 3.9 Source Echeverry-GarLoni (2001) 7 In addition to obligations created by the Constitution and related laws, significant fiscal expansion took place in 1966, and especially in 1997, as public spending was used in an attempt to forestall the effects of an impeniding recession 150 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE highlights the continuing difficulty of the public sector in adapting to decentraliza- tion and to a new role for the state as envisioned by the reforms of the early 1990s. The new spending commitments generated by the Constitution were accompa- nied by a significant easing of Colombia's liquidity constraint as a result of large oil discoveries in the early 1990, the second salient feature of the decade. As a conse- quence, the country would expect about US$3 billion per year between 1996 and 2002, the period of peak production, representing over 25 percent of exports and at the time expected to bring in between 6 and 8 percent of GDP.8 In classic fashion, the projected increase in revenues fueled spending growth. The boom in the oil sec- tor generated growing revenues as well as high expectations in terms of future income flows, and significantly softened the budget constraint at different levels of the public administration. The investment rating enjoyed by the country-as a result of high levels of international reserves, low debt indexes, and the expectations of a petroleum boom-facilitated indebtedness with external capital markets.9 As a result of these forces-namely the spending obligations created by constitu- tional and other legal reforms, and by more ample and easy access to financial resources following the oil boom, and exacerbated by attempts to avoid recession through fiscal expansion in 1996-97-central government spending rose by over 70 percent, from 11 percent of GDP in 1990 to 19 percent in 1999. The NFPS went from spending 24 percent of GDP in 1990 to 37 percent in 1999. Local govern- ments leveraged future oil royalties and transfers, unsustainably expanding their debt. Spending accelerated as the decade wore on, with nearly three quarters of the increase taking place between 1996 and 1999.10 Revenues, despite numerous tax reforms during the decade, did not keep pace (Figure 1), resulting in a continually growing deficits at both the central level (7.1 percent of GDP in 1999) and NFPS level (5.8 percent). Between 1990 and 2000 half a dozen major tax bills were passed. The total impact of the bills through 2001 has been an increase in revenues of about 4 percent of GDP, while central govern- ment expenditures grew by over nine percentage points of GDP. The resulting fiscal imbalance was, however, temporarily obscured by surpluses in the decentralized sectors, principally in social security and the oil sector, and by large privatizations These surpluses allowed the generation of positive consolidated public sector balances in the early 1990s, but which rapidly declined to a deficit. In this respect it is illustrative to note that negative current savings have characterized Colombias fiscal accounts since the mid-1990s. At the general public sector level, 8. In reality, between 1996 and 2000, oil revenues amounted to between 2.5 and six per- cent of GDP, averaging about 3 5 percent of GDP for the period. 9. Spreads on Colombia's sovereign bonds stood at under 200 basis points over US Treasury during 1997. In 2000, they averaged 695 bps, and today they stand at close to 500 bps. 10 The second half of the 1990s also witnessed an escalation in Colombia's internal conflict Security outlays (police and military) rose from 2 4 percent of GDP in 1994 to 3.2 % in 1998. MACROECONOMIC AND FISCAL FRAMEWORKS 151 Figure 1. Central Government Revenues and Expenditure 2 5 - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 2 5-- Total Expenditures 1-5-Tax Revenue _ 910- 1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 current revenues have not covered current expenditures since 1996 For the central government, current savings have been negative since 1994, reaching a peak of -6.4 percent of GDP in 1999 (Figure 2). This has meant that as revenues lag, current expenditures as well as investment are often financed by higher indebtedness In 1999, at the nadir of the recession, economic activity contracted by 4.3 per- cent, the NFPS deficit reached 6.3 percent of GDP, private investment fell by over 60 percent, and private consumption dropped seven percent. One in five Colom- bians was unemployed, and the plunge in demand brought the inflation rate down Figure 2. Central Government Current Savings -2 _ _ ------ c -4 16 199 199 199 199 199 19 -----7 9 9 -8_ I I l 1 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 152 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE from over 16 percent to nine percent in just one year. Imports also collapsed, falling by 25 percent, helping to bring the current account to near balance. In September 1999 the exchange rate band was abandoned in favor of a floating of the peso The financial system entered into crisis as high interest rates and negative economic growth stretched the payment capacity of most borrowers, and arrears in loan port- folios grew. Over the past two years GDP growth has resumed, with the economy expanding just under three percent in 2000, followed by 1.6 percent GDP growth in 2001; the main engines of growth in 2000 were the recovery of exports and of manufacturing. The recovery remains fragile, however, as evidenced by the loss in the momentum of economic expansion in 2001 and early 2002. Since 1999, the Government's adjustment program has led to a gradual reduction of the NFPS and CG deficits to 3.8 percent and 5.8 percent of GDP, respectively, in 2001, helped along by high oil revenue (Figure 3). However, the pace of the adjustment was also significantly slowed in 2001, partially as a result of lower growth, but also due to the difficulty of significantly reducing expenditures and increasing revenues in the short term in an environment of political fragmentation and internal conflict. 1I. Policies for Adjustment In 1999, in response to the severity of the recession, the new government embarked on an ambitious program of stabilization and adjustment. The general thrust of the policies, supported by a three-year IMF Extended Fund Facility, was to tackle both short- and long-term issues: addressing the recession would consist of immediate Figure 3. Growing Deficits - - Current Account 6 A - = ~~~~~~~CG Balance v / \ ~~~~~~~~~~~~NFPS Balance 4-o-- / --X-- -- 24 __ _ \_ --= - _ ___ __ __ 0 - 9 1 990 1 99 1 1 992 1 993 1 994 1 995 1 996 1 997 1 998 1 999 2000 MACROECONOMIC AND FISCAL FRAmEWORKS 153 measures for fiscal consolidation and financial sector restructuring backed by longer- term structural reforms The program was supported by the depreciation of the peso that had taken place in 1998 and 1999, which brought the effective real exchange rate to a level similar to that of the early 1990s, when Colombia's external current account was near balance, prior to the onset of large-scale oil exports.' The program had inherent risks related to the need for timely congressional approval of significant structural reforms, reliance on privatization revenues, and a rapid emergence from the recession. All of these risks materialized to some extent. The first priority was to attack the deficit and fortify the financial system. To address the public sector deficit, actions on both the revenue and expenditure fronts were taken. At the beginning of 1999, to address the decline in revenues associated with the weakening of the economy, a number of measures were introduced These included a widening of the base for the value added tax (VAT) (although its level was later reduced from 16 to 15 percent), higher fuel taxes, and programs to strengthen tax enforcement. The tax administration (Dzre'ccidn de Irnpuestosy Adua- nas Nacionales, DIAN) was reinforced through a program to reduce collection lags, identify delinquent accounts, collect arrears, and reduce contraband."2 Although these measures assisted in increasing revenues, which were further helped by the notable increase in oil prices in 1999 and 2000, a more substantial tax reform was judged to be needed, and one was approved in late 2000. The tax pack- age was expected to yield about two percent of GDP, although given the slower rate of economic recovery, actual yields in 2001 were expected to be lower The package was based on an increase in the VAT from 15 to 16 percent, an increase in the pre- sumptive income tax rate from five to six percent of net assets, and an increase in the financial transactions tax from 0.2 to 0.3 percent. In addition, the financial transactions tax, originally intended as a temporary tax to obtain resources for finan- cial system restructuring, was made permanent by the tax package. The package established some temporary tax incentives for the repatriation of capital, but did rel- atively little to reduce existing loopholes and exemptions. On the expenditure side, due to the inflexibility of public sector spending obli- gations, only relatively limited measures could be taken in the short term These mi- I 1 Changes in the foreign investnment regimie were also implemented at the time to encour- age foreign direct investnment by improving the terms that governed investmnents in the oil sector by lowering royalty payments and scaling back the role played by Empresa Colombiana de Petrcileos (ECOPETROL) in its partnerships with private operators. 12 Privatization pioceeds were to play an important role in revenue generationi in 2000, and were expected to finance nearly the full amount of the combined public sector deficit for the year. This did not materialize (with the exception of the privatization of Carbones de Colombia [CARBOCOL]) because investor interest remained weak due to continued attacks by the guerrillas against electricity installations, the withdrawal by Bogota of the plan to sell its telephone company, and the time required to prepare public banks for divesrnient 6r liquidation 154 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE tially included a tight public salary policy involving a reduction in real wages, although the measure was invalidated by the Constitutional Court in 2000.13 Sig- nificant downsizing of the public sector at the decentralized level proceeded in response to the need to address the worsening fiscal and debt conditions of territo- rial governments Further efforts at public sector downsizing relied on the privatiza- tion or restructuring of public enterprises in the banking and mining sectors. The inflexibility of Colombia's expenditure structure, however, meant that the brunt of the adjustment fell on the central government's investment spending. This declined by over a third to a mere one percent of GDP. Significant reductions in spending would require structural reforms, described below. Avoiding a systemic crisis in the financial sector constituted the other immediate concern of government policy during 1998-99. A multipronged strategy was pur- sued to divest or liquidate public banks, provide mortgage debt relief, grant incen- tives to facilitate the restructuring of private debt, restructure and recapitalize pri- vate banks, and strengthen the regulatory and supervisory framework.14 As a result of the government's interventions, conditions in the financial sector as a whole improved in 2000, and continued to pick up in 2001. The proportion of nonper- forming loans was reduced and capital adequacy ratios of the banking sector increased. Nevertheless, the portfolio of mortgage institutions remains weak due to continuing problems in the housing sector. In response, the government has intro- duced capitalization and restructuring plans to be monitored by the Banking Super- intendency, and has developed a framework for the development of capital market instruments for long-term mortgage financing. The cost of resolving the financial sector's problems in Colombia, though significant, remains relatively low at about eight percent of GDP (in 2000) in gross terms. By way of comparison, the Interna- tional Monetary Fund estimates the costs of other crises at about 55 percent of GDP in Indonesia, and 20 percent in Korea and Mexico. The lion's share of resources in Colombia, over 60 percent, is expected to be devoted to the resolution of public 13. This decision was in turn reversed in late 2001 by a new Constitutional Court 14. In all, four public banks were liquidated or merged. Two private banks are in the process of being prepared for privatization or liquidation, and a number of smaller private insti- tutions were intervened and closed Five private banks were conditionally recapitalized A number of debt relief programs were implemented in 1998 and 1999, financed with public resources, to assist mortgage debtors through the refinancing of loans at lower interest rates and the reduction of loan principals. To facilitate the restructuring of cor- porate debts, credit lines administered by second-tier banks were introduced. The bank- ing superintendency also adopted a temporary, six-month regime defining the conditions under which the classification of restructured loans could be upgraded This was supple- mented by the passage of Law 550/1999, which suspended traditional bankruptcy pro- cedures for five years, and created an alternative, voluntary debt resolution mechanism based on agreements between debtors and creditors A financial sector reform law was enacted in June 1999, which, among other measures, strengthened prudential norms and the power of the authorities to deal with troubled banks in a timely manner MACROECONOMIC AND FISCAL FRAMEWORKS 155 banks, with a smaller, though at about 20 percent significant, proportion going to mortgage debt relief (IMF Selected Issues 2001). The limited maneuvering room available to the government to reduce its expendi- tures or to significantly increase revenues, combined with accumulated problems of fiscal mismanagement, meant that deep structural reforms were going to be needed if a significanit change in the trend of worsening fiscal indicators was to be accomplished. Along with debt service, the two most important areas of expenditure at the central level were transfers to the social security system and to departmental and municipal governments and the government's reform program aimed to address them One of the key obstacles facing the government in its fiscal consolidation efforts was the constitutional requirement that nearly half of all current revenues be trans- ferred to local governments. This meant that the government would keep only one out of every two additional pesos in revenues from enhanced tax collection or increase in tax rates, and was an impediment to expenditure reduction. A constitu- tional reform (Acto Leg7slativo) was therefore passed in 2001 that delinked territorial transfers from the central government's current revenues for a period of eight years, providing instead for a real growth in transfers of between two and 2 5 percent per year 15 At the end of the transition period, transfers would be calculated as a pro- portion of average current revenues of the previous four years. Another important achievement of the constitutional reform was the creation of a single fund, the Sistema General de Participaciones (SGP), comprising the three main types of territorial transfers: the situado fiscal, the participaciones municipals, and the Fondo de Crddzto Educativo (FEC). The latter transfer had acted as a gap-fill- ing mechanism to cover the shortfall in teacher salaries, and its size varied counter- cyclically as current revenues rose or fell. The FEC's inclusion in the SGP should put a cap on this type of transfer given that further discretionary transfers by the central government are avoided. Another important impact of the Acto Legislativo is the limit it places on the central government's operating expenses, with the exception of pensions, which cannot grow by more than 1.5 percent in real terms over the tran- sition period. Simultaneously, to support and sustain this reform, and to help insure against future needs to make further discretionary transfers to local governments, there was the need to improve the control and distribution of resources under the fiscal decen- tralization system given the expected reduction in transferred resources as a result of the Acto Legislatuvo. By the end of the 1990s, transfers had become an important 15. The government's original proposal was to maintain transfers constant in real terms This was rejected by Congress and the two to 2 5 percent real growth compromise was reached Because the government's calculations of fiscal savings associated with the reform were based on projected growth rates of 4 5 percent and above during the transi- tion period, it is now unlikely that these savings will materialize given the lower growth projections, at leasr during the first years of the transition On the other hand, the reform would ensure predictable transfer income for local governments. 156 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE threat to fiscal sustainability, not only because of their direct link to central govern- ment revenues, but also as a result of the transfer of resources having preceded the transfer of responsibilities. The ensuing lack of accountability led to growing deficits and debt at the local level and to a decline in the quality of public services, especially education and health. It was considered critical to complement the Acto LegisLatavo with a reform of the existing law (Law 60/1993), which specified the distribution of resources both by sector and regionally, and which effectively defined the effective- ness and efficiency in the delivery of social services. The new law that accomplishes this (Law 715/2001) promotes the transfer of responsibility by automatically certi- fying larger municipalities to autonomously manage the provision of education serv- ices, incorporates performance goals for the education sector, and facilitates the geo- graphical redistribution of resources and teachers to ensure expanded coverage and higher quality. The reform strategy also emphasized the correction of perverse incentives under which subnational governments and their lenders operated, and began to eliminate discretion on the part of the central government. In the short term, fiscal adjustment and increased control of resources at the decentralized level were to be accomplished by Law 617/2000. This law was enacted in response to the critical fiscal and debt conditions facing many subnational governments, and also to provide relief to enti- ties in the financial system with high exposure to subnational debt. The law provides partial guarantees of territorial debt contracted up to 30 June 2001, under the con- dition that the subnational government concerned subscribe to a fiscal adjustment program with the Ministry of Finance. The law also restricts current spending growth of territorial governments to 90 percent of observed inflations, and provides loans to subnational governments to finance downsizing of personnel. In addition to rationalizing the territorial transfer system and reforming the dis- tribution of funds to promote greater effectiveness in the supply of health and edu- cation, and to encouraging fiscal adjustment at the local level, the government's reform program also envisioned actions to provide a hard budget constraint for local governments through changes in regulations related to territorial debt, and a terri- torial tax reform which would allow local governments to draw on larger revenues to fulfill their responsibilities. With respect to territorial debt, a decree was issued that regulates the borrowing of territorial entities, which should contribute to the elimination in the discretionality in the treatment of debt and therefore to ending unsustainable borrowing and limiting bailouts. On the other hand, while a territo- rial tax reform proposal was presented to Congress in 2001, the proposal was lim- ited to administrative measures and would not have increased tax revenues substan- tially. In any case, the measure is unlikely to be approved by Congress. The looming deficits of the social security sector, and the unavoidable financing pressures implied by them, are the most serious problems confronting the fiscal accounts over the coming years. These pressures are the result of imbalances between contributions and benefits in the various public pension regimes and, more recently, the high operating costs of the health system under the Instituto del Seguro Social MACROECONOMIC AND FISCAL FRAMrWORKS 157 (ISS). The government has attempted to define consensual solutions to bring the sector into balance in the long term, and a controversial and incomplete reform pro- posal was presented to Congress in late 2001.16 This remains a maJor pending reform Some measures to address ISS operating deficits have also been taken, including a renegotiation of the entity's collective bargaining agreement, a restruc- turing of outstanding debt to providers, and a reduction of the backlog in surgeries. It remains to be seen whether these measures are sufficient to address ISS problems. Finally, to provide the overall framework for continued fiscal sustainability, the government has presented a Law of Fiscal Responsibility for Congressional approval. This law is intended to provide a basis for aggregate fiscal discipline through ex-ante constraints and agreements. The law's main elements are: the establishment of a pri- mary balance goal for the aggregate non-financial public sector; the reorganization of the budget system; the reinforcement of fiscal discipline and transparency through inclusion in the budget of contingencies, tax expenditures, and limits for forward budgeting; ensuring that unpaid commitments are included in the follow- ing year's budget; and the regulation of local government indebtedness (see Chapter on Budgetary Institutions). III. Main Issues and Options This section highlights a few of the main issues requiring priority attention over the coming years. The list is selective, not exhaustive. a) Fiscal Sustainability Despite progress on the reform agenda, a number of issues continue to undermine Colombia's economic performance. The first priority is returning the fiscal accounts to a sustainable path, a necessary, though far from sufficient, condition for growth. Sus- tainable fiscal policy is an essential component of strong macroeconomic performance. Fiscal deficits can result in slow growth rates since they often lead to rising interest rates and financial repression and an increase in the current account deficit and, in turn, real exchange rate appreciation. Low and stable fiscal deficits can increase prospects for growth, which in turn allow for further sound fiscal management, thus creating a vir- tuous circle (Easterly, Rodriguez, and Schmidt-Hebbel 1994). Two medium-term scenarios (2002-10) were developed to illustrate the potential impact of continuing the pursuit of adjustment and structural reform versus the case where reforms are abandoned and the fiscal imbalance widens (Table 3) The results are consistent outcomes under expected parameters of the domestic economy and 16. The proposal does not address the reform of the special pension regimes granted to selected subgroups of the public sector. A proposal to address these regimes is to be made by the government dtiring the first legislative session of 2002 158 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE Table 3. Two Medium-Term Scenarios High Case Scenario 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Annual Percentage Change Real GDP 16 2.1 30 33 35 3.7 4.1 4.5 4.5 4.5 Inflation (CPI) 7.7 6.0 6 5 6 0 5.0 4.0 4.0 4.0 4.0 4 0 Percent of GDP National Savings 11.9 12.4 13.2 13.4 138 14.2 14.7 149 152 155 Gross Dom Investment 14.8 15 1 15.8 16 5 17.6 18.1 18 6 19.1 19.6 20.5 Current Account Balance -2.6 -3 2 -3 6 -3 4 -3.8 -3.5 -3 4 -3 3 -3 1 -3 2 NFPS Balance -3 4 -3.1 -3 7 -3 1 -2 5 -2.1 -1 7 -1 7 -1 6 -1.5 Primary Balance 1.1 1.5 1 3 1 5 1.7 2.0 2 4 2 9 3.2 3.1 Public Sector Debt 43 0 43 5 45 0 46 2 47 5 47.9 48.1 48 9 491 50 2 Low Case Scenario 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Annual Percentage Change Real GDP 1 6 1 9 2 2 2 3 2 4 2 1 2.2 2.0 2.0 2 0 Inflation (CPI) 7 7 7.0 6 8 6.9 6.5 6.0 5.5 5 5 5 4 5 5 Percent of GDP National Savings 11 9 12 2 12 4 12.6 12.9 12.9 13 0 13.6 13.5 13.4 Gross Dom Investment 14 8 15.2 15.5 15.7 16 0 15.9 16.2 16.5 16.5 16.4 Current Account Balance -2 6 -2.8 -3.0 -2.7 -2 6 -2 4 -2 3 -2.1 -1.9 -1.8 NFPS Balance -3 4 -3.5 -3.9 -4 7 -5 6 -5.3 -5.9 -6.2 -6 3 -7 0 Primary Balance 1 1 1.4 0.9 0.5 -0.1 0 3 0 1 0.2 0.6 0 8 Public Sector Debt 43.0 43.8 45.4 47.2 48 8 52 9 58 7 65.1 72.9 82.2 international economic conditions. In the reform, or high case, scenario, the nonfi- nancial public sector would enjoy a primary surplus of three percent of GDP in the medium term. Fiscal adjustment and continued declines in inflation would support an increase (depreciation) in the real exchange rate. A more competitive real exchange rate and low domestic interest rates would in turn lead to an expansion of economic activity, and medium-term growth rates would converge to 4.5 percent of GDP Under the low case scenario, fiscal accounts deteriorate due to lagging reforms, and debt service payments grow. The deficit increases from 3.3 percent in 2001 to seven percent in 2010. Growth stagnates, not rising above two percent per year, and investment and savings rates remain low. Given the elasticity of poverty rates to growth, the low case scenario would imply that poverty would remain almost unchanged: sluggish growth of two percent per year would decrease poverty rates to 54 4 percent by 2010, only one percentage point less than today's level (Velez 2001). These scenarios coincide with other studies that have estimated that a significant primary surplus will be necessary over the medium and long term to ensure fiscal sustainability. Easterly and Yuravlivker (2000), in a comprehensive exercise that MACROECONOMIC AND FiscAi. FlUMEWORKS 159 accounts for Colombia's contingent liabilities, calculate that a permanent primary surplus of about 4 percent of GDP will be needed ro ensure the sustainabiliry of fis- cal accounts The Comisuin de Raczonalzzacic5n del Gasto y las Fznanzas Pablhcas (1997) also found that a primary surplus of about 4 percent of GDP per year would be needed to ensure debt sustainabiliry. Posada and Arango (2000) conclude that debt sustainability would be ensured over the next 50 years if primary surpluses between one and 1.5 percent of GDP were generated. Of course, the longer it takes for primary surpluses of these levels to be generated, the larger will be the surpluses required in the future to reach a sustainable level of debt The Debt Sustainability Chapter also estimates that a significant primary surplus will be needed for debt sustainabiliry in the medium term. To avert an explosion of the debt-to-GDP ratio, the combined public sector has to generate a permanent pri- mary surplus of about four percent of GDP.'7This calculation incorporates the pres- ent value of all net public sector liabilities, including public pension payments (of the implicit liabilities) under a no-reform scenario; the scenario also assumes long- term GDP growth rates of 3 5 percent. In 2001, the public sector primary balance stood at 1 4 percent of GDP. In other words, the Colombian public sector will have to generate 2.6 percentage points of GDP in additional savings on a permanent basis. This estimate does not take into account the incoming administration's pro- gram for significantly increasing military and social expenditures over the coming years. This level of fiscal adjustment is a necessary and sufficient condition for ensur- ing fiscal sustainability Even under the above relatively benign scenario, if the primary surplus were kept permanently at four percent of GDP, it is estimated that the total net debt of the public sector would nevertheless continue increasing, reaching about 57 percent of GDP by 2010. To further highlight the vulnerability of Colombia's fiscal path in the absence of comprehensive reforms, one can project the path of public sector debt under the assumption that the primary balance is kept constant at its 2001 ratio with respect to GDP (1.4 percent) Under this scenario, debt would rise from about 46 percent of GDP in 2001 to about 80 percent in 2010 (assuming an annual real GDP growth rate of 3.5 percent) On the other hand, to maintain the debt-to-GDP ratio at today's level of 46 percent of GDP, primary surpluses berween 4 and 6 per- cent of GDP would have to be maintained over the next decade. The magnitude of the required adjustment means that significant measures on both the expenditure and revenue fronts are needed b) Expenditures The challenge in tackling expenditures in Colombia is substantial: a significalit pri- mary surplus needs to be generated, as mentioned above, while simultaneously cre- ating fiscal space for a number of emerging expenditure demands, such as the likely 17 See the Chapter on Public Debt Susrainabiliry and Management for derails. 160 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE push for greater military spending, expanding direct and contingent liabilities, and spending on infrastructure. As has been mentioned, central government spending in Colombia grew rapidly over the past decade, from 11 percent of GDP in 1990 to 20 percent in 2001. This growth was due to new constitutional requirements for increased transfers and greater spending on security and the judicial sector, and was a result of generous col- lective bargaining agreements (for example, teachers, ISS), and a failure to control other expenditures related to salaries at both the centralized and decentralized lev- els. In order of magnitude, the largest expenditure categories at the central govern- ment level are related to (a) territorial transfers, (b) debt service, (c) transfers to the social security system for pension payments, and (d) salaries. Transfers and debt serv- ice payments are also the categories that have experienced the most rapid growth during the 1990s (Figure 4). Looking to the future, and in the absence of additional adjustment measures, all of these spending categories, with the possible exception of salaries, are expected to grow as a proportion of GDP. i. Territorial Transfers One of the priorities of the current administration has been to address the growing burden on expenditures posed by territorial transfers. The emphasis has been on limiting the growth of transfers and delinking them from the central government's current revenues, and rationalizing expenditures by local governments and increas- ing efficiency and effectiveness of these expenditures. As mentioned above, the Figure 4. Growth of Expenditures 0.12- 0.2 -- Interest Transfers 0. I __Wage bill - - - - Goods and Services ._ c,0.08______________ 0 06- - _ c X 0.04 -- 0 I I I I IIIIII I 0.02~ ~ ~ ~~~~~~~ O _ O n IT VI, \0 ts X C, O CN C O CN cl cl Gs CE Cl, \ O X cl Co NC E CE > G MACROECONOMIC AND FISCAL FRAMEWORKS 161 recently approved Acto Legislativo places limits on the real growth of territorial trans- fers, but at the same time imposes a lower threshold on their growth during the eight-year transition period. Thus, as a proportion of GDP, transfers will only decline If the economy grows at a rate greater than two to 2.5 percent, and unless growth performance improves substantially, territorial transfers will continue to pose a burden on expenditures In effect, in this case expenditure adjustment is more the result of rapid growth than of spending cuts. THE MOST1- EFFECTIVE MEANS OF CAPPING TERIJTORIAL TRANSFERS IS 10 ADDRESS THE HIGH COSTS OF PERSONNEL PAYMENTS IN HEALTH AND EDUCATION The bulk of terri- tornal transfers go to financing health and education spending at the local level. In turn, the lion's share of health and education spending is consumed by salary pay- ments. Unless this is dealt with through, for example, the empowerment of munici- palities to fully manage the health and education sectors, the renegotiation of public teachers' salary and promotion scales, the introduction of greater competition through the possibility of hiring private sector teachers, and an accelerated shift in health sub- sidies from supply to demand, it is unlikely that measures in other areas will be suffi- cient to significantly attack the fiscal deficit. The newly approved reform of Law 60 (Law 715/2001) aims to address some of these issues. Continued efforts to rationalize expenditures in education and health will be key both in limiting the danger of pres- sures for further discretionary increases in transfers and, just as Important, in helping ensure the effectiveness and efficiency of these expenditures. The Chapters on Decen- tralization and Health and Education contain further discussion of these themes. IN ADDITION, THE REVENUE BASES OF DEPARTMENTS AND MUNICIPALITIES NEED 1-O BE IMPROVED THROUGH THE PASSAGE OF A TERRITORIAL TAX REFORM BILL. At the local level, unless territorial governments possess an adequate tax base, there is a risk that transfers will remain insufficient to cover their needs, particularly because responsibil- ities are increasingly transferred to them as mandated by Law 715. Under these cir- cumstances, the central government will continue to be faced with an implicit liabil- ity in terms of territorial debt. An illustration exemplifies the magnitude of foregone revenues. the actual land tax rate is about 0.4 percent, although by law (Laws 14 and 44) a rate of 1 6 percent is permitted Moreover, the cadastral valuation (avaluo cata- stral) is typically only 40 percent of the commercial valuation (avaluo commercial), and only 60 percent of the respondents in a survey (1997) indicated that they in fact pay the land tax. This means that municipalities currently collect only six percent of the potential revenues to be obtained from collecting the land tax (World Bank 1998). FURTHER, AN EFFORT TO IMPROVE THE QUANTITY AND QUALITY OF INFORMATION AT THE DEI)ARTMENTAL AND MUNICIPAL LEVELS IS URGENTLY NEEDED. Data availability and reliability at the territorial level is poor An effort is underway (Programa Para el Fortalecimiento del Sistema de Informaczdn Territorial, FOSIT) to make improve- ments in this area, which urgently needs to have higher prioriry. The success of 162 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE efforts to rationalIze subnational expenditures, incorporated into Laws 617/2000 and 715/2001, and the upcoming fiscal responsibility law, depends on the avail- ability of reliable data on revenues and expenditures at the territorial level. Cur- rently, such information is often contradictory, with large inconsistencies among the various sources (Consejo Superior de Politica Fiscal [CONFIS], Central Bank, Con- traloria, and so forth). ii. Debt Service The "bottom line" in fiscal sustainability analysis is the behavior of public debt with respect to GDP as a result of the long-term restrictions faced by fiscal policy. In Colombia, while GDP grew by an average 2.6 percent between 1991 and 2001, interest payment on debt grew nearly five times as fast in real terms, and domestic interest payments grew over eight times as rapidly as GDP This growth, in addition to being the result of the contracting of new debt, has also been due to the effect of the devaluation of the peso and of low growth during the period; it has been calcu- lated that the impact of these two variables on the level of debt between December 1997 and December 2001 has been about 10 percent of GDP (Documento Asesores 2/2002, CONFIS, 7 March 2002). Total public debt service currently makes up one fourth of the central government's expenditures, more than twice the central gov- ernment's investment budget. Of this, internal debt service consumes the larger por- tion, an estimated 13 percent of total expenditures in 2002, up from a mere three percent in 1990. External public debt service declined during the first half of the 1990s, from near 9 percent of expenditures in 1990 to three percent in 1996, but has since risen substantially to an estimated 10 percent of central government spend- ing in 2002. Following Fischer and Easterly (1990), the annual growth of public debt is equal to the growth of the primary fiscal deficit, minus the portion of the deficit that is financed by seniorage income, plus the nominal interest payments on public debt. As a proportion of GDP, public debt falls with inflation and with economic growth. The implication is that debt dynamics depend on the sign and on the difference between the real interest rate and the real growth of GDP: debt primarydeficit seniorage deb AGDP GDP GDP + (real interest rate- GDPgrowth e) GD If the real interest rate is lower than the rate of growth of the economy, primary deficits can be maintained that are larger than seniorage income without increasing the debt ratio. With an interest rate higher than GDP growth, this is not possible. If the primary deficit exceeds seniorage income and if the real interest rate exceeds GDP growth, debt as a proportion of GDP will grow explosively. It has been esti- mated that even the most dynamic economies can expect maximum revenues from seniorage on the order of two to 2.5 percent of GDP. Under such circumstances, if the real interest rate exceeds the growth of GDP, primary deficits that exceed senior- age income cannot be maintained without generating an explosion in debt ratios. In MACROECONOMIC AND FISCAL FRAMEWORKS 163 Colombia, the average real interest rate over the past 20 years has been about five percent, while the average GDP growth has been about three percent. Under such a scenario, and with projections for insufficient primary surpluses if further fiscal reforms are not undertaken, Colombia's fiscal accounts will simply become unsus- tainable. The Debt Sustainability Chapter analyzes this subject. Here, only two points will be highlighted. FIRST, WHILE DEBT MANAGEMEN'I' STRATEGIES MAY HELP TO MARGINALLY REDUCE THE BURDEN OF COLOMBIA'S PUBLIC DEBT, THE KEY IS THE CONT'ROL OF THE T'OTAL STOCK OF (NONFINANCIAL) PUBLIC DEBT AS A PROPORTION OF GDP, WHICH HAS DOUBLED SINCE THE MID-19905 TO ABOUT 45 PERCENT OF GDP IN 2001. It should be remembered that realistically, given the dynamics of Colombia's projected future fiscal balances, a reduction in the stock of debt is highly unlikely This underscores the importance not only of fiscal prudence, but of the efforts to generate growth lev- els that are closer to Colombia's level of potential GDP growth, estimated to be about 3.5 to four percent per year. While fiscal prudence is a necessary condition for sustained growth, it is far from sufficient, stressing the importance of a return to sig- nificant levels of investment, to enhanced productiviry and competitiveness, and to an emphasis on the creation and maintenance of human capital. All of these are top- ics addressed at length in other accompanying Chapters SECOND, THE SIGNIFICANT GROWTH OF DOMESTIC DEBT IS A RELATIVELY RECENT DEVELOPMENT THAT NEEDS TO BE MONITORED CLOSELY Net domestic debt of the NFPS as a proportion of GDP increased by 50 percent between 1998 and 2001. A significant proportion of this debt is concentrated in relatively few hands, including domestic pension funds, ISS, the Caja Nacional de Prevision Social (CAJANAL), and the financial system While there has been a concerted effort by the administration in recent years to limit its reliance on the domestic financial market in order not to crowd out the private sector, Colombia's continued limited access to external capital markets and its large projected financing requirements in the medium term make it likely that the domestic market will be required to absorb significant amounts of public paper. Greater attention to this issue is needed. iii. Pensions The issues of pensions and pension reform are taken up in another chapter, so they will be only briefly mentioned here in terms of fiscal importance. Today, pension- related transfers constitute the third-largest expenditure category for the central gov- ernment, at just over two percent of GDP in 2000, and its largest liability. Without a pension reform that addresses the issues of the transition period, it is projected that these transfers would at least double by 2010, and may triple, depending on the shape of the reform. The pension reserves of the ISS began to be drawn down in 2001, and estimates indicate that reserves will be exhausted by 2007, at which time the govern- ment will be faced with making the necessary transfers to cover ISS pensionis 164 COLOMBLA THE ECONOMIC FOUNDATION OF PEACE FOR THE FISCAL IMPACT OF A REFORM TO BE FELT IN THE NEAR TERM, A REDUCTION OF THE TRANSITION PERIOD FROM THE OLD REGIME TO A REFORMED ONE IS NECES- SARY. While the 1993 Social Security law was a major step forward in that it recog- nized the contingent liabilities of the system and created a new, actuarially improved one, it suffered two major problems. The transition period between the old and the new regimes, during which certain beneficiaries could maintain the benefits of the old regime, at 20 years, was too long. Since the old regime was not financially sound, the contingent liability of the public sector is still growing. The excepted regimes are the other problem-defense, teachers, and ECOPETROL workers kept special retirement benefits, and their contributions are well below what is required to finance those benefits. Reforming the special regimes and bringing them in line with the general system is necessary for longer-run actuarial soundness, although it should be noted that in the near term this will imply larger fiscal outlays by the gov- ernment to cover its contribution as employer. iv. Public Sector Salaries As a result of political resistance, the reduction in the public sector's wage bill has been tackled through wage reductions and freezes rather than through a contraction in the size of the public sector.1" This approach is unsustainable and undesirable because it affects the quality of the public sector. It is also not politically feasible in the longer term. EVEN THOUGH SAVINGS IN ABSOLUTE TERMS MAY BE MINOR, A REFORM OF THE PUB- LIC SECTOR THROUGH LIQUIDATION OF ENTITIES, MERGERS, AND SO FORTH, IS NEEDED IN ORDER TO CREATE FISCAL SPACE, IF ONLY FOR REAL WAGE GROWTH AND IMPROVEMENTS IN QUALITY, EFFICIENCY, AND EFFECTIVENESS. In this regard, a seri- ous review of expenditures in the sectors of justice and defense is required. These two areas account for fully 70 percent of the central government's wage bill (and constitute an equal proportion of spending on goods and services. Given general security conditions in the country, these outlays have not often come under serious scrutiny, nor have they been characterized by accountability or transparency. Never- theless, given the likelihood that expenditure in these areas, particularly defense, will increase in coming years, such a review is needed. c) Net Worth and Illusory Adjustment Following Easterly (1999), fiscal adjustment is considered illusory if it results in the reduction of assets of the state. This raises the issue of the reduction in the level of public investment in recent years in Colombia, largely as a result of the magnitude of other expenditure obligations and their inflexibility. 18 An exception has been the apparently large layoffs, reportedly on the order of 65,000, undertaken at the territorial level, although reliable information on this is scarce MACROECONOMIC AND FISCAI FRAMEWORKS 165 In addition to the question of the magnitude of public spending, the issue of its inflexibility is important. While a number of significant measures have been taken to tackle expenditure reduction, including a real reduction in some public sector wages, upper limits to the growth of general operating expenditures as included in the Acto Legislativo and Law 17/2000, and restrictive budgets, Colombia has tended to rely on GDP growth for fiscal correction This is partly due to the difficulties posed by reducing inflexible expenditures. Three highly rigid categories, debt serv- ice, transfers, and salaries, grew to repiesent 86 percent of central government spending in 2001. As can be seen in Figure 5, this has resulted in the squeezing of investment outlays, which have in effect been treated as a residual.19 The result has been a decline in central government investment, which by the end of the 1990s had fallen to less than half its 1990-97 level. Private investment has not only been uniable to fill the resulting gap, but has declined dramatically over the past five years to a current level of under 8 percent of GDP. INFORMATION REGARDING COLOMBIA'S INVESTMENT GAI IS LACKING, AND I-1-S AVAIL- ABILITY WOULD MAKE AN IMPORTANT ANALYTICAL CONTRIBUTION An investment gap has Important effects on productivity and production costs, with effects on export competitiveness and foreign trade, and on the profitability of capital and investmenit. Through all these channels, Colombia's growing investmenit gap will lead to slower growth unless it is addressed.20 d) Revenues While new expenditure obligations were created in the 1990s, the formulation of a coherent strategy of revenue generation met with less success. Between 1990 and 2000 six major tax bills were passed (Figure 6). The total impact of the bills through 2001 has been an increase in revenues of about four percent of GDP, while central 19 It is important to note that information on public investment suffers from inconsistency of definition and from dubious reliability, particularly at the local government level Investment by local governments ostensibly makes up around 2 5 percent of GDP The bulk of this investment is in the social sectors, financed by intragovernmental transfers. Outlays are often misclassified as investment spending in order to comply with the requirements and earmarking of transfers. A significant portion of this type of spending would not fall under the definition of fixed capital formation 20 A related theme within the framework of public sector net worth and the stock approach to fiscal adjustment are Colombia's liabilities, both direct and contingent These have been estimated at 180 percent of 1997 GDP, and include public pension liabilities (by far the largest), and bailoLits of financial institurions, territorial debt, infrastructure con- cessions, and the achievement of peace and its concomitcant expenditures (Echeverry-Gar- zon and Navas-Ospina 1999) Similarly, a number of public sector "contingent assets" are estimated at about 55 percent of GDP, comprising oil, gas, and coal reserves, and the electromagnetic spectrum 166 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE Figure 5. Composition of Expenditures 100~ [ 7 80 _---- 70 - 60- U50 - 40 - - 30 --- -- 20 - - - 10 60-80 80-84 84-92 92-00 D Operating U Investmeint F] Interest Figure 6. Major Tax Reforms, 1990-2000 7- - 20 5 I- 15 law law 0 3 338/97 633/005law 8810 1~ ~ ~ ~ ~ ~~~~~fn - -- cs~~~~~~~~~~~~~~ ° G law Ta- Rvn l 49/906/92 1 - GDP Growth - Tax Revenue Growth MACROECONOMIC AND FISCAL FRAMEWORKS 167 government expenditures grew by over nine percentage points of GDP. The tax reforms of 1990 and 1992 were motivated by structural reforms of the economy and by the need to reduce distortions in the tax structure, while later reforms, in 1995, 1997, and 2000, originated in fiscal crises and the need for stabilization. The approach has been an incremental one: a tendency toward a unification of tariffs, the widening of bases, especially that of the VAT, and the simplificationi of tax adminis- tration; the use of tax retention and prepayment; stimulation of domestic and for- eign investment through reduction of rates and granting of tax incentives; shifting the tax structure toward indirect taxes, which are easier to levy; and stimulating domestic capital markets (Alonso, Olivera, and Fainboim 1998). This incremental nature has ultimately resulted in a tax structure that is excessively complex. A COMPREHENSIVE TAX REFORM Al THE NATIONAL LEVEL IS REQUIRED. Colombia's tax structure has deteriorated in recent years, with a significant erosion of the tax base through increasinig incentives and exemptions. A widening of tax bases through the elimination of various exemptions and a simplification of the VAT structure (partially accomplished in the most recent tax reform) is needed. An overly complex tax system has created incentives for noncompliance, and tax evasion in Colombia is high by international standards This is exacerbated by the existence of a large informal sector, which contributes only marginally to fiscal revenue. Future tax reforms should aim to expand the tax base rather than increase rates, reduce ear- marking, and improve the administration and collection of taxes (see the Chapter on Taxation Issues for more detailed options). The government has commissioned a comprehensive review of Colombia's revenues (Mzszdn de Ingresos), the recommen- dations of which will be made public in the near future. e) The Central Bank Colombia's macroeconomic stability and management are challenged by institutional weakness. While stability in the "rules of the game" is acknowledged as being key to econotnic growth, it faces a formidable obstacle in Colombia's extremely detailed 1991 Constitution and in the activeness of the Constitutional Court in interpreting and enforcing its prescriptions. This has also meant that even small alterations in eco- nomic institutions require constitutional amendments, and are thtis faced with lengthy and highly politicized discussions between the executive, legislature, and judi- ciary. While it is beyond the scope of this Chapter to comment on the myriad insti- tutions that touch upon economic life, we will briefly discuss the Central Bank because it is a key institution for macroeconomic management. The desirability of an independent Central Bank is by now a widely accepted tenet of economic policymaking It is thought to lead to greater stability in mone- tary policy and to lower inflation In Colombia, Central Bank independence was granted in the 1991 Constitution Nevertheless, as testimony to Colombia's peren- nially consensual policymaking, the Constitution also decreed that this independ- 168 COLOMBIA THE EcONOMic FoUNDATION OF PEACE ence was to be constrained in a number of ways. While inflation reduction is stated as the Central Bank's primary objective, the Constitution also states that it is to be attained through consensus and discussion with the government. The Constitution stressed, in particular, the need for the government to maintain an active role in the choice of exchange rate policies and regimes, implying that the government would have an important role in managing monetary policy. Constitutional Court rulings have further muddied the waters by questioning the precedence of the Central Bank's inflation targets. Moreover, over the years, the Central Bank's independence has been threatened outright by Congress, and more subtly, particularly in recent times of economic recession, by calls for it to engage in monetary emission to "sup- port growth." The consequences of less-than-full independence of the Central Bank were evi- dent in 1996 and 1997, when its monetary stance was particularly loose, contribut- ing to financial turbulence. Even today, the government is actively engaged in exchange rate policy through its monetization of proceeds from sovereign debt issues on international capital markets; movements in the exchange rate throughout 2001 were influenced not by Central Bank actions but by announcements from the Min- istry of Finance regarding the timing of repatriation of these proceeds. The ability of a sitting president to name at least one third (two of six members) of the Board of Directors may compound the political nature and timing of decisions. IN SUM, THE EVIDENCE IS THAT THE CENTRAL BANK ACTUALLY NEEDS A GREATER, RATHER THAN LESSER, DEGREE OF INDEPENDENCE TO MANAGE MONETARY POLICY. The influence of any one administration on the Board can be diminished by further staggering nominations and by precluding those who have served in the current administration from serving on the Board. IV. Conclusions As has become apparent over the past decade, Colombia's fiscal problems are not merely cyclical, but structural. Continuing fiscal adjustment and institutional changes are needed to address these problems. Continued efforts on both the rev- enue and expenditure fronts will be required to ensure a return to fiscal sustainabil- ity, particularly in the face of expanding liabilities from, for example, public pen- sions and increasing pressures for expanding military expenditure, to name but two. It should be borne in mind, however, that whlie revenues can be increased through a comprehensive reform that makes the structure less complex and more efficient, the margin for increasing revenues significantly in the short term, without negatively affecting growth, is likely to be small. Passage of a tax bill that does not correct the deficiencies of the current regime will only necessitate, as it has in the past, further rate increases within a year or two as "tax fatigue" sets in following a temporary rise in revenues. In the effort to render the tax structure more efficient, MACROECONOMIC AND FISCAIL FRAMEWORKS 169 serious considerationi needs to be given to a reduction or reassignation of Colombia's high payroll taxes Given the structural character of fiscal deficits in Colombia and the need to gen- eiate significant primary sLirpluses, major efforts in expenditure reduction are fun- damental. Expenditure adjustment has often been inefficient and inequitable, because it has at times concentrated on short-term measures with relatively low political costs, such as de facto investment cuts, payroll freezes, or wage increases below inflation These are unsustainable in the longer term and form part of what has been called an "illusory fiscal adjustment" as the assets of the state deteriorate, undermining future growth. While the inflexibility of expenditures makes spending reduction a difficult challenge given the large combined burden of debt service pay- ments, transfers, and public sector salaries, combined with emerging obligations and liabilities, spending cuts are imperative to achieving sustainability in the fiscal accounts This will necessarily require further politically difficult reforms. A number of options are available and have been considered, though not yet implemented, in Colombia Primary among these is addressing pension transfers through a cut in the transition period between regimes. Other avenues include the review of expenditures in the )ustice and defense sectors, which consume fully 70 percent of the central gov- ernment's personnel and goods and services outlays. Regarding territorial transfers, as these are defined in the Constitution, the degrees of freedom are limited, although an emphasis on addressing the high proportion of health and education spending that is dedicated to payroll and benefits would help to ensure that future demands for additional transfers are limited. Changes during the 1990s fundamentally altered economic institutions, with perhaps the biggest being the acceleration of the drive toward decentralization. While this responded to a national political desire to bring government closer to the people, it also created the impetus for larger expenditures within a loosened budg- etary constraint In response to widespread severe fiscal imbalances and unsustain- able levels of debt at the local level, the emphasis in recent years, understandably, has been on the restriction of decentralized resource assignment and its increased man- agement from the center (provisions in Law 715/2001 are an example of this) This has negative longer-term consequences because Colombia will remain in a halfway house between a centralized and decentralized state, where territorial governments will have little incentive to responsibly manage resources the control of whilch con- tinues to reside at the central level, with important fiscal consequences. The enhancement of service delivery, one of the main justifications for decentralization, may be compromised at the very time that human capital improvements are key to enhance Colombia's competitiveness and growth prospects. Rather than backtrack- ing on decentralization under the guise of fiscal restraint, local governments need to be empowered to manage central transfers and to raise own revenues within the con- fines of fiscal responsibility. The Decentralization Chapter addresses these issues and gives examples of successful efforts, including reforms in education in Bogoti and other municipalities. 170 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE The government has presented a fiscal responsibility law for discussion and approval by Congress. This is a commendable effort and follows on a number of experiences with similar laws in the region. For such a law to provide an effective constraint in Colombia, improvements in data collection and accounting and budgeting need to be made, particularly ensuring realistic revenue projections. Spe- cific fiscal targets and explicit sanctions in the event that these targets are not met are essential if the law is to be truly effective. Moreover, within any formulation of fiscal responsibility laws, the rules governing the accumulation of resources and their use in the Petroleum Stabilization Fund (FAEP) and royalties allocations should be revisited. This is particularly important because Colombia projects major oil discoveries over the next decade. As international evidence shows, it is far easier to establish such rules before the arrival of the resources when a "voracity effect" is certain to take hold. While the FAEP has accumulated relatively significant resources over the years (about US$1.5 billion), these have nevertheless fallen short of expectations and have not had an impact on curbing expenditures Macroeconomic and fiscal stability is only one pillar for development; the others are well-functioning markets, stable laws, an efficient and effective bureaucracy, good infrastructure, and abundant human capital. While fiscal rectitude and stable monetary policies may be some of the necessary conditions for a resumption of robust economic growth, they are not sufficient. Growth depends on engaging in activities that increase the net worth of the country. Investment, a boost in external competitiveness, and increased productivity are necessary. With Colombia's large fis- cal obligations and limited resources, the country can i1ll afford to have one of the highest labor costs in the region, where payroll taxes and parafiscal payments con- stitute over half of the payroll. Instability in the rules of the game, frequent tax regime changes, Constitutional Court rulings that fundamentally alter economic relations, and repeated bailouts of various sectors of society have compounded the uncertainty associated with Colombia's internal conflict, and have resulted in record-low private investment levels. To achieve the desired outcomes within tight budgetary constraints, the gov- ernment has to improve the efficiency of public spending. A demand for resources that surpasses supply means that there is an increasing need for priori- tization of expenditures and for an enhanced emphasis on accountability and transparency. References Acosta, 0. L. 2001. "Gasto Puiblico Social y Arquitectura Financiera." Bogota. Alesina, A. 2001. "Institutional Reforms in Colombia." Washington, D.C., Inter- American Development Bank. Alonso, J. C., M. Olivera, and I. Fainboim. 1998. "La Sostenibilidad de la Politica Fiscal en America Latina: el Caso de Colombia." Bogota: FEDESARROLLO. MACROECONOMIC AND FISCAL FRAMEWORKS 171 Clavijo, S. 1998. "Polltica Fiscal y Estado en Colombia." Bogota: Banco de la Republica, Universidad de los Andes. Comisi6n de Racionalizaci6n del Gasto y de las Finanzas Publicas. 1997. El Saneamiento Fiscal, Un Compromiso de la Sociedad, Tema P Diagndstico y Princi- pales Recomendaciones. Bogota Ministry of Finance and Public Credit. Dollar, D., and A. Kraay 2000. Growth is Goodfor the Poor. International Monetary Fund Seminar Series, No. 2000-35:1-44, March. Easterly, W 1999. "When is Fiscal Adjustment an Illusion?" Policy Research Work- ing Paper #2109. Washington, D.C.: World Bank Development Research Group. Easterly, W, C. Rodriguez, and K. Schmidt-Hebbel. 1994. Public Sector Deficits and Macroeconomic Performance. New York: Oxford University Press. Easterly, W, and D. Yuravlivker. 2000 "Treasures of Time Bombs~ Evaluating Gov- ernment Net Worth in Colombia and Venezuela." Washington, D.C.. World Bank. Echavarria, J. J., C. Renteria, and R. Steiner. 2002. "Decentralization and Bailouts in Colombia" (revised version) Bogota: FEDESARROLLO. Echeverry-Garzon, J. C. 2001. "Memorias de la recesi6n de fin de siglo en Colom- bia: flujos, balances y politica anticiclica." Boletines de divulgacidn economica, 7. Bogotr: National Planning Department. Echeverry-Garzon, J. C., and V. Navas-Ospina. 1999. "Confronting Fiscal Imbal- ances Via Intertemporal Economics, Politics and Justice: The Case of Colombia." Bogota National Planning Department. Fischer, S., and W Easterly. 1990. "Economics of the Government Budget Con- straint. World Bank Research Observer, 5:127-42, July. Garcia G., J., and S. Jayasuriya. 1997. "Courting Turmoil and Deferring Prosperity. Colombia Between 1960 and 1990." Washington, D.C.: World Bank. Gavin, M. 1997. "A Decade of Reform in Latin America: Has It Delivered Lower Volatility?" Working Paper Green Series, No. 349. Inter-American Development Bank, Office of the Chief Economist Washington, D.C. Gavin, M., R. Hausmann, R. Perotti, and E. Talvi. 1996. "Managing Fiscal Policy in Latin America and the Caribbean: Volatility, Procyclicality and Limited Cred- itworthiness." Working Paper 326. Inter-American Development Bank, Office of the Chief Economist. Washington D.C. Herrera, S., G. Perry, and N Quintero. 1999. "Output Fluctuations in Latm Amer- ica: What Explains the Recent Slowdown?" Washington D.C.: World Bank. IDB (Inter-American Development Bank). 1997. Economic and Social Progress in Latin America, 2001 Report Latin America After a Decade of Reform Baltimore: The Johns Hopkins University Press Lozano, 1. 2001. "Colombia's Public Finance in the 1990s: A Decade of Reforms, Fiscal Imbalance and Debt " Borradores de Economia #174 Bogoti: Banco de la Rep'iblica, Subgerencia de Estudios Economicos Posada, C. E., and L. E Arango. 2000. "Podremos Sostener la Deuda Pubhca?" Bor- radores de Economia, #165. Bogota: Banco de la Repiblica, Subgerencia de Estu- dio Econ6mncos. 172 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Velez E., C.E. 2001. "Fiscal Sustainability, Economic Growth and Poverty." Wash- ington, D.C.: World Bank . 2002. Colombia Poverty Report. Washington, D.C.: World Bank. World Bank. 1998. "Colombia Economic and Social Development Issues for the Short and Medium Term." Washington, D.C. Zapata, J. G., O.-L. Acosta, and A. Gonzalez. 2001. "Evaluaci6n de la descentral- izaci6n municipal en Colombia. Se consolid6 la sostenibilidad fiscal de los municipios colombianos durante los afios noventa?" Archivos de Economia, #165. Bogota: National Planning Department. 6 The Tax System This Chapter was written by Jose Gil-Diaz with the contributions of Vicente Fretes Cibils and Steven Webb. I. Introduction In the 1990s, public sector outlays, fiscal deficits, and the public debt rose sharply in Colombia The importance of oil revenues increased, but the allocation of resources became more difficult because of the already existing earmarking restrictions on the use of public resources, and the decision to channel most of the oil revenue outside the budget. As the size of the nonfinancial public sector increased sharply, it became an important factor in helping explain the better macroeconomic performance before 1990 than after. Reflecting these outcomes, in recent years the fiscal deficit and the sustainability of public debt have been sources of concern for Colombian policy- makers, private agents, multilateral institutions, and foreign private creditors. The crisis of 1998-99 was a breaking point in the economic history of the coun- try, growth was negative (-4 percent) and the public sector deficiti was more than 5 percent of the gross domestic product (GDP). Under the ad)ustment program approved in the fourth quarter of 1998, public expenditures have been contained, and tax collection increased, boosted by changes in the design of the main taxes. However, these changes may have introduced new distortions into the system. II. Background Total taxes increased from 16 1 percent of GDP in 1995 to 18.7 percent of GDP in 2001 (Table 1) The amount collected at the different institutional levels followed different paths The national government, which accounted for over 60 percent of the total, and I The public sector includes the nonfinancial public sector and the Central Bank. 173 174 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Table 1. Colombia-Structure of Taxes 1995 1996 1997 1998 1999 2000 2001 Percent of GDP TOTAL 16.1 16.8 15.8 17.5 16.7 17 4 18.7 Of which by type of government National Taxes 10.3 10.8 98 11.3 107 12.0 13.5 Taxes Collected by Local Government 2.3 2.5 2.7 2 7 2 7 2.7 2 6 Departments 10 10 10 10 0 9 10 10 Municipalities 13 15 17 17 18 17 1.6 Social Security Contributions 1.2 3 5 3 5 3.3 3.5 3.3 2 7 2.6 Of which by type of tax Income Tax 4 5 4.4 5.0 4 9 4 8 4.9 5.8 Tax on Banking Transactions 0.6 0 7 Consumption (VAT, etc) 6 1 6.9 7 1 6.8 6.6 7.1 74 Taxes on International Trade 1.0 0 9 1 0 1 2 0.9 1 0 1 1 Social Security Contributions 3 5 3 5 3.3 3.5 3.3 2 7 2.6 Other 1 0 11 1 1 1.1 1.1 1.1 1.1 Percent of Total TOTAL 100.0 100 0 100.0 100.0 100.0 100 0 100.0 Of which by type of government National Taxes 64 0 64 2 62 0 64 6 64 1 69.0 72 2 Taxes Collected by Local Government 143 14.9 171 154 162 155 139 Departments 62 60 63 5 7 5 4 5 7 53 Municipalities 81 89 108 9.7 10.8 98 86 Social Security Contributions 1,2 21.7 20.9 20.9 20.0 19.8 15.5 13.9 Of which by type of tax IncomeTax 27.9 26.2 31.6 28.0 28.7 28.2 31.0 Tax on Banking Transactions 3.4 3 7 Consumption (VAT, etc) 37 9 41 1 44 9 38 8 39 5 40.8 39 6 Taxes on International Trade 6 2 5.4 6 3 6 8 5.4 5 7 5 9 Social Security Contributions 21.7 20.9 20.9 20.0 19.8 15.5 13.9 Other 62 6.6 70 63 66 63 59 Memorandum items I Social security contributions and other payroll contributions are collected by institutions outside the central government 2 They include pensiois contributions anid health insurance Source Estimates based on data provided by CONFIS, IMF, and High Commission on Taxes municipalities, increased their collection. National taxes increased 3.2 percent of GDP, and taxes collected by municipalities increased 0.3 percent of GDP; in the same period, the taxes collected by departments were quite constant, at around 1 percent of GDP At the national level the most important taxes are the income tax and the value added tax (VAT) (Table 2). They provide 80 percent of the total collected by the central gov- THE TAX SYSTEM 175 Table 2. Colombia-National Taxes 1995 1996 1997 1998 1999 2000 2001 Percent of GDP Income Taxes 4 0 3 8 4.4 4.3 4 2 4 3 5 2 Corporate 2 2 2 2 2.7 2 5 2 4 2 5 3 0 of which from oil activities Personal 1 8 1 6 1 7 1 9 1.8 1 8 2 2 Tax on Banking Transactions 0 6 0.7 Value Added Tax 4 1 4 7 4 8 4 6 4 4 4 9 5 2 Excise taxes 0.6 0.7 0.6 0.5 0.5 0.5 0.6 of which from gasoline and APCM 06 06 05 05 05 05 06 Taxes on International Trade 1 0 0 9 1 0 1 2 0 9 1 0 1 1 Other 06 07 07 07 07 07 07 Total 103 108 98 113 107 126 135 Percent of Total IncomeTlaxes 38 9 35 0 27 6 38 2 39.3 35 8 38 5 Corporate 21 2 20.5 27 6 21.9 22 4 20 8 22 2 of which from oil activities Personal 17 7 14 5 17 0 16 3 16 8 15 0 16.3 Tax on Banking rransactions 5 0 5.2 Value Added Tax 39 8 43 6 48 9 40 6 41 1 40 8 38 5 Excise taxes 5 8 6 5 6 1 4 4 4 7 4 2 4.4 of which from gasoline and APCM 5.8 56 5 1 44 47 42 44 Taxes on International Trade 97 8.4 102 106 84 83 8 1 Other 5.8 6.5 7.1 6 2 6 5 5.8 5.2 'l'ax on Banking Transactions 100.0 100 0 100 0 100 0 100 0 100.0 100 0 Note Contributions on the payroll other than social secuLrity contributions are included as taxes Source Estiiiiares based oni dara provided by CONFIS, IMF, atid Higlh CommissioIn o Taxes ernmenit. In 1998, a tax on financial transactions vwth a rate of 0.2 percent was introduced as a temporary measure with the aim of financing the resolution of the banktng crisis that emerged that year. This tax became permanent in 2000 with a rate of 0.3 percent The oil industry is an important source of public resources for the central gov- ernment, Empresa Colombiana de Petroleos (ECOPETROL), and territorial entities. Resources are captured by a regime of royalties2 and the income tax on 2 There is a general regime of royalties on mining activities. Subnational entities receive roy- alties for mining production that go to producer departments (47.5 percent), producer municipalities (12 5 percent), and port muLlnicipalities (8 percent) The rest (30 percent) is transferred to the National Royalties Fund, whlich allocates resources in the producing regions primarily for purposes of investmiient The most important royalry is on oil exploitation 176 COLOMBIA: THE ECONOMIC FOUNDAION OF PEACE ECOPETROL and the private enterprises that share oil exploitation with ECOPETROL. The central government receives the income tax on the oil activity and its participation in ECOPETROL profits. Moreover, there is an oil financial fund (Fondo de Estabzilzaczdn Petrolera), which is fed with contributions from royal- ties and ECOPETROL, aimed at smoothing the use of oil resources. The fund is managed by the Central Bank. Since the beginning of the 1990s, taxes on foreign trade have been reduced, reflecting the trade reform carried out in the first part of that decade; the amount collected in 2001 was only 1 percent of GDP Social security contributions (pension contributions and health insurance) have lost relative weight in favor of the income tax, taxes on consumption, and the new tax on banking transactions. The tax collection of territorial entities averaged 2.6 percent of GDP from 1995 to 2000. Taxes on liquor, beer, and tobacco provide the bulk of the amount col- lected by departments (Table 3),3 while the main taxes collected by municipalities are a property tax (impuestopredial unificado) and a business tax (impuesto de indus- trza y comercio). Colombian tax collection differs from the experience of developed countries in the level and composition of tax collection (see Table 4). In particular, the rel- ative weight of the Colombian income tax of total taxes is below the percentage of this tax in the total collected by developed countries,4 where the personal income tax is the most important item. At the same time, the composition of Colombian tax collection differs from that of developing countries included in Table 4. The income tax has a higher weight in Colombia than in other devel- oping countries. In both cases, taxes on consumption are higher than the income, corporate, and personal tax, and the relative weight of the personal income tax is less than 10 percent of the total. For developed and developing countries, the rel- ative share of taxes on foreign trade in Table 3. Colombia-Taxes Collected the total is small. by Departments (1999) The present tax system is the result of decades-long reform.5 (See the Appen- Percent of the Total dix at the end of this Chapter for a sum- Liquor 29.8 mary of previous tax reforms.) Impor- Tobacco 14 7 tant lessons about present distortions Beer 25.5 and administrative difficulties can be Surcharge on gasoline 11.7 drawn from previous tax reforms.6 The Other 18 3 different reforms were the response to Total 100.0 pressing fiscal problems at the time of Source High Commission on Taxes. their implementation. 3. Colombia has not followed the international experience, no excise taxes are collected by the national government. 4 More than 25 percent of the Colombian income tax is levied on oil activities. 5. The changes introduced in the first part of the 1990s are analyzed in Shome (1995). [IHr TAX SYS I EM 177 Table 4. Comparative Composition of Tax Revenue (Average 1995-97) Developing OECD Countries Countries Western All America Pacific Europe AU Hemisphere Colombht Percent of GDP Income Taxes 13 9 15.3 15.7 13.5 4 8 3 3 4.6 Corporate 3 1 3.0 4 3 2 9 2 6 2 3 2.9 Personal 108 12.3 114 106 22 10 17 Consumption 10 2 5 7 6 9 11.3 6 0 7 1 7.2 General 6.6 3.7 4.3 7.3 3 6 4.8 Excises 3 6 2 0 2 6 4.0 2 4 2.3 Interniational Trade 0 3 0.3 0.6 0.3 3.5 2 6 1 0 Social Securiry Contributions 9 5 6 1 3.5 10 8 1.3 2 5 3 4 Total 33 9 27 4 26 7 35.9 15 6 15 5 16 2 Percent of Total Income Taxes 410 558 588 376 308 213 28.4 Corporate 9 1 10 9 16 1 8 1 16 7 14 8 17 9 Personal 31 9 44 9 42.7 29 5 14 1 6 5 10 5 Consumption 30 1 20 8 25 8 31 5 38 5 45 8 44 4 General 19 5 13 5 16 1 20 3 23.1 31.0 Excises 10 6 7.3 9.7 11.1 15.4 14 8 International Trade 0 9 1 1 2.2 0 8 22.4 16.8 6.2 Social Security Contributions 28.0 22.3 13.1 30 1 8.3 16 1 21 0 T'otal 1000 100o 0 1000 loo o 1000 loo loo 0 a Table I Source Tanzi anid Zee (2000) Law 75 of 1986 and the subsequent five tax reforms approved during 1990-2000 (Law 49 of 1990, Law 6 of 1992, Law 223 of 1995, Law 488 of 1998, Law 663 of 2000) define the present Colombian tax system. The laws of 1990 and 1992 were approved in the context of the reform process carried out by the Gaviria Adminis- tration (1990-94). Instead of helping with the reform process, the changes inserted during 1995-2000 were aimed at collecting more resources to close fiscal deficits, without taking care of the overall taxation system. The most important advances of the reforms approved in the 1990s were the uni- fication of the income tax rates for incorporated organizations and corporations, the unification of the rate for corporations and the marginal rate for natural persons, the introduction of a general regime of withholdings tax payments, and the payments through the banking system. 6 In this Chapter, a tax reform is a law that changes the tax basis and/or rates, or introduces new taxes 178 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE III. The Tax System: Issues and Policy Recommendations The Ministry of Finance has estimated that, beginning in 2003, the primary balance would have to be increased by at least 1 percent of GDP to maintain in 2010 the present ratio of public debt to GDP (50 percent) (CONFIS 2002).7 However, these calculations include neither an increase in public expenditures related to the armed conflict, nor any enlargement in public investment reflecting the need to increase the supply of public goods and services. New estimates indicate a gap of at least 2.3 percent of GDP instead of I percent,8 resulting from the need to increase (a) mili- tary expenditure (0.5 percent annually); (b) public investment to maintain the pres- ent coverage of public goods (0.5 percent annually); and (c) public expenditure for building penitentiary facilities (0.3 percent annually). Policy actions aimed at eliminating or reducing the fiscal gap would have to enhance economic growth9' 1 because the private sector is concerned about public deficits. The approach taken in this Chapter is that the needed fiscal adjustment would have to be shared by an increase in taxes and a reduction in expenditures. The Chapter discusses the means of raising the collection of the central government at least 1.5 percent of GDP annually. Some of the changes proposed, based on effi- ciency reasons, could however reduce the amount actually collected; the proposed tax increase has to be enough to absorb these losses. The suggested tax adjustment is framed by the following principles: Public expenditure is a more efficient tool for income redistribution than tax- ation. In general, redistribution through taxation is carried out by increases in taxes on the use of physical capital. Consequently, the minimum gross return demanded from a private project increases,jI reducing the growth rate of out- put and income. 7. The exercise assumes that the measures taken in the last two years (tax reform approved in 2002, law 617 and the constitutional reform on transfers to territorial entities) will produced 0 8 percent of GDP annually 8 These estimates are based on the preliminary findings about fiscal sustainability of the High Commission on Taxes. 9. There is an extensive literature on the quality of fiscal adjustments. In general, there is support for reducing current expenditures instead of increasing taxes. However, the tests provided by different authors have not been conclusive because of the statistical difficul- ties of separating causes and effects The fiscal adjustment of Ireland in the 1980s pro- vides an example of these difficulties; the fiscal consolidation occurred during the period of the soaring U.S. dollar The favorable competitiveness effect of the real depreciation might have overcome the negative output effect provided by the negative fiscal impulse. 10 International experiences about fiscal ad)ustments based on tax increases are not conclu- sive. The Clinton adjustment (U S. 1992) was successful; the De La Rua adjustment (Argentina 2000) was not. THE TAX SYSTEM 179 * Exemptions and exceptional regimes make administration more difficult and promote evasion and inequlties. * Developing countries share the experience that the increases in tax rates, above a certain threshold, boost informal markets Instead of raising rates, the adjustment primarily has to be based on expanding the tax base and elimi- nating exemptions a) National Taxes The two most important taxes in Colombia, the income tax and the VAT, present dis- tortions. The successive revisions introduced since 1990 have eroded their base (Shome, Haindl, and Schenone 2000), and facilitated significant amounts of evasion. There are several studies about evasion in Colombia. Based on the figures on income distribution, Steiner and Soto (1999) estimate that the corporate income tax evasion rate was 14 percent in 1995, and that the personal income tax evasion rate was 24 percent. Accordingly, the evasion rate for the whole income tax was 20 per- cent. Using the input-output matrLx, they also compared the potential with the actual VAT. For 1991, the evasion rate was 28 percent. In light of the earmarking of taxes and oil resources, the allocation of public resources trough the budget has become a matter of concern. Table 5 provides infor- mation on tax earmarking for 1999. Following are recommendations aimed at increasing tax collection and reducing distortions. They are based on Shome, Haindl, and Schenone (2000) and extensive discussions with Ulpiano Ayala i. Income Tax The main problems identified in the income tax are the large number of distorting incentives included in the corporate income tax, and the several exemptions granted by the personal income tax. CORPORATE INCOME TAx. The corporate income tax is levied on the world income of national firms, and on income gained by foreign firms in Colombia. The general rate is 35 percent. There is also a minimum presumptive tax on net wealth with a rate of 5 percent; this tax is not neutral because it creates a bias in favor of larger debts 12 The Colombian law allows firms to contract with the administration the maintenance of the present tax rules for 10 years, paying an additional rate of 2 per- cent. The remittances of dividends pay a special tax with a rate of 7 percent How- ever, gains are exempted if they are invested in the country There is also a credit for supporting the employment creation. I I This issue has become relevant regarding globalization For a discussion sce Famzi (1995) 12 The tax may create conditions to make back-to-back operations for taxpayers with resources held outside Colombia, thereby reducing the net wealth 180 COLOMBLA THE EcONOMic FoUNDATION OF PEACE Table 5. Colombia-Central Government-Earmarked Taxes (1999) Earmarked Portion as Tax Percent of the Total Objective Income tax, VAT, tariffs, Transferred to other public institutions gasoline, stamps, financial 24.5 to support education and health of low- transactions income population Income tax, VAT, tariffs, Transferred to other public institutions to gasoline, stamps 20.0 support housing, sewerage, sports and health, and education infrastructure Financial transactions 75 5 Restructuring the financial sector Surcharge gasoline 100 0 Maintenance of roads Liquor and tobacco 100 0 These taxes were transferred to the depart- ments with the aim of financing health and education expenditures Source Shome, Haindl, and Schenon (2002) The Colombian law had established a complete adjustment for inflation aimed at introducing in the taxable base the gain and losses created by inflation. However, in 1998, the adjustment on inventories was eliminated, leaving a crippled system. A number of entities and activities are exempted; the main exemptions are the nonprofit organizations, the public services enterprises (water, sewerage, electricity, telephones, and gas), enterprises located in free zones, livestock funds, collateral funds organized by financial institutions, publishing firms, and new firms operating in areas of natural disaster. Moreover, capital gains resulting from operations with financial values are exempted. In 1998, 140,000 firms paid 2.3 percent of GDP under the corporate income tax. The bulk was provided by big taxpayers (the unst of large taxpayers manages 1600 taxpayers with legal residence in Bogota). The annual payment with the tax return accounted for 50 percent of this amount. The number of firms and entities exempted and the sources of income exempted explain the difference between the potential and the actual collection. Moreover, there are items that can be deducted from the amount of the rax or the tax base. Recommended policy changes include: * Reducing in stages the corporate rate down to around 32 percent, in order to increase the competitiveness of Colombia as a destination for internationally mobile investment. * Eliminating exemptions for firms with operations located in the Paez River and other areas of natural dssasters, and for particular sectors in publishing firms, lotteries, liquor producers, and cooperatives. THE TAX SYSTEM 181 * Eliminating the exemption for capital gains resulting from operations with financial assets * Setting slower rates of depreciation for some categories of fixed assets (build- ings and long-lived machinery) and reintroducing the inflation ad)ustmenlt on1 inventories * Introducing in the tax base the income generated by enterprises of public serv- ices, the livestock funds, and funds organized by financial institutions. * Reintroducing the presumptive income tax on gross assets as minimum income tax; if the tax return establishes a higher tax, the presumptive tax becomes a payment in advance. * Eliminating the credit for employment generation, compensating with reduc- tions in wage and payroll taxes. * Eliminating the option to pay in order to limit future tax liabilities and audits * Reviewing the treatment of charity credits. Finally, the tax on gross assets is neutral regarding debts. Consequently, adopting this tax could be a more efficient channel to increase taxation on income or property. ii. Personal Income Tax Collection of the personal income tax has improved mainly because of the increase in tax collection efficiency. This tax on gross assets is neutral regarding debts Adopting this tax could also be more efficient if it is implemented with increased amounts of taxes withheld at the income taxation source for different categories of income (such as fees, financial gains, and rents), The amount withheld from wages and payments accompanying annual tax returns provides minor amounts under this tax.'3 The rates of the personal income tax (20, 29, and 35 percent) are similar to those applied in the region, but the initial rate (20 percent) is higher in Colombia. There is also a min- imum presumptive tax on net wealth with a rate of 5 percent. An important number of potential taxpayers are outside the scope of the per- sonal income tax. The threshold for a salary exempted from taxation is much higher in Colombia than in other developing countries. Thus, the high threshold exempted reduces the number of taxpayers; more than 90 percent of wage earners are exempted. Also, this exemption reduces the effective tax rate, the effective rate for the wage earners who fill out and submit tax returns is only 0.8 percent, while the effective rate for natural persons averages 12 percent. As a result of various rates and thresholds, similar amounts of income can generate different amounts of personal income tax depending on the types of income including in the income tax base Two issues have to be highlighted about the design of this tax. First, the method of withholding income provides the bulk of the amount levied by the personal 13 Annual payments under the personal income tax are around 0 2 percent of GDP, 350,000 tax returns were submitted 182 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE income tax-more than 80 percent of the total. The main sources of withheld income include formal sector salaries, fees, financial yields, and rents from real estate. Second, the collection obtained with annual tax returns is very small, between 0.1 and 0.2 percent of GDP Compared to international experience, the number of exemptions of the income tax is very high in Colombia. Natural persons with income below three times the average per capita income are exempted, and wage earners pay the tax on only 70 percent of their income (30 percent exempted). Consequently, this tax presents a fairness problem. A stylized description of the personal income tax collection is the following (as percent of GDP): 0.2 percent comes from the annual tax returns submitted by 350,000 taxpayers; tax returns account for 0.6 percent with retaining of 0.3 percent on salaries and 0.1 percent on other sources of income While the retaining in the source provides 0.3 percent from labor payments and 1.4 percent from other sources of income. As a result, the amount collected is 1.9 percent of GDP. The effective tax rate for wage earners without tax returns is less than 0.2 percent. However, other sources of income are subject to higher tax rates. Recommended policy changes include: * Enlarging the base of the personal income tax by eliminating the 30-percent exemption for wage earners, 14 and introducing a general exemption for all tax- payers'5 equivalent to two times the minimum wage.16 • Introducing new rates of 15, 25, and 35 instead of the present rates of 20, 29, and 35. e Reintroducing the presumptive income tax on gross assets as minimum income tax; if the tax return establishes a higher tax, the presumptive tax becomes a payment in advance. * Introducing a common definition of salaries paid and earned for purposes of the corporate and personal income taxes and for purposes of payroll taxes. iii. Value Added Tax The key problems with the VAT are the transactions exempted and various rates, including zero rates for operations other than exports (material for primary schools, books and scientific magazines, etc.). The implicit VAT and the regime for capital goods add other difficulties in managing this tax. The design and administration of the VAT are complex, with numerous exemp- tions, various rates, an implicit VAT aimed at protecting privileged domestic activi- 14. This change will have effects on wage earners with salario integraL The elimination of these effects requires changes in the labor code. 15. The amount of the general exemption will have to be coordinated with the VAT reform if the "first best approach" is followed (see the section about the VAT) 16. This exemption is fixed in a number of minimum wages. THE IAX SYSTEM 183 ties,'7 and a special procedure for the VAT on capital goods. The productivity of this tax is low compared to the international experience. As an indication of the narrow base and evasion, the VAT collection as a ratio to GDP has been below a third of the VAT rate. With the aim of strengthening efficiency and fairness, a reform has to enlarge the base and bring all rates in line with the general rate. The optimum reform of the present system, aimed at increasing collections and reducing distortions, would eliminate all exemptions and establish a single rate of 16 percent. To protect low and medium income people, targeted subsidies and a general exemption in the personal income tax should be introduced. There should be a zero rate for exporters, other zero rates should be removed. The VAT on cap- ital goods should be credited against the VAT. A national excise tax regime for lux- ury goods other than beer and tobacco should be introduced with a flat rate of 30 percent. As a transition the recommended policy changes include: * Eliminating the rates of 8 and 10 percent. * Expanding the base of VAT to include paper and printing, beer and tobacco, and personal services, transportation, and construction * Eliminating the implicit VAT on selected imports. iv. Tax on Financial Transactions The tax on banking transactions, introduced in 1998, has become an important source of funds. Its low administrative cost is an advantage. The Pastrana Adminis- tration has defended this tax on the basis that it could capture tax revenues from the informal economy and illegal activities, including drug traffic and paramilitaries. However, the banking sector resisted the introduction of the tax because it discour- ages financial intermediation. The composition of monetary aggregates has changed in favor of cash since the introduction of the tax reflecting the actions taken by pri- vate agents aimed at avoiding this tax. In spite of the distortions, this tax could be maintained as a mean of collecting resources. However, the loopholes of the regime and the shift of people are eroding the effectiveness of this tax to collect more resources pretty fast. v. Taxes on International Trade Taxes on foreign trade have been reduced since the beginning of the 1990s, reflect- ing the trade reform carried out in the first part of that decade. The amount col- lected in 2001 was only 1 percent of GDP. Rather than an increase in the amount collected, further changes in trade policies would be oriented to reducing the possi- bilities of trade diversion involved in the Andean Agreement, and preparing the country for its entry into the American Free Trade Zone initiative. 17 The implicit VAT was introduced in 1998 The Appendix ro this chapter provides a description of this tax and the distortions involved 184 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE vi. Social Security Contributions Social security contributions have averaged 3.2 percent of GDP; their relative weight in the total collection has been around 19 percent. These taxes on payrolls have suf- fered an important reduction since 1998, reflecting the unemployment increase in the aftermath of the 1998-99 crisis, and consequently the reduction in the tax base. Because the level of payroll taxes is very high in Colombia,18 there is no room to increase rates. Payroll rates will push up unemployment because of the negative link between taxation and employment in the formal labor market.'9 Further reform in these areas should be focused on the cost of using labor,20 the financial imbalances and iniquities of the pay-as-you-go pension system, and the misallocation of resources in the health insurance scheme. vii. Taxation on Oil Activities The new royalty system, introduced recently, seems to provide flexible rules bal- ancing risks and reward for the investor and the government. Because oil is a non- renewable resource, the most important issue is how the resources provided by the oil production should be divided between present and future generations and between the national and subnational governments. Official projections for the medium term assume that a new oil field, with a capacity similar to Cusiana, could begin production in 2004-05. With the aim of preserving resources for future generations, the resources provided by other new fields could be used to repay public debt. b) Subnational Taxes The national government has transferred excise taxes on the consumption of liquor, beer, and tobacco to the departments. The departments also manage these taxes with the exception of beer, which continues under the administration of the central government. There are other minor taxes; some of them overlap with sim- ilar taxes managed by municipalities (such as the tax on motor vehicles, taxes on gambling, and the slaughter tax). The national government needs to streamline its laws regarding subnational taxes, eliminating exemptions and special regimes. 18. Payroll taxes (as percent of the payroll) are (a) pension contributions, 13.5 percent; (b) health insurance, 12.0 percent; (c) severance fund contributions, 8.3 percent; (d) acci- dent insurance, 2.5 percent; (e) training, 2.0 percent; familiar welfare and other allowances, 7.0 percent Total 45.3 percent (36 2 percent paid by employers and 9.1 per- cent paid by employees). 19. The National Directorate of Planning has estimated that the supply of jobs is reduced between 0.25 and 0 35 percent for each percentage point of increase in payroll taxes For the present size of the labor market this means the destruction of between 40,000 and 57,000 jobs 20 A labor reform may involve a reduction in the level of payroll taxes. THE TAX SYSTEM 185 The main taxes managed by the departments (liquor and tobacco) have low elas- ticity and are subject to high evasion With regard to the tax on liquor, there is room for improvements. The monopoly in the supply of liquor established by some departments has seriously distorted the liquor market. As a result, the size of the market and the potential amount of the tax collection have been reduced.2" A reform aimed at increasing the tax revenue levied on liquor would have to eliminate monopolies and reduced tax rates. Municipal taxation rests on two main taxes. a property tax (impuesto predial unificado) and a business tax (impuesto de industria y comerco). In addition, in recent years, the surcharge on the price of gasoline has become an important source of revenue for municipalities (Table 6) The property tax has a fair rev- enue-raising potential. To reach the potential, the transfer of technologies to identify properties and assess their values is a key element. The contrlbuczdn de valorarzzacidn, which partly transfers the cost (and benefit) of specific public projects to the direct beneficiaries, is another tax with good prospects for increas- ing revenues. In addition to these taxes, municipal governments operate several small taxes (a slaughter tax, a tax on public performances, gambling taxes, stamp taxes, and a tax on the extraction of sand, gravel, and stones). In spite of the improvements that the municipalities have made as a whole, they are very dependent on the transfers provided by the national government under the fiscal decentralization. The relative weight of these transfers in total receipts is extremely high especially for small municipalities. Fiscal decentralization has been under pressure because of evidence of the dis- torted incentives including in the system Territorial entities may have been inhib- ited from developing their own tax base because of the reliance on transfers provided by the central government It has also been noted that the quality of expenditures has not improved. The capacity of territorial entities of running deficits and bor- Table 6. Colombia-Taxes Collected by Municipalities (1999) Percent of the Total On property 35 Industry and commerce 38 Surcharge on gasoline 13 Other 14 Total 100 0 Source High Commiiissioni on Taxes 21 The Pastrana Administration has presented a draft bill reforming the taxation on liquor Analysis of the market and a reform proposal can be found in Oxford Center and FEDESARROLLO (2001). 186 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE rowing freely in financial markets, and the possibilIty of the central government bail- ing out local governments, have become matters of concern. To correct these prob- lems, the Congress approved both a constitutional reform (2001), changing the sys- (em of transfers, and Law 617 of 2001 establishing tight fiscal rules for departments and municipalities. These new rules have reduced the need to share national taxes with subnational entities. c) Tax Administration In most countries, both developed and developing, there is room for improving tax administration. The main objectives of a tax reform strategy would be to change the behavior of the average taxpayer, and to raise the compliance rate of the taxpayer population (Baer and Toro 2000). These main objectives must also drive the tax reform, including the Colombian. The multiple exemptions of the income tax and the VAT have created some of the difficulties faced by the Colombian tax administration. Furthermore, the implicit VAT and the VAT on capital goods have added more problems. On one hand, a new tax adjustment would improve the capabilities of the administration to enlarge the tax bases and eliminate exemptions. On the other hand, the administra- tion would be worse off under some proposals recently advanced. For instance, the administration would be overloaded if the tax on financial transactions were cred- ited to the income tax. In recent years, the Colombian administration has carried out a number of pro- grams aimed at improving taxpayer compliance and reducing evasion. In this regard, Law 488 of 1998 improved the administration through the designation of the Directorate of National Taxes and Customs (DIAN) as an autonomous institution for budgetary and administrative purposes, and the inclusion of a fiscal police with the aim of supporting tax collection. An assessment about the budgetary autonomy for the DIAN provided by law 488 is needed. Law 488 also introduced the auditing benefit, however which facilitates evasion. Firms can fix the amount of income tax they pay for a number of years with an increase of 30 percent in the tax declared. In return, firms receive as benefit, after a period of time following the submission of the tax return, exemptions from tax auditing, which is normally carried out by the tax administration. The strategy pursued by the DIAN addresses the main issues on tax administra- tion in the areas of collection, control, cashing, and efficiency indicators. The suc- cess in increasing taxpayer compliance hinges on the balance between the different aims of the plan. Identification of taxpayers, control of taxpayers responsible for submitting tax returns and paying taxes, and services supporting taxpayers will increase adherence to the rules. In this regard, important advances have been introduced under the Public Finan- cial Management Project (PFMP) prepared with the support of the World Bank. Various steps have been taken to help taxpayers fulfill their obligations. These THE FAx SYSTEM 187 include establishinig special units to provide services to taxpayers, importers and exporters, including facilities for large taxpayers to file tax returns. The impact of these efforts has been reflected in the increase in the number of tax and customs dec- larations filed. The PFMP has provided the DIAN with tools to manage the heavy information flow associated with 4 million tax declarations filed annually. A new taxpayer current account system has been implemented, along with a system that updates the current account by incorporating administrative decisions. The system can create new products to improve the enforcement capability of Colombian tax administration. It would be important to explore how the programs carried out under the PEMP could be continued and deepened. The plan has to be accountable, establishing indicators for the efficiency of the management. These indicators have to cover monetary targets, such as the recovery of overdue taxes, and other nonmonetary goals, such as deadlines for actions of the administration and efficiency ratios for controlling tax returns and payments. In the area of collecting, actions aimed at refining and updating the catalogue of taxpayers will continue as a key element of the plan At the same time, the procedures to detect taxpayers with overdue tax returnis and payments have to be strengthened. As an important complement, in the unit for large taxpayers, targets about the num- ber of taxpayers with overdue tax returns and payments would have to be redefined. The annual budget establishes targets for the DIAN collection and for the dif- ferent units. The plan to control compliance is articulated around these targets. This area may improve developing actions focused on the main taxes, the income tax, and the VAT. Concerning the income tax, the procedures used for the administration to cross-check data from different sotirces show good potential for increasing revenues. In the case of the VAT, there is room for enhancement of the billing procedures, an effective control on VAT returns, and cross-checklng of data. Finally, in the area of collecting tax arrears, actions could be enhanced establish- ing targets on the administration portfolio of overdue credits. At the same time, it is important to maintain the financial cost of taxpayers' overdue debt in line with market rates. 188 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Appendix A Summary of Previous Tax Reforms 1. Reforms Approved by the Gaviria Administration (1990 and 1992) To a considerable extent, the changes in public finances that took place in the 1990s reflect political commitments prescribed in the new Constitution and the trade reform carried out by the government during 1990-94. These changes, which sought to increase social fairness and efficiency, involved increases in public expen- ditures without a parallel increase in revenues. As a result, new revenues were needed to support the new constitutional pact and the openness process. The aims of the reform approved in 1990 were to substitute the lost revenues resulting from the trade reform, and promote private savings. On one hand, the VAT rate was increased from 10 percent to 12 percent; the tax base was expanded, including some services with a rate of 4 percent. At the same time, the rates of 20 percent and 35 percent on luxury goods were maintained. On the other hand, cap- ital gains resulting from operations with financial values were exempted. Investment funds were exempted from income tax. In the 1990s taxes on profit remittances were reduced to 20 percent, and in 1996 they were reduced to 12 percent. In addition, the reform introduced a fiscal amnesty for financial capital held outside Colombia. The changes introduced in the VAT and the income tax increased the collection 2 percentage points of GDP. Law 6 of 1992 introduced new changes in the VAT and the income tax. This law expanded the VAT base and increased the rate from 12 percent to 14 percent; the rates on luxury goods were to be 20, 35, and 45 percent. The VAT on capital goods could be deducted from the income tax.22 This change distanced the Colombian VAT from the consumption-type VAT, the easiest VAT model for administrative purposes. Since the approval of Law 6, the administration had to define capital goods and control the breakdown of purchases between inputs and investment. The rate of income tax was increased from 35 percent to 37.5 percent because a surcharge was added. Beginning in 1996, the tax on profit remittances was reduced gradually to 7 percent. A temporary contribution on mining activities (oil, gas, coal, and nickel) was created. This surcharge could be credited to the income tax. The rate of 14 percent for the VAT and the surcharge on income tax were established for the period 1993-97. 22. Compared to a consumption-type VAT, this scheme provides a once-and-for-all increase in the collection because tax returns and payments are bimonthly for the VAT and annual for the income tax. Tl-lE TAX SYSTEM 189 II. The Reform of 1995 The main objectives of the 1995 reforms were to establish as permanent the VAT rate of 14 percent, and substitute the special contribution of the income tax for new rates A new VAT rate of 16 percent was established beginning in 1996 To improve VAT administration, some taxpayers became retainers of 50 percent of the VAT included in their purchases. This reform enlarged the income tax base and included public enterprises and public services enterprises as taxpayers under this tax. If the rents under the income tax were used to capitalize the firms or finance new invest- ments, they would be exempted for eight years. Law 223 eliminated the surcharge on income tax and established new rates for corporations (35 percent) and persons (20, 29, and 35 percent) A presumptive tax was created on gross assets with a rate of 1.5 percent, and the rate on the presump- tive tax on net wealth was increased from 4 percent to 5 percent. Taxpayers have to pay the higher alternative, and the payment was going to be considered an advance on the income tax. The temporary contribution on mining activities was substituted by a permanent surcharge with gradual reductions until 2000. The new surcharge could not be credited to the income tax. In addition, the law established general rules for the taxes collected by the departments on beer, liquor, and tobacco The results achieved by this reform did not meet expectations because additional revenues were not collected. Generous exemptions from the income tax for invest- ments in the area of the Paez River might explain the disappointing result. With the aim of providing help for the region of Paez River, Law 218 of 1995 granted exemp- tions for 10 years on income tax for investments made in the departments of Cauca and Huila Also, the imports of capital goods for investments in these departments received tariff exemptions. Law 345 of 1946 introduced a compulsory investment in Security Bonds for cor- porations and persons with a wealth over COP 150 million. III. The Reform of 1998 The main objective of this reform was to reduce the fiscal deficit that in the central government was 5 percent of GDP. The principal changes were the elimination of exemptions for an important number of imports, and the VAT rate was reduced from 16 percent to 15 percent. The reduction became effective at the end of 1999. At the same time, Law 488 established a rate of 10 percent on some goods and serv- ices. Again, the VAT on capital goods was changed; the tax would have to be cred- ited to the income tax instead of being deducted from it. This reform created a new form of VAT, the "implicit VAT," as a way of protect- ing domestic activities from imports exempted from the VAT. In customs, the gen- eral rate is levied on imports, with some exemptions. In addition, imports exempted from the VAT pay an implicit VAT. The similar local production is also exempted 190 COLOMBLA THE ECONOMIC FOUNDATION OF PEACE from the VAT. Consequently, the VAT included in their purchases of inputs becomes a cost. Using domestic cost structures, the administration calculates for a number of products the implicit VAT as the VAT that these products would have paid had they not been exempted. The implicit VAT became a burden to adminis- ter Decree 1344 of 1999 established the implicit VAT for 78 tariff items with rates between 0.7 percent and 9.6 percent. Since the number of exempted goods is higher, it is unclear how the items were selected The collection under the implicit VAT is 0.1 percent of GDP.23 This reform introduced changes in the design of the income tax. As a support for job creation, a deduction from the amount of the tax was introduced. The pre- sumptive tax on gross assets and the inflation adjustment for inventories and pur- chases made in the fiscal year covered by the tax return were eliminated. This law provided more resources for the departments and the capital through a new surcharge on ACPM. Fifty percent of the collection under this surcharge was for the departments and the city of Bogota; the other 50 percent was earmarked for the improvement of national roads. Using special powers, President Pastrana signed a decree creating a temporary tax on financial operations at the rate of 0.2 percent. IV. The Reform of 2000 In 2000, the Pastrana Administration presented a new tax reform with the aim of reducing the growing fiscal deficit. To increase revenues the tax on financial trans- actions became permanent and its rate was increased to 0.3 percent, and the VAT rate was increased again from 15 percent to 16 percent. Some services previously exempted from the VAT were included in the base of this tax. In 2001, taxes col- lected from the main taxes (the income tax, the VAT, the tax on financial operations) increased. Some writers have identified the tax ad)ustment as the main cause of the low economic growth that year. Law 663 eliminated the VAT on capital goods as a deduction from the income tax. Since the inception of this law, firms have included the VAT on capital goods in the cost of the investment. Therefore, a portion of the tax is included in the annual capital amortization credited to the income tax. 23 Recently, the Andean High Court ruled that, under the Andean Agreement, the levy of this tax on imports from the other Andean partners is illegal. THE TAX SYSTEM 191 References Baer, Katherine, and Juan Toro. 2000. "Colombia: una estrategia para aumentar la eficacia de la administraci6n tributaria." Washington, D C: International Mon- etary Fund. CONFIS (Consejo Superior de Politica Fiscal) 2002. "La deuda publica colom- biana " Marzo Oxford Center and FEDESARROLLO (Fundacion para la Educacion Superior y el Desarrollo) (2001) "Taxation on Liquors-A Proposal for Reform " Shome, Parthasarathi. 1995. "Comprehensive Tax Reform-The Colombiani Expe- rience." Washington, D.C.. International Monetary Fund. Shome, Parthasarathi, Erick Haindl, and Osvaldo Schenone 2000. "Colombia-La politica tributaria 1995-99 y los pasos a seguir." Washington, D.C. Interna- tional Monetary Fund, June. Steiner, Roberto, and Carlos Soto. 1999. "Cinco ensayos sobre tributacioni en Colombia." FEDESARROLLO Tanzi, Vito. 1995. "Taxation in an Integrated World." Washington, D C. The Brookings Institution. Tanzi, Vito, and Howell H. Zec. 2000. "Tax Policy for Emerging Markets Devel- oping Countries." Washington, D.C.: International Monetary Fund. 7 Public Debt Sustainability and Management This Chapter was written by Mario Adolfo Cuevas, Clemente Luis Del Valle, and Vicente Fretes Cibils. I. Summary Over the last seven years the level of total gross public debt in Colombia has been rising steadily in absolute terms and as a proportion of GDP. In addition, public debt service has risen significantly. Both the level of the debt and the debt service are worrisome, and to avoid a debt crisis the government must further strengthen its primary fiscal balance-from a primary surplus of about 1.4 percent of GDP in 2001 to a permanent primary surplus of at least 4 percent of GDP in the medium term. To achieve this, the government will have to implement policies to strengthen the fiscal "fundamentals" (through cutting public sector expenditures and increasing public sector revenues). Moreover, the government must complement the fiscal adjustment with improvements in public sector debt management (through strengthening debt management capacity and debt portfolio management). II. Background The total gross stock of Colombia's public debt,2 both internal and external, increased from about 27 percent of GDP in 1994 to about 58 percent of GDP in I This chapter has benefitted from comments provided by several colleagues We would like to hank them and especially Jorge Garcia-Garcia, who gave us useful suggestions on an earlier draft 2. There are several types of public sector liabilities other than "debt." The following terminology is adopted throughout this paper (a) direct-explicit liabilities are liabilities established by law or contract and include "full faith and credit" debt, expenditures 193 194 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE 2001.3 This resulted primarily from rising fiscal imbalances at different levels of gov- ernment throughout the decade (see Figure 1). The steady growth in fiscal imbal- ances first became visible in the central government accounts, but this was initially masked by surpluses in other parts of the public sector. Figure 1. Public Sector Balances 4 F1 NFPS F Central Government 2 - 0 X -4 - _ -6 ---- _ -- _ _ ~ _. -- _ _ _ - _ - _- _--- __- X1 _ -8 C', 0', 0' \'T, C' ', C, OD, C~' C~ C', C, C', C', C, C' C' C' C', C. C > g R > g 4 > % > o° ocR l, Z C _ _. ~ _ C-, N rq Source IMF and WB Staff estimates prescribed by budget law, and claims for services rendered (the timing and amount of these liabilities may nevertheless be affected by contingencies); (b) direct-implicit liabili- ties are liabilities on which it is presumed, with good probability, that the government will make good, but without a legal obligation to do so; (c) contingent-explicit liabilities are recognized in legally binding documents, but the extent and timing of payment hinges on uncertain future occurrences; and, (d) contingent-implicit liabilities refer to an expecta- tion that government will accept a liability without having a legal obligation to do so. For more details on the classification of public sector liabilities, see Budgetingfor Fiscal Risks, by Allen Schick, World Bank, September 1999. 3. In net terms, the total stock of public debt represented about 46 percent of GDP in 2001. Net debt is total gross debt minus the stock of Treasury bonds (TES) held by all public sector entities The evolution of debt aggregates in the last decade is discussed with refer- ence to gross aggregates because estimates of net debt in Colombia are available only from 1999 onward PUBLIC DEBT SUSTAINABILITY AND MANAGEMENT 195 The central government accounts started the decade of the 1990s with a small surplus. However, since 1992 the central government has posted a series of increas- ing deficits, which peaked in 1999 with a deficit of 7.4 percent of GDP. Total expen- ditures increased from about 15 percent of GDP in 1992 to nearly 20 percent of GDP in 1999. By contrast, total revenues fell from about 15 percent of GDP in 1992 to 12 percent in 1999. Behind the deterioration in the central government accounts are some of the institutional changes brought about by the 1991 Consti- tutional reform, and the resulting impacts of the economic policies implemented throughout the decade The 1991 reforms transferred to subnational levels of government considerable discretionary spending powers, but not the responsibility to fund this spending. As a result, the central government financed through fiscal transfers subnational expen- ditures over which it had little control. In addition, the central government had often negotiated large increases in the salaries and benefits of employees of subna- tional governments (for example, teachers and health sector personnel) which sub- sequently had to be funded through additional fiscal transfers. Moreover, recogni- tion of central government implicit and contingent liabilities arising, for example, from the public pension funds (the Cajas Publicas) and the financial sector (as evi- denced in the 1999 restructuring of the financial sector) put upward pressure on expenditures. Other reforms (although correct for increasing efficiency in resource allocation) negatively affected fiscal revenues (for example, trade liberalization led to a fall in tariff revenues). More important, several reform packages aimed at strengthening tax revenues were only partially successful. The economic slowdown compounded fiscal weaknesses because tax revenues fell in tandem with sluggish economic activity. This led to increased reliance on nonrecurrent revenue-generating measures, such as pri- vatization, and more significantly on increased public sector debt to finance the cen- tral government deficit. By the mid-1990s it became evident that the public sector accounts were rap- idly deteriorating. The public sector deficit rose steadily-basically tripling from just over 2 percent of GDP in 1995 to 6.4 percent in 1999. Large central gov- ernment deficits, growing deficits at the subnational level, dwindling social secu- rity surpluses, and volatile oil revenues all contributed to the deterioration in the pubhc sector balance. This is evidenced by the continued erosion of primary bal- ances in the second part of the decade, when the public sector primary balance went from a very small deficit in 1996 to a deficit of 2.7 percent of GDP in 1999 4 4. Total public sector revenLes have remained at around 27 percent of GDP since the mid- 1990s, while total expenditures increased from )ust under 30 percent of GDP in the mid- 1990s to about 33 percent of GDP by 1999 Throughout this period, the Central Bank posted a small average surplus equlvalent to 0 28 percent of GDP. 196 COLOMBLA: THE ECONOMIC FOUNDATION OF PEACE The increased debt service augmented the vulnerability of public sector finances.5 Total public debt interest service increased from about 2.2 percent of GDP in 1994 to about 4.5 percent of GDP in 2000. Total interest service on internal debt increased more than proportionately-almost threefold-from 1 percent of GDP in 1994 to about 2.9 percent of GDP in 2000 (see Figure 2). In 2001, total interest service remained roughly stable at 4.5 percent of GDP, with a fall in domestic inter- est payments to an estimated 2.3 percent of GDP being offset by an increase in external interest payments. In this context, both the level and structure of public sector debt changed. In the early 1990s, the government reduced its total debt from about 40 percent of GDP in 1991 to about 30 percent of GDP in 1994-mostly through debt retirement and buy-back operations, financed in part by high oil revenues. External debt fell sharply from about 32 percent of GDP in 1991 to about 14 percent of GDP in 1996. How- ever, as the public sector accounts began to deteriorate in the mid-1990s (particu- Figuire 2. Interest Payments of the Public Sector 50- 1 External ! Internal 4.0 --- - - - - U 3.0--- ; 20 1994 1995 1996 1997 1998 1999 2000 2001 Source IMF and WB Staff estimates 5. The real average rate of interest paid on domestic government debt has been high and volatile. It has averaged 7.3 percent since 1997 (for one-year TES securities), peaking at around 20 percent in 1998 and falling as low as 2 2 percent more recently It follows that the steady increase in the government's interest burden cannot be explained by high real interest rates alone, although this factor cannot be neglected as an important cost source. PUBLIC DEBT SUS FAINABILITY AND MANAGEMENT 197 larly central government accounts), debt levels increased. Throughout this period, central government debt increased more than proportionately, thus taking an increasing share of public debt. of the total gross public debt outstanding in 2001, about 40 percent of GDP was central government debt (up from about 16 percent of GDP in 1991) Asizeable portion of the change in the level of total debt occurred through an increase in external debt, which reached about 26 percent of GDP in 2000, bringing the share of external debt in total debt close to the levels of the early 1990s. Most of the increase in the external debt after 1996 has been driven by cen- tral government borrowing-other levels of government have faced statutory and procedural restrictions on external borrowing.6 However, both the central govern- ment and subnational entities have continued to issue domestic debt to finance their increasing expenditure program throughout this period. As a result, domestic gross public sector debt rose steadily, from about 14 percent of GDP in 1994 to about 27 percent of GDP by 2000 (see Figures 3 and 4). This period is also characterized by a drastic change in the sources of finance (see Figures 5 and 6) At the beginning of the 1990s most new financing was coming from nonmarket sources (for example, bilateral and multilateral agencies, and the Figure 3. Gross External and Domestic Debt of the Public Sector 70 - nI Externial 60 - l * Internal 50 - _ 40 0 re : 30 20 I 0 10 - 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Sources IMF, MoF Contralonra and WB Staff estimates 6 These restrictions are discussed in The Legal Framework and Governance subsection of this Chapter 198 COLOMBLA THE ECONOMIC FOUNDATION OF PrACE Figure 4. Gross Public Debt by Debtor 70 - ______________________________ 7 - Central Government 60 -3-0- Sub-National Government LI Other Piblic Sector 50 - O 40 ---/ 30 20 10 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Sources IMF, MoE, Contraloria and WB Staff estimates Central Bank). However, during the 1990s, as a result of a transformation of the overall public sector Institutional framework, the government relied more on the financial market (both internally and externally) as the main source of funding- half of domestic debt and two thirds of external debt are now market based. More specifically, the implementation of the Constitutional reforms of the early 1990s led to elimination of Central Bank financing of the fiscal deficit. As a result, the central government had to tap the domestic market on a more competitive basis to finance its expenditure program 7 The domestic debt of the central government continued to rise through a large expansion in issuance of Treasury bonds (TES, see Annex I). This increase resulted mainly from (a) the need to finance growing deficits, and (b) an attempt to limit the exposure to currency risk and to make only limited use of additional external financing.8 Tapping the domestic market was also 7. In 1990, debt to the Central Bank amounted to 50 percent of the total internal debt of the public sector-falling to less than I percent by the end of the decade. The Central Bank has nevertheless continued to contribute to financing the government through transfers of its balance sheet surpluses (for example, in 2001, these transfers represented about 0.7 percent of GDP). 8. In the mid-1990s, the authorities expected large foreign currency inflows resulting from exploitation of the recently discovered oil fields (Cusiana). This expectation also created incentives for increasing reliance on the local capital markets. PUBLIC DEHI SUSTAINABILITY AND MANAGEMENT 199 Figure 5. Domestic Debt of the Central Government by Creditor 18- 16-C] Central Bank 16 __ 14-] Bonds a Other 12- 10- O5 __ _ _ _ _I 4-~ 2 1980 1985 1990 1995 1999 2000 2001 expected to foster development of the domestic capital market through lengthening of the yield curve and deepening the market.9 The sources of external financing also changed considerably throughout the decade (see Figure 6). Whereas in 1990 official debt amounted to 66 percent of the total stock of public external debt, this had fallen to only 39 percent by 1999. The share of commercial bank debt fell relative to the total stock of public external debt, from 32 percent in 1990 to 23 percent in 1999. In contrast, sovereign bonds gained considerable ground over this period, increasing from about 2 percent in 1990 to over 50 percent of the total external debt of the public sector by 2001 This resulted mainly from Colombia's good access to international financial markets and its bor- rowing strategy In contrast to the case of central government debt, the stock of subnational gov- ernment debt has remained relatively stable since 1991-fluctuating between 7.4 and 9.2 percent of GDP. The borrowing requirement of subnational public entities has been somewhat limited, in part because financing gaps have often been met through direct fiscal transfers from the central government (as has already been men- tioned). In fact, the external debt of subnational entities has been continuously falling, mainly as a result of the restrictions on external borrowing by subnational entities, that is, they require Congressional approval to issue external debt However, 9 The share of TES bonds of longer maturity-five years or longer-increased from 12 per- cent in 1995 to 37 percent of the outstanding stock of TES bonds in 2000 200 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Figure 6. External Debt of the Central Government by Type of Creditor 25 - D Multilateral * Other Official 20 - _ - _ _ * Commercial Bank * Bonds 15 10 _ _ 1980 1985 1990 1995 1998 1999 2000 2001 this has been fully compensated by an increase in domestic debt. In 1991, the domestic debt of subnational levels of government amounted to about 3 percent of GDP, rising to about 5 percent of GDP by 1995, and to about 6 percent by 1999 (see Figure 7). Domestic commercial bank lending to subnatnonal governments has been the main source of the increase, either directly or through rediscounting of Financiera de Entidades Termtorzales (FINDETER) loans. This has partly resulted from (a) the explicit or implicit credit guarantees from the central government, and (b) loopholes in the implementation of the Ley de Semdforos, aimed at regulating subnational government borrowing, creating "perverse" incentives for both subna- tional governments and banks to expand credit rapidly. Recent fiscal adjustment efforts have made good progress in restoring a fiscally sustainable path. In 2000, the public sector deficit was almost halved, to -3.5 per- cent of GDP; moreoves, the primary fiscal balance showed a surplus (although small) of around 0.9 percent of GDP-for the first time in several years. However, there are still large and growing implicit and contingent public sector liabilities (aris- ing mainly from the pension system in the Social Security system, the financial sec- tor, and subnational levels of government), which are not yet fully reflected in the fiscal accounts, threatening Colombia's fiscal stability in the future.10 10 The nature and size of this type of liability is discussed at greater length in other Chapters. PUBLIC DEBT SUSTAINABILITY AND MANAGEMENT 201 Figure 7. Debt Structure of Sub-National Entities External * Internal a6- 0 U 4 2 0~ 0 - I I I II I I I I 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Sources Contraloria anid WB Staff estnnates III. Issues, Diagnosis, and Policy Recommendations a) The Primary Balance and Debt Sustainability The current level and structure of Colombia's pLiblic sector liabilities are risky and vulnerable. Should current trends continue in the absence of comprehensive policy reforms, particularly in Social Security, public sector debt is likely to become un- sustainable To analyze the sustainability of public sector debt, we calculate the pri- mary fiscal balance necessary to keep public sector debt on a sustainable path" I using a model that includes (a) the present value of all net public sector liabilities (in- cluding implicit and continigent liabilities from Social Security); and (b) key eco- nomic parameters and projections in a base case scenario (see Annex 11) for a 11 This is the level of rhe primary balance expressed as a ratio to GDP that if generated "per- manently" by the government, woLild keep the debt-to-GDP ratio sustainable However, the debt-to-GDP ratio does not necessarily "stabilize" or reach a "constant" value during the projection period, because the steady state debt-to-GDP ratio is achieved only in the long term 202 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE detailed discussion of the model and Table 1 for a summary of key assumptions and projections) .12 According to the analysis, to avert an explosion of the debt-to-GDP ratio, the public sector has to generate a permanent primary surplus of about 4 percent of GDP.R3 This adjustment can only be done through a combination of measures to simultaneously (a) cut public sector expenditures, and (b) increase public sector rev- enues. (For details on the specific policy actions and measures to achieve both (a) and (b), see Chapters on the Macroeconomic Framework and the Tax System.) Such level of fiscal adjustment is a necessary condition for ensuring fiscal sustainability and avoiding the outcome of other countries (for example, Argentina) that have delayed strengthening the "fundamentals," particularly the fiscal accounts, through financial engineering (see Annex III). Moreover, by comparison, for the last six years the average primary balance-to- GDP ratio was -0.37 percent (that is, a deficit). During this period, the debt-to- GDP ratio grew very rapidly (the average expansion rate of the debt-to-GDP ratio was about 6 percent per year), thus suggesting that the current path of the debt-to- GDP ratio in the absence of adjustment is not sustainable. It follows that in the absence of comprehensive reforms, keeping debt on a stable path requires a sus- tained fiscal adjustment in the primary balance of at least 2.6 percent of GDP (to bring the primary surplus to 4 percent of GDP, from about 1.4 percent in 2001). If the primary surplus were kept permanently at 4 percent of GDP, it is estimated that Table 1. Key Assumptions and Projections* (percent) 2002 2003 2004 2005 2006 2010 Real GDP Growth (p a) 2.0 2.5 3.0 3.5 3.5 3 5 Inflation (p a.) 6.0 5.0 4.0 4 0 4.0 4.0 Nominal Exchange Rate Depreciation 8.0 7.0 6.0 4 0 3.0 2.0 Real Interest Rate 4.6 5.5 6.2 5 8 5 7 5.2 Total Public Sector Debt (of GDP) 48.2 51.2 53 9 56.6 57.6 567 Domestic 20.8 22.3 23 7 25.0 25 5 25.1 External 27.4 28.9 30.3 31.6 32 1 31 7 *Base case scenario 12. In the base case scenario, it is assumed that some fiscal policy reforms are implemented, but more comprehensive reforms (for example, reforms to the pension system in the Social Security System) are only partly implemented. 13 Primary balances are defined here with respect to the consolidated public sector, includ- ing Social Security and subnational government entities For consistency with other chapters, we consider that seignorage revenues accrue to the government "above the line" (through the reported Central Bank balance). l'UBLIc DEBTI SUSTIAINABILIrFY AND MANAGEMENTI 203 total net debt of the public sector would nevertheless continue increasing, reaching about 57 percent of GDP by 2010 (see Table 1). The vulnerability embedded in pLiblic finances in the absence of comprehensive reforms can also be shown by projecting the path of the total net debt of the public sector under the assumption that the primary balance-to-GDP ratio is kept constant at the 2001 surplus level (just over 1.4 percent of GDP)."4 As can be seen in Figure 8, total net debt would rise from about 46 percent of GDP in 2001, to about 80 percent of GDP in 2010 (assuming real GDP growth of 3 5 percent per year). In this scenario, the debt-to-GDP ratio would be rising at an average annual rate of about 8 percent per year, implying an acceleration in the rate of increase of the debt- to-GDP ratio from the 1994-2000 level (when the debt-to-GDP ratio increased at an average annual rate of about 6 percent). Figure 9 shows the path of the primary balance required to stabilize the debt-to- GDP ratio at the 2001 level (about 46 percent of GDP). As can be seen, the pri- Figure 8. Debt-GDP Ratio Primary Balance Constant at 2001 Level 200- - 900 Debt/GDP Ratio 80.0 1 75- ____________ ___ Qc Primary Balanice / -70 0 1 50- ----------60.0 500 12 5 _-__ __ ^ _ 40.0 1.00- I I I I I I I I 300 0 0 0 0~~~. 0 00 0~C C' EN N mN N IN IN IN mN m Source World Bank estimates GDP growth at 3 5 percent p a 14. If the 2001 primiary balance was ad)usted to filter out cyclical or noise components, the resulting "structural" primary balance would in fact be lower, since unusually high oil rev- enues improved the fiscal accounts that year With the primary surplus kept constant at the lower, adjusted level, the debt-to-GDP ratio would increase even faster Similarly, the size of the fiscal adjustment required for debt sustainability would be greater than 2 6 percent of GDP 204 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE Figure 9. Debt-GDP Ratio Primary Balance Set to Keep Debt-GDP Ratio Constant 70 - -60 6.0 -- - - -- - - - -- - - - _ _ _ _ - Primary Balance r8 5 0 / - 50 Debt/GDP Ratio (2001) 4 0 > 3.0 -- 40 1.0- I I I E -30 ~~ m m > u SD lIO \ X0 C", O o o o oo o o o o - o o 0 0 0 o 0 o 0 o N m 'i m (N (N (N (N N N Source World Bank estimates GDP growth at 3 5 percent p a mary surplus would have to be maintained in the range of 4 to 6 percent of GDP In the next decade to keep the debt-to-GDP ratio constant (higher than the 4 per- cent of GDP required to keep debt on a moderately rising but still sustainable path). Table 2 shows the sensitivity of these calculations to changes in the rate of GDP growth; for example, if GDP growth is 2.5 percent per year in the long term (instead of 3.5 percent), the permanent primary surplus that keeps the debt-to-GDP ratio along a sustainable path is 6.2 percent of GDP (instead of 4 percent, as before). The sensitivity of these calculations to the recognition (explicitly) of liability payments arising from Social Security (mainly pensions) in a scenario of limited structural reforms is also shown in Table 2. For example, if GDP is assumed to grow at a rate of 3.5 percent per year and Social Security "payments" (of the implicit liabilities) are excluded from our calculations, a permanent primary surplus of 0.8 percent of GDP would be sufficient to keep the debt-to-GDP ratio on a sustainable path. This should be contrasted with the 4 percent primary balance-to-GDP ratio required once Social Security (implicit liabilities) are included explicitly as flow-payments in these calculations.'5 15 Should comprehensive Social Security and other reforms take place, the present value of contingent and implicit government liabilities could fall sharply, implying that the per- manent primary surplus required to keep the debt-to-GDP ratio on a sustainable path would be smaller than 4 percent of GDP PUBLIC DEBT SUSTAINABILITY AND MANAGEMENT 205 Table 2. Permanent Primary Surplus: Sensitivity Analysis* (percent) GDP Growth Rate 1.5 2.5 3.5 4.5 Including Social Security 8 0 6 2 4 0 1 3 Excluding Social Security' 1.7 1.2 0 8 0.2 Permanent primary surplus as a percentage of GDP required to ensure a non-explosive debt path -Hypothetical scenario that excludes the future fiscal impact of implicit pension liabilities The impact on debt ratios of slower long-term GDP growth can also be illus- trated with reference to a slow-growth scenario (see Figure 10). If it is assumed that the primary balance quickly converges to 4 percent of GDP during the projection period, but that GDP growth is slower than in the base case (1.5 percent annually, instead of 3.5 percent), ceteris paribus projected total debt would rise to almost 70 percent of GDP in 2010 and keep rising thereafter-instead of stabilizing at about 57 percent as in the base case Furthermore, since the permanent primary surplus that generates a nonexplosive debt path with long-term growth of 1.5 percent annu- ally is 8 percent of GDP (when pension liabilities are included in the calculations) it is clear that in the slow-growth scenario, a primary surplus of 4 percent of GDP is not consistent with a stable debt path. Figure 10. Debt-GDP Ratio Primary Balance Constant at 4 percent of GDP Slow Growth Scenario 7.0- -70 Debt/GDP Ratio 60 - - ____ _______ -60 5 0 - 50 - E 40 XL. Primary Balance 40 3 0 ---- --- ------…-____ 20- I I I I I I -30 o o v' O, O N so 0-, o >w >a r t N r°4 mN C J IN Source World Bank estimates GDP growth at 1 5 percent p a 206 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE The average real interest rate has also impacted the fiscal accounts in the past, con- tributing to the overall fiscal deficit. For example, during 1995-96 and 1998-99, from the strictly financing needs of the public accounts, about half of the increase in the overall financing needs of the public sector was due to increased domestic interest pay- ments. Put differently, about half of the increase in total debt in those years could be attributed to financing current interest payments on domestic debt. During 1995-2001 (a period of rapid expansion of public debt stocks) the real interest rate on domestic public debt averaged about 8.5 percent. By contrast, with fiscal adjustments, the average real interest rate on domestic debt could fall to a range of 5 to 6 percent. We also estimated the sensitiviry of permanent primary surplus to changes in average real interest rates, ceteris partbus (Table 3). For example, if Social Security payments (of implicit liabilities) were included in the analysis and real interest rates amounted to about 5.5 percent over the long term (instead of 5.2 percent as in the base case), the permanent primary surplus required to restore fiscal sustainability would be 5.1 percent of GDP (instead of 4 percent of GDP as in the base case).'6 In sum, the analysis shows that Colombia needs to further strengthen its fiscal accounts in order to increase the primary fiscal balance and ensure fiscal sustain- ability in the medium and long term. This can only be achieved through policy actions that will cut public expenditures and increase public revenues. This is nec- essary because, even under a relatively benign macroeconomic scenario but in the absence of comprehensive reforms, an adjustment of at least 2.6 percent of GDP in the primary surplus from the 2001 level is required to restore fiscal sustainability. If this adjustment in the primary balance were carried out, the debt-to-GDP ratio would nevertheless continue rising, reaching about 57 percent of GDP by 2010, but this ratio would likely be in a nonexplosive path. b) Vulnerability and Management Risks in the Debt Structure To complement the fiscal adjustment, Colombia needs to strengthen its debt man- agement capacity to increase its efficiency and effectiveness for managing a vulnera- Table 3. Permanent Primary Surplus: Sensitivity Analysis* (percent) Real Interest Rate 4.5 5.0 5.2 5.5 Including Social Security 1.7 3 4 4.0 5.1 Excluding Social Security 0 3 0 6 0.8 1 0 *Permanent primary surplus as a percentage of GDP required to ensure a non-explosive debt path -*Hypothetical scenario that excludes the future fiscal impact of implicit pension liabilities 16. It is worth noting that the analysis is "partial equlhbriu"' in essence, and explores the impact on the permanent primary surplus required to keep debt on a sustainable path, of changes in individual economic variables-that is, all else remaining the same. PUBLIC DEBT SUSIAINABILITY ANI) MANAGEMENT 207 ble and complex debt portfolio Indeed, Colombia's debt portfolio is highly vulner- able to unexpected changes in both real interest and exchange rates.7 The high vul- nerability to real interest rates is caused by the relatively large share of debt with short maturities and high refinancing risks. In addition, growth in foreign-currency- and exchange rate-linked debt, has led to an increase in vulnerability to exchange rates. Such exposures could create short-term liquidity problems for the government at times of strong unanticipated adverse movements in real interest or exchange rates (see Table 4) The rapid growth of debt stocks to finance growing expenditures in the past led also to a deterioration in the structure of the portfolio. Management of the debt portfolio has been constrained because (a) domestic markets are shallow and incom- plete; (b) access to international markets has been limited, particularly in the last few years;'8 and (c) weak debt management capacity. Underdeveloped and incomplete domestic markets led to a reliance on relatively expensive financing sources and shorter maturity instruments.19 Moreover, the maturity of new external government bonds has been shortened to between three and five years.20 Given the limited development of domestic markets and the absence of hedging instrumenits, the government is exposed to liquidity and refinancing risks. Although the value of domestic bonds redeemed every year has not increased significantly rel- ative to the size of the portfolio, the shares of floating and indexed domestic debt Table 4. Short-Run Impact of Changes in Key Variables on Public Sector Debt Service* Impact on Debt Type of Change Service as Percent of GDP Domestic real interest rate increases by about 75 percent of its standard deviaton" 0 5 COP depreciates by 10 percent relative to USD 0 4 Wit)h debt portfolio held constant as of end-200I Direct short-run impact Impact Couild vary in a dyniamilic setting, if impacts fromil new borrowjiigs and debt rollovers are factored into the calculations -Tbhe standard deviarion at a niontisly frequency was about 1300 basis points over 1997-2001 Source WB Staffcalctlations 17 Dtie to lack of comprehensive informationi on public debt other than central government debt, this discussion is centered on the latter Unless otherwise stated, all references in this subsection are to gross central government debt, and the discussion excludes other gov- ernment liabilities 18 Colombia lost investment grade rating in 1999 19 Real interest rates on domestic debt have averaged about 8 5 percent since 1995, com- pared with about 5 percenit (in real terms) on external debt 20 Constraints on access to internationial markets were eased in 2001 owing to a strLctured operation that used a World Bank guarantee and that allowed the government to again tap the 1 0-year segment of the market. 208 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE instruments grew from 13 percent by the end of 1996 to 48 percent by the end of 2001, thus significantly increasing the interest and exchange rate exposures. Given the high volatility of domestic real interest rates, managing this exposure is a key challenge to be addressed (see Figure 11, showing the volatility of Colombia's real interest rates). High domestic real interest rate volatility has translated into highly variable real coupon payments on public domestic debt-the average coupon (expressed in real terms) increased from close to 5 percent in 1998 to nearly 12 per- cent in 1999, falling steadily to levels close to 8 percent by 2001 (see Figure 12). The increase in real interest rates on domestic debt experienced during 1998-99 played an important role in the growth of public debt stocks in those years, as new debt was contracted partly to finance the rise in the cost of servicing public debt. In the mid-1990s, when the government started issuing bonds in domestic mar- kets, the strategy for mitigating the refinancing risk built up in the domestic port- folio was to issue longer-term external debt (10 years or more), increasing therefore, the average maturity and smoothing the external debt profile. However, this strat- egy could not be sustained for long in the context of large and growing fiscal imbal- ances, and the deterioration in access to international capital markets. As a result, the external portfolio also has a concentration of refinancing risk (although relatively less than in the case of the domestic portfolio). More specifically, in the external market, debt redemptions are clustering between 2003 and 2005, whereas in the domestic market, debt redemptions are clustered in 2002 and 2003. In 2001, for the Figure 11. Domestic Real Interest Rate Volatility* 0.25 - Domestic Real Interest Rate 0 20 - -- - - - - - _ _ _ _ _ _ _ _ _ _ _ 0 15 - - - - --- - _ _ _ 0.10-- -- ------ 0.00 - 000 < 0.00 ~ ~ ~ ~ 0 il\ Coo CZ N~~~~I *3-month standard deviations. Sou4rce. WB Staff calculations PUBLIC DEBT SUSTAINABILITY AND MANAGEMENT 209 Figure 12. Average Coupon in Real Terms on Domestic Public Debt 12- 19 ;1- - -- - - - - -.- --________ ___ ___ } X_ -___ 10- _ ____----____/ \___- I____ 8- 6( - --- -- - - - - - - - - - - - - - -- - - - - - - - 5-I I I II 1995 1996 1997 1998 1999 2000 2001 Source WB Staff calculations first time the government carried out a debt exchange operation in domestic mar- kets, aimed at smoothing out the near-term redemptions schedule. Additional oper- ations of this nature have been implemented in recent months with relative success, however close to 47 percent of domestic debt still matures in the next four years, thus suggesting that important refinancing risks still remain in the portfolio. Exposure of the portfolio to foreign exchange volatility could, in the future, become a serious problem. First, with the abandonment of formal exchange rate tar- gets and recent adoption of an inflation target by the Central Bank, the volatility of the exchange rate could be increased. Second, given the structure of the debt port- folio, the overall impact of the exchange rate on debt service could have a greater impact than interest rate effects because the former affects about half of the "princi- pal" of the portfolio in addition to the interest rate component. Third, the manage- ment of foreign currency risk is more difficult because there are no derivatives mar- kets for the government to hedge its currency exposure.21 The government has partially maiiaged this risk through matching, to the extent possible, assets and lia- bilities in terms of currency structure. Nevertheless, the accelerated increase in the 21. This is to be expected Since the government can choose to set or influence the value of domestic currency through various policy instruments, issues of moral hazard immedi- ately arise. In the absence of comprehensive monitoring and supervision mechanisms available to potential counterparts, the availability of hedging instruments is generally very limited 210 COLOMBIA. THE ECONOMiC FOUNDATION OF PEACE foreign stock of debt has outpaced growth in international reserves and exports, increasing foreign exchange exposure (see Figure 13). In addition, there is an increas- ing proportion of domestic debt indexed to the exchange rate. Thus, the govern- ment's debt management strategy in the future will have to focus on managing the exposure of the total debt portfolio denominated in foreign currency or linked to the exchange rate. Effective debt management has also been limited by weak institutional capacity and lack of an appropriate governance framework.22 Moreover, the lack of well-defined for- mal benchmarks used in day-to-day operations for the overall debt portfolio23 has also constrained efficiency in debt portfolio management. See Annex I for further discus- sion of key vulnerabilities in the government debt portfolio, and Annex IV for a more detailed technical discussion of portfolio and funding strategy benchmarking. c) Key Government Liabilities and Potential Impact on Debt Management Explicit, implicit, and contingent liabilities arising from subnational government entities, the Social Security and pension systems, and the financial system, among Figure 13. Public External Debt Ratios 100 - Exports/Debt Ratio 75 - -= YSi 5 50- _ , ,,- -- Reserves/Debt Ratio 25 -_- 0- I I I I 1996 1997 1998 1999 2000 2001 Source WB Staff calculations 22 Issues relating to the institutional framework and the need for thorough risk management analysis will be discussed in subsequent sections 23. No formal domestic benchmark has been produced A benchmark for the external port- folio has been produced but it has never been implemented PUBLIC DEBT SUSTAINABILITY AND MANAGEMEN T 211 other factors, have increased the vulnerability of fiscal accounts and are likely to worsen debt exposures. The existing framework for debt management faces severe limitations on handling these types of liabilities because no entity has the mandate nor the capacity to register and quantify them. Subnational government entities are currently exposed to refinancing risks, because most of their debt has maturities of between three and five years. In addi- tion, debt servicing is mostly linked to the 60-day deposit rate, creating substantial exposure to interest rate risk.24 To complicate matters further, a new law requires the government to provide a formal guarantee that could cover up to 40 percent of the outstanding debt and 100 percent of the new debt of departments and municipali- ties that negotiate a debt restructuring and fiscal adjustment program. This would shift the problem of debt servicing and financing from the subnational level to the central government.25 In addition, the recent crisis of the financial system, and in particular of the Housing Savings and Loan Associations (CAVs), has also led to an increase in gov- ernment liabilities (funded through the issuance of government debt as per Law 546, and Fondo de Garantia de las Instituciones Financieras [Financial Institutions Guarantee Fund, FOGAFIN] debt instruments with the guarantee of the central government). l' he combination of these liabilities (both direct and contingelnt) rep- resents an increase of nearly 7 percent of GDP in total government liabilities in 2001 alone.26 Furthermore, the public sector has also accumulated large pension lia- bilities (for example, in Social Security). Calculations of pension system liabilities vary, but there is agreement that at the end of 2001 the order of magnitude of the net present value of these types of liabilities amounted to about 200 percent of GDP. The government also faces other contingent liabilities arising from the efforts during the last decade to involve private sector participation in infrastructure, which included power purchase agreements and "minimum traffic" toll road guarantees. In 2000, a law was passed by the Congress requiring the central government and other public sector entities to quantify all implicit and explicit risks associated with guar- antees, and to appropriately budget for them. More recently, the public telecom- munications company provided guarantees to the private sector to foster an expan- 24. Interest rate exposure already induced a major debt-servicing problem for subnational entities in 1998 and 1999, after real interest rates on domestic debt increased substan- tially (the applicable nominal interest rate reached almost 37 percent per year in mid- 1998, when inflation averaged just under 19 percent). 25 Most large and medium-sized municipalities (excluding Bogora) had already renegotiated their debt by end 2001 However, these restructurings mostly addressed refinancing risks by smoothing the debt profile, but did not deal with interest rate exposures (most munic- ipal debt carries floating interest rates) At the same time, the central government has now provided a more formal guarantee on municipal debts. 26. Recent reforms of the supervisory and regulatory framework for the financial sector may have lowered the relative magnitude of contingent liabilities. 212 COLOMBLA THE ECONOMIC FOUNDATION OF PEACE sion of telecommunications infrastructure; limited monitoring capacity meant that no provisions were made by the government for these liabilities at the time of budget preparation. Information on these liabilities remains sparse because no entity currently has the mandate nor the capacity to register and quantify the magnitude of all implicit and con- tingent liabilities of the public sector in Colombia. The government has recendy started adopting a more systematic approach to the management of these types of liabilities. In 1997 the Department of National Planning (DNP) conducted, for the first time, a study to quantify the value of contingent and implicit liabilities.27 Law 448 of 1998 provided a legal framework that introduced the obligation to quantify and budget for the risks associated with the public sector's contingent liabilities. This law also author- ized the creation of a Guarantee Fund that would facilitate the financial management of resources budgeted by different public sector entities to deal with contingent liabili- ties, mandating the Ministry of Finance to oversee the operations of the overall scheme A small unit within the General Directorate for Public Credit at the Ministry of Finance was created in 2001 to fiilfill the Ministry's responsibilities in this area.28 Although the scope of operations of the new unit is still limited and its authority insufficient to fully address all issues pertaining to management of the public sector's contingent liabilities, its creation is nonetheless an important step in the right direction. d) Polziy, Legal, and Institutional Framework for Central Government Debt Management i. The Legal Framework and Governance The lack of a comprehensive and coherent conceptual and legal framework for debt management, combined with overlapping responsibilities among various institu- tions of the government and the Central Bank, substantially reduce the effectiveness and efficiency of the country's debt-management system. Weaknesses and gaps in the legal and regulatory framework reduce transparency and accountability. The scope of debt management is unnecessarily restricted-no entity currently has the legal mandate or capacity to comprehensively manage all types of explicit, implicit, and contingent liabilities of the public sector, as part of a unified strategy. In fact, Colombia's debt policy has been driven mainly by the need to finance the fiscal deficit, and consequently, debt decisions have been taken with limited systematic analysis of risks, and without a strategic framework for debt management. 27. The 1997 DNP Report estimated that total public sector contingent liabilities amounted to 154 percent of GDP, with pension and infrastructure-related liabilities accounting for 140 percent and 8 percent of GDP, respectively. 28. See Manejo de Pasivos Contngentes en el Marco de la Disciplina Fiscal en Colombia, 2002, by Jorge Enrique Cardona, XIV Regional Seminar on Fiscal Policy, 28-30 January; San- tiago, Chile I'UBLIC DEB-T SUSTAINABILITY AND MANAGEM[NT 213 In 1995, a special risk management task force was created within the Direcci6n General de Cr&dzto Pzubl2co (General Directorate for Public Credit, DGCP) to carry out Colombia's first risk analysis and to provide guidance on the introduction of a comprehensive risk management framework The concept of risk managemenlt was gradually introduced into the operations of the DGCP in the second part of the 1990s. In 1996, a Comitt Asesor de Deuda Pzublica (Advisory Committee for Public Debt, Debt Committee) was informally created to advise the Minister of Finance on debt management issues, including the approval of portfolio and issuing (funding) strategy benchmarks.29 The risk management unit of DGCP was designated as the Secretariat of the Debt Committee. The Debt Committee approved an initial set of foreign debt portfolio benchmarks in May 1997, and a first strategy review was car- ried out in December 1998. The Debt Committee has met only a few times and no portfolio benchmarks have been established yet for either domestic debt or for the portfolio as a whole. Moreover, the lack of formality and enforceability characteriz- ing this framework constrains management of debt portfolio risks, and prevents needed changes in the formulation of public debt policy in Colombia. As a result of these gaps, the only "formal" guiding policy for debt managemelnt remains almost exclusively dictated by the fiscal dimension. While there is no single framework law for public debt management (see Box 1), the governance structure for debt management is also deficient. There is no clear sep- aration of accountabilities and responsibilities between the higher-level bodies responsible for policy design and approval (the Minister of Finance and the Debt Committee), and the levels in the Ministry responsible for policy execution (the DGCP and Treasury). As a result, the Minister needs to formally approve (by means of a resolution) the execution of many individual operations carried out by the DGCP and Treasury.30 Given the dynamics of capital market operations and the lim- ited time a Minister can allocate to this process, in fact, the decisions are taken by the DGCP and Treasury but the formal accountabilities are not properly established ii. The Domestic Debt Markets In the 1990s, Colombia underwent a series of important institutional changes in the domestic capital markets, including the privatization31 and internationalization of banking services, the creation of a public debt market, and the reform of Social 29 The Debt Committee is chaired by the Milister of Finance, and members include the technical Vice-Minister of Finance and the head of Direcci6n General de Tesoreria It also includes the head of the Monetary Policy and Foreign Exchange Operations Unit of the Central Bank as an observer. In tmrn, the DGCP is invited to participate as an observer at the Internationial Reserves Committee of the Central Bank. 30 In the case of domestic Treasury debt auctions, the Minister is consulted about at what rates to declare an auction vacant. The Minister is also regularly consulted on the terms of the international bond operationis 31 Most of the privatized banks had been nationalized in the 1980s during the financial crisis 214 COLOMBIA THE ECONOMic FOUNDATION OF PEACE Box 1. Legal Framework for Debt Management in Colombia The legal framework for public debt management in Colombia is complex: * Issuance of Treasury bonds is regulated by a law that sets a ceiling consistent with the budget approved every year. * The Treasury (not DGCP) is authorized to issue short-term paper for cash management purposes, but not for budgetary financing purposes. o A 1992 Central Bank resolution gives the Central Bank authority over the issuance of short-term domestic public debt. The resolution also establishes that the Central Bank must approve the financial terms of all international public debt operations. * In 1995 Congress gave the Finance Ministry increased flexibility with inter- national bond operations. Congress would approve a multiyear quota for the gross value of operations that can be executed without requiring case-by-case authorization by Congress. The annual authorization to issue external debt was subsequently changed from a gross to a net basis. * The Finance Ministry has authority over debt issuance by public entities. Subnational governments need additional approval by Congress to issue for- eign debt. * The Ley de Semaforos of 1997 links the total indebtedness of a public entity to its debt servicing capacity. * A 1998 law establishes that any public entity issuing any type of contin- gent-explicit liability should estimate the magnitude of the liability incurred and its capacity to service the liability. Security that led to the creation of private pension funds. Despite some progress, domestic debt markets remain shallow and underdeveloped. Key obstacles to their development include (a) the use of a complex, nontransparent benchmark for the minimum rate of return of pension funds (established by law); (b) the presence of conflicting incentives that do not stimulate efficiency gains in the mutual fund industry; (c) limited know-how of mutual fund management techniques; (d) defi- ciencies in the current market makers system: liquidity is concentrated in very few points of the curve, the current ranking is too biased to the longer term instruments and some of the market makers (particularly the security houses) may be exposed to serious duration risk that could affect the capacity of the system to support the mar- ket in the difficult times (e.g: increases of the interest rate) (e) money markets are very inactive affecting the secondary market activity of the government bond mar- ket (f)cumbersome procedures for the authorization of securities-lending transac- tions that increase transaction costs; and (g) the use of an old and possibly biased methodology for calculating the Interest Rate Index (DTF) rate, which might limit PUBLIC DEBT SUSTAINABILITY AND MANAGEMENT 215 its value as a benchmark for the wholesale market for bank asset operations (loans) and contributes to the segmentation of the markets. iii. The Capacity of the Debt Office The institutions currently responsible for cash and debt management, namely the DGCP and the Treasury, are facing technological and human resource gaps that restrict the efficiency and effectiveness of debt management in Colombia. These gaps are major areas of concern, in view of the increase in the value of the debt portfolio, its growing complexiry, and the reliance on financial markets for funding strategies. Initially, the DGCP was primarily a back-office operation responsible for the timely service of government debt. It was also responsible for negotiating agreements with international official financing institutions.32 In 1993, the Ministry of Finance decided to stop refinancing operations with commercial banks and instead decided to access international capital markets directly. A small team was then created to support front-office operations in international markets, reporting directly to the director of the DGCP. In the early 1990s, the DGCP also housed a unit in charge of managing the debt portfolio of the government with the Central Bank, the asset portfolio of loans negotiated with subnational and other public institutions, the portfolio of explicit guarantees provided by the government to other public institu- tions, and local bond issuance operations by subnational entities. Following the 1991 constitutional reform, it was determined that Central Bank financing of the fiscal deficit would be phased out. The domestic debt management unit was therefore reoriented toward the supervision and management of other domestic liabilities. When large-scale Treasury bond operations began, a new task- force reporting to the Deputy Director for Planning was established using vacancies from different areas of the DGCP, but with no formal structure within the organi- zation. The Risk and Financial Planning unit (the middle office) was also created as a task force in 1995, but without a formal structure. In 1999 a decree was issued to formalize these previous institutional changes. This decree was initially challenged and it was only in 2001 that the partial reforms envisioned in the 1999 decree were introduced, the units responsible for risk management functions and international capital markets were transformed into individual deputy directorates. The DGCP employs a staff of about 100 and is divided into five deputy direc- torates (dealing with domestic capital markets, international capital markets, tradi- tional external debt loans, risk management, and operations), and two units report- ing directly to the head of the DGCP (legal and special projects). Modernization of the internal organizational structure of the DGCP has nevertheless been slow. Fur- 32. Although the government was involved in several fully integrated refinancing operations with international commercial banks (the so-called "Jumbo," "Concorde," and "Chal- lenger" operations), the negotiations and structuring of these operations were the respon- sibility of a small group of advisors at the Ministry of Finance This advisory group was dismantled in 1993 216 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE thermore, the current structure remains a collection of the partial, unsystematic reforms that took place in the last decade. This has resulted in lack of transparency and has eroded internal lines of responsibility and accountability. This situation has been compounded by the high turnover rate of heads of the DGCP between 1998 and 2000.33 Moreover, the hiring practices and low-wage structure established by public sec- tor law are becoming more and more binding, as business has shifted from dealing with official financing sources to private, market-based operations. The human resource and skills profile needed for market-based operations at the DGCP is sim- ilar to the profile demanded by higher-paying private financial sector institutions. As a result, junior staff trained at the DGCP often leave the institution to work in the private sector, generating high staff turnover. This problem is also present at more senior levels, also resulting in high management turnover. As a result of staffing problems, management of Colombi's large and complex international and domestic portfolios is frequently left in the hands of relatively inexperienced staff. This situation also erodes the capacity of the Debt Office to identify and take advan- tage in a systematic way of active debt management operations such as debt exchanges and buyback operations. In other cases, limited capaclty may result in increased operational risk in the execution of relatively complex operations like derivatives to cover cross-currency risks (swaps and futures). Unnecessary layers of complexity in the institutional arrangements for debt man- agement operations also increase transaction costs and inefficiencies. Cash manage- ment operations and negotiations with "captive sources" for Treasury bonds are the responsibility of a unit in the Treasury. Although evidence indicates that there is good communication between the DGCP and Treasury,34 there is nevertheless a spurious split of responsibilities with no central accountability. As a result of the rel- atively recent decision by the Central Bank to allow issuance of short-term Treasury bills, the Treasury is in charge of short-term paper, while the DGCP is responsible for medium- and longer-term issues, with no senior "operational" manager respon- sible for operations across the entire spectrum. The analysis of international markets and of active debt management operations is currently carried out using nonintegrated information and technology (IT) sys- tems-a basic back-office system and a multiplicity of spreadsheet-based systems- that cannot be updated quickly and that rely on limited automated functions. This makes the overall process of debt management and operations slow and exposed to high operational risks. It also substantially limits the capacity of the Debt Office to engage in active debt management operations (for example, buybacks, swaps, and exchanges) that require very current and precise information. Although in the last seven years the DGCP has engaged in several (in-house) IT developments, these 33. There have been three DGCP directors in the last three years. 34 There iS a Treasury committee headed by the Deputy Minister to supervise cash man- agement pro)ections and execution, in which the DGCP participates PuBIuc DEBT SUSTAINABILITY AND MANAGEMENT 217 efforts have not produced the expected results.35 The basic back-office operation has been partially automated through a system developed in-hotise and based on out- dated database technology (INGRES), which was introduced in 1995 and modified in 1997. Today there is still no integrated database for external and local debt that could be used to generate a full portfolio analysis on a timely basis. e) Strengthening the Framework for Debt Management Colombia's fiscal imbalances require strong fiscal adjustments. In addition, in light of the large risk exposures and vulnerabilities of its debt portfolio, it requires a transparent pol- icy for debt management. An improved fiscal discipline, together with strong debt port- folio management, including the control of the growth of debt stocks, will help restore fiscal and debt sustainability. The iniltial success of the fiscal stabilization program pro- vides the government with the necessary credibility to implement an active debt man- agement strategy. Actions in this regard will also provide additional comfort to investors and thus help reduce the costs of borrowing and keep liquidity and refinancing risks to a minimum. Failure to address the issues raised by the portfolio structure, and short- comings of the institutional framework and the portfolio strategy for debt management, could significantly delay the full impact of the fiscal stabilization efforts.36 i. Improving the Legal Framework and Governance To create a proper legal and governanice framework for debt management operations, the government should prepare and submit for Congressionlal approval a framework law consolidating all legal and regulatory instruments on debt management. Key principles of the proposed framework law37 should include provisions for (a) separa- tion of policy formulation, implementation, and evaluation functions; (b) creation of a new Debt Office (by merging the DGCP and Treasury) and redefinition of the role and structure of the Debt Committee in light of the proposed new framework; (c) creation of an independent entity reporting directly to the Minister of Finance for the evaluation of debt policy implementation, including the definition of provisions for ensuring adequate checks and balances;38 (d) very clear objectives that include risk-cost tradeoff issues as a central dimension of debt management, (e) use of mod- 35 Part of the problem is caused by a lack of understanding of the Debt Office's own needs, and also by the limited time invested in conveying information to the IT developers. Fur- thermore, senior management has not fully supported the IT development process, shift- ing responsibilities to more junior staff 36 For example, in the late 1990s, the full impact of the fiscal adjustment in Brazil was delayed because, at that time, Brazil lacked an adequate debt management strategy to deal with a vulnerable debt structure. 37. For further disCLssion of the key principles that could underpin a new framework law, see "Guidelines for Public Debt Management," World Bank and IMF, March 2001 38 External auditilig could be used as an alternative to the creation of a new auditing unit. 218 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE ern financial instruments and tools for active debt management; (f) definition of responsibilities and accountabilities at different levels of the government; and (g) sep- aration of debt management from the Central Bank's mandate, particularly with regard to the issuance of short-term Treasury bills to improve the government's cash management flexibility. (See Box 2 for a brief description of the Brazilian experience with reform of the legal and governance framework for debt management.) Within this new framework, the Minister of Finance, with the advice of the Debt Committee, should define a set of portfolio benchmarks and the overall strategy to achieve it. A clear and transparent mandate for the execution of this policy decision should then be given to the Debt Office (or the DGCP). Moreover, the assessment and evaluation of the quality of the execution of debt policy should be conducted by an entity or auditing unit (or external audits) independent of the Debt Com- mittee and the Debt Office. Box 2. Reform of the Legal Framework for the Debt Management Office in Brazil A new institutional framework for debt management has been adopted in Brazil under Decree No. 3782 of April 2001. The Public Debt Office now comprises three main units: back, middle, and front offices. The back office is responsible for the registry, control, servicing, and accounting of both domestic and foreign debts. The responsibilities of the middle office include the development of medium- and long-term strategies aimed at reducing debt costs and risks, macroeconomic monitoring, and investor relations. The front office is respon- sible for the design and implementation of short-term strategies related to bond issuance in international and domestic markets. The new institutional frame- work will result in a substantial improvement in debt management capacity, allowing for the standardization of operational controls, better monitoring of risks, and the separation of functions concerning long-term (strategic) and short-term (tactical) planning. ii. Debt Management Strategy and Portfolio Restructuring To improve debt management and debt portfolio structure, the following specific steps could be considered (a) the Debt Office (or the DGCP) should develop new portfolio benchmarks for the short, medium, and long term;39 (b) the new bench- mark strategies should be presented to the Debt Committee, and submitted for sub- 39. The strategy should distinguish between portfolio and issuing (funding) benchmarking, and should include a formal risk management frarnework as a guiding principle (for exarmple, use the concept of maximum tolerable budget-at-risk-that is, maximum BaR-as a selection criterion). PUBLIC DEBr SUS'I-AINABILITY AND MANAGEMENT 219 sequent approval by the Minister of Finance with the Committee's advice;40 (c) the strategy should include the systematic use of active debt management operations (exchanges, swaps, buybacks, and so forth) to modify the existing stocks and the funding strategy for the new operations; and (d) the Minister of Finance, with the advice of the Debt Committee, should provide clear guidelines to the Debt Office (or the DGCP) for implementation of the strategy. In addition, to facilitate man- agement of the explicit contingent liabilities of the government, the Debt Office (or the DGCP) should develop and implement a plan to improve monitoring (registra- tion and quantification) of these types of liabilities. The portfolio and issuing (funding) strategies should integrate the long-term fis- cal and debt sustainability analysis of the government, with a thorough analysis of portfolio risk exposures. These benchmarks should reflect the government's assess- ment of an "acceptable" level of fiscal risk (budget-at-risk and or cash-flow-at-risk) relative to the expected reduction in cost of funding. To reduce central government debt exposure, a comprehensive restructuring of the debt portfolio is warranted. More specifically, debt portfolio restructuring should address the "bunching" of domestic debt and external debt amortization. The strategy could include the max- imum level of amortizations that the government could "tolerate" under different scenarios of interest and exchange rates, and seek a set of financial instruments and terms that could achieve that benchmark. Given that there are no hedging instru- ments available that could be used to manage foreign exchange risks of the size asso- ciated with the government's portfolio, and that international reserves provide only a limited natural hedge, this type of risk should be managed through the control of the primary issuance of instruments and making strong efforts to deepen the domes- tic debt markets. (See Box 3 for an overview of the Brazilian experience with the modernization of the risk management strategy.) iii. Strengthening the Capacity of the Debt Office To strengthen the capacity of the Debt Office, the Ministry of Finance should cen- tralize operations in a single Debt Office, with expertise in both cash and debt man- agement (merging the DGCP and Treasury), alternatively, within or outside the Min- istry (as a special administrative unit or an institute), in accordance with the new framework law. In the short term, the government should prepare an action plan that includes at least the following key elements: (a) an organizational chart, including a front office, a middle office responsible for strategic debt analysis and risk manage- ment, and a back office;4" (b) a human resources plan, including staffing and training; (c) a comprehensive diagnosis of information systems and plans for improvement; and (d) proposals for the elimination or transfer to other entities of all noncore functions. 40 In some countries (for example, Poland, Sweden, and the United Kingdom) the Committee is actually a board chaired by the Minister of Finance and empowered to take key decisions. 41 The front and back offices should ideally be the same ones for cash and debt manage- ment operations. 220 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE Box 3. Risk Management in Brazil The main types of risks arising from public debt currently being monitored by the Brazilian government are (a) refinancing risk, (b) market risk, and (c) credit risk. Risk analysis is conducted under an Asset and Liability Management (ALM) framework, using a model that allows debt managers to project expected and potential debt costs under different refinancing strategies in the medium and long term. Key debt management indicators, such as average maturity, duration, and debt composition are generated for each strategy, thus allowing senior management to choose the optimal strategy to be pursued. A more pow- erful risk management system, introduced in early 2002, also allows a compre- hensive examination of the characteristics of the assets and liabilities of the pub- lic sector, and the adoption of alternative approaches to "at-risk" models (for example, Cost-at-Risk, Cash-Flow-at-Risk, and Budget-at-Risk) The proposed functional structure consists of a unit responsible for leading trans- actions (the front office), one responsible for analyzing portfolio risks and aligning the actual portfolio with the benchmarks (the middle office) and, finally, a unit in charge of settlements (the back office). The DGCP and Treasury should be merged into a debt management unit to ensure consistency along all the term-structure of government debt, and to eliminate overlaps (especially in the back office and domes- tic front offices). Two possible alternatives for institutional design have been identi- fied: (a) transforming the Debt Office into a special administrative unit under the Ministry's structure (following the example of the tax and customs units), or (b) cre- ating a separate institute under the Ministry of Finance following the structure of institutions like FOGAFIN. There are advantages and disadvantages to both options. Under the first option, constraints on salaries and administrative proce- dures are likely to remain, but the smooth flow of information within the Ministry will be preserved. Under the second option, there will likely be more salary and structure flexibility that will help increase the capacity of the Debt Office, but rela- tionships with Ministry departments-units will likely be negatively affected. With a clearer governance framework provided by a new law, the latter may nevertheless become the preferred option. Noncore business activities currently under the DGCP and Treasury should be phased out or transferred to other units within the Ministry or other institutions like the National Planning Department.42 However, the current arrangements whereby a specific group within the Debt Office is responsible for 42. Noncore business activities include participation in negotiations of multilateral and bilat- eral loans, supervision of the issuance of domestic bonds by subnational government enti- ties, and involvement in the financial design of infrastructure projects and in the process of privatizing public companies. PUBLIC DEBT SUSTAINABILITY AND MANAGEMENT 221 monitoring contingent liabilities should be strengthened and given increased authority to request timely information from other government entities responsible for the management of those liabilities. The new unit should have access to necessary human resources and skills to carry out its functions, and should have the administrative flexibility to act and respond to fast changes and opportunities in the financial market. This requires adlusting the salary level and structure to make it more competitive vis-a-vis the private sector It is also important to establish and institutionalize a comprehensive training program, including specialized courses, scholarships, and internships in public and private sec- tor institutions in Colombia and abroad The proposed reforms of the organizational structure will require a new, fully reengineered information and technology system 43 The Ministry of Finance should also issue comprehensive regulations and guilde- lines for debt management and for Debt Office operational procedures. These reg- ulations and guidelines should clearly define responsibilities and internal account- abilities; ensure that adequate checks and balances are maintained; address issues of actual or potential conflicts of interest encountered by Debt Office staff; define mechanisms and procedures for preserving the confidentiality of information and operations of the Debt Office, and establish a code of conduct to guide the behav- ior of Debt Office staff, consultants, and all other entities or persons rendering serv- ices to the Debt Office. With regard to IT, there are a few important issues that require attention. First, the new structure will require a new business plan to identify user IT requirements prior to engaging in a major upgrading of existing IT systems. A coherent plan is needed for the development of a unified database, analytical software, and market tools that integrate IT information flows among front, middle, and back offices. The improvement of the middle office's analytical software could proceed immediately, however, because it is relatively independent of the new structure, and significant progress has already been achieved. Three areas need further improvement (a) cur- rent VAR-based risk analysis methodologies can be improved by incorporating the concept of budget risk in the modeling, and tools need to be developed for such budget risk quantification and benchmark exercises, (b) tools need to be developed so that external and domestic debt analysis can be integrated; and (c) current tools need increased "automation" to reduce operational risk and training costs for new staff. It is worth emphasizing that, to succeed, improvements in IT will need to be championed by senior management of the Debt Office. iv. Deepening the Domestic Debt Markets To enhance and deepen the functioning of domestic markets, thus facilitating the management of public debt, the Ministry of Finance should draft and submit for Congressional approval a legal amendment to simplify the benchmark for the mm- 43 The decision to introduce the UNDP database (CIGADE) that centralizes all debt oper- ations is an important step in the right direction. 222 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE imum rate of return of pension funds.44 The benchmark for the minimum rate of return of pension funds should be changed to the lowest nonbinding benchmark of either (a) a real rate of return calculated so as to achieve a specific pension replace- ment rate, or (b) an average of the current rate of return on the government's medium-term bonds. In addition, the Ministry of Finance should undertake a study to identify the major regulatory and incentive constraints for increasing the effi- ciency of collective investment vehicles and mutual funds, and to develop an action plan to remove such constraints. The government could also explore and analyze the conditions for establishing a securities-lending backstop facility that, in exchange for cash or securities, lends different securities to an intermediary to support repo and short-selling activities. Finally, the government and the Central Bank should assess the suitability of the current DTF rate calculation methodology and evaluate with the industry the possible introduction of an alternative market reference rate (for example, the London Inter-Bank Operations Rate [LIBOR] type) for secondary market activities. To successfully develop domestic debt (and capital) markets, the investor base must be expanded. The market needs a strong and diversified group of institutional investors (including pension funds, insurance, mutual funds, and foreign investors) that can generate demand for different segments of the yield curve (short, medium, and long term), and generate a liquid secondary market. Reducing transaction and information costs could facilitate this process (see Box 4 for a brief discussion of alternatives for building benchmarks). One important step toward domestic capital market development is the phasing out of government reliance on captive investors. Currently, nearly 50 percent of new issuing is allocated to a group of captive investors-the so-called inversionesforzosasy convenidas (see Figure 1 in Annex V) for a breakdown of domestic public debt by type of holder). These are basically the public pension fund (at the Social Security Institute) and government entities run- ning cash surpluses that, by regulation, have to be invested in a portfolio of govern- ment bonds through direct negotiations with the Treasury. The size and importance of captive sources in government financing have introduced market distortions (seg- mentation) and have negatively affected the development of the investment and mutual fund industry.45 In the case of the pension funds, simplifying complex regulations behind the cal- culation of the minimum yield on pension fund investments could reduce unneces- sary "noise" and reduce information costs. For mutual funds, further market open- ing for foreign financial companies managing mutual, pension, and investment 44. See "Developing Government Bond Markets- A Handbook," 2001, Washington, D.C.: World Bank and IMF 45. Recently, the oil company ECOPETROL was allowed to manage its pension reserves through outsourcing to a group of private fund managers (Fiduciarias) that had given an important boost to the mutual fund industry, in particular the medium-term funds (I to 5 years) PUHI3LC DEBr SUSTAINABILITY AND MANACEMFN'T 223 Box 4. Building Benchmark Issues Through Reopening and Buyback Operations As part of a strategy to build benchmark issues, "reopening" and "buyback" operations can be used to increase the "fungibility" of benchmark instruments by reducing the number of issues outstanding and concentrating remaining issues on fewer benchmark bonds of the desired maturities Reopening the same type of bond (including same coupon and maturity) over several auctions elim- mates the need to always open a new issue, maintains bond issues in their life cycle, and helps build outstanding volume to that desired for the benchlmark. Buybacks can be used to eliminate issues that are nonstandard and not very actively traded, thus aiding in the concentration and standardization of out- standing issues. funds could create competition in the mutual fund industry to diversify and extend portfolio duration. These changes will likely induce consolidation in the industry, which will also reinforce the process of modernizing financial management compa- nies. (See Box 5 for an overview of the international experience with institutional investors and the development of domestic debt markets.) The government could also promote the rating of asset management companies (and of the funds they administer) by local or international rating agencies, reduc- ing information costs for investors and increasing competition and transparency. While most of the above principles also apply to the insurance funds industry, pro- Box 5. Growth and Importance of Institutional Investors Institutional investors, including pension funds, insuranice companies, and mutual funds, have experienced rapid growth worldwide over the last 20 years In the early 1980s, only in a few industrialized countries did institutional investors control assets in excess of 50 percent of GDP By late 2000 nearly all OECD countries had institutional investors with resources in excess of 50 per- cent of GDP Chile and Malaysia, among developing countries, also surpassed the 50 percent benchmark. Private pension fund systems have played an impor- tant role in the development of domestic capital markets; in turn, pension funds strengthen the development of insurance companies, which promotes the avail- ability of annuity products. In Chile, for example, life insurance companies already control assets in excess of 10 percent of GDP. Institutional investors can thus play a unique role in intermediating domestic savings and developing cap- ital markets. 224 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE moting the development of annuity products (rentas vitahczas) through new regula- tion (now under preparation) would further promote the development of domestic capital markets. One important area for improvement relates to the existing relationship of the Debt Office with primary dealers. The system needs to be reformed to introduce liq- uidity along the entire yield curve and not just at a few points in the longer-term segment, as it currently operates. This would require that the valuation procedure put equal emphasis on key points along the curve. The government should also con- sider a rotating system by which each dealer is given a combination of off-the-run and on-the-run points, so that pressure is put along the entire curve.46 This will allow the government to move effectively along the curve depending on expectations and market conditions. An important issue that needs proper government supervision is the level of risk exposure that key market makers acquire when acting as Primary Dealers (PDs) of the government In particular, security houses seem to be taking large risks, given the relatively lenient capital adequacy standards and limited enforcement that they currently face. This situation presents a sharp contrast with that of another impor- tant group of intermediaries-the banks-which are subject to more demanding capital adequacy and supervision standards. This is important for the financial sys- tem as a whole and for the sound operation of the PD system, because a sudden increase in interest rates could conceivably force an important group of PDs to move out of the market-making business, or even to go out of business altogether. Market liquidity and the risk exposure of government would likely deteriorate in such a sit- uation, highlighting the need for Joint action by the Debt Office and the securities market regulator. To support a more active secondary market and facilitate the holding of longer- term instruments, market agents need to be able to access instruments that will promote liquidity and hedge unnecessary risks (such as repos or simultdneas). How- ever, there are constraints on the development of secondary markets, including the lack of standardized and fungible issues. A possible solution to be explored is the introduction of a securities-lending backstop facility that, in exchange for cash or securities, lends different securities to an intermediary. These and similar arrange- ments have been adopted in several countries (see Box 6). The Colombian govern- ment could explore this or other types of solutions to facilitate the development of secondary markets. The introduction of alternative reference rates to the current DTF may con- tribute to an improvement of financial market integration and price signaling within 46. Even in highly developed markets like the United States, the liquidity of off-the-run issues is normally poor compared with on-the-run issues. This, however, does not guar- antee that on-the-run issues keep their benchmark status if, for example, there is a large discrepancy berween original and remaining maturities, or if trading becomes thin despite its being an on-the-run issue PUBLIC DEBT SUSTAINABILI rY AND MANAGEMENT 225 Box 6. The Role of Secondary Market Windows In relatively underdeveloped markets, liquidity risks are high. In this context, a secondary market window may be opened (possibly with government or Cen- tral Bank support) to facilitate market making. Typically, dealers supporting sec- ondary market trading can buy or sell securities through the government-spon- sored secondary market window, thus encouraging dealers to provide quotations to the public, maintaining market continuity, and reducing liquid- ity risks. In the U.K., the Bank of England has operated secondary market win- dows for several types of instruments to encourage the development of a mar- ket for those instruments (for example, gilt-edged and index-linked securities) Examples of developing countries that have introduced similar types of arrange- ments include Iceland, Malaysia, and Thailand. financial market segments (for example, bank deposits and lending, money markets, and government bond markets). The alternative reference rate should be transpar- ent and consistent with the activity of the secondary markets to effectively facilitate the pricing of financial assets and liabilities. For example, a group of financial insti- tutions (market makers) has already been working on a proposal to create a Bogota Inter-Bank Operations Rate [BIBOR], a concept similar to LIBOR). The govern- ment )ointly with the Central Bank should evaluate and explore alternative solu- tions. Finally, promoting the development of retail investors could also strengthen capital markets. In general, there are two main avenues to tap this market. (a) the creation of specialized Treasury bond funds (for example, Spain was very successful in promoting specialized government bond funds as close substitutes for bank deposits), and (b) the direct distribution of securities (through, for example, the Internet). f) Estimated Fiscal Impact Implementation of reforms for improved debt management, complementing the implementation of a comprehensive reform package, is estimated to yield annual average budgetary savings on interest payments of about US$275 million through 2005 (in constant U S. dollars), equivalent to an average of about 0.35 percent of annual GDP. Annual average budgetary savings on interest of US$375 million (in constant U.S. dollars), or an average of about 0.45 percent of annual GDP, are pro- jected through 2010. Savings are estimated relative to a no-reforms scenario where the primary balance-to-GDP ratio is kept at the 2001 level. In this scenario the debt-to-GDP ratio would reach at least 80 percent of GDP by 2010. In the base case scenario, which assumes implementation of a fiscal effort sufficient to bring the pri- 226 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE mary surplus to the level required to keep the debt-to-GDP ratio on a sustainable path, the debt-to-GDP ratio reaches 57 percent of GDP by 2010 (a 29 percent fall in the debt-to-GDP ratio relative to the no-reforms scenario). The more comprehensive the structural reforms effectively implemented and the greater the fiscal effort undertaken, the smaller the debt-to-GDP ratio will be by 2010, all else being equal. We assume that the maximum plausible adjustment effort would keep the debt-to-GDP ratio around the 2000 level (say, at about 46 percent of GDP, a 43 percent fall in the debt-to-GDP ratio relative to the no-reforms sce- nario). If it is assumed that a I percent reduction in the debt-to-GDP ratio leads to a fall of 7 basis points in average interest rates, we estimate that the potential reduc- tion in interest rates from introducing increasingly comprehensive reform packages ranges from about 200 to 300 basis points (which translates into average annual sav- ings of between 0.35 and 0.5 percent of GDP).47 47 This is consistent with a model of the long-run equilibrium spreads estimated by Gold- man Sachs for Colombia. Every percentage change in the external debt-to-GDP ratio leads to a compression in spreads of about 7 basis points Nevertheless, estimating the direct fiscal impact of improved debt management would require a detailed comparison between the proposed new portfolio benchmark and issuing (funding) strategies, and current benchmarks and practices in Colombia. Because no new benchmarks have yet been developed, it is difficult to establish a priori the scope for fiscal savings following improved debt management, subject to newly defined risk limits. PUBLIC DEBT SUS IAINABILI l'Y AND MANAGEMENTI 227 Annex I Key Elements of Vulnerability of Central Government Debt I. Internal Debt The largest share of internal central government debt consists of Treasury bonds (TES).48 At the end of 2000, TES issuance accounted for 86 percent of total inter- nal debt. There are four types of TES issuance: fixed-rate bonds, CPI- and UVR- indexed bonds, and USD-indexed bonds. CPI- and UVR-indexed bonds also differ in that the latter type capitalizes the interest payments while the former type does not. The various types of bonds are also available in different combinationls of matu- rities (see Table A.1.1). Figure A.l. I shows the outstanding value of these four types of instruments since 1996. The shares of floating and indexed debt instruments grew from I1 percent at the end of 1996 to 42 percent of the outstanding stock of TES by the end of 2001. WVR-indexed bonds remain a very small part of the domestic debt portfolio, since most of the increase in the stock of bonds since 1996 took place in CPI-indexed and USD-indexed bonds. Together with this change in structure of the domestic debt portfolio, exposures to the exchange rate and inflation have increased considerably. The duration and modified duration of the domestic portfolio have been rising steadily since 1998, but are still low (remaining at around 2). This is the result of Table A.I.1. Types of Domestic Treasury Bonds Type of Rate Denomination maturity (years) Fixed rate COP 1, 2, 3 and 5 CPI-floating rate COP 5, 7 and 10 (no capitalization of interest) Fixed rate UVR*-indexed 5 and 7 (capitalizationi of interest) Fixed rate USD**-mdexed 2 and 3 *Real Value Units -US dollar 48 Due to lack of consistent informationi on public debt other than central government debt, this dIscussion is centered on the latter Unless otherwise stated, all references are to gross cencral government debt This discussion excludes government liabilities otler than debt (for example, contingenit or implicit liabilities arising from Social Security or sub- national levels of government), since no comprehensive inventory of all governmient ha- bilities exists 228 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Figure A.I. 1. Domestic Central Government Debt by Type of Coupon TES Bonds 14-n Fixed Rate COP 12- L CPI FloatingRate- _ _ _ :]USD Fixed Rate _ :, 10- f UVR Fixed Rate a 8 1996 1997 1998 1999 2000 2001 Source MoF, DGCP high interest rates, which quickly return to investors (in the form of coupon pay- ments) substantial amounts of their investment.49 The half-life of the portfolio increased sharply in 1998 to about 4.5, then fell in 1999 and 2000 to 3.5, and again recovered to 4.5 by late 2001 (see Figure A.I.2). Figure A.I.3 shows the evolution of the domestic debt maturity structure since 1997. At the end of 1997, almost 90 percent of domestic debt would mature within four years. By contrast, at the end of 2001, about 47 percent of total domestic debt is expected to mature over the next four years. This suggests that with a "flattening" of the maturity structure, refinancing risks have improved since 1997. However, there is still an immediate and significant refinancing risk, because 30 percent of the domestic debt portfolio will mature within a year.50 49 Duration can also be thought of as a measure of the time that an investor keeps his money invested For zero coupon bonds, maturity is a straightforward measure, since investor's money is tied to maturity By contrast, with coupon bonds, maturity is an imperfect measure of investment time since part of the value of the investment is returned in the form of coupon payments before the bond matures. 50. Formally, to conduct a more comprehensive assessment of refinancing risk we would need to identify the main holders of maturing domestic government debt and estimate their propensity to agree to offer refinancing However, it remains true that refinancing risk is higher than it would be (keeping the identity of bondholders constant) if the debt profile were "flatter." PUBLIC DEBT SUSTAINABILITY AND MANAGEMENT 229 Figure A.I.2. Half Life and Duration of Domestic Central Government Debt 45- 4 0 __ __ _ ~~~~Half Life<-_ 3 5 ________ 25- 1 5-_/-_ o 5 - ~ _ - - - - - - - - - - - - - _ D- - O 5O- D u r a ti o n M o d ifi e d D u rla t io n 00- I I I I I 1995 1996 1997 1998 1999 2000 2001 Source Mor, DGCP Figure A.1.3. Maturity of Domestic Central Government Debt 50 - 45-- - - - 1997 2, 40-,- 1999 L 35- ,- 2001 8 35---___ 0 ao 30,-__ ___ ____ ____ _ __ _,___ ___,____ 30o -- = -A -= E 5 20 ----…---------- - - 20- 10 0 - I I I I I --r'r 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Maturity (Years) Source Mor, DGCP 230 COLOMBIA. THE ECONOMIC FOUNDAnON OF PEACE I1. External Debt The evolution of the currency composition of external debt is shown in Figures A.I.4 and A.I.5. At the end of 2001, USD, EUR, and JPY debt account for 77 per- cent, 17 percent, and 5 percent of total external debt, respectively. The current benchmark portfolio is USD (83 percent), EUR (13 percent), and JPY (4 percent). The Increase in the share of EUR-denominated debt and decline of USD debt dur- ing 2001 has meant a further departure from the existing benchmark currency com- position. External debt has a larger share of fixed-rate debt than in the case of domestic debt (compare Figures A.I.1 and A.I.6). Fixed-rate debt accounted for about 65 per- cent of external debt at the end of 2001, with variable and semivariable rates accounting for the rest of the portfolio. The coupon structure of the external debt portfolio has remained relatively stable over time, with perhaps a mild trend against semivariable rates and in favor of variable rates. In addition, the external portfolio traditionally has had longer duration (around 3.5 years) than the internal portfolio (compare Figures A.I.2 and A.I.7). In addition, the half-life of the external portfolio is also higher and more stable than the half-life of the domestic portfolio. Together, these characteristics imply a smaller interest rate risk than in the case of domestic debt. Figure A.1.8 shows the evolution of the external debt maturity structure during 1998-2001. Compared with domestic debt, the external debt has Figure A.I.4. Currency Structure of External Public Debt 90 - _______________________________ 80 … _ P 2000 70 -____ _____ ___- 60 D1 2001 _ e Benchmark ,, 50- _ __ _____ ___ 30 _ _ _ _- - _ _ 20- ---__ __ ________ __._.__ 10 - _ - 0 - JPY EUR USD Source MoF, DGCP PUBLIC DEBT SUSTAINABILITY AND MANAGEMENT 231 a relatively smooth maturity structure. As a result, only 11 percent of external debt matures within a year (compared with 30 percent for internal debt). However, rela- tively more significant external refinancing needs will arise in 2003, when 16 per- cent of the current external debt will mature. As can be seen in Figure A.1 8, the pro- portion of the external portfolio maturing in the one-to-three-year range increased considerably in 2001, relative to the levels of 1997 (36 percent versus 27 percent, respectively). Correspondingly, the share in the portfolio of longer maturities has fallen markedly from the levels of earlier years. External refinancing risk has thus been rising steadily in the last five years. Figure A.I.5. Currency Structure of External Public Debt 25,000- -i USD U JPY 2 EUR U Other o 15,000- 5,000 0 - I I I I I I I 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Source MoF, DGCP 232 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Figure A.I.6. External Central Government Debt by Type of Coupon 20 - _____________________________ liFixed 15- 20 10 -_ ___ __ _. t' 00 CO C) C,o :7~~~~~~~~~~~~7 Source. MoF, DGCP Figure A.I.7. Half Life and Duration of External Central Government Debt 100 - COL Public Revenues/Total Public Debt 75- \ v bC ARG Public Revenues/Total Public Debt 50 ---- - - . - - - 25- I - I 1996 1997 1998 1999 2000 2001 Source MoF, DGCP PUBLIC DEBT SUSTAINABILITY AND MANAGEMENTI 233 Figure A.1.8. Maturity Structure of External Central Government Debt 16- 0 6 - -- - - - - IU 4 -1999 2 -___ - 2001 _____ _ -Benchmark 0 - I I I I I I 1 2 3 4 5 6 7 8 9 10 Maturity (Years) Source MoF. DGCP 234 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Annex II Analytical Framework for Fiscal Sustainability I. The Case of Constant Growth and Interest Rates For a given definition of the public sector, let GDPt be the economy's gross domes- tic product (GDP) in period t, D, be the net stock of public sector debt, PS, the pub- lic sector's primary surplus, X, the amount of seignorage financing, g the (constant) rate of growth of GDP, and z the (constant) average interest rate on public sector debt. In the standard framework for the analysis of debt sustainability,51 the increase in debt is a function of the level of debt multiplied by the interest rate, and of the government's primary surplus adjusted for seignorage financing of the budget. This is expressed as: D,= - Dt = 0D, - PS, - X, For consistency with the analysis conducted in other Chapters, our definition of "government" is the Consolidated Public Sector. Since in Colombia direct Central Bank financing of the fiscal deficit has been phased out and institutional arrange- ments are such that the independence of the Central Bank is credible, we assume in our projections that the likelihood that the government would resort to inflationary financing of the deficit is negligible. However, the balance of the Central Bank is included above the line (as a government revenue item directly included in the cal- culation of the primary surplus), as is done in other Chapters. In this case, the increase in the debt stock can be simplified to: Dt + 1 - Dt = iDt - PSt Rearranging terms and dividing both sides of the equation by GDP+ = (1 + g)GDPt, we obtain: Dt + I 1+i Dt 1 PSt GDPt+i 1+gGDPt 1+gGDPt To simplify this expression, debt-to-GDP and primary-surplus-to-GDP ratios are expressed in lower case: a' 1+i 1 dt+1 =-d,I dt ps, l+g l+g s 51 See, for example, "The Economics of the Government Budget Constraint," 1990, by Easterly and Fischer, The World Bank Research Observer 5 (2). PUBLIC DEBT SUSTAINABILITY AND MANAGEMENT 235 This difference equation describes the behavior of the debt-to-GDP ratio given constant interest and GDP growth rates, and given a sequence over time of primary surpluses. It can be shown that if: Lirln' - 1 i K dv = 0 holds, then the debt-to-GDP ratio is nonexplosive. Applying this condition to the difference equation given above, yields. d, _ This states that to avoid a nonexplosive debt-to-GDP ratio, the pubic sector must generate a sequence of primary surpluses (relative to GDP) such that the net present value of the sequence is worth at least as much as the debt-to-GDP ratio at time t. If we further assume that future primary surplus-to-GDP ratios are constant, and mak- ing use of established results on the summation of sequences, it can be shown that. ps = (I - g)d, This expression can be interpreted as saying that a stable debt-to-GDP ratio is obtained when the ratio of the primary surplus to GDP is equal to the difference between the interest rate and the economy's rate of growth, multiplied by the cur- rent debt-to-GDP ratio. a) Relaxing the Assumptions: Time- Varying Growth and Interest Rates The constant interest and growth rates assumption is restrictive. Assume instead that interest and growth rates vary until period Tand remain constant thereafter.52 The difference equation previously derived to describe the behavior of the debt-to-GDP ratio over time can be generalized to: l+ g 1+g, A solution to this equation can be found by imposing a modified boundary con- dition to account for time-varying interest and growth rates. It can be shown after some rearrangement that d) =(- R,,p, where RT I+ I 52. This discussion is based on Debt Sustainabshty in Latin America, Deutsche Bank, Global Markets Research, December 2000. For a technical discussion of debt and deficit financ- ing issues, including of the government budget constraint under rtine-varying interest rates, the reader may also refer to Section 3.4 in Lectures on Macroeconomics, by Blanchard and Fischer, MIT Press, 1989 236 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE By assumption, after period T, 1 + g remains constant. Thus, 1+: d , = J(-) RI vPSv + RI,T( ) (I1 + i )v PS This implies that: PS p T- (1 l+ I+i +~RIT i~ l+ (1+j)I R1,v +R1,T-1 (1+i) - This expression yields the permanent primary surplus-to-GDP ratio that is needed to ensure the sustainability of public finances, under the assumption that interest and growth rates are allowed to vary until period T. Interest rates are assumed to remain constant thereafter. The permanent primary surplus-to-GDP ratio calculated above can be compared with recent (historical) primary surpluses, to estimate the fiscal effort required to ensure the sustainability of public finances. However, if important cyclical or noise components are present in historical primary surpluses, it may be necessary to first filter out these components before calculating the size of the fiscal effort (if any) required to restore fiscal solvency.53 II. Incorporating Implicit Liabilities in the Analysis In Colombia, as in other countries, the public sector carries various types of liabili- ties that are not explicitly incorporated into measures of public sector debt. The ana- lytical framework described above can be extended to estimate the permanent pri- mary surplus that is required to maintain a country's fiscal solvency, taking implicit liabilities into account. Letps'ss and ps'NSs be the public sector's primary surplus (rel- ative to GDP) with and without the projected flows associated with implicit liabil- ities, respectively. The value of the stock of implicit liabilities can be calculated as follows Let, d,,, = l+i d, - I p5NSS _ 1 pS5s 53. For a discussion on this subject, see "C6mo Armar el Rompecabezas Fiscal," by Talvi and VWgh, Inter-American Development Bank, 2000. PUBLIC DEBT SUSIAINABILITY AND MANAGEMENT 237 Imposing boundary conditions, it can be shown that: (I+ i IE +I) P' + s) Letting p, be the net present value of the reduction in pension system and other balances, relative to current balances: p4 T =- !rj1J ) +ir (pS'SSpPSI) v-1 Then, d, + PI,( + I + r) P' From this equation, we can derive an expression for the constant primary balance required to keep the debt-to-GDP ratio along a stable path. In our calculations, this expression is also generalized to the case where growth and interest rates are allowed to vary over time. Just as before, these results can be compared with historical pri- mary balances to estimate the magnitude of the fiscal adjustment required to ensure the sustainability of public finances. 238 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Annex III External Liquidity and Fiscal Vulnerability Issues in Colombia-A Contrast with the Argentine Case The recent crisis in Argentina was partly caused by persistent fiscal imbalances, with repeated attempts at fiscal adjustment having been delayed or altogether derailed by domestic politics. The rigid monetary system that Argentina had until recently, led to a substantially overvalued real exchange rate and thus to external trade imbalances which, coupled with a deteriorating external financial environment and persistent strength of the U.S. dollar in international markets, helped precipitate a fiscal financing crisis that ultimately led to the largest sovereign bankruptcy in history. This Annex briefly explores key differences and similarities between Colombia and Argentina in three aspects where vulnerabilities may arise (a) the capacity of each country to generate external liquidity, (b) the capacity of their respective pub- lic sectors to command the necessary resources to service their debt (which includes the resources needed to purchase foreign exchange reserves and service external debt), and (c) what the public sector does with the resources at its disposal and its consequences for growth and sustainability. I. Colombia's External Liquidity Position is More Comfortable than Argentina's A key difference between Colombia and Argentina is the relative importance of the exter- nal trading sector in overall economic activity. In the last five years, total exports and imports of goods and services averaged about 37 percent ofGDP in Colombia, compared with only 29 percent of GDP in Argentina. Besides a relatively small contributon to over- all economic activity, Argentina's external trading sector has also tended to generate greater external imbalances (current account deficits) than has been the case in Colombia. The average curtent account deficit in the last five years in Argenuna amounted to 3 percent of GDP, versus an average deficit of 2.3 percent of GDP in Colombia. Another key cls- tinguishing feature is Argentina's high dependence on external resources to finance eco- nomnc activity, with long-term capital inflows averaging about 8 percent of Argentina's GDP in the last five years; by contrast, Colombia's long-term external financing needs averaged about 2.5 percent of GDP during the same period. Short-term financing is also much more important in Argentina, with a 20 percent average participation of short-term debt in total external debt stocks since 1996, versus 9 percent in Colombia.54 54. Differences in need for external financing between these two countries cannot be explained by the relative scarcity or abundance of domestic savings. Gross national sav- ings averaged 13 7 percent of GDP in Argentina, compared with 14.1 percent of GDP in Colombia, in the last five years. PUBLIC DEBT SUSTAINABILITlY AND MANAGEMENT 239 The picture that emerges is one where Colombia's higher propensity to trade and to stay relatively closer to balance in the external current accounts, is associated with lower long- and short-term external financing needs.55 By contrast, Argentina's smaller trading sector and persistent current account imbalances have been associated with large external financing needs, including a high propensity to maintain comparatively large stocks of short-term debt. These features have a bearing on each country's over- all capacity to accumulate foreign exchange reserves relative to their respective need to acquire external liabilities (including public sector liabilities), with consequences that carry over to various indicators of solvency and the external liquidity of the public sec- tor. A few of these indicators have been selected for discussion below. Figure A.III. 1 compares the evolution of the ratio of foreign exchange reserves to the stock of public external debt in each country. As expected, international reserves cover a higher proportion of public external debt in Colombia than in Argentina. Clearly, there was a marked deterioration of this ratio in Colombia through 1999, when the ratio began to stabilize and improve. By contrast, Argentina's more pre- carious position did not improve over the same period, and instead deteriorated fur- ther in 2001. Other indicators confirm that Colombia's external liquidity conditions have been systematically more favorable than Argentina's. Despite a persistent deterioration Figure AJIII.L. International Reserves and Public External Debt Stocks 75 75 - - - - - --__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ COL Reserves/Debt Ratio 50 - 1996 1997 1998 1999 2000 2001 Source WB Staff calculations 55 We note but do not discuss the potenitial inipact that Colombia's more flexible exchange rate system may have had on the country's relatively sound externial performance 240 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE since 1996, Colombia's exports still cover a higher percentage of public external debt stocks than is the case in Argentina, suggesting that the export sector does indeed play a key role in making available foreign exchange resources to the public sector, possibly reducing (although only indirectly) the need to subsequently acquire exter- nal liabilities (see Figure A.III.2). Colombia's comparatively favorable external liquidity conditions also become apparent by inspection of a few debt service ratios. Colombia's external public debt service to international reserves ratio has remained broadly stable in the last few years; by contrast, Argentina's public debt service to international reserves ratio dete- riorated very rapidly beginning in 1999, reaching nearly 100 percent in 2001 (see Figure A.III.3). The ratio of external public debt service to total exports tells a sim- ilar story, with Colombia maintaining relative stability and Argentina showing a marked deterioration since 1999 (Figure A.III.4). We may conclude that, on balance, Colombia's capacity to generate international liquidity is relatively comfortable compared to Argentina's, including with particular reference to historical external public debt and debt-service levels. However, having a relatively satisfactory level of foreign reserves does not guarantee that a particular sec- tor of the economy (in this case the public sector) necessarily commands the resources to purchase foreign reserves and service its debt, both foreign and domestic. The discussion now turns to the nature of selected aspects of the relationship between the public sector and the rest of the economy, and more specifically to the revenue-generating capacity of the public sector relative to the size of its liabilities. Figutre A.III.2. Total Exports and Public External Debt Stocks 100 - - COL Exports/Debt Ratio 75 > ~~~~ARG Exports/Debt Ratio ' 25 0 _ _ _ - - 0 - II I II 1996 1997 1998 1999 2000 2001 Source WB Staff calculations PUBLIC DEBT SUSTAINABILI ITY AND MANAGEMENT 241 Figure A.III.3. Public External Debt Service and Total International Reserves 100 - ARG Debt Service/Reserves Ratio - ' 75 _, t 75 -- C - -- - - - - - - - - 25_COL Debt Service/Reserves Ratio 2 5 - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 0- 1996 1997 1998 1999 2000 2001 Source WB Staff calculations Figure A.III.4. Public External Debt Service and Total Exports 75 - ARG Debt Service/Exports Ratio 50----------------- ------ U ,_____- 2 5 _ ,_-_ _-__ = _ __-_- COL Debt Service/Exports Ratio 0- II I I II 1996 1997 1998 1999 2000 2001 Source WB Staff calculations 242 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE II. Similar Public Revenue and Expenditure Trends in Colombia and Argentina Despite the differences already discussed in capacity to generate foreign exchange, the public sectors in Colombia and Argentina are displaying similar trends (and imbalances), which result in growing vulnerability and raise sustainabihty issues. For example, in both countries, interest and total debt service payments represent a growing proportion of government expenditures. Moreover, when measured against public sector revenues (for example, tax revenues), interest payments loom as an increasing burden on government Finances in both Colombia and Argentina (see Figures A III.5 and A.111.6). This is important insofar as it indicates that servicing a growing public sector debt also requires increasing the amount of resources available to the public sector on a recurrent basis. It is also worth noting that at these levels of commitment of public revenues to interest payments (25 percent of tax revenues are used to pay interest on existing debts), Argentina had already entered into a serious fiscal crisis. The stock of total public sector debt (on a gross basis) in Argentina and Colombia, relative to the revenue-generating capacity of the public sector, was also at similar levels by the end of 2001, following a sustained deterioration in Colombia during the late 1990s (see Figure A.III.7). Furthermore, total gross public debt in both countries is approaching similar levels relative to the size of their economies, with a rela- tively faster rate of increase in Colombia than in Argentina in the last few years (see Figure A.II1.8). In both countries, the underlying cause of growing total debt stocks has been a large and persistent fiscal deficit, fueled primarily by increasing public current expenditures. Colombia's fiscal deficit averaged 4 percent of GDP during 1997-200 1, while Argentina's averaged 3 percent of GDP over the same period (see Figure A.III.9). Current public expenditures in Colombia rose from under 22 percent in 1997 to about 24 percent of GDP in 2001, and in Argentina from over 21 percent in 1997 to about 24 percent of GDP in 2001. Current public expenditures in both countries increased as a result of higher debt service costs and an expansion in other types of current expenditures (for example, salaries and transfers). Efforts by both countries to restore fiscal balance by increasing public revenues have been moderately successful (see Figure A.III.10), but such efforts have thus far been insufficient to match the pace of growth in public expenditures. Further- more, in the case of Colombia, total public sector revenues at about 29 percent of GDP are already higher than in Argentina (and higher than in many other coun- tries in the region), suggesting that there is already a limited upside to the fiscal effort that can be carried out on the revenue side in Colombia. More specifically, both in Argentina and Colombia total tax revenues already amount to about 19 percent of GDP. This leads us to a discussion of the quality of fiscal adjustment and its implications for growth and fiscal sustainability. PUBLIC DEB I SUSTAINABILI lY AND MANAGEMENT 243 Figure A.III.5. Public Interest Payments and Tax Revenues 30 COL Total Iiterest Payments/Tax RevenueS 20 - - - ---------- - 15 ]-- ----- …- ' - - ARG Total Interest Payments/Tax Revenuies 10 5 -- __ _ __ ___ -_ -- - - - - - - - - -_ __ _ _ -_ _ _ _--- __ - _ - 0 -II I I I 1996 1997 1998 1999 2000 2001 Source WB Staff calculations Figure A.III.6. Public Interest Payments and Total Revenues 20 - ARG Total Interest Payments/Total RevenIes ' 5 - I I I … I 1 996 1 997 1 998 1 999 2000 200 1 Sourte WVB Staff calculations 244 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE Figure A.III.7. Public Debt Stocks and Total Revenues 100 - COL Public Revenues/Total Public Debt 75 - _ _ _ _ __ ru ARC Public Revenues/Total PublcDebtD 5 0 _---- - __- _-- _-- _--- _--- _ _ _- _- __ _ _ _ 25- I I I I 1996 1997 1998 1999 2000 2001 Source WB Staff calculations Figure A.III.8. Gross Public Debt and GDP 16 - _ __ __ _______ = a 14 - - 97 pw \t5 -' 12 - - -- - - c3 4 30 1 5 6 7 8 9 10 u 6 2001- ---- -Benchmark 1 2 3 4 5 6 7 8 9 10 Source WVB Staff c-alculations PUBLIC DEBT SUS1AINABILITY AND MANAGEMEN T 245 Figure A.III.9. Consolidated Public Sector Balances 0- 0 0= -3 - -C COL Overall Balance -5 - - - . . - ARG Overall Balance -6 ~ 1996 1997 1998 1999 2000 2001 Source: WB Staff calcuilationis. Figure A.III.10. Total Public Sector Revenues 35 D COL Total Revenues M ARG Total Revenues 30 -- ------- 2 25- - 20 - 10 - 0 1996 1997 1998 1999 2000 2001 Source: WB Staff calculations. 246 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE III. Fiscal Adjustment in a Dynamic Context Attempts at further increasing public revenues by raising taxes may turn out to be counterproductive in the context of weakening private investment and high domes- tic uncertainty. Private investment has fallen steadily in Colombia, basically halving from levels of 14 to 15 percent of GDP in earlier years, to about 8 percent in 2000. Colombian private investment performance is currently weaker than it has been in Argentina, where private investment has also fallen from earlier levels of about 18 percent to 14 percent of GDP more recently. Since in Colombia public investment has also been falling as a proportion of GDP (and of public sector expenditures),56 and given that the public sector displays a high propensity to use revenues to finance current expenditures, it may be argued that in the absence of comprehensive public expenditure reform, further increasing the resources available to the public sector could further weaken overall investment performance in Colombia. In this connection it is worth noting that weakening investment performance also feeds back into an analysis of sustainability and vulnerability of public finances. A lower propensity to invest economywide (which is associated with higher public sector participation in the economy)57 implies slower GDP growth in the long run, thus increasing fiscal risks by reducing the prospective tax base. Moreover, slower GDP growth would also put upward pressure on real interest rates, as public debt stocks would become more important relative to the size of the economy and thus of domestic savings in the long term (compared with a higher GDP growth sce- nario). Both slower long-term GDP growth and higher real interest rates have a neg- ative impact on the sustainability of fiscal accounts. This suggests that the degree of success of measures to improve fiscal sustainability in Colombia (for example, by raising the public sector primary balance on a permanent basis) is not invariant to the quality of public expenditure, the manner in which public revenues are adjusted, and overall public sector participation in the economy. 56. Public investment in Argentina has had a limited downside, since it amounts to only 1 0 to 1 5 percent of GDP. However, it is worth noting that given the small investment com- ponent of public expenditures in Argentina, basically all recent increases in public rev- enues have been used to finance current expenditures. 57. Given the public sector's high propensity to consume the resources at its disposal, higher public sector participation in the economy may also be associated with a fall in national savings, which in turns reduces the availability of internal resources and increases depend- ence on foreign capital to finance growth PUBLIC DEBT SUSTAINABILITY AND MANAGEMENT 247 Annex IV Debt Management Strategy and Portfolio Benchmarking I. Background Debt benchmarking, which attempts to identify the optimal debt portfolio structure and funding strategies, is the analytical foundation for sovereign debt management. The basis for debt benchmarking in Colombia today is the cost-risk tradeoff analy- sis known as Value-at-Risk (VaR) analysis Colombia is one of only a few develop- ing countries where VaR modeling is being integrated into day-to-day debt man- agement decisions This Annex assesses the current benchmarking methods used by the Dzreccid%n General de Cr&dlto PNblco (DGCP), and related debt management issues. It also provides recommendations for further analytical improvements so that the consistency of debt benchmarking is improved and further incorporates debt sustainability analysis. II. An Overview of Current Methodology Benchmarking in Colombia consists of cost-risk tradeoff analysis across different debt structures and funding strategies, each of which has an implied VaR. The opti- mal debt structure and funding strategy is chosen to minimize the VaR. Specifically, VaR is calculated as the difference, discounted by an appropriate discount yield curve, between debt service costs in a risky scenario and under an expected scenario. Under the expected scenario, the variables that determine debt service flows exhibit the behavior "normally" expected by DGCP. The risky sce- nario can be projected either as a specific scenario thought to be risky, or as a sce- nario following a certain stochastic process. The discount curve is used to calculate the present value of the cash flows in both the expected and risk scenarios. The VaR associated with one underlying variable is the difference between the present value of debt service under the expected value of the underlying variable, and the "risk scenario" of the variable.58 For example, to calculate the VaR due to interest rate movements, it is assumed that the interest rate follows a geometric Brownian motion process given by dS = 58 The analysis can be generalized to the case of several underlying variables For example, the VaR of a portfolio of 2 underlying variables is equal to .JV2,1 + ,2a2 PM 2vW2CF102 22 22 2 248 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE pdz + o-dz, where dS is the change in the interest rate variable, z is a random draw- ing from a standardized normal distribution, dt is the (infinitesimal) change in time, m is the average change of S, and s is the standard deviation of S. Therefore, the value of the interest rate variable in the "risk scenario" under 95 percent confidence level is given by: S(t) = Soe( 2 (1) - - Yci+1 645yFT = Soe'1e 2 (2) --2 1 645CrFi = E[S,]e 2 (3) It follows that the VaR at time t due to the movement of interest rates is calcu- lated as the discounted value of the debt service under risky scenario S(t), minus the debt service under the expected scenario E(St). When implemented, S(t) is calcu- lated using the equation above. E(St) is chosen as what the authorities believe is the most likely market rate at time t.59 III. Assessment A few issues arise regarding the way this methodology is applied in Colombia, mainly relating to the definition and calculation of the expected scenario. First, the definition of debt risk is not coordinated between DGCP and the unit in charge of macroeconomic programming. The DGCP relies on an internal risk benchmark, which is calculated as the debt service under an expected market scenario. The risk is then defined as the deviation of the debt services under alternative scenarios from the benchmark debt service. Subsequently, funding strategies are chosen so that risks can be minimized. By contrast, the macroeconomic programming unit views risks in terms of fiscal sustainability. As a result, a funding strategy that the DGCP deems not risky, can in fact be quite risky from a macroeconomic programming point of view, and vice-versa. Second, there are problems in the calculation of the risky sce- nario. Equation (3) is used by the DGCP, with E(St) taken as exogenous (based on the DGCP's views on the market). There is some inconsistency in such a calcula- tion: if geometric Brownian motion for the variable is assumed, equation (1) should be used directly to calculate S(t). Third, analysis needs to be based on the same assumptions used by both the DGCP and macroeconomic programming. Currently each unit has its own set of assumptions. 59 When a forward market exists, the forward rate is used as the approximation for E(St). PUBLIC DFBI SUSTAINABII.LIT AND MANAGEMENT 249 IV. Recommendations It is recommended that a unified view on debt risks be developed by the DGCP and the macroeconomic programming ullit. Ultimately both units should be concerned with budget risk, that is, the magnitude of the risk that debt service can brng to the budget. The macroeconomic programming unit can conduct ani analysis of the government's balance sheet, in particular, of revenue and expenditure patterns. Based on such analysis, a maximum "budget-at-risk" (max-BaR) can be derived, which can be defined as the maximum tolerable risk (the amount of debt service) that the budget can sustain. The DGCP can then use this max-BaR as a risk benchmark-the DGCP should choose only funding strategies anid portfolios that imply smaller BaRs than the max-BaR. A three-step approach is recommended for conducting a unified benchmarking analysis which combines sustainability and vulnerability analysis STEP 1: Derivation of the Efficient Portfolio Set Given the government's primary budget forecasts and, hence, borrowing require- ments for the next few years (for instance, 5, 10, or 15 years), sequential steps are fol- lowed to identify an optimal benclhmark portfolio. At this stage, the borrowing (issu- ing) program is assumed to be time-invariant over the next few years, and the efficient portfolio set five years from now is derived over various combinations of borrowing programs. To guarantee the feasibility and optimality of the benchmark portfolio, the current sovereign debt structure should be taken as an initial condition, and official primary budget deficit forecasts should be used as inputs to compute the yearly bor- rowing requirements. The efficient portfolio set can then be obtained by conducting simulations over the future evolution of interest rate term structures. To simulate future term structures, either a historical method (bootstrapping) or simulation (Monte Carlo) method calibrated by actual data, could be used. STEP 2: Derivation of Feasible Portfolio Set from Exogenous Medium- Term Targets In this step, medium-term target duration, currency composition, max-BaR, and the shape of debt profile for both external and domestic debts, are exogenously estab- lished by conducting macro, finanicial, and budget analyses. Ultimately, the Ministry of Finance should establish medium-term specific targets that will be supplied as inputs to identify an optimal benchmark portfolio, from the efficient portfolio set derived in Step 1. Given the medium-term targets, a feasible efficient portfolio set will be identified by the maximum tolerable 95 percent BaR. STEP 3: Identification of Optimal Benchmark Portfolio within Feasible Set The third step is to identify an optimal benchmark portfolio from the feasible set. Given the feasible efficient portfolio set, a portfolio which minimizes the gap with 250 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE the medium-term targets set in Step 2 above will be chosen as the benchmark portfolio. There is also a need to separate portfolio benchmarking and issuing program (or funding strategies) benchmarking. As has been mentioned, the aim of debt portfolio benchmarking is to identify the optimal debt portfolio with a structure that provides an optimal cost-risk tradeoff, consistent with a country's economic goals. The risks here should include refinancing risk, budget risk, and other types of risk, in addi- tion to the typically discussed interest rate and exchange rate risks. Issuing program benchmarking aims at identifying the optimal path to reach the portfolio bench- mark, taking as the starting point the current debt portfolio structure. Currently, these two benchmarklng exercises are not properly differentiated. In standard portfolio benchmarking, each possible debt portfolio (for example, defined by a maturity structure) is examined for its cost-risk behavior under differ- ent risk scenarios over many years The structure of each debt portfolio is static over time The issue here is to find a static debt portfolio that performs best in the expected economic environment. The objective of issuing program (funding) benchmarking is to identify an optimal path of issuing programs to reach (from the current debt portfolio) the portfolio benchmark. It deals with practical issues such as how quickly one can reach the portfolio benchmark and what is the depth of debt markets and its implication for debt issuing. Each path of issuing programs is exam- ined for its implied costs and risks. Essentially, what the DGCP currently does is to try to accomplish portfolio benchmarking and issue program benchmarking simultaneously. Following the DGCP approach, the structure of a debt portfolio is defined by the issuing pro- grams. A path of issuing programs over several years determines a debt portfolio whose structure will change over time The cost-risk tradeoffof such debt portfolios will be analyzed under different scenarios and the one with the best cost-risk com- bination is then chosen as the benchmark. But the chosen portfolio is neither a benchmark portfolio nor a benchmark-issuing program in the conventional sense. Instead, the DGCP should carry out two rounds of cost-risk tradeoff analysis, one for portfolio benchmarking and the other for funding benchmarking. The objectives of these two types of analysis are different and should not continue to be confused. PUBLIC DEBT SUSTAINABILITY AND MANAGEMENT 251 Annex V Development of Debt Markets In the last decade Colombia has undergone a series of important institutional changes in domestic debt markets. These included the privatization and interna- tionalization of banking, the creation of a public debt market, and the reform of Social Security that led to the creation of private pension funds. In this context it is possible to point to various major issues that have characterized the development of debt markets in Colombia: (a) the rapid development of the public debt market; (b) the relatively unsuccessful development of private financing through capital markets both in stock markets and over-the-counter; (c) the instability in the recent past of the financial sector; and (d) the impact of fiscal and monetary policy, and of insti- tutional reform. Regarding the public debt market, profound and systematic changes occurred throughout the 1990s. In 1993, the government created Treasury bonds (TES), con- ceived as a tool for internal debt financing and as an instrument for the development of capital markets In 1995, an auction system for TES was launched. In 1997, with the introduction of the "market makers" program, a series of important lstitutilonal innovations started in debt markets, including the designation on merit of a group of financial institutions to provide primary and secondary liquidity to TES issuance,60 the standardization of the TES,61 increased transparency in the auction process through a pre-announcement of financing needs, set rules for auction pricing, and the introduction of electronic transactions at the Central Bank and the stock markets. These reforms, combined with the government's growing financing needs, con- tributed to an increase in the outstanding stock of public debt in primary and sec- ondary markets. From 1997 to 2000, the volume of Treasury bonds traded over-the- 60 Governments often designate as primary dealers a number of intermediaries with an obli- gation to assist in the development of markets, often acting as market makers. Having a "market makers" program facilitates the transition away from captive sources of financ- ing to a market-based system. Eligibility criteria for being a primary dealer usually include financial capacity, management and technical skills, market presence, and will- ingness to cooperate with the authorities (for example, providilg information) 61 A variety of debt instruments may allow debt managers to satisfy che needs of many investors However, the availability of too many products can also produce market frag- mentation, that is, for a given potential level of market activiry, active trading of any one product may not be viable International experience has shown that the benefits of prod- uct standardization far outweigh any benefits from diversification In Brazil, more than 250 different types of federal securities are reported to exist, but only about 30 are deemed to be "relatively active" and 10 are regarded as "liquid " In Korea, only six types of domes- tic government securities are known to be issued, but until recently Treasury bonds issued on different dates were treated as separate Issues, thus causing market fragmentation (Developing Government Bond Markets A Handbook, 2001, World Bank and IMF) 252 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE couLiter increased from COP19,400 billion (or 16 percent of the 1997 GDP) to COP87,800 billion (or 51 percent of the 2000 GDP). During this period the vol- ume of daily transactions multiplied fourfold. Despite these significant changes, the major holder of domestic central government debt remains the public sector itself, with about 50 percent of the outstanding stock in the hands of public sector enti- ties (mainly the Social Security pension fund) (see Figure A.V.1). Successful devel- opment of private debt markets requires a much more diversified investor base (see Box A.V. 1 for a description of the Spanish experience promoting investment funds). The limited size and liquidity and the high risk of Colombian debt markets con- tinue to impose major constraints for efficient debt financing of the public sector. Colombia lags behind countries with similar and sometimes lower levels of eco- nomic development. In addition, for similar degrees of risk as measured by price- standard deviations, Colombia offers relatively higher returns than Brazil, Chile, and Mexico, for example. From 1998 to 2000 Colombia faced serious shocks to its financial system. As aggregate consumption contracted, financial institutions that focused on the con- sumption credit market suffered. The housing mortgage market also has been adversely affected by the deteriorating economic environment. In this environment, Figure A.V.1 Holders of Central Government Domestic Debt* Trusts 122% Private Pension Commercial Banks 9% Funds 7% S 4 ~~~~~~Deceval 5% Banrep 6% _ Non-Finiancial Enterprises 9%/ Insurance Companies 1% Public Sector 50% Other Financial Corporations 1%_ *TES Bonds as of 2001. Source: MoF, DGCP. PUBLIC DEBI SUSTAINABILITY AND MANAGEMENT 253 Box A.V. 1. The Spanish Experience with Specialized Investment Funds Spain was very successful in promoting specialized government bond funds as close substitutes for bank deposits. In the early stages of development, the Span- ish government nurtured constructive relationships with banks and persuaded them that income lost from lending spreads on deposits would be more than offset by gains in the form of fee income earned through managing investment funds. A Joint public-private publicity campaign was launched to elicit interest in new investment vehicles. Favorable tax treatment in contrast to deposit accounts, also made these investment funds attractive to retail investors. In addition, under a special agreement with the Spanish Treasury, privately man- aged, specialized government bond funds (called FondTesoros) were introduced in 1990. These specialized bond funds were important in promoting liquidity and depth in secondary government bond markets. the Constant Purchasing Power Unit (UPAC) system, a tool designed to guarantee the value of savings channeled to the housing sector, came under pressure through a combinationi of rising interest rates, a widening maturity gap between savings and loans, and the collapse of real estate prices. 8 The Financial Sector This Chapter was written by John Pollner. I. Background The Colombian financial crisis began in 1998, although financial system weakness was already evident starting in 1996, after the credit boom during the first half of the 90s. Credit performance of all borrowers was severely affected, and falling real estate values, in particularly squeezed mortgage debtors. Local and regional govern- ments also contributed to debt default and high indebtedness, further weakening bank portfolios. The credit boom came to an end in early 1996 as a result of a slow- ing down of gross domestic product (GDP) growth, reflecting domestic political uncertainty, weakening international confidence, and a tightening of credit policy in an effort to dampen domestic demand growth. Growing market concerns about the sustainability of the large fiscal and external deficits resulted in the emergence of sig- nificant exchange rate pressures. These accelerated in 1998 in response to the tur- moil in the international capital markets. As in other Latin American countries, the increase in country risk and the deteriorating international environment led to a substantial increase in the yields of Colombian international bond issues followed by the loss of Colombia's investmenit grade in 1999. In turn, the efforts by authori- ties to stabilize the exchange rate led to a sharp increase in real interest rates, and an abrupt contraction of real lending flows. At the same time, real GDP growth decel- erated abruptly in 1998, and turned negative during 1999 Debt restructuring programs for the private corporate sector and sub-national governments, therefore, were crucial as part of the government's economic emer- gency program adopted in 1998 which included, among other things, funding the deposit insurance fuid (Fondo de Garantias de Instituciones Ftnancteras, FOGAFIN) in order to compensate depositors of failed banks and finance workout schemes for intervened banks and financial institutions The fiscal cost was in part financed by 255 256 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE a tax applied to all financial system transactions. FOGAFIN also established a cap- italization credit line for private financial institutions, provided that such institu- tions took corrective measures to improve their financial health. To avoid fiscal losses, the credit lines were lent to bank shareholders who purchased FOGAFIN bonds and added these to bank assets so as to increase net capital. Since the bonds were low risk-weighted assets, this also improved balance sheet solvency though without a positive cash flow effect. Prior to these adjustments, loan loss provision- ing requirements were increased to first determine the actual capital gap. Foreign exchange exposure was not a major issue in Colombia, since banks were well hedged in terms of matching the currency of their deposits and assets. However, the dollar loan portfolio generated credit risks in borrowers that had only domestic currency incomes, a factor that affected repayment capacity and loan performance fol- lowing the rapid depreciation of the Colombian peso. The situation of public and intervened financial institutions was by far the worst of the system, most of them being deeply decapitalized or outright insolvent, with much higher loan delinquency ratios and lower liquidity levels than private banks. The rate of deterioration of the portfo- lios of public banks had been much higher than that of private institutions and, unlike them, continued during 1999. The cooperatives and the savings and loan institutions (CAVs) also experienced substantially more severe difficulties than private banks due to the acute deterioration of consumer loans and mortgages, and a Constitutional Court ruling which required them to compensate borrowers for past years of payments due to a change in an interest rate indexing procedure which was ruled invalid. As lending to the private sector halted during the crisis, banks and financial insti- tutions purchased more government securities to fill their balance sheets. Govern- ment securities on bank balance sheets increased its share from about 10 percent of assets before the crisis to almost 30 percent currently, a phenomenon which was also consistent with the increase of government indebtedness to meet expenses, given the economic slowdown and decline in tax revenues. Bank earnings were negative dur- ing 1998-2000, only to be reversed with minimal positive net income in 2001 (see Table 1). Banking solvency ratios showed the greatest decline during 1999 at 9 per- cent year end, and rose thereafter, reaching 14 percent in 2001. The financial crisis was also manifested in the level of nonperforming loans in the banking system, which rose from approximately 7 percent in early 1998 to a peak of 20 percent by end 1999. With the tightening of prudential regulations, loan loss provisions began increasing, from 28 percent in mid-1998, reaching 51 percent of the total non performing portfolio in mid-2001. Table 1. Financial System Return on Assets (percent) 1997 1998 1999 2000 2001 (est.) 1 2 -2.5 -3.5 -2 3 +0.9 Source Superintendencia Bancaria THE FINANCIAL SECTOR 257 II. The Financial Crisis Management Strategy Starting in November 1998, with the issuance of a financial emergency decree that provided an initial policy framework, the Colombian authorities took a number of important steps to address financial sector distress. The decree contained measures to alleviate pressures on mortgage debtors, increase protection of depositors in finan- cial cooperatives, clarify the extent of the future safety net in that sector, and assist- through FOGAFIN credit lines-in the recapitalization of viable financial interme- diaries and the liquidation of bank assets. The restructuring of the savings and loan banks (Corporaciones de Ahorro y Vivienda, CAVs) was a crucial element of the program because these institutions were the most insolvent with over 20 percent of its loan portfolio non performilng due to borrower defaults caused by interest rate risks and falling real estate values used as collateral To deal with the CAV problem, the government issued a number of measures, including inducing mergers of CAVs with financial groups based on capital shortfalls, enforcing capital adequacy, and provisioning rules while providing capitalization assistance contingent on owners contributing fresh capital, developing an interest rate hedging instrument, and preparing the ground for the establishment of a new mortgage securitization agency. Additionally, the government took proactive steps using an innovative combina- tion of policy measures to quickly shield the financial system from a potential col- lapse, which in the end was successfully averted. These measures included: * INTERVENTION AND CLOSURE OF FINANCIAL INSTITUTIONS. The Banking Superintendency intervened 10 institutions that were to be liquidated, and FOGAFIN took over three to be restructured and resold to the private sector in the immediate future. * A DEBTOR RELIEF PROGRAM The government launched a debtor relief pro- gram that included (a) refinancing at preferential interest rates to homeown- ers who maintained a good payment record but whose standing balance on indexed housing loans went up more than 20 percent in nominal terms, (b) refinancing for housing borrowers who had arrears for three months or less on their mortgage obligationi, (c) mortgage payment relief for low-income home- owners that lost their jobs. * A RECAPIrALIZA`I1ON PLAN. The government announced a recapitalization plan under which the government extended credit lines through the deposit insurance agency, FOGAFIN, to enable financial institutions to write off impaired assets. Under one of these programs, credit lines were lent to the shareholders, who purchased bonds issued by FOGAFIN and capitalized their banks with them. This consequently did not Imply cash disbursements for the government. The cash flows from FOGAFIN to the beneficiary institutions and the shareholders, thus reflected only the differential between the interest rate paid on the loan provided by FOGAFIN and the bond that was issued. 258 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Another program started in 2001 for mortgage banks, also provided for credit lines directly to the banks as well as to shareholders. * THE REFORMED FiNANCIAL SECTOR LAW A new law was passed by Congress in mid-1999, the main provisions of which were: * An increase in the minimum capital of financial institutions and raising the risk weighted capital ratio to 9 percent * The strengthening of procedures for the timely detection of troubled insti- tutions and the introduction of "prompt corrective action" triggers by the Superintendency * The introduction of time limits for different procedures undertaken by the Superintendency, such as the decision on whether to liquidate or close an institution, intervention without liquidation, and liquidation proper * The tightening of clauses under which a financial institution may be intervened * New powers to FOGAFIN to expedite mergers and acquisitions, using other banks as trustees to manage credit portfolios, transferring assets and liabilities of the troubled bank to another (using the initial carving out of good and bad banks' assets and liabilities), making loans to financial insti- tutions and purchasing assets to facilitate these transactions, and promot- ing securitization of portfolios to expedite asset transfers. A modification to the banking law, in process of being approved, would also provide pow- ers to the Superintendency of Banks to authorize asset and liability trans- fers as part of financial institution improvement/recovery plans. * LENDER OF LAST RESORT FACILITIES. The Central Bank increased the amount of repo transactions and eased access to the rediscount facility in order to pro- vide liquid resources to those financial institutions that suffered withdrawals. III. Restructuring and Recapitalization of the State-Owned and Private Banks The government's strategy with respect to the state-owned banks was meant to iso- late them from the rest of the financial system in order to restructure or sell them under controlled conditions while avoiding an additional factor of potential conta- gion to the already weak private banks. Public banks were formally "officialized," meaning that the management oversight and ownership stake were taken over directly by FOGAFIN with a view to clean up the balance sheets, recapitalize the banks, and sell them or liquidate them. The Caja Agraria was closed and a new Banco Agrarzo created out of the Colvalores Leasing Company which was part of the Banco de Desarrollo Empresarial; following a reform in the statutory mandate of the institution. Although the assumption that the remaining banks could be sold in one or two years for a multiple of what they could be sold at present was probably real- istic if economic and market conditions were favorable, the potential further deteri- THE FINANCIAL SECTOR 259 oration of the loan portfolio under less favorable conditions and the operating costs of keeping deeply troubled banks open, created some risks. Moreover, for banks that had been experiencing severe management difficulties and weak control systems (which was often the case for public banks), it could be particularly difficult to redress ingrained habits and instill a new management culture under such uncertain conditions and governance. The fact that the rate of deterioration of the portfolios of public banks had been much higher than that of private banks, suggests that their management and debt repayment cultures in public bank clients adversely affected their capacity to man- age loan portfolios effectively under difficult market conditions. The postponement of a prompt ex ante regularization of their condition resulted in sizable further loan deterioration and large operating losses derived from oversized staffing and infra- structure, low efficiency, and poor loan management. This created difficulties to pri- vatize the banks promptly and increased pressures to keep them open. Under subse- quent FOGAFIN oversight and intensive supervision by the Superintendency of Banks, these institutions recovered substantially after downsizing, and generated positive returns by 2001. Nevertheless, a review of the experience might consider whether it was optimal to keep the banks open as institutions or whether it might have been preferable to have considered at the outset, the sale of their potentially good assets and deposits to sound financial institutions (and if needed, with a gov- ernment risk-sharing agreement with the purchasing institution), while transferring all unviable assets to a liquidation company and recognizing sooner the eventual fis- cal liability. a) FOGAFINs Crisis Response Instruments FOGAFIN's operations in coordination with the Superintendency of Banks man- aged to isolate the weakest banks from the rest of the financial system, and thus pre- vented contagion. Those banks which were intervened underwent heavily scruti- nized improvement programs with FOGAFIN's capitalization support, and in about half the cases for the state-owned banks, the strategy was to continue their opera- tions and eventually sell them off. FOGAFIN's financial instruments included: (a) the purchase by FOGAFIN of stock issued by an institution that was unable to recapitalize, (b) long-term repo operations for the purchase and repurchase of credit portfolios, (c) credit lines funded by bonds to financial entities under rehabilitation programs or to facilitate mergers or acquisitions, and (d) a capital guarantee which was treated as capital in the recipient entity (which reflected a contingent liability for FOGAFIN whereby it had an obligation to inject true capital if the bank became illiquid) The resolution process, particularly for the state-owned banks, has been some- what arduous: more than three years after Intervention, the final resolution status of many of these banks is not clear, and as the time draws out, the more costly option of outright dissolution looms larger. While the government has successfully been 260 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE able to gradually reduce and sell off the balance sheets and physical infrastructure of thcse banks, its investment in this process may nccd to continue for some time to achieve its end target of privatization or extinction. However, the "officialization" of these institutions also implicitly means that FOGAFIN stands behind all their lia- bilities despite the requirement for rehabilitation and eventual sale or disposal of assets. This role might therefore reduce some of the incentives for a prompt resolu- tion of these banks, particularly if remaining management and staff feel that the State will fund them until a suitable buyer of remaining performing assets is identi- fied. As shown further below, the liability base of such banks could have significant fiscal impact if the sale or liquidation of their assets does not materialize. In its support of private banks, fewer risks remain. However, though the expe- rience has been generally favorable, if additional banks were not to repay the capi- talization credit lines to FOGAFIN, then FOGAFIN would take on the share own- ership which was meant to be financed by the credit line. However, FOGAFIN would need to continue paying the interest on the bonds used to fund the credit line. This would hypothetically result in a negative cash flow for FOGAFIN until the entity was sold or liquidated. Fortunately to date, only one bank has not repaid promptly. Table 2 shows the support of FOGAFIN's capitalization program to restore tech- nical solvency of state-owned banks following the intervention and officialization of these banks At end 2000, after the ongoing disposition of assets and deposits of the inter- vened state-owned banks, Col$13,197 billion in assets and Col$8,165 billion in deposits remained on the balance sheets, amounting to 8 percent of GDP and 5 per- cent of GDP, respectively. While FOGAFIN still intends to dispose of these assets through market mechanisms and absorption by the private sector, it remains unclear as to whether, in the end, a large portion of depositors' compensation may need to be assumed by the State as a fiscal outlay or additional financial support to FOGAFIN. Table 2. Capitalization Support Provided to State-Owned Banks, 1998-2001 Bank Amount (Col$ billions) BCHi 1,546 Banestado 1,240 Bancaf62 1,137 Granahorrar 332 IFI 400 Banco Agrario 150 FES 45 Total 4,850 1 Intcludes CISA BCH 2. Includes CISA Bancaf6 Source FOGAFIN THE FINANCIAL SECTOR 261 Payments for deposit insurance of all intervened private financial institutions since the start of the crisis have amounted to Col$73.46 billion, equivalent to under 0.5 percent of the government's revenue base. The assets and liabilities of the private intervened institurions since 1998, before any liquidations or deposit payouts, amounted to Col$1,106 billion and Col$1,520 billion, respectively. The liabilities (including all deposits) of such institutions represented I percent of the average GDP over that period. Table 3 shows the funding provided under the FOGAFIN support program (that is, capitalization credits provided to private banks) for the private banks that entered into the capitalization program To date, government funding for FOGAFIN has provided it financial sustain- ability, although maintenance of this sustainability depends in large part on the per- formance of the credits it has extended under the capitalization and other support programs Nevertheless, FOGAFIN is partially hedged against private banks, given that it has a large stock of bond liabilities used to fund its credit lines, without any transfer of cash (see Table 4) b) The Development of Regulatory and Supervisory Strategies Swift action by the Superintendency of Banks was taken and will need to be con- tinued to enforce proper loan classification and provisioning requirements, includ- ing reducing the overreliance of banks on collateral property as a substitute for cash provisions, particularly given depressed real estate values. When capital deficiencies are detected, financial institutions should be given short deadlines to comply, if needed, with FOGAFIN support. The joint monitoring and supervision of bank Improvement Plans by FOGAFIN and the Banking Superintendency has been effective to prevent unforeseen deterioration of such institutions. Those that do not comply should be immediately closed or intervened to ensure that only banks with Table 3. Private Bank Capitalization (Credit Line) Program, 1998-2001 Bank Amount (Col$ billions) Colpatria 212 Superior 94 Cr6dito 60 Interbanco 43 Um6n Colombiano 28 Cofinorte 20 Coltefinanciera 28 Multifiianiciera 3 Credinver 2 Confinanciera 3 Megabanco 286 Total 779 Source FOGAFIN 262 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Table 4. FOGAFIN Balance Sheet as at end 2000 and end 2001 (Col$ billions) Col$ bilions Col$ bilions Assets 2000 2001 Liabilities and Equity 2000 2001 Banking support Bonds issued 5,414 5,781 operations 1,941 2,207 Investments and repos 844 1,580 CAF credit line 218 202 Government payables Other liabilities 232 212 and capitalizations 4,308 4,047 Granahorrar portfolio 413 374 Total liabilities 5,864 6,195 Other 90 22 Net equity 1,732 2,035 Total assets 7,596 8,230 Total liabilities & equity 7,596 8,230 Source FOGAFIN sufficient capital and capacity to extend credit populate the financial system. Capi- talizatson support should be limited to situations meant to avoid systemic problems and to institutions with sufficiently good prospects for solvency. The Superintendency has upgraded its comprehensive risk management system (based on CAMEL indicators'), but also added modules for supervising financial groups (consolidated supervision) to determine consolidated solvency positions, and evaluating market risks. The government recently approved a decree that will weight specific market risks (interest rate, foreign exchange, maturity) requiring an alloca- tion of risk-weighted capital. This will likely reduce the current risk-weighted capi- tal ratsos of banks by a couple of percentage points, although as a whole it will likely not push any additional bank into insolvency. The Superintendency is also success- fully implementing a prompt corrective actions regime with associated matrixes of sanctions to be applied against banks exceeding prudential limits. Management per- formance risk is also being carefully evaluated with a view to setting weights for management-related factors in assessing the overall stability of banksng institutions. Additional reforms being undertaken include directives for banks to develop their internal value-at-risk models in line with the upcoming reforms in regulation and supervision as proposed by Basle Colombia also has some unique characteristics due to the impact of the drug trade and guerilla violence that, while not having directly affected the recent finan- cial instability, bear some mention. These factors impact the financial sector via the laundering of illicit funds, which are not only generated from illegal trade but also can be destabilizing to financial institutions depending on what actions (regulatory, enforcement, criminal) are taken by the authorities. The government has, however, taken commendable diagnostic measures to begin tracking and encumbering such illicit monetary assets and developing and approving an anti-money-laundering legal framework. Through the screening of money transfers, deposits, and withdrawals 1. Capital/Assets/Managemenr/Earnings/Liquidity, an international prudential and regula- tory diagnostic tool for assessing financial soundness of banks. THE FINANCIAL SECIOR 263 that are deemed out of proportion to corresponding business enterprises, the track- ing of illicit flows has been more systematically developed. This has also been sup- ported through the use of technology to map, both geographically and organiza- tionally, the links among various business concerns and their potential correlation with subversive activity or drug trafficking. Because such measures require the coop- eration of all financial institutions, some of which may be under threat by the very clients who generate laundered funds, the road ahead to perfecting such monitoring tools will be challenging. However, with the cross-referencing of various financial system sources and industry links within the country, such initiatives will be able to continue progressing in the identification of laundered funds within the banking system IV. Fiscal and Economic Costs in Relation to Financial Sector Restructuring To implement the financial sector emergency measures and ensure that FOGAFIN had sufficient resources to implement them via prompt payment to depositors, financial engineering schemes, credit support, and capitalization measures, the gov- ernment allocated Col$5 trillion in net fiscal funding, representing 4 percent of GDP and 25 percent of the central government revenue base. The government also allocated Col$600 million directly to FOGAFIN to build up its reserves 2 To finance additional costs of the overall financial institutions emergency support pro- gram, a financial transactions tax of 0.2 percent (applied to all transactiol flows ini- tiated by banks and financial institutions) was implemented, which raised Col$1 3 trillion. Since 1998, the total cash resources directly sourced or administered by FOGAFIN, which have been provided to support financial institutions, has amounted to Col$1.9 trillion, equivalent to 8 percent of the 2000 government rev- enue base and 1. I percent of GDP. Total support including noncash resources (that is, FOGAFIN securities) has reached Col$8.7 trillion, of which support to state- owned banks represents 73 percent of that amount (see Table 5). After considerinig all government cash and debt outlays net of (i) financial tax proceeds (paid by the private sector), (ii) recoveries from asset liquidations, (iii) repayments of government loans, and (iv) funds used to capitalize FOGAFIN (net of losses), the total estimated fiscal costs for financial system support represented almost 4 percent of GDP. Since the crisis began in 1998 a number of financial institutions were liquidated or began the process of liquidation after other alternatives were considered unfeasi- ble The cumulative asset value of those liquidated institutions was approximately Col$560 billion, or 0 6 percent of average GDP over the period. 2 By 1998 FOGAGIN had used up some Col$650 million in own funds and incurred an additional debt of Col$500 million, in orher words, it had reached a capital deficit situation 264 COLOMBiA THE ECONOMIC FOUNDATION OF PEACE Table 5. Total FOGAFIN Support to Financial Entities 1998-2002 (Col$ millions) Resources Cash Bonds Total Prvate Sector Dep Insurance Funds 283,599 1,575,310 1,858,909 Econ Emergency Funds 184,865 320,601 505,465 Public Sector State Bank Fund 0 4,784,012 4,784,012 Dep. Insurance Funds 642,322 0 642,322 Econ Emergency Funds 779,998 133,027 913,025 Total 1,890,784 6,812,950 8,703,733 Source FOGAFIN Although some of the above contingent fiscal rssks were inevitable given the urgency of the measures taken to avert an all-out financial sector crisis, the gov- ernment had already, in other programs, taken action to limit fiscal budgetary con- tingent liabilities. For example, central government guarantees, whether for debt payment of third parties in the public sector, risks associated with public invest- ment cash flows, or unpredicatable revenue streams from productive public enter- prises, the government's potential obligations were budgeted according to the probable expected values of realizing such guarantees. The government had thus already explicitly taken into account actuarial principles of event occurrence to require probability-adjusted budgeting to hedge such contingent obligations and prefund them.3 a) The Corporate Sector The corporate sector was also a key component of the financial sector rescue pro- gram since it addressed the resolution of debtor problems. It also benefited from emergency measures by way of Law 550 of 1999, which streamlined debt restruc- turing procedures among debtors and creditors and avoided lengthy judicial processes, which in the past had drawn out the eventual settlement and agreement on corporate rehabilitation and debt restructuring plans. The previous Concordato system involved judicial decisions and usually took years for a final settlement to 3 In a similar initiative, the government is considering the development of an insurance- based product to transfer government risk to the market to hedge against losses to gov- ernment property, which may occur due to natural catastrophes, such as earthquakes. Such an initiative would complement the government's quasi-risk-management program already established under the calamity fund budget, although using more innovative insurance methods and possible secunritzation of insurance risk (see Natural Disaster Man- agement Chapter by Pollner, J ). IHE FINANCIAL SECTOR 265 take place. Concordato negotiations generally resulted in creditors pushilng compa- nies overly quickly toward liquidation. The new restructurinig agreement gives votes to each creditor based on claims at stake, but requires all creditors to vote on a con- sensus approach. With the new extrajudicial debt restructuring agreements, all stakeholders (finan- cial institutions, commercial creditors, tax authority, employees, suppliers, cus- tomers, and so forth) are required to vote on a common restructuring plan. How- ever, each creditor weights its voting share by the value of its claims at stake. Based on the data as of end 2001, 586 companies had applied for debt restructuring agree- ments, with just under 100,000 employees affected. The asset value of these com- panies amounted to 6 percent of GDP. An additional 29 companies are scheduled for sale. Under the prior Concordato (judicial) system, since 1998, 322 companies were involved in restructuring negotiations, with total assets estimated at 1.3 per- cent of GDP. For those companies not considered solvent under both the Con- cordato and the new debt-restructuring program, there have been 574 liquidations since 1998, with assets equivalent to 1.5 percent of GDP (see Table 6) These data reflect the fact that the total affected in the corporate sector in terms of assets inca- pacitated since 1998 reached almost 10 percent of GDP (average GDP over those years). In summary, it should be pointed out that the government's rapid response and policy approach was instrumental in averting a systemic financial collapse given the macroeconomic, regional, and global environments. The prompt introduction of a new legal framework for banking capital adequacy, portfolio concentration, and consolidated supervision of financial groups was critical for early diagnosis of the problems. At the onset of the crisis, the establishment of progressively stringenit prompt corrective actions for banks facing possible insolvency prevented what might have been fiscal outlays in multiples of what was actually disbursed. The fast imple- mentation of bank capitalization schemes coupled with incentives for risk sharing by owners and more modern procedures for the prompt exit or merger of weak banks in the system was crucial to avoid the proliferation of "zombie" institutions that would have only postponed more serious problems raising the overall cost of the cri- sis. Therefore, the government's record in tackling this precarious situation was com- mendable. The following recommendations are meant to reinforce the policies already taken, and to sustain the progress achieved Table 6. Assets as Percent of GDP of Companies/Corporations in Distress (percent) Companies Under Concordato, Companies Under Companies Under 1998-2001 Restructuring Arrangements Liquidaton 13 6 1 5 Source Superintendencia de Sociedades 266 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE b) Pending Policy and Institutional Reforms- The Banking Sector While innovative policy tools and procedures were initiated and rapidly deployed during the financial crisis, the evolution of these processes still requires additional incentives and policy measures to ensure a proper and final resolution. The process has not yet been completed in terms of assuring a sustainable and stable condition for a number of banking institutions, and risks remain including: * The dependence on an uncertain economic recovery to achieve adequate bank capitalization and returns, in particular for the recapitalized private banks and CAVs, some of which continue to remain highly vulnerable to macroeco- nomic shocks; * The risk that shareholders will not be able to fully recapitalize banks if a recov- ery does not materialize, thus requiring a potential second round of govern- ment support; o The latent currency devaluation risk which could further damage debtor repayment capacity and thus portfolio quality, requiring additional infusions of capital in the banking system to avoid bank insolvencies; * The banking sector's dependence on government securities, which, while pro- viding needed balance sheet liquidity, also raises the risk of increased exposure to an uncertain fiscal sector, * The risk of the inability to privatize the remaining public or officialized banks with large ensuing contingent fiscal liabilities for both the State and FOGAFIN. A new law of Financial Ad)ustment (Ley de Ajuste Financiero, Modificaczones a la Carta Organica), recently approved by the lower house of Congress, will address some of the gaps to fully implement the financial resolution policies and supporting mechanisms for carving out assets and liabilities of banks. Some of the key gaps requiring additional specification to permit the conclusion of the processes initiated, however, would involve: * Improvements and increasedflexibilty in the banking resolution measures used to restructure or dispose of weak banks using market agents and instruments While the prior reforms of the banking law permitted such mechanisms, the good bank/bad bank approach facilitated by securitizations of portfolios held in trust, was effectively not used. To avoid the languishing of financial institutions under the State's umbrella, if privatizations do not materialize, increased incen- tives such as forced auctions should be applied, to allow a more prompt carv- ing out of good assets with matching deposits from intervened institutions.4 4 A good example is the mechanism under Article 35 Bis of Argentma's Banking Law. See "Resolving Bank Failures in Argentina," de la Torre, Augusto (2000). THE FINANCIAL SECTOR 267 Given the recent developments and the establishment of the legal basis for asset securitization in Colombia, this mechanism should be exploited to trans- fer asset-based securities overcollaterized by underlying credit portfolios, to market participants under auction procedures. Such tools would avoid lengthy due diligence audits and provide a more liquid instrument with which to share risks between the government and the private sector. The establish- ment of trustees for this purpose would also be required in order to limit lia- bility to banks assuming these portfolios, which should reflect a multiple of the face value of the security issued by the trust. Any remaining asset gap to match transferred deposit liabilities would be filled by a State/FOGAFIN security yielding market rates Reducing and Extinguishing Potential Fiscal Liabilities: To improve the prompt privatization or liquidation of financial institutions considered long term insol- vent, and to wind down weak institutions that have no earnings prospects while compensating depositors, additional options for the liquidation or asset sale proce- dures should be considered using a multibank portfolio approach managed compet- itively by firms. Time-bound performance measures and compensation fees should be applied, with rotation of liquidators to provide incentives for achiev- ing maximum returns in stipulated time periods Definition of time-bound lim- its for the disposition (sale or liquidation) of intervened banks should be set, particularly for banks where investmenits have already been made in recapital- ization and improvement of their financial condition. The Central de Inver- siones, S.A (CISA), the designated asset management agency for the govern- ment has successfully recovered almost one third of assets taken on, and regularized another third to reach performing status, however, its operational and legal requirements to maximize liquidation value while minimizing time elapsed, sometimes generates differing objectives which should be ranked and given priority within defined policy targets Particular attention should be focused on the State-owned banks (Bancafe, Granahorrar, Banco del Estado, Fundacidn para la Educacidn Superior, Financiera [FES], and Instituto de Fomento Industrial [IFI]) anld two private banks currently intervened and under FOGAFIN's watch. Failing successful liquidation or sale of assets in the fore- seeable future, the Government should include in its medium term budgetplan, the fiscal amortization and payment of any remaining orpotentialfiscal contingent lia- bilities and obligations to the officialized banks and FOGAFIN (to protect its liquidity needs), generated under the bank restructuring and support schemes. While banking prudential regulations are strict and include additional provi- sions for market risks in conformance with Basel standards, completion of these best practices should include the implementation of a matrix of prompt corrective actions associated with a schedule of sanctions and penalties, based on the quantification, weighting and incorporation of deficient corporate man- agement and sub-optimal risk management practices as key triggers for invoking corrective measures and penalties. These should be included in the already 268 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE developed schedule of penalties by the Banking Superintendency for exceedance of prudential limits. To support the long term implementation of these practices, the Superintendency should be provided with resources to strengthen its supervision procedures, provide staff training, upgrade its knowledge of risk management analysis and modernize its systems for evalu- ating and monitoring banking risk. * In terms of managing the debtor dimension which is linked to economic reac- tivation, while at the same time addressing the banklng-Financial sector issues, continued attention to the corporate sector will be required in order to. * Prevent deterioration of companies as a main future strategy by using information tools capturing corporate data, to establish early warning indi- cators and detect structural weaknesses ex ante; * Increase the efficiency in implementing restructuring agreements, and implement electronic tribunal modalities online to expedite decisions among debtor companies and their creditors, and * Reform the bankruptcy law to remove obstacles to restructuring and final liquidation of corporate assets to permit smooth exit, restructuring, and liquidation mechanisms. * To ensure the prudential management of non-credit institutions, the next key priority is the reform of solvency standards and consolidation of the insurance sector, which is currently overdimensioned (in terms of number of players given market size), with 50 insurance companies in the industry. Priority should be given to upgrading insurance industry regulations, including the proposed adoption of actuarially sound solvency margins particularly to ensure more adequate technical reserve provisioning for both health insurance and general insurance risk exposures, presently not well captured under cur- rent norms. The upgrading of regulations and guidelines for the trust indus- try will also be required, particularly with the advent of new products to be managed, such as asset securitizations. V. Reform of Second-Tier Public Banking Institutions In 2000 the government issued an executive decree to begin the process of merging the second-tier banks. However, opposition to the decree generated a court case, which culminated in Colombias Constitutional Court ruling that for the mergers to take place, a new law was needed because each state bank had been established under a specific law and charter. a) Characteristics of Government Banks in Relation to a Merger Proposal Of the banks discussed, Financiera Energetica Nacional (FEN), the energy sector bank, has the largest outstanding loan portfolio, followed closely by the Banco de THE FINANCIAL SIc:rOR 269 Comercio Exterior de Colombia (BANCOLDEX), the export bank. The next largest is the Instituto de Fomento Industrial (IFI), followed by the Fondopara elFinanciamiento del Sector Agropecuario (FINAGRO) and then Financiera de Desarrollo Territorial (FINDETER). The total second-tier lending portfolio in Colombia is approximately Col$4 5 trillion, or US$2 0 billion In terms of operating efficlency (staff costs over income), the most costly of the banks are FINDETER (10 percent ratio) and IFI (9 percent ratio). The other banks hover closer to 2 percent on this indicator. BANCOLDEX, the most commercially successful of the second-tier State banks, is legally a commercial bank, unlike most of the other banks. Its cuirent equity is approximately $500 million. If the State banks were merged, including BAN- COLDEX, they would represent from 20 percent to 25 percent of the entire bank- ing system, which could raise the issue of concentration under goveinment owner- ship. The advantage of an aggregated new bank, however, would be its common treasury function, which could imply funding efficiencies, although a disadvantage of this could be that the government might forcibly place its bonds in such a bank's treasury portfolio FEN, the energy sector bank, is certainly a candidate for being merged given that it is operating at a loss due to failing regional utilities that have received subsidies in the form of State guarantees Its portfolio is not growing, and this bank is becom- ing obsolete. A merger of FEN would entail transferring a passive portfolio for even- tual collection, but no new anticipated lending. IFI, the national development bank that in the past offered both first- and second-tier loans, and is now solely a second tier institution, represents the major problem in the merger proposal given that it has a substantial capital gap due to seri- ous defaults generated from its prior first-tier portfolio. IFI requires a substantial capitalization, and the assumption that a merger with the other banks would pro- vide this is highly problematic since it would only create a new bank that was highly fragile, but represented a significant postion of the financial system IFI estimates that it gross current liquidation cost (that is, asset-liability gap if liquidated) would be Col$700 billion. FINAGRO, the agricultural-agroindustry second-tier bank, has the most obvi- ous financial subsidy in its funding base, since the entire banking system is forced to purchase FINAGRO securities at rates well below the commercial bank deposit rates, presumably for FINAGRO to direct credit to the agricultural sector FINA- GRO, however, has been successfuil in reorienntng its end products, given changing financial circumstances and institutionial financial structures For example, in the past FINAGRO lent through specialized agricultural banks, many of which no longer exist. Therefore FINAGRO had to tailor its products to the commercial banking system in order to survive. Some such prodticts have included guarantee instruments versus direct loans. An additional product developed by FINAGRO was the pooled community loan (crid'ito asociativo). This was lent via first-tier banks, for example to a growers' association, and the loan was guaranteed by a claim on the association's sales revenues from contracts already entered into. 270 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE FINDETER, which lends to municipal governments for public infrastructure, is considered too politicized by many observers. While FINDETER would like to access capital market funding through, for example, securitized bonds backed by their portfolio, its current delicate financial state and its prospects for being merged (while opposed by the management) would put such initiatives on hold. b) Pending Policy and Institutional Reforms-Second- Tier State Banks Ironically, the bank with the best prospects of surviving on its own commercially without many problems is BANCOLDEX, the export bank. This bank, however, is also the one best suited to being the management center of a new institution com- posed of merged banks, because it is the least politicized, most profitable, and best managed, with the least government interference. However, an ill-designed merger would likely deplete much of its surplus capital in order to keep the other acquired lines of business solvent, particularly those of IFI. The following reform parameters should therefore be considered: * A key issue to resolve initially is how to fund the equity gap of IFI. Even con- sidering the surplus capital of banks like BANCOLDEX, this would be grossly insufficient to cover the approximately US$0.3 billion needed to make IFI minimally solvent. Therefore a successful merger proposal would imply the need for in)ection of capital by the government, funded via long-term instruments such as a bond issues or multilateral support. Some efficiencies in staffing, while not approaching anywhere close to IFI's capital gap, could be achieved by merging the funding-treasury functions and back office and accounting tasks into a common office. * The financial-legal structure of the new bank could also entail different processes and options. The simplest but least desirable approach would be a straight merger of all the banks. A more promising option would be for a new law to have the charters of each bank expire and establish one government holding company for various rediscount lines of business, which would oper- ate as separate trusts to avoid the mixing of unrelated portfolios. Such a struc- ture would permit the rationalization of costs and management, and the con- tracting of independent restructuring expertise to downsize the aggregate portfolio size of the second-tier operations. This could be accomplished by first defining a reduced target of first-tier clientele and sector(s), followed by the separation of each bank's portfolio into "good" and "bad" assets prior to merger, so that nonperforming portfolios could be dealt with and disposed of while simultaneously winding down operations with first-tier intermediaries in highest default A key precondition, as already identified, would be to elim- inate and wind down first-tier lending (for example, the IFI case) as a prod- uct line, given the high risks to the government and the potential for engag- ing in deposit taking TilE FINANCIAL SECTOR 271 * A more radical though increasingly accepted reform of the State banking sec- tor would be eliminating state-owned banking and substituting it with budg- etary resources for funding technical and investment support for the targeted sectors, such as agriculture. The rationale for this approach is that State bank- ing institutions should, first, not put depositor funding sources at risk, and if not funded by deposits (tlat is, donor funds), these would imply government liabilities anyway. State banking institutions also pose large liabilities to the government if they are operating at a loss or are technically insolvent The budgetary approach avoids any government contingent liabilities from bank- ing operations and ensures that investment resources or technical resources are channeled in conjunction with existing private credit programs (or provide more security to such programs) by extending direct government assistance in sectors with potential productive capacity. Essentially, it makes little sense for the government to lend to noncreditworthy subjects, nor does it follow that the government should lend to creditworthy subjects who can be serviced by the private sector. * A portfolio audit of all the banks should be conducted prior to determining the market asset values and liabilities in each institution. Only once the actual aggregate capital gap of the various banks can be determined more precisely, could a viable financial structure and realistic downsizing strategy emerge. VI. Capital Market Development and the Public Debt Market Accelerating the development to deepen the markets for long-term debt and securi- tized assets and derivatives, and promoting an ordered development of the large insti- tutional players in such markets-including insurance companies, mutual funds, and pension funds-constitutes an important alternative mechanism to facilitate the absorption of risks by market participants. The development of the market for cor- porate stocks and securities will lessen the reliance on banking intermediatioll, thereby reducing debt leverage and allowing macroeconomic and financial shocks to be directly absorbed by investors rather than debtors or financial intermediaries. A comprehensive strategy for capital markets and financial intermediaries should be developed At present, there exists an excessive fragmentation in Colombia's financial markets. Capitalization of the stock market amounts to only 13 percent of GDP, one of the lowest for the more developed economies of the region The inte- gration of the three stock exchanges in Colombia will help to increase market liq- uidity. The new securities law, now under discussion in the Congress will also help reduce obstacles into the issuance of securities by companies and provide incentives for capital market activity, such as providing the legal basis for developing securitized transactions, one of which has been recently implemented. To facilitate securitiza- tion, a standardized mortgage instrumenit needs to be created The recent launching of a securitization firm now provides the legal and institutional mechanism to jump- 272 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE start the securitized asset market, which should help promote longer-term securities backed by credit or other asset portfolios. The government has continued to progressively and successfully develop the basis for a capital market, particularly through its government securities market. The insti- tutionalization of market-making agents in the financial sector and the regular sched- ule of auctions implemented by the government has succeeded in creating a liquid and deep capital market in government paper. More recently, the government imple- mented short-term issues of bills and notes and longer-term bonds, so as to complete the maturities available on the government debt yield curve while providing a full range of risk-free benchmarks to support private market securities pricing. The addi- tion of issues of short-term maturities (60 days to 180 days) will also help set the zero risk benchmark of the yield curve for short-term corporate paper. At the subsovereign level, the Bogota District government recently began to issue municipal bonds on the fixed-income market, backed solely by the credit of the municipality-district. To date, therefore, the government debt market has been instrumental in gener- ating both primary and secondary capital market activity in the fixed-income sector. Within the existing range of maturities, government treasury instruments now cover the yield curve from short-term up to 12-year maturities. The share of dollar- denominated securities represents 10 percent of the total, and inflation-indexed securities represent 22 percent. Institutional investors including banks, pension funds, and corporations all invest in government Treasury securities. The success and liquidity of this market are commendable. However, with the increasing need for government debt financing, the success of this market may have started to crowd out private financing by absorbing the available demand in the domestic capital market. The government securities market currently amounts to Col$33.4 trillion (US$15 billion). This amounts to about 20 percent of GDP.5 In addition, over the last three years the banking sector has increased its share of asset holdings of government secu- rities from 10 percent to 30 percent. While this reflected an aversion to extending credit, it nevertheless shows that the government debt market takes up a large part of the capacity for financial investments in the private sector. Given the fiscal stance of the government, this high exposure represents a relatively high risk. PENDING POLICY AND INSTITUTIONAL REFORMS-CAPITAL MARKETS. The govern- ment may need to reassess its domestic debt financing strategy both for fiscally sus- tainable considerations (given that the domestic interest rate is substantially higher than external rates) and in terms of its effect on the development of the private cap- ital market. The following should be considered in the reform strategy: Access to the fixed-income market by nonsovereign issuers should be further encouraged in order to diversify funding sources. With the institutions now set up for asset securitizations in Colombia, encouragement should be given 5 Year 2000 figure used because this is the latest available full-year figure. TIHE FINANCIAL SECTOR 273 to deepening the market for securitized mortgage instruments once the terms of mortgage loan contracts become more standardized. * In terms of the private market both for fixed-income and equity issues, the new securities law should be approved without delay in order to implement transparency principles in the reporting of corporate financial information as the basis for securities issuance The passage of the law will only start the reform process, however. Following passage, the key medium-term challenges to effect institutional reforms in the area of capital market development will necessitate the following: * Implementation of streamlined regulations to facilitate efficient securities issuance procedures to allow firms to take advantage of capital market pric- ing opportunities. Ground rules need to be set for the preregistration of com- panies to qualify them for securities issuance and have company prospectuses prepared in advance. The existing lengthy legal procedural requirements to comply with issuance steps would thus be substantially removed. * The capacity of the Superintendency of Securities needs to be substantially fortified to provide it with the skills to detect and investigate cases of mar- ket fraud, manipulation, or insider trading. Prompt administrative )ustice procedures need to be fully implemented to efficiently address indemnlifi- cation of losses suffered by investors * Accounting and financial reporting norms required internationally for the issuance of securities to provide reliable investor information need to be implemented. This, along with the corporate governance requirements of the new securities law will be key to instilling market confidence and increasing both the demand and supply of private securities (equity and fixed income) * Market crisis contingency plans need to be developed in the event of mar- ket disorder or breakdown, particularly at the incipient stage of new prod- uct launches. The authorities should develop procedures for the orderly closing or managing of markets if and when disorder sets in, due either to exogenous factors or to factors related to market manipulation at the domestic level. THE PENSION SECTOR The transfer of all workers from the public pension system to private pension funds should constitute a primary reform objective of the gov- ernment (see Pension Reform Chapter). If such a reform cannot be undertaken in the immediate future, measures should be introduced to limit the scope for back- and-forth switching between the public and private pension funds and restrict the benefits of the public pension plan. It becomes particularly critical to migrate the pension system to the private sector given the growing and unsustainable pension liabilities of the public sector (see Table 7). Under the public pension system, numerous regimes exist Besides the territorial-regional regimes, most of which have been mismanaged and are currently 274 COLOMBIA- THE ECONOMIC FOUNDATION OF PEACE Table 7. Public Pension System Liabilities Liability Value Government Level Percent of GDP (in 2000 Col$ billions) National 131 244,413 Regional-territorial 38 71,105 Total 169 315,518 Source Direcci6n de Seguridad Social, Ministerio de Hacienda y Cr6dito Piiblico unsustainable from a financial-actuarial viewpoint, there exist a number of specialized regimes for distinct professions in the public sector (the military, energy companies, and others), which offer higher benefits to affiliates. The government plan is to begin dis- solving these and rolling them into one common public pension regime with equal ben- efits for all. In addition, there still exist seven national level cajas depension, which were allowed to remain under transitional arrangements during the last wave of pension reforms (Ley 100). These funds, while not having significantly different benefits than the main public pension social security scheme (prima media) are managed by separate institutions for particular groups (public employees, communications companies, Con- gress, and others) and, with the exception of the fund for Congressional members (FONPRECON), none are open to new affiliates; thus most of these funds are in their wind-down phase. One of the largest funds, Caja Nacional de Prevision Social (Govern- ment Employees Pension Fund, CAJANAL), has had an account established, the Fonde de Pensiones Publicas (FOPEP), to better account for and fund upcoming liabilities and to incorporate the pension accounts of any other public sector entity. Despite the apparent proliferation of funds at the national level, the critical chal- lenge is at the territorial level where individual municipal or other special-purpose public pension funds have a variety of regimes, and most are technically insolvent. To redress this situation, the government recently instituted a territorial pension reform that sets up FONPET, a territorial pension account in which all territorial pension liabilities are to be aggregated and funded. While the central government allocates some funds to FONPET, the reform expects the territorial level govern- ments to adjust their fiscal policies and to be held accountable for funding their pen- sion schemes. Any funding gap on the part of territorial (departmental or munici- pal) governments would be deducted from the earmarked transfers provided by the national government to the territories. Table 8 outlines the current fiscal funding provided by the territorial and national governments to FONPET. To reduce the future cost of fiscal outlays to fund the public pension schemes, a number of reforms are being proposed, including: * Increasing the contribution rate (salary based) from 12.5 percent to 14 per- cent, with an additional 2 percent phase-in, during later years. * Reducing the share of administrative and insurance expenses from 3.5 percent to 2.5 percent. While this provision is directed at the private pension sector in THE FINANCIAL SEC FOR 275 Table 8. Estimated Fiscal Funding of Territorial Pension Obligations (Col$ billions) 2000-05 As percent of As percent of 2000 Payments 2000 GDP Government Revenue National Fiscal Contribution 1,490 0 9 6.4 Territorial Funding 3,784 2 2 16 Source Direcci6n de Segturidad Social, Minoisterio de Hacienda y Credito l'ublico (Ministny of Finiance) order to reduce its administrative charges, it would affect the public system as well. * In terms of benefits, reducing the pension amount from the current 65 per- cent to 85 percent of salary, to a level that would start below 65 percent * In terms of benefit accrual, the months of work and years of age required to obtain full benefits would both be increased. Some aspects of draft reform legislation, however, are less progressive and should be reconsidered with urgency. These include the provisions for (a) forcing all public sector employees to switch to the old social security pension system regardless of to which system they are currently affiliated, (b) reducing the accrued benefits on recognition bonds for those affiliates wishing to transfer to the private system, (c) charging Admintstradora de Fondos de Pensiones (Pension Fund Managers, AFPs) a fee to set up a fund to finance the minimum pension guarantee, and (d) reducing AFP commissions by regulation. PENDING POLICY AND INSTITUTIONAL REFORMS-PENSION SECITOR While the share of the national public pension scheme is the largest, from an actuarial point of view, the territorial pension funds are much more insolvent and thus require critical action. The Ministry of Finance has already made significant efforts to obtain the requisite employment history information to use as the basis for calculating the actu- arially fair value of the pension liabilities at the territorial level However, a number of institutional reforms are required in the effort to regulate and enforce the transi- tion from an essentially insolvent territorial scheme to a funded one The critical steps in the second generation of pension reforms therefore include: * Consolidation of the various public pension regimes, including the specialized regimes, the national cajas, and the territorial (departmental, mullicIpal, and sectoral) cajas, the latter falling under FONPET A sustained institutional effort in the regulation and control of the operation of the consolidated funds and assuring consistency in the accumulation of benefits. * To adequately determine the funding needs of FONPET, all municipal and other autonomous entities need to be reviewed and analyzed to determine 276 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE what type of pension structures, funding, and benefit rights are currently in force. * Invoking judicial procedures to reverse benefits offered under illegally estab- lished pension regimes, an exercise that would generate substantial fiscal sav- ings but may risk counterlawsuits with regard to "acquired rights" arguments. * Strong coordination between the social security and central budget units in the government to assure accountability from departmental and municipal governments and to withhold, when necessary, portions of the territorial transfers when territorial level governments are not funding their actuarially calculated obligations In terms of the reform proposals affecting the private pension system, some cur- rent proposals presumably meant to fill the liability gap in the public system would nevertheless be severely detrimental to the private system and to any possibility of developing the private capital markets in Colombia. In particular, the following con- siderations should be addressed with a view to putting in place a reform which addresses the fiscal problems of the public system without sacrificing the integrity of the private pension system and its key role in the domestic capital market and in assuring fully funded pension schemes: e Any proposal to force public sector employees to be affiliated with the social security system reflects unwarranted state interference in artificially raising the funding (temporarily) of that system, and also constitutes a dramatic reversal of policies aimed at private sector development from both the affiliate and the AFP points of view. This provision should thus be eliminated entirely because in the long term it would generate larger fiscal problems, given the eventual need for the social security system to pay out those new pensions (with high benefits), which have been forced in. * Similarly, any reduction in accrued benefits allowed in recognition bonds issued (meant for transfer to the private system) will only force more affiliates back to the public system and generate larger pension liabilities for the gov- ernment in the future. Any such provision, therefore, is not recommended. • The funding of the minimum pension guarantee should first establish an equalization of terms under which affiliates from either system can invoke the minimum pension, given that currently the conditions under the public sys- tem are much less stringent in terms of weeks of contributions made com- pared to the private pension system. Again, as above, such disincentives to the private sector will only encourage affiliates to return to the public system and increase fiscal liabilities dramatically. * The proposal to force reductions in AFP commissions appears artificially designed with the purpose of financing the minimum guarantee fund without taking into account the market's cost structure or comparative private pension commissions in other countries with similar affiliate populations. ITHE FINANCIAL SECTOR 277 Thus, a reexamination of the proposed pension reforms should aim to transfer more affiliates to the private system (to reduce future fiscal liabilities) rather than to temporarily postpone the funding gap in the public system by channehlng affiliates in the private sector AFPs to the public system. Implementation of any such policies would risk increasing the fiscal deficit bubble dramatically in the long term, while weakening the entire private pension structure and the prospects for developing any significant capital market in Colombia. VII. Conclusion Colombia's pending reforms, needed to assure future financial sector growth and sta- bility, tend to integrate a number of interlinked subsectoral issues In particular, the resolution of banking system weaknesses is also connected to the development of capital markets and instruments to help progress toward finality in this area. In turn, both the capital markets and financial systemic management issues have impacts on fiscal resources and the use of government debt. The fiscal dimension becomes even more critical in light of the state of public pension structures, which are in need of immediate reform. In light of these factors, the reform of the Colombiani finanicial sector in all its aspects requires the simultaneous coordinated efforts of the relevant regulatory and supervisory agencies in the banking, capital markets, and pension sectors. 9 Pension Reform This Chapter was written by Hermann von Gersdorff I. Executive Summary The current condition of Colombia's pension system constitutes a major risk to fiscal sustainability over the medium and long term. The public pension system is technically insolvent and the yearly imbalances of the public sector worker regimes are increasing. The estimated net value of pension system liabilities amounted to about 200 percent of GDP in 2000-up from an estimate of about 150 percent of GDP in 1997. Moreover, the Treasury's transfers to finance the deficits for public sector worker regimes increased from about 0.8 percent of GDP in 1991 to 1.3 percent of GDP in 1995, and to 2.3 percent of GDP in 2000. The causes of the current pension system insolvency and liquidity prob- lems include generous pension benefits, pension benefit guarantees, perverse incentives for reserve management, and recent Constitutional Court rulings expanding pension benefits. These causes are rooted in the slow pace of reform and the large number of workers exempted from the reform. This has resulted in pension expenditures still growing at a fast rate, particularly the pensio n expen- ditures for public sector workers. To confront the problems of a financially bankrupt and fragmented defined ben- efit (DB) pension regime, Colombia undertook a first generation of reforms of the Social Security System (enacted by Law 100 of 1993). The main elements of the reform included (a) introduction of a mixed defined benefit (DB) and defined coIn- tribution (DC) pension system-workers could choose between the DB regime run by ISS or the DC regime run by the Admznistradora de Fondos de Pensiones (private pension fund managers, AFPs); (b) increases in contribution rates paid by all active workers (that is, all public sector workers would now contribute), and reduction of pension benefits for younger workers; (c) insolvent pension funds for public sector 279 280 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE workers at the national and regional level would be closed and pension payments would be taken over by the Public Pensions Fund (FOPEP) and specialized regional entities; (d) introduction of explicit redistribution through additional contributions to support pension benefit programs for the poor; and (e) government recognition of pension rights accrued until 1994 through "pension recognition bonds"-that is, the government would assume the repayment of the corresponding implicit pension debt (IPD).i The law, however, exempted the pension funds for Empresa Colom- biana de Petrdleos (ECOPETROL, with 6,000 members), teachers (300,000 mem- bers), the armed and police forces (200,000 members but only half with pension rights), and members of Congress. Furthermore, the law allowed for an extended transition to the new system, not scheduled to be completed until 2014. One of the major government objectives for defining a second generation of pen- sion reform is to reduce Colombia's IPD. New entrants will at least not increase the current IPD as a result of the reform process. The reform proposals maintain the current situation of concurrent defined benefit and defined compensation regimes, deepen the reform envisioned in Law 100 of 1993, and strengthen the pension sys- tem, including reforming at least one of the exempted pension regimes. The reform program will seek to harmonize or introduce a single set of unified pension rights and obligations for all new entrants, while increasing coverage of and improving services to all participants. Key actions in the government's pension reform program include undertaking technical work and the preparation of legislation to fund the minimum pension guarantees through an insurance scheme; and closing entry into any pension fund except the Instituto de Seguros Sociales (Institute of Social Security, ISS) and AFPs. The reform also envisions, among other things, a deadline for a final switch of cur- rent members of the system between regimes, and increased eligibility requirements in terms of age and number of contributions for a minimum pension guarantee. To deepen the reform envisioned in Law 100 of 1993 and strengthen the pension sys- tem, the government's reform program should include (a) centralizing the adminis- tration and control of transfers of public pension regimes into a single agency, (b) improving the databases required for the recognition and recertification of pensions, (c) prefunding new accrual rights to extraordinary benefits, and (d) creating a Uni- fied Registry of Contributors (URC). In addition, the reform of the exempted regimes-for example, the armed forces-should include key features such as (a) individual pension savings accounts, (b) a close link between contributions and ben- efits, and (c) the application of the new regime to all new entrants. All of the reforms mentioned above will have a positive impact on the financial situation of the public pension system and on the required transfers from the gov- ernment Treasury in the medium to long term-that is, in 8 to 20 years. The only reforms that will have an impact in the short term are those that accelerate the tran- 1. The implicit pension debt is the value of all pension rights accrued by workers plus the net present value of all pensions already granted PENSION REFORM 281 sition-that is, apply the new pension system to most of those that are still under the rules of the old pension system until 2014, and to those that are in the exempted regimes, like teachers and the workers of ECOPETROL. II. Introduction Colombia's social insurance system has been beset by all the ills of publicly managed, defined benefit, pay-as-you-go (PAYG) pension schemes that are encountered in other developing and Organization for Economic Cooperation and Development (OECD) countries. For over 50 years Colombia's pension system has been charac- terized by uneven contributions and benefits across different pension regimes, low coverage, very little funding of pension liabilities, and rising fiscal transfers to cover cash deficits Colombia started a pension plan for central government workers in 1946 with the creation of the Caja Nacional de Prevision Social (Government Employees Pen- sion Fund, CAJANAL). Subsequelltly, other national pension funds and cajas were created for central government employees working in specialized agencies or profes- sions tO cFeate more generous pension regimes Regional governments, departments, and municipalities started additional cajas and pension funds. Another segment of the public sector pension system is comprised of state-owned enterprise funds estab- lished by public sector enterprises for their workers with different contributions and benefits. Mandatory old age insurance for private sector workers was formalized in 1967 through the Instututo de Seguros Sociales (Institute of Social Security, ISS), which was already providing other services. In 1993, Colombia's fragmented pen- sion system comprised three categories of pension institutions for public sector workers: (a) CAJANAL, (b) 55 other national pension funds and cajas, and (c) close to 1,000 regional pension funds and cajas. Another group of pension institutions comprised decentralized government and public enterprise pension funds. ISS and some funds in private enterprises covered private sector workers The pension system was based on social insurance principles Originally established as partially funded regimes, the pension systems' reserves were rapidly drawn down due to generous benefit eligibility conditions coupled with high evasion of contribu- tions. The financial pressure on the pension system, however, is not yet due to anl aging population, since Colombia is still a demographically young country. In addition, the system covered only about 35 percent of the economically active population. In 1993 Colombia undertook a major reform of the Social Security System. Among other reforms, it introduced a privately managed defined contribution regime with defined contribution (DC) and individual capitalization, and reformed and scaled down its public defined benefit pension regime by reducing somewhat preferential treatment for individual groups, raising the retirement age, and tighten- ing eligibility conditions. The new, tighter rules and eligibility conditions are being phased in very slowly. Therefore, there is still a strong incentive, particularly for 282 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE older workers, to remain in the defined benefit regime. Moreover, various groups of workers, such as workers in the oil sector, teachers, and members of the armed forces, kept their own pension regimes, which they will continue to do even after completion of the 1993 reform in 2014. The somewhat limited extent of these reforms, however, reflects the political dif- ficulties of passing a radical pension reform in Colombia. The DC regime was orig- inally meant to replace the existing DB regime. However, due to the resistance of trade unions, the social insurance bureaucracy, and other groups, the DC regime was offered only as an alternative to the existing DB regime, and there is no obligation for new labor force entrants to join the DC regime. The government of Colombia is planning to deepen the 1993 pension reform to respond to the increasing liabilities and deficiencies of the current system. There is a comprehensive reform proposal and also a draft law before Congress that attempts a more limited reform. This Chapter will focus mainly on the reform of the old age pension system, and will only briefly touch on disability, both common and occu- pational, and survivorship pensions. Another important issue that is neither ana- lyzed in the required depth nor resolved is the financing of the pension liabilities that have already been accrued. This Chapter describes Colombia's pension system, identifies the fiscal requirements of the current system, and concludes with some proposals for reforming the system. III. Colombia's Pension System a) The 1993 Pension Reform Although there were a large number of funds and regimes and inequality in benefits in the pension system, there was also low coverage. In 1993 only 35 percent of the economically active population of 14.3 million were part of the system. ISS had 3.5 million active members and 265,000 pensioners. Some 1.1 million public sector employees were provided for by the government. Benefits provided by the old system varied widely among different regimes and were generous on average, but even more so for public sector employees. Retirement ages were 50 for women and 55 for men in the public sector, and 55 for women and 60 for men in the private sector. Pensions relative to wages during the last year of work life were mostly 75 percent for the public sector and 45 to 90 percent for pri- vate sector employees with as little as 10 years of contributions to the pension sys- tem. However, many different special pension regimes provided more generous con- ditions than those described above, such as early retirement and replacement rates well in excess of 100 percent of wages. In CAJANAL alone there were 17 diffetent pension regimes in 1993. Colombia's defined benefit pension system was de facto a pay-as-you-go (PAYG) scheme, the reserves for which were close to zero. CAJANAL's reserves actually were PENSION REFORM 283 zero because contributions for public sector employees generally were zero. Contri- butions to ISS were 6.5 percent of wages until 1993, when they were increased to 8 percent of wages. ISS reserves of COP650,000 million covered only a small fraction of its pension liabilities. The pension regimes for public sector workers already had a cash deficit, and in the absence of reserves the central government was forced to make yearly transfers to the public pension regimes of about I percent of GDP Hav- ing had to finance the public pensions for years, and the upcoming need to finance ISS pensions, combined with all the other problems of the pension system, led the government to initiate a pension reform that resulted in Law 100 of 1993. b) Law 100 of 1993 The main elements of Law 100 are: * A significant increase in contribution rates paid by all active workers (public sector workers now contribute) and reduction of pension benefits for younger workers. * Introduction of a mixed defined benefit and defined contribution pension system. All workers can choose between the DB regime run by ISS and a DC regime run by private pension fund managers (AFPs). * Mandated closure of pension funds for public sector workers at the national and regional level and payment of the pensions to be taken over by the FOPEP and specialized regional entities. * Explicit redistribution through additional contributLions for a program to sup- port pension benefits for the poor. * Government recognition of pension rights accrued until 1994 through pension recognition bonds, implying governmenit repayment of the corresponding IPD. * Regulation and supervision of ISS and AFPs by the Superintendency of Banks * Regulation of the transition to the reformed pension system and of the excep- tions to the general pension system. Participation in one of the regimes is mandatory for workers. Workers who do not express a preference are automatically enrolled in the DB regime. However, workers are permitted to switch back and forth between the public and the private regimes every three years. In contrast to most other reforms in Latin America, the generous pension benefits of the old pension system were kept for women age 35 and older and men age 40 and older or with 15 years of service in 1994. Only the young and those Joining an AFP submit immediately to the new pension parame- ters. The continuation of low retirement ages and generous benefits results in a high fiscal cost, creates incentives for corruption and fraud, and introduces a strong bias against the DC regime (for example, women joining an AFP can retire only at age 60 instead of 57). 284 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE The new contribution rates amount to 10 percent of wages. Workers also pay 3.5 percent of wages to cover the premium for disability and survivors' insurance. In the private system about 1.87 percent is for insurance and 1.62 percent is for the AJFPs' management fees. All members who earn more than four times the minimum wage must contribute an additional 1 percent of their wage as a solidarity tax (that is, a tax used to subsidize the contributions of some poor workers). This revenue is sup- posed to be matched by budget transfers and used to subsidize the contributions of targeted poor groups in an attempt to extend the coverage of the system. Entitle- ment to contribution subsidies, however, is not automatic, but is determined on a case-by-case basis. The beneficiaries of the subsidies are required to join the regime managed by a nonprofit organization. Only ISS qualifies as a nonprofit organiza- tion, and the 200,000 beneficiaries have joined ISS. While the Law 100 reforms reduced overall net pension liabilities, it increased some of the costs of the pension system. For example, it increased the cost of the minimum pension, increased the cost of higher pensions by changing indexation, and increased nonsalary labor costs substantially (see Table 1). The increase in labor costs, particularly, reduced or even eliminated the benefits normally associated with pension reform of reducing the perception of the tax element in pension contribu- tions. In light of Colombia's high nonsalary labor costs, the reform will most likely not result in better incentives to participate in the pension system, or increase the willingness to join the formal system. Figure 1 shows that accounting just for pay- roll taxes for social services, Colombia has the third-highest contribution rates in Latin America-lower than those of only two countries with more mature systems and older populations. However, the increase of contribution rates for all, and higher retirement ages for younger cohorts decreed by Law 100, were financially very effective; they reduced the IPD for 1994-2025 by about 37.6 percent of GDP. Table 1. Colombia: Total Labor Costs (as percent of wage) 1990 1996 Separation payments 8 3 8 3 Retroactive separation payments 4 2 4 2 Vacations 6.7 6.7 13th month's salary 8.9 8 9 Pensions (worker) 2.2 3.4 Pensions (employer) 4.3 10.1 Health (worker) 2.3 4.0 Health (employer) 4.7 8.0 Training (SENA) 2 0 2.0 Family Welfare (ICBF) 3 0 3.0 Compensation Funds 4.0 4 0 FSOPEN I ARP 2 2 Total 52.6 64.6 (+1) PENSION REFOPM 285 Figure 1. Payroll Taxes for Social Insturance and Health in Latin America, 1995 50 0- 45 0 - --- For all Social Insurance 40 0 - - * Of which Pensions_ 350 -_ __ 300 _ o 25.0- - _ __ _ _ _- a 20.0- - ----- c. 15 0 -_ - 10 0 - --- 50o 0 0 E~~~~~~ c E i:Ur I U 0~~~~~ The minimum pension is equal to the minimum wage at the moment the pen- sion is granted. It is adjusted either by the consumer price index or the change in the minimum wage, whichever is larger. The minimum wage is very high in Colom- bia. In 1994, it was about 72 percent of the average salary in the seven largest cities. Workers in the transition regime require only 10 years of contributions to be eligi- ble for the minimum pension; low-wage workers have a very strong incentive not to contribute afterward. Younger workers in the DB regime need to contribute 20 years, and members of the DC regime are required to contribute for at least 22 years, to be eligible for a minimum pension. The government would pay the minimum pension once the pensioner in the DC regime runs out of funds for his or her pen- sionl, and the government guarantees the availability of funds in the DB regime. To reduce the likelihood of having to pay the minimum pension in the DC regime, Law 100 introduced a minimum rate of return requirement for pension funds. The law mandated that the minimumn rate of return be tied to marker returns. This is being implemented through a minimum rate that is determined as a combi- nation of the performance of a synthetic reference portfolio and the average per- formance of the AFPs. Otherwise, the regulations for the investment of the pension funds' assets are less stringent in Colombia than in other countries. The institutional setup for the transition of the reform is complex and will last until 2014. All public sector workers changing jobs and new entrants to the labor market are required to join either ISS or an AFP, but have the chance to switch to 286 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE ISS or an AFP every three years Public sector workers may remain members in the pension fund to which they belonged before the 1993 pension reform. However, if the fund is insolvent it has to be liquidated and the workers have to affiliate with ISS or an AFP. If the public sector worker joins ISS he or she will be able to retire under its old regime if he or she retires before April 2014. If the worker joins an AFP he or she is immediately subject to the new pension system. Only solvent funds are sup- posed to retain their active members until the last pensioner dies. Funds declared insolvent may try to constitute reserves to reach solvency and continue managing the regime for their members, but new members are not allowed to join. The pen- sion funds for the armed and police forces, teachers, and oil workers (ECOPETROL) were exempted from the requirements of Law 100. c) Current Structure of the Pension System As a result of Law 100 the pension system has the following regimes and entities: * ISS manages the new DB regime and the transition regime for men who were older than 40 or women who were older than 35 or had more then 15 years of service in April 1994 and chose to join ISS or to stay in ISS. ISS has about 4.5 million members, but only about 2 million are contributing About 200,000 are public sector workers and 200,000 are beneficiaries of the sub- sidy to contributions. * The six AFPs manage the DC regime. As of January 2001, AFPs had 4 mil- lion members, of whom 1 68 million came from ISS, 73,000 came from pub- lic sector worker pension funds, and the rest were new members of the pen- sion system. However, only 1.6 million were actually contributing. * All pension funds for public sector workers that have not been declared insol- vent continue to manage their regime for their active members who have not switched to ISS or an AFP, and continue paying the pensions of their retirees. Funds declared insolvent may not continue managing active members, and just pay pensions until they are liquidated and the payment of pensions is taken over by a specialized entity. These funds have about 467,000 active members and about 334,000 pensioners * The funds for ECOPETROL (6,000 members), teachers (300,000 members), and armed and police forces (200,000 members, but only half with pension rights) were exempted from Law 100. d) Evolution of Coverage Based on the membership and labor force figures in Table 2, the coverage of the pen- sion system in Colombia is in the middle range of Latin America, as can be seen in FiguLres 2 and 3. In terms of pension coverage, of the population over age 60 (which constitutes 6.5 percent of the total population), 30 percent receive a pension (that PENSION REFORM 287 Table 2. Labor Force Statistics as of January 2001 Total population 41 4 million Working-age population 31 15 millioon Employed 16 49 million Economically active 19 73 millioni Unemployed 3 24 million is, about 2 percent of the total population). In addition, Colombia has a support scheme for the elderly poor that are not entitled to a pension. This support is given through a small program (about 0.05 percent of GDP), REVIVIR, a program estab- lished by Law 100 in 1993 and with beneficiaries identified by the municipalities, and many even smaller programs run by municipalities The REVIVIR program pays up to 50 percent of the minimum wage to poor elderly who are older then age 65 (age 50 if handicapped). Between 1997 and 1999 REVIVIR had about 90,000 beneficiaries out of a pool of about 320,000 potential beneficiaries e) Instituto de Seguros Sociales The 420,000 pensioners of ISS are only about 20 percent more than the number of pensioners from the public sector; however, Colombia devotes almost 70 percent of pension expenditures to public sector employees compared to a typical 20 percent in developed economiiies. In 1998, Colombia spent 4.5 percent of GDP on pensions, of which only 1 38 percent was expended by ISS. This is clear evidence of the gen- erosity of the regimes for public sector workers. Part of the explanation is that, con- trolling for education and experience, Colombia consistently has the highest public sector wage premium among Latin American countries, typically between 15 and 20 percent. To equalize contribution rates between the DB and DC regimes, the contribu- tion rate for the DB regime was raised from 6.5 percent in 1993 to 13 5 percent of wages This implied a strong increase of revenues for the DB regime Of the 13.5 percent of worker salaries that ISS receives, it allocates 10 percentage points to old age insurance and 2 3 percentage points to disability and survivorship insurance. The remaining 1.2 percentage points are for its own expenses. Any surplus from the 1 2 percent is to be deposited into ISS reserves. Total assets under ISS management amounted to about US$2 4 billion in July 1997, which was more than twice the amount of funds managed by the AFPs. In 2000, ISS reserves grew to about US$2.8 billioni, equivalent to about 3.1 percent of GDP, and the reserves of the AFPs grew to US$4.3 billion. The year 2000 is likely to have been the last year in which ISS reserves grew. This is because, after several years of surpluses as a result of the pension reform, an increase in the number and size of pensions is resulting once again in ISS deficits. The reserves of the DB regime are supposed to be managed by competitive trust companies and have to meet min- Figure 2. Pension Benefit Recipients as Percentage of Elderly Population, 1980-99 9 oo 80 0 - 70 0 - _ ___ __ O60 0 _-_ 5 00 a4 40.0- 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 30.0-__\ 2 40.0 - ____ m Cl- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~____r,1 0 a 10.0 - 0_ a: 0.0 c 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 0 Argentina - Dominican Republic - Mexico z - Bolivia iO El Salvador ° Venezuela tr Colombia Honduras A Chile Figure 3. Contributors to Publicly Mandated Social Security as Percentage of Economically Active Population, 1980-99 z I z 70.0 60.0 -- - _ _ _.__ _-____ _ O 50.0- 40 0- vO 30.0- lu20 0- - 100 0 0 - I I I I I I I I I I I I I I I I I I I 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 _ Argentina X Dominican Republic - Mexico U Bolivia : El Salvador O Venezuela A Colombia + Honduras A Chile \O 290 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE imum rate-of-return requirements. Since no regulations have been issued for the minimum rate-of-return guarantee, this provision is not applied and, consequently, the reserves are not invested in a diversified portfolio. During the last few years the new pensions granted by ISS have been much higher than the old pensions. This has led to an increase in the average pension from 1.53 minimum wages in 1997 to 1.67 in 2000. The explanation could be that the new retirees have contributed longer than previous retirees, that the changes in the indexation of the pension base are starting to have an impact, or that the process of granting pension benefits has changed. This is happening while there are still 44,000 men and 39,000 women that will retire with less then 1,000 weeks of contributions, the requirement for the younger cohorts of workers. f Private Pension Fund Managers The main risks for the government's Treasury arising from the DC regime and the individual accounts are the generous minimum pension guarantee and the option of switching back to ISS. Workers will switch back to ISS if the pension they get there is higher then the one they would get from their savings and, therefore, they would necessarily get a subsidized pension. By definition, if the capital accumulated by the worker is not enough to buy an annuity equal to the pension provided by ISS, ISS will be subsidizing the pension of this worker. The switchback option has to be elim- inated-at least the option of returning to ISS-because it is a large contingent lia- bility and because the government is already providing a minimum pension guar- antee. Although the problem is clear, it is very difficult to obtain a reliable estimate of the cost of this switching option. Similarly, while it is simple to estimate the cost of the minimum pension guarantee for an individual worker, the total cost of the minimum pension guarantee is difficult to estimate and very sensitive to the model and the assumptions. For example, even when there is agreement about the number of contributions made over the worker's life, the distribution of these contributions has an important impact on the cost of the guarantee. g) Transition Regimes All female public sector workers older than 35 years of age and all male public sec- tor workers older than 40 years of age or with more than 15 years of service can retire under their old pension regime. This right was granted until 2014 because it was assumed that by then all of these workers would have retired. In 1995, there were about 935,000 government employees in Colombia, (excluding those employed in municipalities that are not department capitals), of which 483,000, or 52 percent, were members of the unreformed pension funds By 1999 public sector employment had grown to 1,194,000, and there were still 467,000 active workers in the unreformed pension funds for public sector workers. The number of pen- sioners from these funds had grown to 334,000 in 1999. PENSION REFORM 291 None of the national pension regimes-which number around 55-is allowed to receive new members. However, it appears that the number of members in these regimes is not falling as fast as the number of new pensioners and the number of workers transferring out of the regimes would indicate. This could lead to a transi- tion problem In 2014, there could be public sector workers that might receive dras- tically reduced pension benefits just because they get to retire a month after their colleagues. The perception of an inequity could lead to the introduction of a new transition to phase in the new parameters for this group of workers at an even later date It would be fair and necessary to revise the transition and phase it in so that it is completed by 2014 at the latest for all workers in the transition. Among the public pension regimes that continue to operate are CAJANAL and FONPRECON. CAJANAL has 17 different regimes for public sector employees. The initial reform target was that all CAJANAL functions would have been taken over by ISS in 1995. However, ISS declared itself incapable of absorbing them. CAJANAL has assumed an important role in the transition by assisting in the calcu- lation and recognition of pensions for public sector workers. However, this role and the fact that it is still collecting contributions for its active members are at odds with the legislation. The pension liability for current pensioners of CAJANAL is about 14 percent of GDP. FONPRECON is for members and employees of Congress. This regime was supposed to be closed to new entrants by Law 100, but it is still accept- ing new members and has exempted itself from several of the rules that apply to the pension system. Most noticeably, FONPRECON has exempted itself from the max- imum pension limit that is equal to 20 minimum wages (about US$2,400). The pen- sion liability for current pensioners of FONPRECON is about 0.9 percent of GDP. In 1996 about 250,000 active workers and pensioners belonged to the system of territorial pension regimes, which provide generous benefits with minimal or no contributions. Just a few territorial pension regimes have active members anymore; now they just recognize and pay pensions. They have many data problems, and it may prove difficult to establish the work history of their members. To issue the pen- sion recognition bonds for the previous members of these regimes, the work history needs to be reliable Very little is known of the aggregate pension expenditures and obligations of these regimes. However, the average pension of regimes which were liquidated and transferred to ISS (111,000 workers) and to the Fondo de Pensiones Territoriales (National Fund for Regional Pensions, FONPET) (62,000) was 3.4 minimum wages in 1996-equivalent to more than twice the average ISS pension. The Superintendency of Pensions is supervising the liquidation of insolvent territo- rial regimes. Of the 1,050 regional funds and calas counted, only about 252 are properly established funds; the rest were just accounts kept in the regional entity. Of these, 79 are under liquidation and their assets and liabilities are being transferred to a regional entity that will carry out the payment of pensions and pension recog- nition bonds. All territorial regimes are insolvent; only the ones of Annoqufa and the University of Cauca have declared themselves solvent and are allowed to oper- ate. There is no deadline to liquidate the insolvent regional funds and cajas. 292 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE The central government has been rightly reluctant to intervene in the almost 1,000 regional pension funds. They have taken this position to reduce the probability of hav- ing to assume the funds' liabilities. However, because these funds are insolvent, the gov- ernment is at risk of assuming these liabilities anyway. In the meantime, these funds are not properly winding down their activities, are not providing the information about the labor history of their members, are granting many unjustified pensions, and are not accumulating adequate reserves to cover their liabilities. A number of specialized funds have been set up to cover the outstanding liabilities, but they are still very small. The supervisors are aware of many problems but do not have the authority to act on this information; they just note the status of things. It is necessary to designate an individ- ual or institution with the authority over these funds to stop the growth of pension lia- bilities and diminish the potential burden on the government's budget. To ensure that the territorial entities would be in a position to pay their pensions and redeem the pension recognition bonds, FONPET was created at the end of 1999. It will be funded by contributions from the territorial entities (including pri- vatization proceeds and share of specific taxes and revenue) and by a share of the transfers of the national government to these territorial entities, lottery income, pri- vatization proceeds, and other sources of funds. h) Exempted Regimes In Colombia there are over 300,000 public sector teachers, equivalent to 26 percent of public employees (see Table 3). Of these, about 258,000 are in the teachers pen- sion regime and the remainder are members of municipal regimes. The powerful teachers union has won favorable conditions both in terms of salary and pensions that go beyond the already favorable conditions (relative to Latin America) of pub- lic sector employees, and managed to keep their regime exempt from Law 1000. The usual justification for the exemption of teachers from the 1993 reform is that the more generous pensions are compensation for lower salaries. However, there is no evidence of a wage penalty for teachers. After adjusting for work hours, public sec- tor teachers earn more (about 25 percent more) than other comparable private sec- tor professionals. The benefits of the public teacher regime vary depending on the age of the teacher. Teachers that are now under age 39-112,932, or 37 percent of the total-qualify Table 3. Exempted Regimes Liabilities Aetive Workers Retirees (percent 1999 GDP) Public Teachers 303,000 61,848 19 0 Armed Forces & Police 220,000 93,300 39 0 ECOPETROL 6,000 7,947 2.4 Total 529,000 163,095 60.0 PENSION REFORM 293 for a retirement pension at age 55 for females and 60 for males, and can continue to work and draw a salary together with the pension. At age 65 they can draw an old- age pension that is calculated based on the salary of the last year. The replacement rate is 75 percent of this salary. Teachers age 39 to 48-125,592, or 42 percent of the total-will be able to draw a "grace pension" at age 50, and a retirement pension at age 55, while receiving their salary if they continue to work. At age 65 they get an old-age pension. Teachers age 49 to 65-64,701, or 21 percent of the total-get the same benefits as the previous group. During 1999 about 26,000 of these teachers received two pensions in addition to their salary. Forty thousand received a grace pen- sion and 34,790 received a retirement pension in addition to their salary Members of the armed and police forces regime contribute 8 percent of their salary without a contribution by the employer (the government). Police officers at the managerial level contribute only 6 percent (see Table 4). Pension benefits go from 50 percent of the salary base after 20 years of work to 95 percent of the salary base The salary base, in addition to the salary, includes other nonsalary income. In 1999, ECOPETROL was required to start building a reserve to cover its pen- sion liabilities. Thanks to high oil prices during 2000, ECOPETROL deposited COP2.7 billion, equivalent to about US$1.4 billion. It is expected that by 2007, ECOPETROL will have covered 70 percent of its pension liability. The question remains whether instead of diverting resources, which would otherwise go to the national Treasury, workers should make a contribution out of their salaries to finance their generous benefits. i) Collective Agreements Different sources of pension liabilities for the government are the collective or labor agreements in the public enterprises. These agreements grant additional or comple- mentary benefits to the workers of the enterprise, as is the case with ISS workers. The 25,000 regular ISS workers (there are 20,000 additional workers with a differ- ent contract and not the same rights) have negotiated a labor agreement that pro- vides them with extraordinary pension benefits. In summary, through these benefits they can retire five years earlier then regular workers and receive a pension equal to their last salary. Currently, ISS spends as much on its active workers as on the pen- Table 4. Contributions to Exempted Pension Regimes (as a percentage of salary) Worker Employer (percent) (percent) Regular Worker 3.375 10 125 Armed Forces & Police 8 0 0 Police-Managers 6 0 0 Teachers 0 3.0 ECOPETROL 0 0 294 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE sion benefits of its own retired workers. It is problematic that ISS is financing these benefits through revenue from the contributions of all workers instead of income generated by ISS itself. In addition, ISS is not prefunding the liabilities it is accru- ing with active workers. ISS's pension liabilities are valued at about US$2 billion, with no relief in sight because the collective agreement requires that every time a public agency negotiates a more favorable agreement, this new agreement becomes retroactively valid for ISS workers. The burden of benefit payments became so large that it endangered ISS's ability to operate and pay salaries. To avoid a crisis, workers agreed to a temporary reduction in benefits in October 2001. The government as owner of the public enterprises is responsible for the liabili- ties of the enterprises toward their employees if the enterprises cannot pay them. The liabilities arising out of all the different collective agreements have not been fully estimated, but are about US$10 billion. j) Pension Recognition Bonds Colombia has been experiencing problems calculating and issuing recognition bonds for past contributions. In contrast to other Latin American countries that have reformed their pensioni systems, Colombia is faced with a special difficulty due to its extremely fragmented system for public sector employees and the fiscal decen- tralization of the country. Unlike Chile, which unified the different systems under an umbrella organization, Colombia has kept all of its different public pension insti- tutions, with the result that no one institution is solely responsible for issuing all the bonds. While those bonds for which the central government is responsible are being issued without major delays, the regional governments are still not fully complying with their obligation to issue bonds for affiliates switching from regional public sec- tor funds to the new DC regime. Problems also arise with the recognition bonds of regional pension funds that are due to workers who switch to the public ISS system as more and more regional public funds are closed. The option of switching back and forth between the public and private systems will further complicate manage- ment of recognition bonds. The bonds are cashed in and the proceeds go to the pub- lic system when affiliates that switched first from the public to the private system return to the public regime again. IV. Fiscal Requirements of the Current System a) The Model The model used in this Chapter to evaluate the state of the pension system and the impact of the reforms was built by the Departamento Nacional de Planeacidn (National Department of Planning, DNP). It is an actuarial/financial model where macroeco- nomic variables are exogenous. It covers the following entities: (a) ISS, (b) AFPs, (c) PENSION REFORM 295 funds for public sector workers at the national level, (d) funds for public sector work- ers at the regional level, (e) the teachers fund, and (f) the regime for armed and police forces. The time horizon of the model is 2000-50, and it does not include financing of deficits and the interest cost of the government's debt issued to cover the deficits. A key issue concerning the information sources is that the information about the members of the pension funds for public sector workers does not come from the accounts and records of the funds themselves, but has been drawn from the House- hold Survey and from the Ministry of Finance. A key assumption is that there is no new entry into the DB regime. The assump- tion is less drastic than it appears in the reform scenario because the minimum requirement of the new parameters for new members is that contributions and ben- efits are actuarially balanced; that is, that they do not increase the liabilities of the system. In the no-reform scenario it leads to an underestimation of net pension lia- bilities. Other Important assumptions are that coverage remains constant, GDP grows at 4.5 percent, the discount rate is 4.5 percent, private wages grow at 1 per- cent, and the return on investments is 6 percent, all in real terms. The ISS actuarial unit has also produced a model, but it covers only ISS and not the other regimes. The results of the ISS model are more optimistic than the results of the DNP model in the sense that they show a lower IPD, and financial flows of ISS improve faster in response to changes in parameters. This Chapter will use only the DNP model because it is more comprehensive than the ISS model. b) Pension Liabilities The fiscal requirements and future cash flows due to pension reform are difficult to estimate in Colombia because cost projections are complicated by the fact that affil- iates may switch between the new and the old systems. In addition, the data for many of the regional pension funds have not been processed and it is not clear who will have to cover their unfunded liabilities, and data for funds for public sector workers at the national level are still incomplete. The fragmentation of the pension system results in an inability to track contrib- utors and pensioners. This inability to track all the information could be driving the increase in pension liabilities because new liabilities are being discovered, and the large number of regimes makes it easy to create new unjustified liabilities. Net debt position of the pension system was estimated at 140 percent of GDP in 1997, and now it is estimated at almost two times GDP (see Table 5). Defaulting at least par- tially on these liabilities is not only socially and politically difficult, but also uncon- stitutional; Art. 350 of the Constitution states "public social spending will have pri- ority over any other spending." Exceptions, lack of ad)ustments, extension of regimes, imbalance of benefits, implicit guarantees, weak incentives for reserve management, incentive bias toward short-term cash flows, minimum pensions, and recent court rulings expanding ben- efits have all led to growing pension liabilities. 296 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE Table 5. Pension Liabilities, by Regime Entity Percentage of GDP Year ISS 65.0 2000 Teachers 19 0 2000 Armed forces & police 39 0 2000 CAJANAL (only pensions) 14 0 1999 Communications 3 2 1997 Ports 3.0 1998 ECOPETROL 24 1997 ISS as employer 2 2 1999 Railroads (only pensions) 1.6 1999 Congress (only pensions) 0.9 1999 Others 64 1999 Regional* 31.0 1998 Regional universities 2 8 1999 Total 198.0 2000 * Includes estimate of pension bonds liability The national Treasury has been funding the deficits of the regimes for public sec- tor workers for a long time, but the pace is growing (see Table 6). In 1991 the trans- fers were 0.84 percent of GDP, and during 2000 they increased to 2.26 percent of GDP. The switch option between the parallel regimes frustrates efforts to quantify transition costs, raises the operational costs of both regimes, retards development of the private pillar, and renders the public pillar vulnerable to strategic abuse. Even after the 1993 reform, the parameters of the PAYG pillar remain overly gen- erous. Vesting requirements, the benefit formula, and the minimum pension guar- antee in the PAYG pillar have not been adjusted to corresponding parameters in the DC pillar. Figure 4 clearly makes the point that the Colombian pension system is not sus- tainable in its current form. The financing requirements of all DB regimes will grow for at least the next 20 years. Without reform the annual financing requirements will reach 6 percent of GDP by 2020. The most immediate problem arises from the old regimes for public sector workers, and is compounded by the problems of the fund for teachers. In the longer term, ISS grows to become the largest problem. Table 6. National Government Pension Payments (as percent of GDP) Year 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 0 84 0.94 1 01 1.16 1 34 1 62 1.66 1 80 2 04 2.26 0 0 0 - - 200w \ E ) [ __I I 2002 2004I 2006- C. oa2008_ \ 11 i O 2010 t 2012 2014~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 c' 202X- ! I . 2o 016 r 2020 2020 2 i i i20244- / 2028 2032 2034~~~~~~~~~~~~~~~~~~~~ ~- 2036 2028 - 2030 I 0 42032 I H20344 2040 2042 2044 /6Z WNO:a31 NOISNIdI 298 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE V. Reform Requirements a) Overall Strategy Colombia's IPD is at the high end of the international experience and similar to that of countries with mature populations and almost universal pension system coverage (see Table 7). As in those countries, the main source of problems has been the defined benefit regime. Among developing countries, Colombia's IPD is compara- ble to that of Eastern European countries that have much older populations than Colombia (see Table 8). Almost all countries with high IPD have taken or are taking action to reduce it. Even developed countries like Italy have taken radical action instead of a slow approach. Italy moved toward individual pension accounts and cut its IPD from more than six times GDP to its current level of about three times GDP. The options for reducing the IPD are basically three: increase revenue, reduce benefits, and take fiscal actions. However, the possible combinations of these options are multiple, and no country has replicated the reform of another country. A guiding principle in most of these reforms has been that at least the new cohorts entering the pension system will be actuarially balanced and, therefore, they will not be the source of additional pension liabilities. The inclusion of the younger cohorts in addition to the new cohorts in the pension system reform also helps to reduce the IPD. Experi- ence has shown that reductions often are not permanent, and reforms also now include an element of automatic adjustment of benefits to changes in economic and/or demographic conditions. Individual savings accounts have this feature, and in different forms are being introduced in all reforming countries. Colombia's new pension reform requires three elements: (a) reform of the pen- sion system itself; (b) identification of financing options for the accumulated liabil- ities; and (c) additional reforms that make the pension reform more acceptable and sustainable, so that it involves more then just a cut in benefits. The following pro- posals concentrate on the first element. The contribution to the second element- financing options-is limited to proposals to reduce the liabilities toward active Table 7. Estimates of Implicit Pension Debt and General Government Debt, Selected OECD Countries (percent of GDP) Country OECD IMF General Gross Debt Canada 121 94 96 France 216 265 48 Italy 242 357 129 Japan 162 166 83 United Kingdom 156 117 46 United States 113 106 69 West Germany 157 221 50 PENSION REFORM 299 Table 8. Implicit Pension Debt Estimates, and Pension and Public Debt Data (percent of GDP) IPD Public Pension Fund Pension Discount Rate: Discount Rate: Country Debt Reserves Expenditure 4 Percent 5 Percent Brazil 0 7 330 275 Costa Rica 3 6 2 121 97 Croatia 25 0 11 201 175 Ecuador 0 1 63 51 Estonia 3 0 9 189 163 Korea 9 10 0 33 26 Kyrgyz Rep. 0 7 185 154 Lithuania . 0 7 155 134 Malta 38 0 5 234 194 Mauritius 35 16 3 47 42 Moldova 0 8 159 136 Morocco 70 2 1 32 26 Nicaragua .. 3 2 131 104 Philippines 53 4 1 107 81 Poland 42 0 12 261 220 Romania 0 6 256 214 Senegal 1 2 75 65 Turkey 45 0 5 146 123 Ukraine 0 9 328 279 Uruguay 19 0 14 222 193 = Not included Source Author's calculations Public debt figuLres from Government Finance Statistics Yearbook (IMF 1999) workers by, for example, reducing the length of the period during which the worker receives his or her benefits. The only recommendation regarding the third ele- ment-additional reforms-is to redirect the additional contribution of workers earning more than four minimum wages from the current subsidy to contributions to the support of the very poor old. b) Reform Options The reform proposed in this Chapter is along the lines of the current dialogue in Colombia, but is different and goes beyond the reform proposed by the draft law before Congress. It maintains the current situation of concurrent DB and DC regimes, and aims to address the fiscal problem and to be fair to all parties. There has been a convergence in the reform proposals of different parties, and several of these proposals satisfy the target of not creating additional net pension lia- bilities by the new cohorts entering the pension system. There is agreement on clos- ing entry to all regimes except to ISS, AFPs, and the armed forces to improve the 300 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE equity of the pension system and to lower the long-term deficits of the system. In additlon, some proposals result in very low liabilities for those cohorts that are fully subject to the reforms. An example of such a proposal is a reform that changes the current system by introducing retirement ages of 60 for women and 65 for men, a replacement rate of 50 percent after 1,300 weeks of contributions, and increasing the contribution rate by 1 percentage point. To explore whether such a system is in balance, a simulation with a cohort of 1,000 representative women and 1,000 representative men is run to check whether the reserves of such a system last until the last member of this cohort dies Figure 5 shows that a system with these or similar parameters is bal- anced in the long run; that is, the reserves are drained about the time the last mem- ber of this cohort dies. This reform proposal is not only balanced for future genera- tions, but if implemented together with an acceleration of the transition of Law 100, it could reduce the IPD by 40 to 50 percent of GDP. The reduction in the IPD is borne mostly by the younger generations. In a sim- ulation of the impact of the reform, together with an acceleration of the transition that leaves untouched men that in April 1994 were age 50 or older and women that were age 45 or older, that introduces a faster transition for men age 42 to 49 and women age 36 to 44, and fully introduces the reformed system for those men age 41 and women age 35 or younger, the impact on the IPD of ISS and the national pension funds for public sector workers is the following: * As expected, the IPD for workers that are not touched by the reform remains constant at 58 percent of GDP. * For workers that are subjected only gradually to the reform, the IPD drops by almost half, from about 19.5 percent to 10 percent of GDP. * For workers that are fully subject to the reform, the IPD drops from about 41 percent of GDP to 4 percent of GDP. Leaving older workers and pensioners untouched by the reform also has the effect of delaying any major improvement in the flows of the pension system. The impact on the flows becomes significant about only eight years after the reform. Only then will a critical mass of workers be in a position where they continue to contribute to the system instead of collecting benefits. Changing the transition as defined by Law 100 is a critical element of the new reform Almost all the short-term impact of the reform on the flows of the system will result from an acceleration of the transition. In addition, because most of the beneficiaries of overly generous benefits are in this transition, the acceleration of the transition not only has a significant impact on the liabilities of the system, but also contributes to the fairness of the pension system. Colombia's pension system has very large fiscal problems and needs all the rev- enue it can collect to address them. However, the proposed increase in the contri- bution rate is problematic. This Chapter has shown that Colombia already has one 0 o o 0 0 0 o 0 0 0 X 0 1 999- 2005- 2008- 2o8 I 2014 - 2017- 2020 -. 2023 - 2026- 2029 - 2032 -. 2035 -b 2038 - 2041 - 2044- 2047- 2050 - 0 2053- 2056- 2059- 2062- 2065- 2068- 2074 2077 2080 _ _ _ _ _ _ _ _ _ _ _ _ _ IK V%-O N)O13 NOINNi1d 302 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE of the highest social security contribution rates in Latin America, and already has nonsalary labor costs of 66 percent of the salary. While it could be argued that an additional percentage point would not have a significant impact when the totals are already so high, there are strong indications that the trend to informality is growing. Therefore, there is a risk that additional labor taxes might actually reduce total rev- enue and further reduce the number of contributors and coverage. It would be worthwhile to explore the option of raising revenue through fiscal measures, includ- ing an increase in the value added tax. There are many different parametric reforms that can achieve the same objective of eliminating the creation of new liabilities while reducing existing liabilities, with- out imposing unacceptable hardship on pension system members. Table 9 presents examples of what would happen to the implicit pension debt of ISS if some para- meters were changed one at a time. From these examples it becomes clear that the biggest reductions can be achieved by increasing the retirement age. An increase of one year in the retirement age during 2002 reduces ISS debt by almost 12 percent of GDP, but doing it in five years still has a large impact. The timing of the reform becomes critical for the decision of increas- ing the number of weeks of contributions required to retire. For ISS a smaller but fast increase achieves the same impact as a much larger increase three or five years later. c) Specific Reforms IPD reduction is not the only objective of the reform. One of the main objectives of the pension reform is to unify workers' pension rights and obligations, regardless of the regime they are under. This objective is certainly achievable for new workers. However, this is more difficult for those workers that are currently under a transi- tion regime, or are or will be in an exempted regime. By 2014, the problem could have been resolved for those in a transition regime, but not for the other cases. Table 9. Impact of Individual Parameter Changes on the ISS Deficit IPD as IPD Reduction as Reform Percent of GDP Percent of GDP Current ISS IPD 64.97 0 Increase age I year in 2002 53.02 11 95 1 year in 2007 55.78 9.22 Increase number of weekly contributions to. 50 in 2002 63.2 1 77 per year up to 150 63.87 1.1 30 per year up to 300 63.27 1.7 Reduce replacement rate at 1,000 weeks of contribution by 1 percentage point 63 35 1 62 Increase contribution by I percentage point 62 96 2.01 PENSION REFORM 303 To respect the acquired rights of workers and make the system fairer, a promls- ing approach would be for all workers in these regimes to start paying an additional contribution into an individual savings account for the additional benefits they will receive This additional contribution would be equivalent to the actuarially fair con- tribution required for a new entrant to the regime to get the additional benefits of the regime. Workers who retire under their current regime would transfer the funds in their individual accounts to the entity that pays their pension In this way their acquilred pension rights are respected, but there is a cost to early retirement or a higher pension. Workers could choose to continue to work and to retire under the new general regime instead of their special regime. In this case the worker could use the funds he or she saved for an increased pension, for withdrawals until he or she reaches the pension age required by the new regime, or could even receive part of it as a lump sum. Workers (particularly younger ones) could also choose not to make the additional contribution and just be members of the general pension regime. A similar approach could be followed for collective agreements. Based on Law 100, regulations could be issued that requlre establishing funded trusts for all future benefits by all pLiblic enterprises, with extraordinary pension benefits arising out of collective agreements. Parties could still freely negotiate the agreements, but enterprises would have to deposit the actuarially required funds in an external trust and finance them out of the current budget. The right of workers to negotiate labor agreements or collective agreements would be ulltouched; however, the actual cost of the demands for higher pension benefits would become more explicit, and man- agers of public enterprises would have to immediately face the consequences of their decisions instead of leaving the problem to future managers or to the govern- ment's Treasury. As noted, the level of Miiimum Pension Guarantee is high and the guarantee is costly Although difficult to estimate, expected liabilities arising from it could be about 5 percent of GDP Just for the capitalized regime. In ISS the average salary of its members is somewhat higher than in the AFPs, but required contribution periods are shorter Liabilities arising out of the guarantee in ISS could be similar, and the total cost of this guarantee could reach 10 percent of GDP. Changes to the guaran- tee are constrained by a Constitutional Court ruling. However, the law reforming the pension system should try to find language acceptable to the Court that allows for a minimum pension that is equal to the minimum wage when granted, but is indexed to the consumer price index afterward. This would reduce the cost of the guarantee, but would also address the distortion that the current ruling creates in the market for annuities. In the DC system when a worker retires he or she buys an anluity For workers with pensions close to the minimum wage there is a high risk for the insur- ance company. If during the life of the annuity the minimum wage were increased above the value of the annuity, the insurance company would have to increase the value of the annuity It is not clear what the outcome of this situation would be, but insurance companies will certainly charge for this risk, resulting in lower pensions for workers. The same risk of an unexpected increase in the minimum pension exists for 304 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE disability and survivorship pensions. This risk increases insurance premia, and some estimates of this increase explain about a third of the premium by this risk. If a change in the Minimum Pension Guarantee is not possible, eligibility requirements, both in terms of age and number of contributions, should be increased. Requirements should be the same in the defined benefit and the capital- ized regimes. In the DB regime they need not be the same as for a regular pension because the guarantee represents an added benefit for which additional requirements are justified. For example, if a noncontributive benefit for those older than 67 years of age is instituted, the requirement for the minimum pension could be 67 years of age In this way, if the worker ends up not qualifying for even a pension, he of she could qualify immediately for old-age assistance. The additional 1-percentage-point contribution to subsidize contributions of the poor is not achieving its objectives. Some childcare providers (madres comunitarias) may be a positive exception. However, the subsidy scheme is not leading to a sub- stantial increase in the probability of subsidy beneficiaries receiving a pension. Pen- sion coverage in terms of the number of old people receiving a pension is more important than the number of contributors. Therefore, the additional contribution should be devoted to finance the old-age assistance mentioned above. Another alter- native is that the additional contribution be used to finance the minimum pension. The contributions of the madres comunitarias should be paid either by the commu- nity or by the Instituto Colombiano de Bienestar Familiar (ICBF). As mentioned, the institutional requirements for the implementation of Law 100 are complex. For example, the law envisioned that the insolvent national and terri- torial public pension funds (cajas) that were not transformed into a pension admin- istrator would send their affiliates to ISS, and pensions would be paid by FOPEP, FONPET, and territorial funds. Only 2 national funds and 12 regional funds have already been liquidated, and only 5 more at the national level will be liquidated in the near future. This is creating the potential for further increases in liabilities. Slow progress in building and documenting the work history of members of old regimes, and reduced capacity to adequately control recognition of new pensions granted under all the different regimes, potentially increases liabilities. Although the unification of the pension system will address many of the problems, in the meantime a Social Security Department in the Ministry of Finance has been cre- ated to defend the interests of the government Treasury during the process to avoid waste and fraud, and to accelerate implementation of the previous reform and, even- tually, the next reform. The new Department has expanded ongoing control efforts as evidenced by the number of inspections and number and amount of contested pen- sions. The Department should have sufficient authority and resources to build and secure databases required for recognition and recertification of pensions in national and territorial funds (cajas). The work of the Department would be facilitated if the discussion of the budget of an entity with a pension fund (caja) were linked to the sat- isfactory provision of information required to build work histories and the database for the recognition of pensions. PFNSION REFORM 305 Evasion of contributions and underreporting of income are growing problems in the pension system that have been worsened by the economic downturn. This is an important problem for ISS, but also affects the government through the minimum pension it has to pay for pensioners of the DC regime. If workers do not contribute enough, even high returns will not be enough to avoid this liability. In addition, other social services like health are affected and place demands on government resources. The most direct solutions to this problem are (a) improved tax adminis- tration, because the main force behind evasion is the attempt to avoid paying taxes; and (b) labor reform. However, other effective measures could be taken in the mean- time For example, the creation of a URC would improve the possibility of controls not only by allowing consistency checks, but also by forcing the different entities to improve their own databases. The URC could also pave the way for more efficient means to collect contributions. The international trend now is increased use of the Internet and electronic payment that allows the interested parties to directly control the collection instead of building large, centralized, costly agencies that lack the incentives to do adequate enforcement. Continuing to have a pension system that includes a DB regime run by ISS will require ISS to become a more dynamic and efficient organization. The mismanage- ment and rhe problems in the health branch are affecting ISS's pension administra- tion through a worsening of image, loss of members, and diversion of scarce man- agement capacity, and may lead to illegal cross-subsidies in addition to the current bias in the allocation of ISS's fixed costs. Ensuring the survival of the DB regime under ISS will require separating the four branches of ISS into four separate and independent businesses. The four separate businesses would be (a) health provision, (b) health insurance, (c) workers' compensation, and (d) pensions. Only more spe- cialized entities will be able to compete with the private sector and provide the serv- ices that the members of the social security system deserve and pay for Further dete- rioration of ISS services will lead to a further decline in coverage. d) Fiscal Impact of Pension Reform Figure 6 presents the impact of only one of the many different ways in which the pension system can be reformed. The impact on flows is large and the reduction of the IPD is about 50 percent of GDP. However, if the pension reform is compre- hensive, the figure could be underestimating the impact of the reform because it shows only the impact of the changes in parameters. The figure does not account for the impact of changes on incentives, administration, and controls. e) Pension Reform in 2002 Draft Law The draft law introduced to Congress in early 2002 includes some valuable elements that go in the right direction. Under many scenarios the new parameters could bal- ance the system for new entrants to the pension system. The draft law addresses a sig- 306 COLOMBIA THE ECONOMIC FOUNDAnON OF PEACE Figure 6. Total Current Deficit of the Pension System With and Without Reform, as a Percent of GDP 120 - Deficit Pension System 198% E Without Reform 10.0 ----..-O-_-___ __ ___ _______ - With Reform 80- 8.0 --- -- -- ----- 6.0 - ___ 4.0 _ - 2. Deficit Pension System = 149% 0 .0 -.I I I I I I I I I I I I I I I I I I I I C-, crz v- I GE _ ', N _ Of 'n t\ N _ -) tE t C n _ A Vr t C- - 3NNmr 0 N N N r N N mN r' ' rq m r nificant fraction of the pension problem. To complement this, the government plans to address in a separate reform the problems in the exempted and special regimes with a large fiscal impact in the short and medium term. Over the long run, 40 years, the pension system is likely to be more sustainable. The improvement will come at the cost of higher labor market distortions unless the government adopts the planned labor reform, and will affect workers' choice between the private and public pension scheme. Therefore, an additional pension reform will be required in the near fiuture The government's reading of the draft law is that the Minimum Pension Guar- antee Fund that wiii be created will cover the minimum pensions arising from old age, disability, and survivorship contingencies. That is, the government will cover the difference between the workers' funds and insurance coverage and the minimum pension level. Approval of this is very important because it removes the political risk caused by the level of the minimum pension from the calculation of insurance pre- miums and annuities. This should increase the lower pensions, reduce the insurance premium, and Improve the soundness of the insurance industry. The fiscal impact of the draft law is likely to be positive during the next 10 years. The improvement in the fiscal situation during the next two decades relies mostly PENSION REFORM 307 on the increase in contribution rates and on diverting workers to ISS. The increase in contribution rates provides new funds to ISS and other defined benefit schemes, and reduces the government liabilities arising from the Minimum Pension Guaran- tee. The impact could reach 0.1 percent of GDP per year during the next 15 years. However, the higher labor taxes could increase unemployment, reduce coverage, and affect revenues if not complemented by labor market reform. The draft law also implements the Old Age Poor Fund (OAPF) created by Law 100, but this will be a slow process because it will be subject to availability of funds and the sources iden- tified by the law are small. In terms of fiscal impact it should be negligible unless the courts transform this benefit into an entitlement. CONTENT OF THE DRAFT LAW. Changes before 2014: * Contributions increase by 1 percentage point when the law is approved, fol- lowed by increases of 0.25 percent per year during 2005-08. Increased con- tributions of workers in AFPs fund a newly created Minimum Pension Guar- antee Fund that will be managed by the Treasury. Increased contributions of workers in ISS fund payment of ongoing benefits. Comment- This will have a positive fiscal impact but there will be increased labor market distortions. An ongoing policy of high pension benefits even for new entrants to the labor force makes the higher contri- butions necessary to balance the system for new entrants Deeper cuts in benefits would allow lower contributions. Unless the draft law is revised to make clear that there will be a minimum pension guarantee for workers contributing to the private pension even in the event that the Minimum Pension Guarantee Fund manages to run out of funds, this will create a new bias against the privately managed pension system because low- income workers will prefer to join ISS to have the guarantee that is offered in ISS in any case. * The maximum charge for the AFP management fee and survivorship and dis- ability insurance is reduced from 3.5 percent of salary to 2.5 percent. The 1 percent difference goes to the Minimum Pension Guarantee Fund Comment: This provision reduces fiscal liability arising from the mini- mum pension guarantee, although there are expenses associated with this revenue However, it endangers the commercial viability of the private pen- sion fund managers because the potential reduction in insurance premiums arising from changes to the survivorship and disability insurance should be about 0.6 percent of salaries and will take time to materialize The pension fund managers will face losses until they achieve the efficiency gains and the premiums fall to their new level. It also amounts to higher labor taxes for workers in AFPs because the potential reduction in fees and insurance premium have already been captured by the government and will not go to the worker's account. 308 COLOMBIA: THE ECONOMIC FOUNDATION OF I)EACE * Access to benefits from survivorship and disability insurance now requires also having contributed 200 weeks during the last 5 years, in addition to the cur- rent 26 weeks during the last year * Comment: This is an important action that helps address the weakness of current legislation. * OAPF implementation is to provide income to those over age 65 that do not have a source of income. The benefit is to be about 40 percent of minimum wage, and about 500,000 beneficiaries are expected. * Comment: In strict terms this is not a pension issue but an old age poverty alleviation or social assistance measure. The level of new expen- ditures will depend on how OAPF is set up, how fast it becomes opera- tional, qualification criteria, and benefit levels. There is the risk of creat- ing a new entitlement in Colombia; until now there has been no right to this assistance, and the small assistance programs were restricted to the availability of transfers from the Treasury and of local funds. Interna- tional experience shows that this kind of program tends to have explosive growth. There is no good reason why the new expenditures should be financed out of labor taxes instead of general revenue. Implementation of the OAPF should be subject to the availability of resources in the Trea- sury and not subject to the introduction, increase, or earmarking of labor taxes. * Pensions will be taxed but at a lower rate than salaries to take into account that they will not be able to claim any tax deductions. After 5 years revenue from these taxes will help fund OAPF. * Comment: This is the correct action, but revenue from these taxes should not be earmarked to OAPE * An additional 1 percent contribution is extracted from those with income of 20 minimum wages or higher. Revenue from this contribution is earmarked for OAPF. * Comment: This is an additional labor tax that will further promote under- reporting of labor income and evasion through, for example, consulting contracts. * From the 4 percent contribution to compensation funds, I percent is diverted to OAPF. * Comment: It is positive that existing labor taxes are being reviewed and found to be in excess of what is needed to fund the activities for which they are earmarked. However, instead of using this reallocation to fund new expenditures, it would be better to cut labor taxes or use them to avoid increases in other labor taxes like pension contributions. * Formula and parameters of pension recognition bonds for those who join or joined AFPs are equalized to the bonds for those who join ISS. In addition, pension recognition bonds for workers with salaries above 10 minimum wages will be calculated using a cap of 10 minimum wages. PENSION REFORM 309 * Comment: Reduces government liabilities arising from the bonds. How- ever, it reduces the benefit of joining the privately managed pension regime for an important segment of the market of AFPs. * Switching between AFPs and ISS is allowed only during the first 10 years after first employment Current workers that have 8 or more years in the labor force will be able to switch one time during the two years after passage of the law. * Comment: This is an important action that limits government liabilities and reduces arbitrage among systems. It would have been more positive if frequency of changes had not been increased and the period during which switching is permitted were shorter. * New civil servants may only join ISS if they just joined the labor force or come from private employment, even if they already have an AFP account. * Comment- The requirement sets a bad precedent in terms of directing work- ers to pension regimes, and again reduces the market of the private system. Already workers with subsidized contributions may only join ISS, and civil servants have to enroll In the Fondo Nacional delAhorro for their severance payment accounts. The only objective of this requirement is to improve cash flow of ISS and reduce the size of the transfers required in the near term. Changes in and after 2014: * The retirement age increases to 63 for men and 58 for women in 2014 In 2020 retirement ages increase to 65 and 60. * Comment: This is an important and correct action that helps the sustain- ability of the system Introduction is slow and there is the risk that the increase of two years could be contested. A gradual and faster introduction would be an improvement. * Number of weekly contributions required to retire and to have access to retire- ment in ISS increases by 35 weeks per year during 2014-18 and by 40 weeks during 2019-20, to a total of 1,250. The replacement rate remains at 65 per- cent of the average salary of the last 10 years. * Comment: This is an important and correct action that helps the sustain- ability of the system. The impact of the reform is limited by introduction in the distant future. Another concern is that it seems to reconfirm the pol- icy of hligh benefits even if it requires high contributions. * Workers who reach retirement age but do not comply with the new require- ment of weeks of contributions and have more than 1,000 weeks of contri- butions will receive a reimbursemenit equal to their contributions with a real rate of return of 4 percent. Workers with less than 1,000 weeks receive the contributions they made without rate of return. * Commentr Reimbursement in monthly installments is a better alternative to limit the number of workers that receive a lump sum and then benefit from OAPF. * The right to a minimum pension in the AFPs is acquired with 1,350 weeks of contributions and the retirement age. 310 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE This IS a correct action that lowers government liabilities but reconfirms the policy of favoring ISS in the rules of the game. VI. Conclusion The government is carrying out a number of actions and even presented a new draft pension law for discussion in Congress in early 2002. For example, the URC is oper- ating and it includes all ISS contributors, the national pension information system is in place and includes the pensions of about 400 pension funds, all pensions of eight public pension funds at the national level are being paid by the single agency, and a consolidated estimate of all pension liabilities is updated quarterly. However, even if the draft law is passed, the outstanding agenda in the area of pension reform is large. It requires (a) alignment of the vesting and minimum benefit parameters between the two pension regimes, (b) indexing minimum pension guarantees to inflation instead of the minimum wage, (c) phased integration of special pension regimes (starting with the pension plan for teachers) into a single national system, (d) completion of the phasing out of insolvent institutions, (e) calculating and issu- ing of pension recognition bonds, (f) regulatory adjustments to improve the effi- ciency of the private pension fund management industry, and (g) measures to limit the subsidy to the retirement accounts of lower-income workers. Most likely not all elements of the draft law will be passed by Congress, and the outstanding agenda will need to include some other crucial reforms like eliminating the option to switch between systems every three years, and very likely measures to restore the balance in worker's choice between the public and the privately managed pension scheme will be required. An immediate action plan on pension reform for the first year of a new govern- ment should include the following: o The new Social Security Department in the Ministry of Finance should keep the pension liabilities information gathering and control activities ongoing and institutionalized. The Department should have an adequate structure and sufficient authority and resources to expand and more efficiently carry out the ongoing information and control activities, including building and securing databases required for recognition and recertification of pensions in national and territorial funds (cajas), monitoring capitalization plans of pension funds and FONPET, monitoring membership of pension funds, issuing general procedures and norms for recognition of pensions, and issuing norms to implement Law 100. In particular the Department should ensure that ongo- ing control efforts, like in FONCOLPUERTOS (the pension fund of port workers), are expanded. o The government should link discussion of the budgets of national and terri- torial public pension funds (cajas) and entities responsible for their liabilities PENSION REFORM 311 to provision of informationi required to build work histories and databases for recognition of pensions. * Deadlines should be defined for the liquidation of all insolvent regional pen- sion funds and for the transfer of all remaining national public pension regimes into a single entity. * Pension Revisions Law should be approved that will allow for the suspension of pensions granted illegally, even during ongoing court proceedings. * A national pension information system should be created that includes infor- mation about the regime, years of service, entity, and so forth, for each pen- sion granted. * Regulations should be issued to require the establishment of funded trusts for all future benefits by all public enterprises, with extraordinary pension bene- fits arising out of collective agreements; parties could still freely negotiate the agreements, but enterprises would have to deposit the actuarially required funds in an external trust and finance them out of the current budget * At least one of the exempted regimes should be reformed in a way that at least brings it closer to the general pension regime, particularly by establishing indi- vidual savings accounts and a close link between contributions and benefits, reducing the flow of required transfers in the medium term, and eliminating them over the long term. Benefits should be equivalent to those in the regimes managed by ISS and the AFPs, and any benefits above those should be funded by additional contributions that are actuarially fair at least for new entrants. * The government should prepare an updated comprehensive pension reform proposal that includes the results of ongoing consensus-building efforts and fully considers the availability of new data. 10 Housing Finance This Chapter was written by Loic Chiquier. I. Executive Summary Despite several important policy and regulatory measures taken by the government of Colombia, the mortgage lending industry has not yet been restored to soundness. This could soon lead to further severe impacts on the financial system, the construc- tion sector, labor markets, economic growth, and the social protection of households. The ex-savings and loan institutions remain highly vulnerable to both market and credit risks, particularly due to a contaminating culture of nonpayment Imme- diate priorities are facilitating the issuance of mortgage securities and restoring mort- gage collateral strength, while introducinig consumer protection regulations. II. Genesis and Evolution of a Crisis Since 1972, specialized savings and loan corporations, called Corporaciones deAhorro y Vivienda (CAV), had dominated construction and mortgage markets, with the privilege of providing more affordable mortgage loans and attractive savings denom- inated in a constant purchasing power index (UPAC). Only since 1990 have com- mercial banks been allowed to compete by providing mortgage loans, and only since 1993 have they been allowed to provide UPAC-remunerated deposits. This liberal- ization has increased the cost of CAV funds, because UPAC-savings became less attractive than other products, such as CDs.i The UPAC formula has been changed several times to reflect that reality. Since 1994, it has been linked to Deposit Rate 1 Consequently, savings accounts in UPACs declined from 73 percent of total deposits in 1993 to 32 percent in 1998 313 314 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Index (DTF) rates (average cost of 90-to-180-day term deposits in the market), and in 1999 it equaled 0.74 DTF. On top of that, there was a boom in poorly designed credit products between 1995 and 1997, when the level of nonperforming loans was already rising and hous- ing prices started to decline for upper segments.2 A steady deterioration of the port- folio was visible before the brutal interest rate shock in 1998. Some amortization schemes permitted an excessive capitalization of interests for the benefit of temporary payment relief.3 As the nominal outstanding debt grew, CAVs also faced growing liquidity issues. When market real rates jumped to unbearable levels in 1998, both nominal and real mortgage rates reflected this rise with no moderation,4 as margins were also steep, with very adverse impacts on credit repayments. This was amplified by rising unemployment-20 percent-and the bursting of the housing price bub- ble. The CAV industry faced a severe solvency and liquidity crisis. Even higher levels of nonperforming loans-30 to 40 percent-were recorded as loans to developers. The impact was dramatic on insolvent households and financial institutions, and on the rapidly declining construction industry, with negative impact on low-qualified labor markets. The wealth and social protection of households was equally affected. III. Government Packages of Reforms In 1999 the Constitutional Court reacted to this situation by ruling that the drift of UPAC from its initial purpose (constant purchasing power index) was illegal, and that UPAC portfolios should be converted into inflation-based Unidad'de Valor Real (UVR), at a fixed real rate (that is, no capitalization of interest in real terms). By the end of 1999, the residential mortgage debt represented a large proportion-31 percent-of the total credit of the banking system, and was largely concentrated on the CAVs. The government then passed Housing Law 546, on 23 December 1999, which converted the existing mortgage debt and accordingly wrote off part of the debt (about US$1.2 billion, or 15 percent of the overall debt) funded by taxed financial institutions. The Housing Law introduced broader reforms of the whole housing finance system, including: o The conversion of CAVs into banks regulated and supervised as such5 2. In real terms, -40 percent between 1994 and 1999 for units exceeding $64,000 in Bogota. Market values of social housing units depreciated less. 3 Some schemes allowed debtors to pay 10 percent out of a 35 percent nominal rate. The resulting capitalization expanded mortgage debt, but subsequent debt repayments had to increase accordingly, later rising to unbearable levels. 4. Nominal mortgage rates reached 50 percent in July 1998, and real rates reached 28 per- cent in October 1998. 5 Except for a few transitory measures for which a three-year grace period was granted to make arrangements HousiNG FINANCE 315 * The standardization of mortgage loans into a few eligible LVR-denominated schemes, excluding any further capitalization of interests * A new social housing policy accompanied by various upfront and tax-related subsidies, and the requirement that banks dedicate at least a quarter of their mortgage production for social housing (Vivienda de Interes Social, VIS) loans * New secondary mortgage market instruments and institutions (mostly securi- tization agencies and mortgage bonds). Because the first package of relief measures designed in 1998 produced no visi- ble progress, the government more actively pursued its housing finance policy by improving prudential regulations of mortgage markets, and creating the following special support tools: * Purchase from restructured banks of nonperforming loans * Recapitalization credit lines * A special fund against market risks * A guarantee fund for social mortgage securities * Income tax exemptions for borrowers, lenders, and investors, complemented by upfront subsidies for social housing. But the main challenges ahead remain * Preventing further deterioration of primary mortgage markets * Developing the long-awaited capital market funding of mortgage markets * Improving the housing finance infrastructure, starting with real collateral * Developing a nondistorting, targeted, and efficient social housing policy (a subject not treated in this Chapter). IV. Main Sectoral Issues a) A Contaminating Deterioration of the Mortgage Portfolio The mortgage market remains concentrated at 95 percent in seven banks, where mort- gage loans represent the majority of their assets, which increases their vulnerability. The largest mortgage lenders are Davivienda and Granahorrar,6 followed by Conavi. The WVR conversion was expected to improve the portfolio quality, and in the following weeks the average nonperforming loan ratio for mortgage loans fell from 20 percent to 12 percent. But this level has since returned and stabilized at the high level of21.8 percent, as illustrated in Figure 1. 6 After being transferred to productive assets from the liquidated Banco Credito Hlpote- cario (BCH) (1 5 trillion peso mortgages) 316 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE I I \1,'AlN10-A to-d.?s IX _ I O-jef e~ ~ ~ ~~~~I fOX-XE-doS 'I !~~~~~~~~O-JLpv lO-uvf I' OO~~~~~00AON 00-d,,S 0 I' ~ ~~ ~ 00-in! 0 ~ ~~~~~~~~~~~~~~66-AoNj 66-dzS~ 0 86-Jflf~~~~~~~~~~ 866-En 6'6-uEf 0~~~~8-o cr1 r1I ~~~~~~~~~J - -6d, HOUSING FINANCr 317 Three years after the crisis, the mortgage industry has not exited the tunnel The main reasons are: * A continuing economic recession. * Further expectations of State bailouts, aggravated by further rulings of the Constitutional Court that have capped mortgage credit rates; and several measures in favor of borrowers against the lenders have fueled a culture of nonpayment among households. * Delays in restructuring two public banks (Bancafe and Granahorrar), which are capitalized and refinanced through the Fondo de Garantias de Institu- cones Financieras (FOGAFIN) or through its asset managenmenit company (CISA) by purchasing the defaulted mortgage loans.7 The hemorrhaging of housing loans turning into trouble has slowed down, but not stopped. Priva- tization and/or the portfolio resale are recommended. CISA needs to improve its policy of servicing banks in order to improve mortgage debt recovery A new and recent negative trend is the apparent contamination of private banks by the nonpayment culture (a consistent story across banks, including Davivienda), as illustrated in Figure 2, which shows rising default loans (the persistent economic recession-with an 18 percent unemployment rate-cannot itself explain this rise). b) A Contracting Mortgage Stock and Production of New Loans Outstanding mortgage debt remained stable in real terms after 1997, before starting to contract after 1999. Before the UVR conversion, the mortgage stock was about 17 trillion pesos, compared to a reduced amount of 12.5 trillion pesos by Novem- ber 2001. This means a sharp contraction expressed as a percentage of GDP from 11 percent to 5.5 percent in four years. The contraction is less of a problem than the collapsed new production, except that it also implies a loss of banking profitability and less potential for secondary mortgage markets The contraction is due to the joint effects of: * The amortization process of the UVR-denominated stock; most loans have a residual maturity of less than seven years. * A rundown production of new loans, because of a constrained housing demand in a recession economy, collateral fear of lenders fueled by Court rul- ings, a limited credit affordability due to high market rates and noncompeti- tive margins,8 and liquidity constraints of the ex-CAV banks (see Table 1). 7 D- and E-classified loans (more than six months late) for a total amount of 2 57 trillion pesos 8 Mostly due to nonperforming loans, weaker collateral, market risks, expensive origillation, and servicing functions 318 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE A\ ,' r~~~~~~~~~~o-daS 5 I \ 6o-inf IO-ul?f 00-AON c i -8d6-162- - - *%i I- -b ' Sg g A, 66, en PS S-X % ~~~~~~~66-uEf 5 I ~~~~ 8 6 - d z S _ _ _ _ _ _ _ _ _ _ _ 8 6 - u p f *~~~~~~~~~~~~~~a - C0 HOUSING FINANCE 319 Table 1. Recent Evolution of Mortgage Lending Yearly Production Credits (Thousand Billion Pesos) 1999 2000 2001 Mortgage loans 1 39 0 69 0 78 Total loans 36 73 37 3 38.16 Proportion mortgage (percent) 3 8 1 8 2 Mortgage percent GDP 0 93 0 34 0.34 Evolution mortgage (percent) -50 +13 On the positive side, the contraction has permitted the more intenisive provisioning of bad loans, and helped to restore the liquidity and capitalization of distressed CAVs converted into banks Housing market prices have also adjusted to the new economic condition, but urban prices depreciated in real deflated terms in 2001, with some adverse impacts on the wealth and social protection of households If the situation were to last longer, the construction sector and economic growth would be further affected. The renewed development of residential mortgage markets is needed because the real estate sector is showing some recent encouraging signs (higher number of con- struction licenises and real estate transactions during the second half of 2001). A mortgage production level of 0.34 percent of GDP is insufficient to finance the revival of the construction sector. The major impediment ahead is the increased sys- temic risk related to mortgage markets because of a growing culture of nonpayments. c) Legal, Regulatory, and Institutional Environments Pursuant to the Housing Law and reacting to the past abuses of capitalized mort- gage interests, further rulings of the Constitutional Court have weakened housing as collateral for the lending institutions in the present judiciary system by * Forbidding out-of-court foreclosure proceedings by fear of unprotected households versus predatory banks * Granting rights to borrowers of mortgage debt restructuring, open for two months at the beginning of every year, in case of repayment diffictilties according to "objective" conditions, as determined by the Banking Superin- tendency (Superintendencia Bancaria) * Capping the credit interest rate of mortgage loans at the "lowest" of compa- rable market rates (to be periodically adjusted by the Central Bank Council). These rulings have fueled legitimate fears on the part of banks and dampened their interest in resuming any significant production of housing loans. The high lending margins have not significantly declined because of the resulting perceived credit risk In turn, it is also likely to increase the costs of market mobilization of resources through mortgage securities. 320 COLOMBIA: THE ECONOMIC FOUNDATION OF PEACE Home ownership as a protection against poverty and as a worthy form of collateral has been hurt, as have building activities. The overreaction of the Court has produced effects that actually run contrary to its protective goals. The government should urgendy take measures to protect mortgage holders against predatory lending behav- lors and restore a sounder balance among judges, lenders, and borrowers. A national information campaign may be useful to announce a consistent package of measures. Debt restructuring should be limited to cases of an unpredictable loss of bor- rower income followed by some restored credit solvency. Bad faith borrowers should expect no bailout, but enforced foreclosure. Alternatives to foreclosure are to be explored, but not to the detriment of a foreclosure made unpractical through the judiciary process. Revised and acceptable forms of out-of-court foreclosure proce- dures should be worked out by the government in order to fully respect the rights of borrowers, and should be implemented with an adequate training program. The system of a capped interest rate for mortgage loans preferably would be revised in a manner compatible with Constitutional provisions to avoid predatory practices, but without necessarily being determined as of the lowest of other market rates. At least the adjusted level of the cap should not produce any retroactive effects on the existing portfolio. New regulations should be explained to better inform mortgage consumers about their rights and duties, including the importance of the mortgage pledge, the deliv- ery of simple, standardized, and comparative summary features of loan proposals, and some common contractual features. It could be developed through a formal reg- ulation (like in the United States) or through a self-regulatory code of conduct between lenders and borrowers (like in Europe). Lenders may even commit to fully exploring alternative solutions before moving to foreclosure proceedings, provided borrowers signal early any unexpected problem in ability to pay. Another destabilizing source of volatility in credit repayments comes from the cur- rent definition of the UVR index, which needs adjusting. It currently derives from the latest monthly inflation rate, and should be derived from the inflation cumulated for the last year, in order to reduce the very strong seasonal effects of the consumer price index in Colombia for the last five years. This would provide borrowers more stable monthly debt service in relation to their income, which in turn could help improve overall portfolio performance. Moreover, mortgage-backed securities (MBS) institutional investors also generally prefer securities indexed by a moving average inflation index (as shown by their preference for IPC-based rather than UVR-based Treasury notes), because of a more steady evolution of the net balances and a result- ing better liquidity through secondary markets. A study is needed to ascertain the legal feasibility of applying such change to the existing loan portfolio, with possible retroactive applications.9 The timing of any change should be decided in order to 9. Such as the possible retroactive effect on the existing portfolio of any change of the UVR index or the interest rate ceiling, or further debtor's procedural appeals in the courts, or any unpredictable further decision of the Constitutional Court. HOUSING FINANCE 321 avoid any brutal increase of repayments for debtors. Care should also be taken to avoid sending a wrong signal about the stability of a new housing finanice system, and to prevent simultaneous use of multiple definitions of UVR. The reform would need appropriate safeguards in order to avoid market confusion and disruption Overall, a strong package of several initiatives along the lines above will likely be needed to moderate the prevailing uncertainties and tensions in the housing finance sector and weaknesses in the mortgage markets The government could also consider a more proactive policy involving the creation of a special fund cofinanced by the State and the lenders (from their margins paid by borrowers and from part of the foreclosure proceedings) to reimburse a small part of the credit amortization of bor- rowers with good payment records. More credit competitiveness is also still needed in Colombia and could be achieved by encouraging (a) all banks-not only the erstwhile CAVs-to make loans for housing, (b) the issuance of mortgage bonds as a funding tool competing with the securitization company, (c) the modernization of mortgage servicing at banks, and (d) the unbundling of loan origination and loan servicing functions Other measures that could improve the overall infrastructure for housing credits include: * Housing appraisal rules could be better adjusted to comparative market trends * The credit reporting system should not be limited to current records * Reduced delays in property registration. PRUDENTIAL REGULATIONS AS APPLIED TO HOUSING FINANCE INSTITUTIONS AND ACTIVITIES. Since July 2000, strengthened loan-loss provisioning rules have been applied to banking activities related to housing finance. Most important, only a lim- ited and declining (over time) percentage of the estimated collateral value can now be deducted from nonperforming debt,'" in contrast to former full deduction of an artificially inflation-appreciated collateral An additional prudential regulation that has been tightened relates to provisions against the net exposure of banks to market and interest rate risks. This new assignment is particularly relevant for mortgage loans because of the interest rate index mismatch between the converted UVR-based long-term loans and the dominant DTF-based liabilities. The additional capital requirement is being phased in gradually starting in January 2002. Initial estimates by mid-2001 suggested a gap of Col$770 billion (of which about Col$230 billion is in private banks). Overall, banks have reacted positively to these measures through recapitalization efforts under the close scrutiny of the Banking Superintendency. As illustrated, loan- loss provisioning has improved considerably (from 11.4 percent to 36 percent), although it may not be sufficient. 10 Seventy percent until 18 montlhs of late payments, 50 percent until 24 monthls, 0 per- cent after two years 322 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE Several banks have been struggling to meet higher capital requirements, however, with an average ratio of 8.3 percent by January 2000, improving in October 2001 to between 10 percent (the minimum new requirement) and 15.6 percent (the high- est ratio recorded). At least two banks met the required level by borrowing special lines from FOGAFIN (see next section). These corrective measures were appropriate and efficient, but they cannot work as preventive tools to restore the overall soundness of the mortgage industry. FOGAFIN RECAPITALIZATION LINES. To help banks comply with their minimum 10 percent capital adequacy, since 2001 FOGAFIN has been offering some sub- ordinated credit lines, mostly for capitalization purposes. The provisions of this sec- ond-round recapiralization program were tailored to more readily meet the needs of institutions with a heavy volume of housing loans. The most attractive line offers lengthy terms (up to 9 years), including a four-year capital grace period, 2.5 years of capitalized interests, a moderate price (DTF + 3 percent), and guarantees limited to overcollateralized shares (133 percent) of the recapitalized bank. Given this low risk- adjusted rate, FOGAFIN should consider strengthening the collateral requirements under this program by requiring collateral with lower-risk assets. FONDO DE ESTABILIZACION DE LA CARTERA HIPOTECARIA AS A TRANSITION MEA- SURE. To facilitate prudent risk management by housing banks, the government also established an interest rate swap fund (a public hedge facility called the Fondo de Estabdlizacion de la Cartera Hipotecaria, FRECH). This new instrument is one of the special transition measures allowed under the new Housing Law, put in place by decree11 and administered by the Central Bank. It is designed to enable banks to hedge part of their market risks introduced by the conversion of UPAC into UVR- denominated loans. Although FRECH was a welcome initiative, care should be taken to make sure that the operations of the fund are financially balanced. The mechanisms of recip- rocal payments between banks and FRECH are triggered by the evolution of DTF above UVR compared to a fixed 4.9 to 7.5 percent collar,12 as applied to up to 40 percent of the eligible portfolio (outstanding balance up to 31 December 2000).I3 The option can be exercised at any time during four years, but this period may be extended by further four-year cycles up to a total of 12 years. These features appear reasonable enough to maintain the fund equilibrium, and thus avoid any further 11. Resolution 1163, 14June 2001; extended through Resolution 2868, 24 December 2001. 12. With a market DTF rate below the lower threshold, the bank pays FRECH the differ- ence. Between the extremities of the band, the fund is neutral Above the high level, the FRECH pays back the bank with the observed difference, without any limit. New loans originated after 2001 are not eligible. 13 Beyond that, the applied band is 5 9 to 7.5 percent, as an incentive for the bank to look for other solutions HOUSING FINANCi 323 State contingencies. But unless real short-term market rates are perceived to rise sig- nificantly, banks see no incentives for using this fund that could only hedge their catastrophic market risks. They will still face significant losses if DTF market rates rise moderately (so far DTF has stayed below 4 percent in real terms) The FRECH can (and should) therefore play only a limited hedging policy role. The contractual deadline was extended by another year (to 31 December 2002) A longer life period is recommended for the scheme, together with a requirement that the collar values'4 be regularly market-adjusted without forecasting an evolution of interest rates related to past monetary policy. GUARANTEE FUND FOR MORT-GAGE SECURIrrIES. As mandated by the Housing Law,'5 FOGAFIN is implementing a guarantee fund for timely payment-capital and inter- est-of eligible mortgage securities issued by banks, fiduciaries, and securitization com- panies, issued in the next two years for the sole purpose of funding Vivienda de Intere's Social (VIS) social housing credits. This initiative is expected to assimilate VIS securi- ties as public debt (at a lower price and "repo-able"), thus reducing the funding cost of VIS social loans (capped credit rate at UVR + 11 percent) and placing new mortgage securities amnong reluctant investors.'6 FOGAFIN has prepared preliminary standard- ized rules and actuarial estimations. International experience shows that advanced risk- monitoring procedures are needed'7 to avoid an inactive fund or further State losses. The proposed premium expressed as a yearly spread applied to the outstandinig debt is 1 percent for all eligible MBS and 0.4 percent for all eligible mortgage bonds, with eligible mortgage securities meeting the minimum conditions equivalent to a BBB rating. The fund has direct access to significant but limited budgetary support (US$200 million). The use of the external rating as a pricing factor was abandoned because of the difficulty on the part of relatively inexperienced rating companies in Colombian mortgage securities about mortgage markets (appraised housing market values, use of credit bureaus to screen credit applications, and so forth). There is another serious moral hazard issue, because issuers may challenge the rating agencies to provide overestimated ratings in order to pay cheaper premiums, but the credi- bility of the ratings agencies would then be less sanctioned by market investors, who would be repaid by the public fund anyway. Banks and securitization companies view the MBS premium as too expensive- above a 36-billion-peso threshold-implying a subsidy difference of 64 billion pesos, notwithstanding indirect subsidies related to the regulatory treatment of these securities. This real level of indirect subsidies is even higher if actuarial estimations 14 Established from a 1 0-year moving historical data average, whichi may not faithfully indi- cate forward rates, given the introduced changes in financial and monetary policy 15 Already two years after the law has been enacted. 16 Many nonperforming loans, depressed housilig prices, ineffective foreclosure procedures 17 As if combining products and skills from Ginnie Mae, the Federal Housing Administra- tion, and Fannie Mae in a more uncertain and risky context 324 COLOMB3A: THE ECONOMIC FOUNDATION OF PEACE are taking into account: (a) the moral hazard factor of a public fund within a mort- gage culture of nonpayment; and (b) the aggravated credit risk related to a deterio- rating legal, political, and macroeconomic environment. The premium could also be differentiated according to the quality of the port- folio (loan-to-value ratios, seasoning, enhancement, servicing standards) and the unsecured bank rating when applied to mortgage bonds. Otherwise, we may face a situation of a subsidized guarantee to a mediocre MBS, and an overpriced guaran- tee to a sound mortgage bond. For the following reasons, creating a separate legal structure may be preferable to just making it a department at FOGAFIN: • Such insurance activities should be regulated and supervised independently. o FOGAFIN acts as an intermediary agent of the nation and should be finan- cially insulated from the related risks of the Fund; a separate structure would reflect this separation role and facilitate the estimation of related budgetary liabilities. * The Fund may have to guarantee mortgage bonds issued by a bank owned by FOGAFIN. The Fund should also be constituted as if it were to be run for more than the scheduled two years. International experience suggests that a lobbying effort can be expected to result in the continuation of the program, particularly if subsidized. The Fund should invest in a sustainable structure, implying staffing and training, insti- tution building, investment, and outsourced experts for some functions (servicing contracts, selection and monitoring of risks with adequate incentives). More detailed servicing instructions, debt recovery procedures, pricing mechanisms, sanctions, and incentives are needed. The recent experience of the Hong Kong Mortgage Company represents a useful model and reference. There have been several requests to make the MBS instruments "repo-able" at the Central Bank. Given the reluctance of the Central Bank to go this route, an inter- mediate option might be to first swap the MBS instruments through the FRECH into Treasuries without haircuts, and then use them for "repo" operations with the Central Bank. This has the disadvantage, however, of deviating the FRECH from its principal original objective. SECONDARY MORTGAGE MARKETS. Securitization companies were expected to per- mit banks to gain access to capital market funding and, therefore, reduce their expo- sure to liquidity and market risks. Since the new vehicles and funding tools were cre- ated by the Housing Law, no MBS has been issued because of regulatory and operational delays, costly credit enhancements related to a vulnerable mortgage industry, high market interest rates, and a lack of benchmarks. On the other hand, 78 percent of the portfolio is still performing well, including many seasoned loans that resisted adverse stress conditions. They also meet securiti- HOUSING FINANCE 325 zation standards because their cash-flow pattern has been standardized by the Hous- ing Law. Recently several housing finance institutions (ex-CAVs) jointly established and capitalized a securitization company (Hitos) with the objective of accelerating the development of secondary mortgage markets. Hitos currently has about US$110 million in capitalization (including International Finance Corporatioi participa- tion) Its first MBS program, of Col$500 billion, is expected shortly, after the pur- chase of seasoned pools from sponsor banks and from FOGAFIN. The Superintendency of Securities (Superintendencia de Valores, SV) Imple- mented a regulation imposing a minimum capital of 40 billion pesos on securitiza- tion companies. This could represent a barrier against further competitioi, although it serves as a cushion to offset the current high level of market uncertainties and unknown risk exposure of securitization companies. The SV passed another decree in November 2001 regulating securitzation com- panies, mortgage-backed securities, and mortgage bonds (bonos hipotecarlos) SV has opted for an evolutionary strategy requiring a high level of disclosure of information to investors, and leaving flexibility to securitizatioi companies about their activities and resulting risk exposure." A small core team at the SV has been working inten- sively on these issues. The Ministry of Finance has also exempted securitizatioln com- panies permanently from the 0.2 percent financial tax and stamp duties, and for three years from the presumptive taxes. SV needs to expand its staffing and training by starting thorough oversight inspections, detailing its regulations and, if necessary, implementing prompt correc- tive actions. The regulated company also needs this information to adjust its own procedures More explicit capital adequacy rules should also relate to the net exposure to risks,'9 in order to avoid capital regulatory arbitrage between banks-regulated by the Banking Superintendency-and securitization companies-regulated by the SV. A combination of a ratio-based minimum and a structured approach to internal model evaluation may be usefully developed A number of additional improvements in the rules, regulations, and practices gov- erning home mortgage securiti7ation should be considered to help deepen this market: Securitization companies should also be authorized to issue mortgage bonds if they can be assimilated as "credit institutionis" holding mortgages on bal- ance sheets. 18 Both Collaterlized Mortgage Obligation (CMO) and pass-through mortgage securities are explicitly referred to through a draft regulation On the other hand, the drafted decree about securitization companies mentions that all investors of securities should be equally treated, this should be understood per specific class in the case of a CMO program 19 Which may vary according to diverse pool purchases, funded differently, with various enhancements 326 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE * The SV may want to explore the possibility of mixing eligible mortgage loans with public government debt in order to back MBS or mortgage bonds, for enhancement purposes. This would be useful for non-VIS securities that are ineligible for FOGAFIN guarantees. * Dematerialized MBS are automatically registered without preapproval as long as the prospectus meets the general standards. The specific nature of MBS would legitimize further exemptions from the information required according to the corporate governance section of the capital markets law (code of good conduct). * Despite the differences existing between mortgage securities, information for- mats should be standardized between mortgage-backed securities and mort- gage bonds (both about the corresponding identified mortgage pool and about the issuer and/or guarantor). * SV should also adjust its regulations with the revised Basle principles in mind. For example, the subordinated debt retained by the selling institution as a first loss exposure should be deduced from capital,20 whereas second-loss positions are subject to lower capitalization rules. Another reference is the Australian regulaton."21 A bank or securitization company purchasing back its own securities should not lose the tax exemption reserved to external investors and to VIS originators holding their loans. Otherwise, securitization companies may face difficulty in supporting the liquidity of their papers. The same issue exists for banks holding their own mort- gage bonds. Finally, institutional investors should be encouraged to invest in MBS and mort- gage bonds by: o Low asset risk weighting for banks holding well-rated MBS and mortgage bonds o Relaxed constraints from the synthetic portfolio of pension funds (short-term bias) o Investment ceilings for insurance companies and eligibility to technical reserves o Eligibility of MBS and bonds to the public social security and pension fund (ISS). UNDEVELOPED MORTGAGE BOND MARKETS. A parallel SV regulation allows the issuance of mortgage bonds (bonos hipotecarios), that is, bonds issued by credit insti- tutions using general mortgage pools as collateral. However, this potentially useful 20. Up to a ceiling corresponding to the capital charge as if the portfolio had not been securitized 21. Clear treatment of implicit recourse, and the differentiation of first- and second-loss exposures. HOUSING FINANCE 327 alternative instrument to deepen market resource mobilization has not received offi- cial encouragement, and remains untested. No mortgage bond has been issued The needed capital market funding excessively depends on the development of one sin- gle securitization company The legitimate goal of complementary and competing funding tools could not materialize. Yet, mortgage bonds represent a good funding option for relatively sound banks, as demonstrated in Chile. In Colombia, they would permit both lenders and investors to benefit from available income tax incen- tives similar to those for MBS. Ostensible reasons include the lack of UVR treasury benchmarks, and the distrust of investors moving toward a vulnerable mortgage sec- tor. But the SV regulation also raises several issues. * Existing mortgage pools are so far disregarded as eligible collateral. The imposed "perfect matching" rule between credits and bonds should be widened to accept a limited dose of overcollateralization and the refinancing eligibility of the existing stock. This would help to reduce the market risk exposure of banks, particularly because (a) the production of new eligible loans is limited; (b) their market exposure is significant, especially for VIS loans capped at WVR + 11 percent; and (c) the alternative securitization of their most profitable cred- its is not always the most attractive option. This change would be facilitated by a high degree of portfolio homogeny due to the UVR standardization, and a reasonable residual average maturity (six to seven years) that matches the expec- tations of investors (the term of new loans would be excessive) The SV men- tioned a possible change, but the market perception remains blurred. * The prepayment regime may be improved by proportional allocation rather than a lottery system. In addition, an issuing bank could be permitted to adjust the level of prepayment risk passed to investors, through overcollater- alization or by stripping different payment classes. An issuing bank should be authorized under certain conditions (including a mention in the prospectus) to buy back and trade its bond. * In the case of mortgage loan foreclosure, the accepted substitution by another eligible asset-as in most other countries-would represent an alternative viable solution to the imposed prepayment, in order to keep paying cash flows to investors. * The requirement of bond maturity exceeding the loan term-article 4-does not respect the principle of permanent matching between cover assets and bonds. * The pool of matching assets should be authorized to also include some gov- ernment debt in order to enhance the quality and rating of the mortgage bonds. * The tradable yearly coupon-article 6-should be optional rather than mandatory * The prospectus of a mortgage bond could be less demandinig of the issuer than for a corporate bond (remuneration of the Board, and so forth). 328 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE • The functions of the intervention agent and the investors' representative may usefully be merged within one single agent (this is cheaper). * The SV qualified only one amortization standard (UVR fixed payment) for mortgage bonds, whereas investors may prefer a menu of schemes pre- approved by the Banking Superintendency. * The bankruptcy privilege should be reconciled with the preparation of the close- ending bank resolution, should a large proportion of such high-quality assets be isolated to protect mortgage bonds, thus leaving depositors at higher risk. LIQUIDITY BACKSTOP FOR MORTGAGE-RELATED SECURITIES. In an effort to foster development of primary and secondary markets for MBS, the authorities have also been actively considering the establishment of an MBS "backstop facility" to help provide some minimum liquidity to the institutional investors of eligible mortgage securities. By providing a minimum liquidity, such a backstop facility could serve as a valuable instrument in helping MBS markets take off. However, to avoid poten- tial adverse effects, its design and application should be carefully circumscribed within pre-agreed criteria and parameters. Most important, the backstop should operate purely as a liquidity-enhancing mechanism, and avoid taking on any credit risks per se. Moreover, any true "mar- ket-maker" role should be avoided for such a facility at this stage, given the lack of reliable benchmarks that would be needed to establish strict market-maklng rules and the high level of intermediary risks. An alternative which has been suggested to increase liquidity in the MBS market is to make MBS instruments eligible for central bank "repo" operations under cer- tain conditions. However, this will require a major amendment of the Central Bank law (for example, along the lines adopted by the European Central Bank). Given that the Central Bank may not regard such a change as desirable (at least under the current conditions of volatility in mortgage markets), a temporary expedient would be to allow the FRECH mechanism to swap on a temporary and short-term basis eligible mortgage securities against Treasury securities with some adequate "haircuts" to take into account the associated risks. Separate accounting systems and stringent reserve rules should be used under such an approach to avoid any conflict with the main FRECH purpose Another variant is creating a public fund supporting some minimum demand for end investors, through periodic auctions. Its proceeds could be used to acquire and sell eligible mortgage securities on primary or secondary markets, or to refinance an issuer facing an exercised put option embedded in its mortgage security. The fund should operate according to tight eligibility rules, sufficient capitalization, and sun- set mechanisms It should trade a small portion of securities and adjust prices according to market with sufficient incentives for private agents to gradually replace it. The banks as issuers of their mortgage bonds should also be authorized to trade their papers. 11 Urban Development This Chapter was written by Thakoor Persaud and Alexandra Ortiz. Urban development is crucial to Colombia's goal of overall economic develop- ment. To a large extent, the country has been successful in exploiting the com- parative advantage of its cities as centers of commerce, finance, employment, pol- itics, and culture. Rapid urbanization, coupled with the economic crisis that has beset the country over the last five years, however, has outpaced the government's capacity to provide adequate services, and has brought about an increase in the level of urban poverty, both in relative and absolute terms. From 1996 to 2000, the incidence of poverty in Lirban areas went from 42 8 percent to 51 percent. In absolute terms, the number of urban pool increased from 12 million to 16 mil- lion during this period. By 2000, of the 25 million poor Colombians, 64 percent lived in urban areas. This Chapter presents the main issues accompanying the growth of increasingly complex urban areas in Colombia, and provides an overview of the ma)or challenges facing its urban sector' Unlike other traditional sectors, the urban development sector is a composite of crosscutting sectors as broad as urban governance and urban poverty reduction, and as specific as urban water supply and sanitation. This Chapter first discusses the nature and magnitude of urbanization in Colombia, and the institutionial and legal framework in place to manage it. This is followed by an assessment of urban poverty and the main problems in urban land and housing, water supply and sanita- tion, solid waste management, and transport sectors The last section outlines policy recommendations to help shape priority actions for addressing the key urban issues discussed I This Chapter incorporates the findings of a background study condLIcted by Jorge Acevedo of the University of Los Andes, Bogota, in collaboration with Wigberto Casrafieda and Manuel Salazar. Elizabeth Mehta helped edit the Chapter 329 330 COLOMBA. THE ECONOMIC FOUNDATnON OF PEACE I. Urbanization in Colombia Colombia is a highly urbanized country. More than 70 percent2 of its population of 42.3 million live in urban areas, defined as those with a population of 5,000 or more. Unlike in many other countries in the region, there is a more balanced distri- bution of urban population in Colombia's cities. The largest cities-Barranquilla, Bogoti, Cali, and Medellin-have more than 1 million inhabltants each, and about 40 cities have populations of over 100,000. In recent years, however, the concentra- tion of population in the capital city is becoming increasingly evident. In 1980, 20 percent of the urban population lived in Bogoti, the largest city. By 1995, Bogota was home to 22 percent of Colombia's urban population (World Development Indi- cators 2000). Over the past two decades, the rate of internal migration in the seven largest urban areas of Colombia has shown an overall decline, although significant interur- ban variations still remain (Table 1). In Bogota, for example, there was a sharp decline from about 2 percent in 1982 to about 0.9 percent in 2000; in Medellin and Cali, the decline was not as pronounced. Among the major cities, Bogota has the lowest gross rate of migration. In absolute terms, however, it accounts for the largest net inflow of migrants, followed by Cali, Medellfn, and Bucaramanga, with the lat- ter experiencing the highest in-migration rate, at 3 percent. According to recent esti- mates by the National Department of Statistics (Departamento Nacional de Estadis- ticas, DANE), the number of migrants arriving in Bogota every year remains roughly constant at about 55,000 people. This number would be higher if those fleeing from areas of armed conflict in the country are considered. According to cal- culations from the Red de Seguridad Social, there were about 15,000 refugees who fled to Bogota in 2000.3 Another pattern that has recently emerged is that smaller municipalities on the periphery of large cities now receive a larger share of migrants. While some migrants arrive directly and settle in these areas, as a first step to becoming integrated into the 2. Population projections from Departamento Nacional de Estadisticas (National Department of Statistics, DANE), based on the definition of urban and rural (cabecera-resto). 3. The estimates of desplazados-internal refugees-vary gready according to the researchers, the methodologies, and the time the study was done. Codhes estimates are larger than those of the Red de Solidaridad SociaL According to the former, between 1985 and 1996 the violence in Colombia displaced 856,781 people. Out of 181,010 people displaced in 1996, 23 9 percent (43,261) went to Bogoti. Both methodologies differ substantially: the Reds is an official figure, based on the displaced population written in its registry, figures from the Codhes were calculated for 1996, based on a survey. The calculations of the Red may underestimate the displaced population, because the unregistered households are excluded The calculations of the Codhes show figures closer to the DANE estimates of total internal migration. Quite likely the question about "the motive of migration," as Codhes formulates it, is biased, overestimating the number of forced migrants For this paper the figure of the Red is selected, even though it may underestimate the true figure. Table 1. Gross Annual Rates of Internal Migration to the Main Cities and Some Conurbations, 1982-2000 z Bogotd Cali Yumbo' Medellin Barranquilla Soleda Bucaramanga September 1982. Recent migrants < I year 57,344 26,106 1,824 25,434 21,365 4,318 N/A' Gross annual rate 2.0 2 8 6 3 2.4 3.5 4 7 N/A Total population 2,867,232 932,384 28,967 1,059,754 610,438 91,896 N/A September 1987 Recent migrants < I year 69,180 37,350 1,045 29,817 19,626 3,769 N/A Gross annual rate 1 6 2 6 2 3 2.0 2 1 2.1 N/A Total population 4,846,884 1,431,293 45,738 1,496,246 945,544 181,594 N/A September 1992 Recent migrants < 1 year 74,239 44,263 2,976 34,877 20,428 10,207 N/A Gross annual rate 1.5 2 8 6 0 2 3 2.0 4.3 N/A Total population 4,940,516 1,654,754 49,258 1,516,391 1,015,989 234,924 N/A September 2000. Recent migrants < I year 55,280 39,223 1,692 30,187 10,871 12,122 15,203 Gross annual rate 0.9 1 8 2.1 1 6 8 4 3 3.0 Total population 6,449,980 2,124,496 80,749 1,891,120 1,279,431 280,163 512,083 1 Conurbated to Cali 2 Conurbated to Barranquilla WN/A = Not available Note Gross rate of annual immigration = number of migrants with less than a year of residency/total population '100 Sources Encuesta Naciona/de Hogares (ENH), September 1982, 1987, 1992, and 2000, and consultant calculations 332 COLOMBIA: THE ECONOMIC FOUNDAnON OF PEACE city, others move from the city centers to reduce housing costs and become home- owners or tenants. As a result, many municipalities near large cities face growing demands for urban land, basic services, and other urban amenities. These include Soledad, Malambo, and Puerto Colombia near Barranqullla; Floridablanca and Piedecuesta adjoining Bucaramanga; Yumbo in Cali; and the most prominent case, Soacha, a municipality close to Bogota. These peripheral towns gather around large core cities, together forming func- tional agglomerations and dependencies, which blur municipal frontiers. These urban agglomerations, or metropolitan centers, are expected to increase in both number and size over the next decades, requiring special institutions and policies to effectively address common problems of housing, basic services, land use, planning, environment, transport, and other needs. While the Constitution contains provisions for managing conurbations, imple- mentation is severely constrained by such factors as the difficulty of getting munic- ipalities, especially the smaller ones, to yield key elements that affect their autonomy and independence, and getting larger municipalities to share the costs of responding to demands by new residents. To some degree, Medellin has been rather successful in consolidating its metropolitan area through development planning and infra- structure building. Bogota, on the other hand, has not been able to surmount sev- eral political, jurisdictional, and related difficulties with its neighboring municipal- ities. It is now trying to address these problems through an ad hoc mechanism, the metropolitan region, which is essentially an informal consensual planning arrange- ment. Through this process, Bogoti and its neighboring municipal administrations seek to harmonize their respective Planes de Ordenamiento Territorial (nine-year development plans, or POTs) and to work together to find joint solutions to their urban land, housing, transport, and other basic services needs. II. The Legal and Institutional Framework The political, fiscal, and administrative decentralization, which the government of Colombia began two decades ago, has passed through its first stage and is now being adjusted to respond to various areas of need. With decentralization, city/municipal administrations enjoy a high level of autonomy in managing their own finances, although the major source of income for many smaller municipalities remains con- stitutionally mandated central government transfers. Other positive impacts are wider local participation, accountability, and the strengthening of the legal and reg- ulatory framework of municipalities. Decentralization, however, has also given rise to serious problems, such as absent or lax central government monitoring, poor service quality, and inefficient financial management, resulting in many bankrupt municipalities, both large and small. In addition, other political issues impede the decentralization efforts. Mayors and coun- cilors are elected to serve for three years, but legally are not allowed to serve consec- URBAN DEVELOIPMENT 333 utive terms. Their short term in office disrupts the continuity of decentralized admin- istration and undermines the sustainability of many development programs. Many mayors and other elected local officials are also among the casualties of violence. In response to financial issues faced by municipal governments, the government of Colombia recently introduced two structural reforms: a constitutional amend- ment modifying the rate of increase of national revenue transferred to the munici- palities by the central government, and an amendment to Law 60 of 1993 (Law 715 of 2001) defining clearer distribution criteria along with conditions governing the use of transfers by municipalities. Several other laws were enacted to facilitate the decentralization process The Urban Reform Law (Law 9 of 1989) was designed to provide local governments with such powerful tools as expropriation, the designation of priority land for urban expansion, land banks, land rezoning and urban integration, land improvement tax, and clarification of development and construction rights. The new Constitution of 1991 provided additional impetus to decentralization by consolidating the key role of the municipality and clarifying the role of the Departamento as an intermediary authority between the central government and the municipality. The recently mod- ified Law 60 of 1991 explicitly defines this division of responsibilities. In particular, the duties of the Departamento include planning, technical assistance to the munic- ipalities, policy coordination, and implementation of sectoral programs Municipal- ities are primarily responsible for providing basic services such as water supply and sanitation, education, health, and housing, with special emphasis on satisfying the needs of the population who have Unsatisfied Basic Needs (NBI).4 Law 60 also spec- ifies the formulas for computing the distribution of resources from national revenue among the municipalities, and the manner by which these resources should be used. Law 99 of 1993 established the National Environmental System (SINA) and cre- ated a new Environmental Ministry to manage it. The SINA, which defined the overall framework for environmental management, conferred direct responsibility for urban environmental management on cities of over 1 million inhabitants. City Environmental Management Agencies (EMAs) were subsequently created (Barran- quilla and Cali were the first cities to establish their EMAs) or adapted to the new legal framework in 1994. Although the law established that the EMAs would have similar functions as Autonomous Regional Corporations (CARs) withlin their Juris- dictions, it did not specify whether they would have the same "rights" in terms of managing financial resources. Revenues obtained by EMAs from environmental 4 The indicator of Unsatisfied Basic Needs (NBI) is used to compute the transfer of resources from the central government to the municipalities and departments With this methodology a person or a household with any one of the following features is considered poor inadequate housing, housing without services, critical overcrowding, no school attendance, and high economic dependency The NBI, which is different from other poverty measurements of income, permits direct measurement of the state and local gov- ernment's efforts to improve the coverage of basic services and school attendance 334 COLOMBIA: THE ECONOMiC FOUNDATION OF PEACE licenses or fines are transferred to the city's treasury. The agencies' annual budget is decided based on a separate negotiation with the mayor, creating sustainability issues for the EMAs. Law 99 also established the mechanisms and guidelines for the prepa- ration of municipal environmental plans. Law 134 of 1994 created mechanisms to strengthen citizen participation in local decisionmaking in cities, while Law 388 of 1997 modifies the urban reform law, incorporating elements of recent legislation on environmental, metropolitan, and territorial issues. Law 388 also mandates that all municipalities must formulate and adopt their respective POTs covering land use, land regulation, land development, and urban expansion. To ensure continuity, the law stipulates that the local three- year development plans of incoming administrations must be consistent with the POTs. While the former are short-term plans, POTs have a nine-year planning hori- zon, spanning three administrations. With the above legal and institutional reforms, the government of Colombia seeks to promote an integrated urban planning and development process, and ensure continuity in the administrative framework while broadening local participation in the decisionmaking process. III. Urban Poverty In Colombia, the indicator of Unsatisfied Basic Needs (NBI) is widely used as an indicator of poverty.5 Using this indicator, the last decade saw a significant decline in the proportion of poor people with unsatisfied basic needs (Table 2). Between 1993 and 2000, this indicator dropped substantially from 26.8 percent in 1993 to 16.4 percent in urban areas, as a result of the efforts of the government in widening the coverage of basic services (sewerage system, potable water, energy) and improv- ing school attendance of children and youth. For all indexes, improvements occurred in both rural and urban areas, albeit in relative terms; the overcrowding index, however, did not improve as much as the others. A less optimistic picture emerges from the findings of the World Bank's 2001 "Colombia Poverty Report." After a continuous and significant reduction in urban poverty between 1978 and 1995, economic recession has pushed poverty back to its 1988 levels. Urban poverty decreased at rates close to 1.5 percentage points per year between 1978 and 1995. Unfortunately, the 1999 rates are again close to those of 1988, and even slightly higher for the poverty gap. The recession not only increased the number of poor people in urban areas, but the poor also became poorer, an observation confirmed by the increase in the percentage of extremely poor people in 1999 (Table 3). 5. Over the years, there have been extensive discussions on the strengths and weaknesses of NBI as a poverty indicator, but it continues to be widely used by the authorities and academia. Table 2. Indicators of Unsatisfied Basic Needs by Zone, 1993-2000 c z Percent in Percent in Percent Percent Percent Percent Not Percent High Poverty Misery Inadequate Inadequate Critical Attending Economic < Year Zone (1 NBI) (> 1 NBI) Housing Servues Crowding School Dependeny o 1993 Total 372 149 11.6 105 154 80 12.8 z Urban 268 90 70 72 120 48 8.8 Rural 62 5 30.3 23 7 19.2 24 4 16 4 23.3 1996 Total 260 8.9 86 55 11.1 40 90 Urban 16 9 4.1 4 1 2.6 8.1 2 1 5.3 Rural 48 6 20.7 19 8 12.8 18.5 8 5 18 1 1997 Total 259 8.6 80 5.4 11.3 42 89 Urban 17 8 4.5 4 3 2.8 8.3 2 8 5 4 Rural 46 5 19.1 17 5 11.9 19.0 7 8 17 9 1998 Total 260 8.2 66 47 11.1 4.7 100 Urban 17 4 4.1 3.6 2 0 8.3 2 5 6 1 Rural 47 8 18.6 14 0 11.5 18.2 10 2 19 8 1999 Total 24 9 7.3 6 7 4.0 11.2 3 9 8 8 Urban 17 5 3.9 3 5 2.1 8.6 2 7 5 7 Rural 43 7 15.9 15 1 8.8 17 8 7.0 16 7 2000 Total 230 6.5 67 36 102 32 76 Urban 164 3 5 3 5 1 7 7 9 24 5 1 Rural 400 142 148 8.5 162 53 140 Sources Department of National Planning, Boletfn SISD, nimero 30, DNP-DDS-GCV, based on DANE's Encuesta Nacional de Hogares, September 336 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Table 3. Poverty Indicators and Income Inequality, Urban Colombia, 1978-99 1978 1988 1995 1999 Poverty Poverty rate (percent) 70 55 48 55 Poverty gap (percent) 35 23 19 26 Extreme poverty rate (percent) 27 17 10 14 US$2 per day poverty' (percent) 34 5 3 5 Mean income per capita2 157,080 235,163 294,522 277,469 Income inequality Gin Coefficient 0.473 0 486 0 522 0.545 I Based on Purchasing Power Parity Convertors from World Development Indicators database. 2 1999 pesos, based on monthly household income Source World Bank (2001) Income inequality, measured by the Gini Coefficient, has steadily increased in urban Colombia during the last two decades, a worrisome finding in view of the strong link between income inequality and violence. For the most part, however, social indicators had improved over the past 20 years, with some drawbacks in school enrollment from 1995 to 1999 (Table 4). While the level of poverty is proportionately greater in rural than in urban areas, in absolute terms there are more urban poor in Colombia given the country's high urbanization level. According to the 1997 Quality of Ltfe Survey,6 the expenses of the urban poor are mainly for housing and food. For the poor in large cities, transport and utilities are significant items in their budgets. Among the poorest 20 percent of city dwellers, expenditures are divided among housing (close to 40 percent), food (30 percent),7 and transport (5 percent). In Bogota, the poorest spend 30 percent on housing, 30 percent on food, and 8 percent on transport. IV. Access to Public Services Statistics show that residents in the major urban areas have better NBI indicators today than in the past, which reflects efforts made by various administrations over the past decade to improve their access to basic services. In terms of income, how- ever, the years of economic decline exhibit an alarming national impoverishment process in which urban residents are finding it progressively harder to survive. This 6. The Quality of Life Survey is the most recent research with data about household expense structure. Though the Encuesta de Hogares includes questions on household expenses, it does not go as deep as the former and its data are not as good. 7 Computations by Jorge Acevedo based on the 1997 Quality of Ltfe Survey. URBAN DEVELOPMENT 337 Table 4. Social Indicators, Urban Colombia, 1978-99 1978 1988 1995 1999 Average education > 18 years 6 2 7.7 8.4 8 9 Illiteracy rate' (percent) 5.3 3 3 2 8 2 6 School enrollment (percent) Ages7 to I1 918 94 8 96.5 95 3 Ages 12 to 17 76 9 80 5 84 4 82 2 Ages 18 to 22 31 2 35 8 41.0 36.3 Complete primary school (ages 12 to 17) 67 0 78 7 77.7 89 8 Complete high school (ages 18 ro 22) 21.6 35 3 48 7 59 2 Child labor (percent) Ages 12 to 16 12 0 11 5 9 9 9 5 Ages 12to 14 5.8 50 52 37 Child Malnutrition2 Crime3 Homicides (per 100,000 pop) 26 62 65 594 I For population age 12 and older 2 For population unider age 5. represents nationial data for 1977, 1986, 1995, and 2000 3 Levitt and Rubio (2000) 4 1998 figure Note Urban Colombia represents Baranquldla. BogotA, Bucaramanga, Cali, Manizales, Medellifn, and Pasto Source World Bank (2001) situation is reflected in such indicators as mounting arrears on mortgage payments, high levels of illegal connections to basic utilities, and the rise of other forms of ille- gal activities A gradual impoverishment due to falling incomes does not negate the positive effects of wider service coverage; however, it presents a major challenge to policymakers since it lowers the ability and willingness of the poor to pay for serv- ices and increases the tendency for illegal connections and defaults. Under such a scenario, policies aimed at reducing or eliminating subsidies and legalizing connec- tions in order to make utility companies more self-sufficient may be perceived as aggravating the already impoverished condition of poor households, especially those in which such costs constitute a significant part of their household expenditures. The main challenge is not only to provide adequate levels of basic services, but also to ensure that these services are affordable and that there is an effective demand for services among the poorest. Policymakers are therefore confronted with various efficiency and equity challenges: how to guarantee a rational and efficient supply of services, how to regulate utilities in order to achieve a better tradeoff between eco- nomic and efficiency needs on one hand, and welfare needs on the other; how to make user charges reflect the cost of the service provision-all of which need to be accomplished without placing an unbearable burden on the urban poor. The fact that the cost of providing conventional services to the poor in marginal neighbor- 338 COLOMBIA THE ECONOMIC FOUNDATION OF PFACE hoods is usually much more than in well-established areas is also another factor to be considered in this context. Apart from addressing the service needs of the urban poor, responsibilities for infrastructure and services provision are further complicated as more conurbation areas are formed in major Colombian cities. This poses a daunting challenge espe- cially to smaller cities and municipalities, demanding a more integrated and partic- ipatory framework governing the planning and delivery of services. To shed further light on the many issues involved, a discussion on three main groups of public serv- ices-land and housing, water supply and sanitation, and urban transport-is in order. a) Land and Houszng Housing policy in Colombia involves a complex institutional network of public, pri- vate, governmental, and nongovernmental participants at the national and local lev- els. Among the major actors are the Presidency of the Republic, the Ministry of Development, the Agriculture Ministry, and the Social Solidarity Network, which are mainly responsible for policy design; the Ministry of Economy for financing; various specialized agencies for the provision and management of credit, subsidics, and other financial support, other governmental and nongovernmental organiza- tions for the provision of technical assistance; several Superintendencias for regula- tion and control, and regional and local governments for urban planning and land regulation The fragmentation of responsibility and authority makes it difficult to imple- ment an efficient housing policy, especially in light of the many sociopolitical and economic factors associated with housing. Coordination between and among con- cerned agencies is so minimal as to constrain decisionmaking on urban planning, spatial organization, and the provision of housing and related infrastructure and services. i. National Housing Policy in Perspective Traditionally the government of Colombia has played a prominent role in the country's housing market as both financier and builder of housing units. During the last three decades, the government's housing policies have gone through three cycles. (a) from 1972 to the late 1980s, (b) from 1991 to 1997, and (c) from 1998 to the present. The first phase began in 1972 with the UPAC, a system created in an inflation- ary period to help generate both savings for mortgage loans and demand for such loans at a time when such a market was nonexistent. Through specialized financial intermediaries, the UPAC system opened the door for short-term savers to earn a positive real return on their savings by combining a monetary correction factor with an interest premium. At the same time, it offered mortgage borrowers long-term financing at real positive rates of interest, but with an adjustment mechanism that URBAN DEVELOPMEN I 339 allowed them to make partial payments initially, to capitalize part of their interest payments during the early years of the loan, and to gradually increase their amorti- zation schedules during the later years. While the UPAC system was largely respon- sible for the )ump in housing finance, it nevertheless had many fundamental weak- nesses related to term mismatch, wrongly grounded assumptions about increased borrower payment capacity over time, capitalizationi provisions, and other unfavor- able market conditions. Additional demand incentives were provided to low-income families through such devices as zero down payment and upfront subsidies, but such instruments led to problems during periods of economic decline. The second phase came as part of the new framework of the 1991 Constitution, which upholds "dignified housing" as a constitutional right and lays special empha- sis on Social Interest Housing (Vivienda de Jntre's Social, VIS) for all Colombians. The earlier policy, based on the role of the State as financier and direct producer of housing, was revamped, partly as a result of charges of corruption and inefficiency of public programs. With direct demand subsidy as its main feature, Law 03 of 1991 established a national system for VIS and created the National Institute for Hous- ing of Social Interest and for Urban Reform (Instituto Nacionalpara la Vivienda de Interes Socialy la Reforma Urbana, INURBE) as the State agency in charge of pro- viding subsidies, enabling low-income households to purchase dwelling units sup- plied by private producers In spite of these reforms, 27 percent of the subsidies granted went unclaimed because of low supply of VIS,8 making it clear that not enough attention was being paid to supply-side constraints Moreover, the worsen- ing macroeconomic condition in the country precipitated another crisis. High inter- est rates, along with growing unemployment, resulted in mountilg insolvency as debtors found their unamortized loan balances becoming higher than the value of their underlying assets. By the end of 1998, about 140,000 of a total of 800,000 mortgage borrowers surrendered their units because they could not afford their monthly payments. The government of Colombia was forced to step in and pro- vided debt relief for about 800,000 families at a cost of about US$1.8 billion In 1998 the Constitutional Court further complicated matters by declaring unconstitutional the policy instruments introduced in the 1993 reform Law 546 was enacted in December 1999, defining a new housing finance system, the Unidad de Valor Real (Constant Value Unit, UVR), which has several features of the old UPAC system but without the capitalization of interest. The housing policy during the third phase represents the government of Colombia's serious attempts to strengthen the country's housing finance system, stimulate the entry of large com- panies into the VIS market, create incentives for the poorest families to save, and get the Cajas de Subsidio Familiar to provide more financing for social housing. In spite of these actions, however, the housing finance sector remains weak and, in a wors- ening economic climate could precipitate a serious crisis in the sector 8 Ocampo (1996) cited in Giraldo Isaza (1997), p. 184 340 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE ii. Housing Deficits in Major Cities Housing policies in Colombia are invariably based on estimates of qualitative and quantitative deficits for various target groups. Aided by such estimates policymakers try to measure the gap between the number of housing units available and the num- ber required so that every person is guaranteed access to decent housing. Estimates of resources and investments required to fill the identified gap are then made.9 For the period 1997-2000, estimates of the National Planning Department (Departamento Nacional de Planeaczdn, DNP) show that by 2000, the quantitative deficit was 11.8 percent, corresponding to 1,165,122 units, and the qualitative deficit was 19.5 percent, or 1,914,518 families living in dwellings that could be repaired Around 64 percent of the average housing deficits are faced by households earning less than three minimum salaries. In both urban and rural areas the overall deficits remained constant in spite of multiple housing programs. Urban areas accounted for 76 percent of the country's quantitative housing deficit in 2000 (Table 5). Such deficiency in housing supply, coupled with continuing in-migration, has become an issue that has reached critical proportions in some cities. Qualitative and quantitative housing deficits and the concomitant issues of over- crowding and the poor quality of the housing stock, however, form only a part of the overall urban housing problem in Colombia. Housing deficit estimates do not shed light on several other important issues (for example, the reason for limited housing supply), and only serve to draw attention away from supply-side con- straints. They simply focus on measures to boost housing demand through tempo- rary and unsustainable mechanisms, such as subsidization and credit provisions. iii. Low-Income Housing Issues at the Local Level The shortage in housing supply as evidenced by the housing deficit has an unwel- come consequence of producing informal housing and land tenure, a widespread problem in all large Colombian cities. Informality usually starts by either organized invasions of public lands by several families or by the sale by "pirate developers" of lands that do not meet formal regulatory requirements. In both cases informal settlers receive neither property titles nor even the most rudimentary services from the gov- ernment (laws and regulations prohibit government intervention in these "illegal" areas). With limited access, if any, to basic services such as water supply and sanita- tion, informal settlements are characterized by unhealthy environmental and social conditions. Although reliable data are limited, information for Bogota shows that as many as 240 hectares of land are developed informally per year for a total of 23 per- cent of the city's developed area. Data from Cali show that 13 percent of its residents, or as many as 260,000 people, live under informal property tenure regimes. A closer look at the Bogota case shows that households belonging to the first three socioeconomic strata can afford a maximum of $9,000 for a housing solution, 9. These statistics ought to be read with a critical eye since there are various methodologies, each with their own weaknesses URBAN DEVELOI'MENT 341 whereas VlS housing being offered by the government is in the range of $13,000 to $16,000. Therefore the main issue in low-income housing is the lack ofan affordable sUp- ply of this service for the poorest Only municipalities can act on these supply-side con- straints, while demand-side interventions are best taken by the national government. Two supply-side constraints have the highest impact on the housing shortage (a) land scarcity, hence the high price of serviced land in urban areas, and (b) the rigid- ity of land regulations and the building permit process. The case of Cali is illustta- tive of the first constraint. In January 1993, the municipal government presented to the city council a project to develop 380 hectares of land for low-income housing (Ciudadela Desepaz). The land price per square meter then was 365 pesos (at con- stant prices). Upon acceptance of the project by the City Council, the land price jumped to 571 pesos With the establishment of urban regulations it rose to 1,580 pesos, and later to 1,751 pesos when basic infrastructure was provided. By the rime the project was completed, the price of land was 3,600 pesos per square meter, a ten- fold increase in two years. Besides affecting the price of land directly, land regulations also restrict the effec- tive supply of land by controlling the floor-to-area ratio through minimum lot size, maximum construction density, and setback requirements, among other things. By controlling the floor-to-area ratio, local governments control the consumption of land, the only factor for which poor residents can outbid nonpoor residents. As illustrated in Figure 1, nonpoor residents will always prefer more consump- tion of land and will be willing to bid for it only after a minimum amount L*. Poor residents will also prefer more consumption of land, but can only otitbid the non- poor below L*. By imposing a minimum land consumption at L_, municipal gov- ernments force the demand of poor residents to the right of Lm,n where, for the most part, they cannot outbid the nonpoor. Again taking the case of low-income housing in Ciudadela Desepaz in Cali, regulations established that 15 percent of the total area should be earmarked for community centers, open space, and installations for public telephones, 15 to 20 percent for vehicular streets; 15 to 20 percent for sidewalks and parking spaces (1 for every 10 lots); and 3 percenc as a reserve area. According to the cost figures of private developers, the cost of urbanization oscillated between 32 and 50 percent of the total development costs. With high demands accompanying urbanization and the skyrocketing price of land, the housing solutions intended for families in the first and second socioeconomic strata ended up in the hands of middle-income households. b) Water Supply, Sanitation, and Solid Waste Management Until the mid-1980s, basic water supply, sanitation, solid-waste collectioll, and dis- posal services in the major cities were provided by public municipal companies cre- ated for those purposes In 1987, within the decentralization framework, the responsibility for water supply and sanitation was handed over to municipalities. Table 5. Housing Deficit (National, Urban, and Rural), 1997-2000 ,< Change in Yearly Average National Concept 1997 Percent 1999 Percent 2000 Percent 1997-2000 1 Households 9,155,481 100.0 9,440,151 100.0 9,834,709 100 0 24 2 Households sharing rooms 841,829 9 2 904,281 9 6 925,209 9.4 3.2 3. Households in nonhabitable dwellings* 306,352 3 3 264,545 2 8 239,913 2.4 -7.8 4. Quantitative deficit = (2) + (3) 1,148,181 12.5 1,168,826 124 1,165,122 118 05 5 Qualitative deficit = households in dwellings that can be improved** 1,812,560 19.8 1,973,743 20.9 1,914,518 19.5 1.8 6 Total housing deficit (4) + (5) 2,960,741 32.3 3,142,569 33 3 3,079,640 31.3 1 3 o 7 Households without deficit 6,194,740 67 7 6.297,582 66.7 6,755,069 68.7 2 9 8 Habitable housing stock > (1) - (4) 8,007,300 87.5 8,271,325 87 6 8,669,251 88.1 2.7 -1 Urban rr 1. Households 6,745,782 100.0 6,986,380 100 0 7,284,336 100.0 2 6 0 z 2. Households sharing rooms 766,850 11 4 825,662 11.8 820,432 11.3 2 3 0 3. Households in nonhabitable n dwellings* 98,949 1 5 64,364 0.9 64,066 0.9 -13 5 I 0 4. Quantitative deficit = C z (2) + (3) 865,799 12.8 890,026 12.7 884,498 12 1 0 7 5. Qualitative deficit 0 households in dwellings that Z can be improved** 932,096 13.8 921,157 13 2 908,876 12.5 -0.8 0 6 Total housing deficit (4) + (5) 1,797,895 26.7 1,811,183 25.9 1,793,374 24 6 -0 1 Table 5. (continued) Change in Yearly Average National Concept 1997 Percent 1999 Percent 2000 Percent 1997-2000 7 Households without deficit 4,94,887 73 3 5,175,197 74 1 5,490,962 75 4 3 5 8 Habitable housing stock (1) - (4) 5,879,983 87 2 6,096,354 87.3 6,399,502 87 9 2.9 Rural 1. Households 2,409,699 100.0 2,453,771 100 0 2,550,373 100.0 1.9 2. Households sharing rooms 74,979 3 1 78,619 3.2 104,777 4.1 11 8 3 1-louseholds in nonhabitable dwellingst 207,403 8 6 200,181 8 2 175,847 6 9 -5 4 4 Quantitative deficit = (2) + (3) 282,383 11.7 278,800 11 4 280,624 11 0 -0.2 5 Qualitative deficit = households in dwellings that can be improved** 880,464 36 5 1,052,586 42.9 1,005,642 39 4 4.5 6 Total housing deficit (4) + (5) 1,162,846 48.3 1,331,386 54 3 1,286,266 50 4 3.4 7 Households without deficit 1,246,853 51.7 1,122,385 45 7 1,264,107 49.6 0 5 8. Habitable housing stock (1) - (4) 2,127,317 88 3 2,174,971 88 6 2,269,749 89 0 2 2 Households living in dwellings with simultaneous deficiencies in public services, space, and manerial of the walls ** Households living in dwellings with one or two simultaneous deficiencies in public services, space, or structure Sources Ministr-y of Development and Economics, documento para el Banco Interamericano de Desarrollo, 2001, "D6ficit de vivienda," p 2 DANE-Encuesta Nacional de Hogares, calculos preliminares DNP DDUPRE-SV 344 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Figure 1. Bid Rent for Land: Poor versus Nonpoor Rent $/m2 \ id rent poor:\ Bid rent poor Bid rent non-poor Lmn Size of land parcel per household (mi2) Source Ortiz (I 999) Moreover, the 1991 Constitution promotes the goal of private sector participation (PSP) in providing these services and defined the responsibility of the State to reg- ulate, supervise, and monitor the performance of PSP, and assist in providing secu- rity for private sector players. Law 142 (public domestic services statute), which was enacted in 1994, established the legislative and regulatory framework for strength- ening the role of municipalities in providing quality service, emphasizing efficiency through competition, and the promotion of PSP in service delivery. The Ministry of Economic Development has assumed responsibility for devel- oping policies and plans for the sector since the creation in 1992 of the Dzreccz4n de Agua Potable (Potable Water Directorate) within the Vice-Ministry of Housing, Urban Development and Potable Water.'" The Ministry provides technical assis- tance, defines service norms and standards, and provides technical and financial support to municipalities. The Ministry of Health sets water quality standards, while the Ministry of the Environment formulates environmental policies and regulations to protect water sources. The Ministry of Mining and Energy regulates groundwa- ter development and use. The Water Regulatory Commission (CRA) promotes competition among service providers, controls monopolies, defines tariff-setting methodologies based on stan- dard formulas and on investment plans submitted by the operating companies, and sets service quality and technical standards to be followed by water utilities. The 10. Decree 219 of 15 February 2000 eliminated the Vice-Ministry of Urban Development but maintained the unit. URBAN DEVELOPMENT 345 Superintendency of Domestic Public Services (SSPD) is responsible for monitoring and supervising the adequacy and efficiency of operations, establishing uniform accounting systems, supervising the administration of subsidies, and monitoring the general adminiistrationi of public service companies. The Autonomous Regional Corporations (CARs), as regional implementing agencies of environmental policy, regulate and monitor the use and quality of water resources, and issue environmen- tal licenses i. Present Situation Colombia has achieved significant advances in the coverage of water supply, sewer- age, and solid-waste collection services (Tables 6 and 7). The coverage of these serv- ices in the main urban centers is over 90 percent, a high figure compared to other countries of the region "' In the major cities in 2000, water supply coverage was 95 percent, sewerage 88 percent, and solid waste collection 91 percent. 12 Such relatively high overall coverage levels, however, mask several problem areas. One is the disparity in the coverage of services, leaving out "informal" areas where many among the poorest strata of the urban population live. Another is the gener- ally unsatisfactory quality of water, with the exception of cities of over 500,000 inhabitants that are provided a 24-hour supply of potable water with 100 percent quality (fit for human consumption) 13 In most other cities, not only is the quality of water unreliable, there is also deficiency in the quantity of water made available as supply lasts for only 21 hours per day on average."4 Table 6. Coverage of Water Supply and Sewerage Services in Colombia (percent) Water Supply Sewerage 1985 1993 1997 1985 1993 1997 Total 70 6 79 7 83 30 59 45 63 0 70 Urban 89 2 94 6 97.6 80.7 81.8 80-90* Rural 280 41 1 443 113 144 150 * DNP shows 90 percent, a figure that seems high wheni compared with an 80 8 percenit figure from the Ministry of Development Souirces DNP-UDS-DIOGS, based oni DANE, 1985-93 census EH/93/97 Conisulted DNP's webpage, "Sistema de Indicadores Sociodemiiogrificos para Colombia II Colombia Water Sector Reform Assistance Pro)ect, 2001, Pro)ect Iniformation Docu- ment, World Bank 12 These figures might be overestimated due to the impossibility of getting reliable infor- mation as cities grow and because of outdated cadastre information 13 Potable water and basic health policy, with information from the second national assess- ment of water of the Health Minister Consultation from the Economic Development Minister website. 14 Ibid 346 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE Table 7. Coverage of Water Supply, Sewerage, and Solid-Waste Collection in the Main Cities (percent), 2000 City Water Supply Sewerage Solid waste Bogota 95 1 86.0 Over 95 Cali 95 9 94.0 81.0 Medellin 99 1 93.2 91.2 Barranquilla 94 0 80.0 100 Cartagena 77.8 64.1 100 Cucuta 91.4 89.8 71 8 Bucaran1anga 99.0 98 0 100 Perelra 96 6 93.9 92 0 Average 95.3 88.3 91.0 Source Superintendency of Domestic Public Services (SSPD) The main issue confronting the water supply sector, however, is how to ensure ade- quate service coverage in thefiture. Based on past trends In service coverage, the dif- ficulty of augmenting water supply after a certain level becomes evident (Table 8). Between 1995 and 2000, only Barranquilla showed some real improvement in sup- ply coverage. Cartagena IS a special case: while its population has doubled over the past 20 years (to around 750,000 today), service coverage has not matched the growth in its population. Since about 31 percent of its population live under extreme poverty conditions, water supply coverage in the city is not likely to improve. The combination of technical and financial inefficiencies and weaknesses (high water losses, poor cost recovery, and poor strategic planning) raises serious doubts about the future availability of adequate water supplies for the growing urban agglomerations. If present tiends continue, those living in the most urban- Table 8. Coverage of Water Supply Services in Major Cities City 1995 1996 1997 2000 Bogota 85 98 98 95.1 Call 92 94 93 95 9 Medellin 98 99 99 99.1 Barranquilla 83 83 86 94.0 Cartagena 72 76 83 77 8 Cticuta 91 4 Bucaramanga 98 98 98 99 0 Pereira 97 97 95 96.6 Source Supercifias en m3, No 2 of SSPD, consulted in CRAs webpage 15 Polfitca de Agua Potable y Saneamiento Basico Consulted the Mlnisterio de Desarrollo Econ6mico webpage. URBAN DEVIELOPMENT 347 ized areas (about 38 percent of the population) will face the risk of water rationing by 2016.15 According to studies conducted by Empresa deAcueductoy Alcantarillado de Bogotd (EAAB, the sewerage and water supply utility company in the city), unless substantial investments are made to increase its production and distribution capacity, Bogota will be able to meet projected water demand only until 2014. Most Colombian cities have a combined underground sewerage and drainage system, often with limited capacity and installed in low-lying areas that are subject to periodic flooding in the rainy season. Of the major cities, only Bucaramanga and, to some extent, Bogoti and Medellin, had wastewater treatment plants in operation in 2000 (Sanchez Triana 2000). In most other cities, only about 8 percent of domes- tic wastewater generated receive some kind of treatment (ibid.), a very low level when compared with other countries of similar income in the region Untreated domestic wastewater is thus discharged directly into nearby rivers and waterways, causing water pollution. Aggravating the problem are solid waste, pesticide runoffs, and wastewater discharges from industries The government estimated that in 1991 most industries inside or close to the urban centers dumped their waste into rivers and shores. Of these, close to 40 percent were dumped into rivers feeding the Mag- dalcna River (not including the Bogota River) and 35 percent into the Bogota River The lack of adequate infrastructure to transport and adequately treat domestic wastewater and industrial effluents is increasingly becoming a problem nationwide. As more rivers and streams become polluted, the negative impacts on health can reach serious proportions. In Bogota initial efforts at wastewater treatment imposed very high and unsustainable costs to the city. This is an area that will require high priority attention so that cost-effective solutions are found. Except in Cali and Cucuta, solid-waste collection rates in major urban centers are relatively high-over 90 percent. The problem, however, concerns waste disposal. In Colombia only 14 percent of the municipalities have any control over the final dis- posal (Presidencia de la Repuiblica 1998); most garbage is disposed of in open field dumps without any lining or other measures to prevent runoff and underground seepage. In cities such as Bucaramanga, this practice has led to the contamination of adjacent water sources (Alcaldia de Bucaramanga 1999). ii. Main Sector Problems Water supply and sanitation services are the responsibility of the municipalities and districts through their autonomous public service enterprises (ESPs). The creation of ESPs as public, private, or mixed enterprises was decreed by Law 142 of 1994 to improve services and to cushion them from political intervention. The results so far have been less satisfactory, including low level and quality of service, high water loss, high labor costs, distortions in tariff and subsidy structures, weaknesses in project planning and implementation, a high proportion of fraudulent users, and billing errors Many of these problems can be attributed to political interference at several levels of management within utility agencies Additionally, pressures from very pow- 348 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE erful labor unions have led to costly salary and benefit packages, making it difficult to encourage greater private sector participation or to increase efficiency and pro- ductivity within ESPs. Between 1995 and 1997, the SSPD found in a sample survey of the 16 largest utilities in the country that 91 percent of their operating income was used to cover operating costs (ibid.), leaving a very small margin for expansion and new invest- ment. In some water companies, such as in Cticuta, the operating costs in 2000 exceeded operating income by over 50 percent. While in theory the companies could levy charges to ensure self-sufficiency and sustained operations, in practice they have not been able to do so. Tables 9 and 10 give an idea of the revenue base of utilities in major cities. Law 142 of 1994 instituted a scheme of cross-subsidization under which the commercial and industrial sectors, and the two highest strata of residents could be made to pay an additional contribution of up to 20 percent16 over the value of their consumption to help finance a percentage of the basic consumption of the lower strata. For two reasons, however, the benefits of the scheme do not necessarily accrue to those for which it is intended. First, the socioeconomic stratification used as the criterion for targeting subsidies proved to be a poor indicator of income levels such that many recipients are nonpoor households. Second, the level of subsidy is too high and the population targeted too large such that the cost of service provision is borne by only a small percentage of the population. A case in point is Bogoti, which, of the four major cities, earns the highest revenue from tariffs (though it still lags in updating its tariff schedule). Recently, however, it incurred operational Table 9. Water Supply and Sewerage Utilities: Total Expenses/Operational Income, 1999 and 2000 Result in Result in 2000 1999 City Utility Companies (percent) (percent) C6cuta Empresa Industrial y Comercial de COicuta SA 166.77 162.33 Cali Empresas municipales de Cali "Emcali" 114 24 105.7 Bogota Empresa de acueducto y alcantarillado de Bogota 108 4 114 03 Pereira Aguas y Aguas de Pereira 94 78 110 72 Bucaramanga Compafifa del acueducto metropolitano de Bucaramanga 86.99 102.54 Cartagena Aguas de Cartagena SA. ESP 82.46 80.03 Barranquilla Sociedad de A A A de Barranquilla-Triple A 76 26 78.42 Medellin Empresas piblicas de Medellin EPM 72 02 Source SSPD The table has been arranged by the results in 2000, from the highest down 16. Law 632 of 2000 allowed higher percentages to maintain the equilibrium between income and consumption cost, for which the utilities must follow the methodology expressly defined by the national government URBAN DEVELOI'MENT 349 Table 10. Solid Waste Companies: Total Expenses/Operational Income, 1999 and 2000 Result in Result in 2000 1999 City Utility Companies (percent) (percent) Cucuta Empresa Industrial y Cornercial de Cucuta 166 77 162 33 Medellin Empresas Varias de Medellin 105 72 132 26 Pereira Empresa de Aseo de Pereira 98 61 96 35 Cartagena Ciudad Limpia del Caribe 91 64 96 5 Bogort Consorcio Lime 85 96 80 84 Bogota Ciudad Limpia 83 43 81 89 Bogota Consorcio Aseo Capital 74 08 82 24 Source SSI'D The table has been arranged by the results for 2000, from the highest down deficits With 77 percent of its users belonging to income strata 1, 2, and 3, for which subsidies are provided, and only 9.5 percent from strata 5 or 6, which pay for the services,17 its financial position has weakened. Apart from the large subsidies, which impose a heavy burden on their budget, unaccounted-for water (UFW) also represents a substantial loss to water utilities. In 2000, on average, UFW accounted for about 39 percent of the volume of water pro- duced (Table 11). The reported annual losses of the 16 utilities surveyed are on the order of 600 million cubic meters, a waste valued at US$370 million (Sanchez Tn- ana 2000). The CRA estimates that if the 16 largest utilities in the country could reduce these water losses from 40 percent to 20 percent, the water saved could pro- Table 11. Unaccounted-For Water in Major Cities, Percentage City Utility 1995 1996 1997 2000 Bogoti EAAB 36 33 32 34 Cali Empresas P6blicas de Cali- Emcali- 33 36 30 34 Medellin Empresas P6blicas de Medellin- EPM 35 35 35 32 Barranquilla Triple A 60 55 50 40 Cartagena Aguas de Cartagena 56 54 51 42 Cicuta EIS Ctcuta 50 Bucaramanga Compafifa del acueducto merropolitano de Bucaramanga 29 34 31 26 Pereira Aguas y Aguas de Pereira 43 49 48 42 Source Years 1995, 1996( and 1997, CRA's webpage, based on SSPD, Supercifras en rs3, No 2 17 CRA's webpage 350 COLOMBA THE ECONOMIC FOUNDATION OF PEACE vide service to about 40 percent of their population. With adequate maintenance, average investment cost could drop from 69 percent to 55 percent."8 c) Urban Transport In Colombia, public transport use is high, as evidenced by the large share of motor- ized trips by this mode in some cities (Table 12). The situation in all ma)or cities does not differ markedly. For this reason, this section focuses mainly on public transport issues and less on other (though not the least) transport-related concerns, such as road management, nonmotorized transport, traffic management, and so forth. Overall, public transportation in Colombia's largest cities is characterized by inef- ficiency and lack of an adequate management structure.19 The main issues are (a) excessive service supply, (b) inefficient government regulation, and, as a conse- quence, (c) a high accident rate. These are further discussed below. i. Excessive Service Supply The excess of public transport vehicles has reached critical proportions in some Colombian cities. Cities like Barranquilla and Bucaramanga have less than 240 inhabitants per vehicle, resulting in congestion and low travel speed. An exception is Armenia, which has a somewhat more orderly public transportation system and has fewer vehicles serving more residents (Table 13). The excess of public transport vehicles and road deficiencies result in a high level of traffic congestion and disorder. For example, the volume of passengers in the Table 12. Demand for Public Transportation in Major Cities Share of Publie Population Transport City (thousands) Trips per Day (percent) Bogota (95) 6,500 10,256,000 72 Bucaramanga (Metropolitan area) (97) 920 961,700 87 Pereira (97) 400 695,000 66 Ibague (00) 426 499,000 77 Sources Based upon calculations of Jorge Acevedo from various sector studies 18 CRA "Basis of the new regulatory framework for the Potable water and Basic Health Sec- tor: Presentation Document for Companies." Document presented at the International workshop on "Participative Construction for the new regulation of the aquaduct, sewers, and waste disposal services in Colombia," October 2001. 19 Consorcio Instituto SER de Investigaci6n y Pablo E Bocarejo Ingenieros Consultores, 2001. URBAN DYVELOP'MENl 351 Table 13. Public Transport Vehicles and Their Productivity in the Main Cities, 2000 Public Passengers Passengers Population Transport Inhabitants per Vehicle- per Vehicle- City (thousands) Vehicles' per Vehicle km day Bogota 6,500 22,000 295.0 2 02 432 Medellin 1,938 3,813 508 3 3 31 392 Cali 1,800 4,253 423.5 1 76 400 Barranquldla 1,000 4,201 238 1 N/A 317 Bucaramanga 431 1,833 235 0 2 30 407 Maniizales 350 795 440 0 2 96 409 Valledupar 270 546 495 0 N/A 126 Armenia 250 305 820 0 3 19 495 This iniformiiation comiies from local aLuthorities, November 2000 There might be unregistered vehicles movinig through the city Somiie figures for Bogora are presenlted later in this paper showing the signiificanit imiagnlitude of unregistered vehicles Source Consorcio Instituto SER de Investigaci6n, Pablo E Bocarejo Ingenieros Co0nsUltores, 2001 most congested corridors in Pereira peaks at 7,000 passengers per hour in each direction; in corridors in Bucaramanga, it is 142,000 passengers per day in each direction; and in Bogoti, there are corridors with close to 30,000 passengers per hour in each direction. As a consequence, the speed on these corridors drops to lev- els of 5 to 8 kilometers per hour in Pereira, 6 to 10 kilometers per hour in Cali, and occasionally less than 5 kilometers per hour in Bogota. Low speeds and the con- centration of vehicles caLlse high levels of pollution and noise where most of the users wait for transport. ii. Inefficient Government Regulation Services are privately operated by companies that, while not recipients of subsidies, do not pay for the right to the routes they service. Companies granted route rights by local governments are typically groupings of many affiliates (as many as hundreds or even thousands, depending on the city) who are owners or owner-drivers. In return for a monthly fee, these companiies organize and assign routes and coverage intended to equalize profitability for their members. In theory, the licensing system controls entry, routes, trip frequencies, and fares. In practice, there is minimal enforcemiienit of license terms and companies have no difficulty in creating a new route or withdrawing from existing service. While the system has the semblance of a deregulated system, in practice, there is very little competition among companies since each has monopoly rights (without time-limit constraints) to its awarded routes. The main area of competition lies among the driv- ers and operators who have no job security or social benefits and whose salaries depend on1 the number of passengers transported, a situation which pushes them to commit many infractions and get involved in many accidents 352 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE The main issue thus is the lack of an efficient transport regulatory system, resulting in the concentration of bus routes in major corridors which has, in turn, led to dete- rioration in service quality, and the progressive abandonment of unprofitable routes on the periphery, which in turn has given way to increased informal or illegal trans- port systems charging high fares, as discussed below. As mentioned, the increased number of vehicles has resulted in a diminishing number of passengers per vehicle, which in turn has generated pressure from the operators on the government to raise fares. In all cities there is a flat fare (not related to distance but still considered to be high) the level of which varies from city to city. In some cities (led by Bogoti) the fare system is complicated by several levels, incor- porating factors such as age of vehicle and services offered (Table 14). There is usu- ally a surcharge for nightly operations. Upmarket "executive" and "super-executive" service vehicles do not allow standing passengers.20 iii. High Accident Rate An adverse consequence of all or a combination of the problems discussed above is the alarmingly high incidence of accidents in many Colombian cities, a large pro- portion of which involves public transport vehicles. Table 15 shows the figures for those kllled and hurt in 1999 in traffic accidents in several cities. For comparison, the data are calculated per 100,000 inhabitants for each city, and from calculating the average among the group of cities listed. While there are wide differences in accl- Table 14. Public Transport Fares in Several Cities, 2000 City Type of Service Fare (pesos) Bogota Bus 90 and previous 600 Bus 91 and after 700 Buseta 92 and previous 600 Buseta 93 and after 700 Executive bus 800 Super-executive bus 800 Microb6s 800 Medellin Plain bus 600 Bucaramanga Plain bus 450 Executive buseta 600 Luxury microbis 600 Manizales Buseta gasoline 450 Buseta diesel 450 Bus 350 Armenia Bus 450 Source Consorcio Insntuto SER de Invesagaci6n, Pablo E. Bocarejo Ingenieros Consultores, 2001 20. This is the formal requirement, but in practice, with no enforcement control, there are standees among such passengers URBAN DEVELOPMENT 353 Table 15. Accident Rates in Colombia's Main Cities, 1999 %of Events with Population Accidents in 1999 Rate per O00,000 Inhabitants Pubhc City (OOOs) Dead Injured Dead Injured Transport Bogota 6,500 872 22,427 13 4 345 33 3 Medellin 1,938 447 21,781 23 1 1,124 37 4 Cali 2,000 397 7,265 19 9 363 37 3 Barranquilla 1,000 137 1,191 13 7 119 41 6 Bucaramanga 508 32 1,790 6 3 352 39 0 Manizales 350 149 706 42 5 201 N/A Ibague 426 27 643 6 3 151 60.7 I'oral 12,722 2,061 55,803 16 2 439 N/A Source Fondo de Prevenct6n Vial, 1999, anid calculations by Jorge Acevedo dent rates and in incidence of accidents and number of casualties among cities, there is clear need for stronger transport policies to address this problem The existing transport conditions affect the poor in various ways. On one hand, the excess of vehicles and the lack of efficiency of the system mean high fares. On the other, in many poor neighborhoods on dte periphery, no legal transport service operates. Unli- censed pickup trucks or jeeps, which generally provide alterniative informal services, charge even higher fares, thereby lessening the mobility of the poor and limiting their access to employment opportunities, social services, and other city amenities To alleviate the urban transport problems, two recent projects were undertaken by the government, both of which have had a profound influence on policies in the public transport sector (Boxes 1 and 2). V. Policy Recommendations Following are some of the factors and options which merit greater policy, institu- tional, and fiscal attention by the government of Colombia as it seeks to promote urban growth and development The objective is to offer a set of policy recommen- dations that focuses not on grandiose plans but on strategically important "triggers" that have the greatest potential for improving the quality of life in urban areas, and which can be easily translated into actions at the local level. a) Urbanization In Colombia about 31 million people reside in urban areas and about 12 million in rural areas Based on current projections, the country will have close to 40 million urban and about 10 million rural inhabitants within the next two decades. Apart 354 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE Box 1. The Medellin Metro Work on the Medellin Metro began in July 1984 with a scheduled comple- tion date of October 1989. However, operations commenced only in 1995 after multiple delays, cost overruns, and protracted litigation among the par- ties. While the original project cost was US$600 million, the final cost is now estimated to be about US$2.5 billion, with pending claims expected to add another US$600 million. Initially, the government of Colombia was expected to guarantee the cost of the investment only while the city made the pay- ments. Since the city could not meet its payments, the government of Colom- bia has had to assume the obligation, adding a very significant burden to its fiscal situation. The Metro system comprises three lines linking 25 stations spanning a length of 29 kilometers. It has 42 motorized units pulling 126 passenger wag- ons with a total daily capacity of about 430,000 passengers. However, it cur- rently transports only about 310,000 passengers per day According to the Empresa Metro de Medellin, the annual operating income from fares adds up to US$32 million, and publicity and office rentals add another US$2 million annually. The annual revenue covers operating, administration, and routine maintenance costs. Due to the adverse fiscal situation created by the Medellin Metro, Law 3 10 of 1996 was passed to redefine the government's role in future mass transport projects in Colombia. It stipulates that the national government could assume only partial financing of such projects, between 40 percent and 70 percent of the debt service, and only after a series of strict feasibility analyses and require- ments have been met. Box 2. The TransMilenio System in Bogoti After the Medellfn system began operations, Bogota pushed ahead with separate plans to design, build, contract, and operate a new system of buses on exclusive busways, the TransMdlenio (TM). The TM system, which commenced opera- tion at the beginning of the new millennium, is structured on main corridors, with lanes destined exclusively for the operation of articulated, high-capacity buses. These lanes are physically separated from the mixed, general traffic lanes, which are available for the circulation of private vehicles, trucks, and taxis. The network of main corridors is integrated with feeder routes, operated by buses with lesser capacity in order to increase the coverage of the system. Roads and bus stops have been adopted for the feeder routes, and stations are made easily accessible to pedestrians on the trunk routes. Operation and control are sup- URBAN DEVELOPMENT 355 Box 2. continued ported by a central control center, where the information provided by buses and by the stations is processed, providing continuous monitoring and feedback to the operators. The municipality is responsible for the construction and maintenance of the infrastructure, and the supply and operation of the control center equipment. The private sector, through concession contracts, supplies and operates the buses and the fare collection equipment. The income from the transportation activity must cover all operating, maintenance, and equipment replacement expenses, and the profits of all of the private actors involved in the system. The private sector, through a commercial, trusteeship, receives the income collected from fares and distributes it among the participants according to contractual agreements. TM is expected to be implemented in four stages, with a total of about 388 kilometers of trunk lanes that would cover about 80 percent of public transport routes in the city over the next 16 years. It will have about 4,475 articulated buses costing about US$900 million. The completion of the entire trunk lane system is expected to cost about US$2 billion. The first stage of the system was inaugurated in December 2000. It currently comprises three main corridors extending 38 kilometers of exclusive busways. The current fare is 900 pesos, or about US$0.40, similar to that of the conventional service in Bogota. TM trans- ports about 570,000 passengers per day at an average speed of 26 kilometers. The system has received universal acclaim and has been cited as a paradigm of a modern and effective public transport system. In comparison with the Medellin metro, the impressive positive impact of the TM led both national and local governments to sign new agreements for the latter In an agreement dated November 2000 the government of Colombia has committed itself to contributing US$1.3 billion for the continuation of the TM system in Bogota Funding will be provided over a 17-year period, starting in 2000. The resounding success of TM contrasts sharply with the prevailing public transport system in other parts of Bogota and other cities in Colombia, where congestion, pollution, and long commute times are still the norm. In Bogota, the authorities have taken some short-term measures to help ameliorate such conditions. The pico y placa system seeks to reduce traffic congestion during peak hours by restricting access by both private and public vehicles through their license plate numbers. In doing so, the city was able to achieve a 20 per- cent reduction in their movement without negative impacts on passengers. Efforts are under way to create incentives to discourage the use of old polluting vehicles and to eventually take them out of circulation. 356 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE from overall population growth rates, there wll be many more conurbation areas as a result of rapidly expanding cities and municipalities, forming large urban agglom- erations. These trends should be accorded greater consideration by policymakers when making decentralization and revenue-sharing decisions. As provided in the Constitution, the creation of a strong, intermediate-level authority, such as the Departamentos, to work with municipalities and conurbations should be given high priority. The immediate challenge is to rescue bankrupt munic- ipalities and Departments. At the same time, the projected addition of 9 million more people in Colombian cities calls for stronger decentralized governance. Two priority actions are recommended: • The government of Colombia should offer strong incentives for bankrupt municipalities, municipal entities, and Departamentos to prepare and imple- ment workable plans involving such measures as improved cadastres, more realistic property tax rates, improved tax collection, and enforcement systems that would help improve local capacity to generate resources. • With only three years in office, the term of mayors and councilors is so short as to promote short-term behavior. This term should be extended and/or the local officials should be allowed to serve consecutive terms to ensure continu- ity in development administration. b) Land and Housing On the demand side, lack of income is the main reason why families do not have adequate housing and access to services. On the supply side, the main constraints are the existing disincentives to suppliers, such as complicated, costly, and lengthy processes involved in land and housing development; and norms, standards, and other regulations that are often counterproductive. More often than not, official interventions place emphasis on boosting demand through such artificial devices as temporary subsidies (in grants, interest rates, wage taxes, and so forth). Once this temporary support is withdrawn, housing programs collapse, with consequent wider budgetary and financial implications. * Progressive upgrading via community-based development has proven to be not only an efficient mechanism by which the poor are helped to help themselves, but also a practical solution to problems of property tenure and access to basic services. Major efforts should therefore be taken to scale up urban upgrading in all major cities. This calls for the creation of a more flexible and creative regula- tory environment, allowing for orderly settlements on environmentally suitable sites that can later be regularized. At the same time, rigid control measures must be adopted to prevent occupation of hazardous or ecologically sensitive areas * Urban upgrading alone will not solve the problem of low-income housing. It is necessary to prevent the need for informal development in the first place. URBAN DLVELOIMENT 357 New government interventions should therefore support the expansion of shelter supply for the poor by addressing supply-side constraints To some degree this was attempted by Bogota when it created a municipal entity, Metrovivienda, with the authority to remove some of the ma)or constraints that preclude private suppliers from entering the housing market. While this may prove to be a good strategy, it is at best a short-term measure. The gov- ernment should also work with local authorities to remove ma)or institutional bottlenecks and disincentives as mentioned above so that entities such as Metrovivienda become unniecessary. c) Water Szupply and Sanitation While the coverage rate of water supply, sewerage, and solid waste collection serv- ices is high in the major cities, in absolute numbers, hundreds of thousands of fam- lies are still inadequately served, most of which belong to the poorest income strata. Often, high water, sanitation, and solid-waste coverage ratios mask qualitative gaps and focus attentioni away from very serious deficiencies in services provided. More- over, only about 7 percent of wastewater and about 14 percent of solid waste gener- ated receive any type of postcollection disposal and treatment, posing considerable health and environm llenital risks, pai ticularly to those who live along the floodplains and around city dumps. It is therefore recommended that: * Given the weak financial condition of many municipalities and public water utility companies, necessary actions should be taken by the government of Colombia to improve efficiency within these entities through increased tariffs to reflect the full cost of service provision, includinig investnmenit and operat- ing capital, and through more rationalized and better-targeted subsidies to improve access by the urban poor. * In its effort to expand services, the government should adopt an approach that will rely on more appropriate technology and variable design standards to deal with differential capacity to pay by different income groLips Particularly in low-inconme communities, it should assist in identifying alternative levels of infrastructure with costs that are not beyond their capacity to mobilize com- muniity contributions or pay user charges, and that are appropriate for com- murity-level maintenanice. * More stringent enforcement of existing policies should be ptirsued and stronger regulatory measures adopted to prevent or reduce water pollution problems arising from inadequate wastewater and solid waste disposal. d) Urban Transport Local governmentc agencies in charge of public transport often lack technical, administrative, and financial capacity to introduce required changes in the sector, let 358 COLOMBIA THE ECONOMIC FOUNDATION OF P)EACE alone implement these reforms in a more strategic and efficient way. Building the appropriate policy context in this sector has become an urgent area for the next gov- ernment to address. Given the extent of urban transport inefficiencies as discussed earlier, the following policy recommendations merit priority consideration: o A national urban transport policy should be formulated to promote people- oriented urban transport strategies and to transfer the "good practice" knowl- edge being accumulated from Bogota to other cities. It would also aim at strengthening the institutional capacity of local government agencies respon- sible for urban transport, roads, and public spaces. o TransMilenzo demonstrates how it is possible to implement policies that give priority to public transport over private car travel, using scarce road space and providing quick access from outlying low-income areas, while at the same time improving the image of what used to be perceived as a chaotic, polluting, and unpleasant bus service. It also shows that private sector participation can be facilitated by carrying out a transparent bidding process and instituting clearly defined concessions. Transport studies should be initiated to identify other cities where TransMilenio-type bus reforms could be feasible and desirable and, where appropriate, assist in their implementation. Increased use should be made of the municipal surtax on gasoline for the investments required. o The new urban transport strategies should address not only issues of access, efficiency, and vehicular pollution, but also of traffic safety. The traffic safety strategies to be developed for each city should place emphasis on road safety education, sound road and traffic engineering, and enforcement of traffic rules. References Acevedo, Jorge. 1999. "Comentarios al Plan de Ordenamiento Territorial de Bogota desde la Perspectiva del Transporte." Cimara de Comercio de Bogota. . 2001. "Gesti6n del Transporte Urbano." Curso de Gesti6n Urbana del Banco Mundial. San Salvador, abril. Acevedo, J., and J. P. Bocarejo. 1993. Bases de una Politica Integral de Transporte Urbano en Colombia. Departamento Nacional de Planeaci6n, Bogota. Acevedo J., and W Castafieda. 1998. `Analisis de la estructura y la distribuci6n del gasto en los hogares colombianos, con base en la encuesta nacional de calidad de vida aplicada por el Dane en noviembre de 1997." Departamento Nacional de Planeaci6n, Bogota. Acevedo, J., W Castafieda, and J. C. Salazar. 1993. El Metro de Medellin. Una Ilusidn Costeada por Todos los Colombianos. Fonade e Instituto SER de Investi- gaci6n, Bogota. Alcaldia de Bucaramanga. 1999. Plan de Ordenamiento Terrmtorial del Municipio de Bucaramanga: Documento Diagndstico. Bucaramanga. URBAN DEVELOI'M ENI 359 Alcaldia Mayor de Bogota. 1999. Plan de Ordenamiento Territorial Departamento Administrativo de Planeaci6n Distrital, Bogota. Alfonso, O., and J. V Zamudjo. 2002. "Balance y perspectivas del derecho a la vivienda en Colombia." Cal y Mayor y Asociados S.C. 1997. Estudio de Demanda del Sistema Integrado de Transporte Masivo para la Ciudad de Cali. Fase III. Cali Camara Colombiana de la Construcci6n, Camacol. Boletin de andlisis (4). Docu- mento de Internet. Sin fecha. "Ciudad Bolivar." Documento Internet. (Sin autor, sin fecha.) Comisi6n de Regulaci6n de Agua Potable y Saneamiento Bdsico. 2001. "Bases del Nuevo Marco Regulatorio para el Sector de Agua Potable y Saneamiento Basico: Documento de Presentaci6n a las Empresas." Documento presentado en el Taller internacional "Construcci6n participativa para la nueva regulaci6n de los servi- cios de acueducto, alcantarillado y aseo en Colombla." Octubre. Consorcio Instittto SER de Investgaci6n, Pablo E. Bocarejo Ingenieros Consul- tores. 2001. Reorganzzacidn del Esquema Empresarial de Transporte PzTblico en Colombia. Bogota Consorcio Pablo E. 2001. Bocare)o Ingenieros Consultores, Euroestudios S.A., ETT S A Mejoramiento del Sistema de Transporte Pablico Complementarzo a 7RansMile- nio de Bogotd Bogota. Departamento Nacional de Estadisticas (DANE). 1982, 1987, 1992 y 2000. Encuesta de Hogares, septiembre. 1998. "Proyecciones de poblaci6n." 2002. "Informes de financiaci6n de vivienda." Departamento Nacional de Planeaci6n. 1997. "Avances del Plan de Agua." Docu- mento Conse)o Nacional de Polftica Econ6mica y Social (CONPES) 2902, enero. _ 1997. "La Participaci6n Privada en Agua Potable y Saneamiento BMsico-Polftica y Estrategia." Documento Consejo Nacional de Polirica Econ6mica y Social (CONPES) 2912, marzo. 1998. "Participaci6n Privada en Agua Potable: Seguimiento." Documento Consejo Nacional de Polftica Econ6mica y Social (CONPES) 3006, mayo . 1999. "Cambio para construir la paz. Informe de avance del plan nacional de desarrollo." Documento Consejo Nacional de Politica Econ6rnica y Social (CONPES) 3100. 1999. "Carantia de la Naci6n a una Operaci6n de Cr6dito del Distrito Turisrico y Cultural de Cartagena para el Plan Maestro de Acueducto Alcantaril- lado y de Manejo Ambiental de las Aguas Residuales del Distrito." Documento Consejo Nacional de Politica Econ6mica y Social (CONPES) 3036, mayo 1999 "Garantia de la Naci6n a una Operaci6n de Credito Externo de la Empresa de Acueducto y Alcantarillado de Pereira S.A. para el Plan Maestro de Acueducto y Saneamiento Hidrico de Pereira." Documento Consejo Nacional de Politica Econ6mica y Social (CONPES) 3037, mayo. 360 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE . 1999. "Plan para el Sector de Agua Potable y Saneamiento Basico." Docu- mento Consejo Nacional de Politica Econ6mica y Social (CONPES) 3031, abril. . 2000 "Sistema de Servicio Publico Urbano de Transporte Masivo de Pasajeros de Bogota-Seguimiento." Documento Consejo Nacional de Politica Econ6mica y Social (CONPES) 3093, noviembre. . 2001. Partzcipacwin privada en znfraestructura. perfiles de proyectos. Bogota, octubre. . Boletin SISD, nimeros 28 y 30. Duarte Guterman and Cia Ltda. 1996. Bases de una Politica Nacional en Transporte Urbano Bogota. Economic Commission for Latin America and the Caribbean. 1992. The Transport Systems of Latin American Cities. How They Might Better Serve the Needs of the Poor. Figueroa, Oscar. 1996. Lineamientos para la Definici6n de una Politica de Transporte Urbano para Colombia Ministerio de Desarrollo Econ6mico, Bogoti. Fondo de Prevenci6n Vial. Accidentalidad Vial en Colombia 1999. (Sin fecha). . Accidentalidad Vial en Colombia 2000. Resumen. (Sin fecha). Geotcnica S.A 2000. Estudiopara elDiseFio ConceptualyPredisefio T&cnico delSistema de Transporte Masivo para elArea Metropolitana de Bucaramanga. Bucaramanga. Giraldo Isaza, Fabio. 1997. "La Politica de Vivienda en los Noventa." En CENAC-Viceministerio de Vivienda, Desarrollo Urbano en Cifras, No. 2, abril. Halcrow Fox. 2000. Review of Urban Public Transport Competition. Department for International Development, London. Hidalgo, Dario. 2001. "El Sistema de Transporte Masivo de Bogota." Ponencia pre- sentada al Primer Semimario Internacional de Transporte Urbano "Transmilenio, la Experiencia en Bogota." Bogota, noviembre International Institute for Environment and Development (IIED). 2001. Public- Private Partnerships and Environmental Services for the Urban Poor London. Jouralev, Andrei. 2001. "Regulaci6n de la calidad de los servicios de agua potable y alcantarillado." Ponencia presentada en el taller institucional "Construcci6n Par- ticipativa para la Nueva Regulaci6n de los Servicios de Acueducto, Alcantarillado y Aseo en Colombia." Cartagena, octubre. MacAllister, Edgar. 2001. "Operaci6n del Sistema Transmilenio." Ponencia presen- tada al Primer Seminario Internacional de Transporte Urbano "Transmilenio, la Experiencia en Bogota." Bogota, noviembre. Metro de Medellin Ltda. 2001 Transporte Masivo de Mediana Capacidad Metro Plus. Avance del proyecto y propuesta de gestidn para su ejecucidn en el periodo 2001-2003. Bello, febrero. Ministerio de Desarrollo Econ6mico. 1995. Ciudades y ciudadania: la Politica Urbana del Salto Social." Bogota. . 2001. "Deficit de Vivienda." Documento preparado para el BID, junio. . 2001. "Fortalecimiento de la Direcci6n de Vivienda." Documento preparado para el BID, junio. URBIiAN DEVELOP ENIT 361 2001 "Marco de Referencia para el Sector Vivienda." Documento preparado para el Banco Interamericano de Desarrollo, junio. Ocainpo, Jose Antonio 1996 "Participaci6n privada en la provisi6n de servicios sociales. el caso colombiano " En Fedesarrollo-Instituto SER. Coyuntura Social, (14), mayo Ortiz, Alexandra. 1999. "The Effects of Local Planning Tools on Access to Real Estate Assets by the Urban Poor: Implications for the Design of Urban Projects." World Bank internal discussion paper, June 24. Washington, D.C Piez, Juan Alberto. 2000 Asesoria al Departamento Nacional de Planeacidn a Traveis del Proyecto PPC I En Temas Atinentes a la Regulaci6n del Sector de Transporte Puiblico Terrestre Automotor. Bogota. Presidencia de la Rep6blhca. 1994. Plan Nacional de Desarrollo 1994-1998 El Salto Social. Departamento Nacional de Planeaci6n, Bogota . 1998. Plan Nacional de Desarrollo 1998-2002 Cambio para Construir la Paz Departamento Nacional de Planeaci6n, Bogota. Rodrfguez, Daniel A. 2000 "Planeaci6n de Transporte y Uso de Sueloe eAutomovil- idad o Accesibilidad?" En Ricardo Montezuma, ed., Presentey Fututro de la Movil- zdad Urbana en Bogotd. Retosy Realidades. Veeduria Distrital, Bogoti. Rubianio, Gonzalez, y Granados. 1998 Migraciones Laborales. Sinchez Triana, Ernesto. 2000. Evaluaci6n del Sector Agua Potable y Saneamiento Bdszco en Colombia. PNUD. Ministerio de Desarrollo Econ6mico, Proyecto COL /92/001 Bogota, noviembre. Sarmiento A, and C. Ramirez C. 1998. "El indice de condiciones de vida: una propuesta para la distribuci6n." Steer Davies Gleave. 1999 Dzsefio Operacional del Sistema de Transporte Publico Colectivo de Santa Fe de Bogotd "Transmilenio " Bogota. TransMilemio S A. 2000. TransMilento, Un Sistema de Vida Bogota, diciembre Uni6n Temporal TTC-SYSTRA-GGT. 2001. DiseFio Conceptual del Sistema de Transporte Masivo del Area Metropolitana del Centro Occidente-AMCO (Perezra). Pereira. Universidad Nacional de Colombia. 2000. Estudio para la Elaboracct6n del Plan Piloto de Trdnsito y Transporte del Municipzo de Ibague Ibague. World Bank 1995 Colombia-First Santa Fe Water Supply and Sewerage Rehabil- itation Project. Staff Appraisal Report. Washington, D C. . 1996. Colombia. Bogotd Urban Transport Project Staff Appraisal Report, Report No. 14901-CO. Washington, D.C. 1996 Sustainable Transport. Priorities for Policy Reform. Washingtoni, D C 1997. Colombia-Second Munilcipal Development Project (formerly known as Water Sector Project) Project Information Document. . 1999 Colombia-Cartagena Water Supply, Sewerage and Environmental Management Project. Project Appraisal Document. 1999. Colombia-Cartagena Water Supply, Sewerage and Environmental Management Pro)ect, Report No. PID4507. May 4 Washingtoni, D C 362 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE _._____ 2000. Colombia-Water Sector Reform Assistance APL, Report No. PID9655. November 8. Washington, D.C. . 2001. Colombia-Water Sector Reform Assistance Project. Project Infor- mation Document. Washington, D.C. . 2001. Colombla-Cali: Towards a City Development Strategy, Report No. 20607-CO, August. Washington, D.C. World Development Indicators. 2000. Oxford University Press. INTERVIEWED OFFICIALS AND EXPERTS Departamento Nacional de Planeaci6n Francisco Lozano Maria Constanza Garcia Alfredo Sarmiento Camilo Correal Sandra Alvarez Ministerio de Desarrollo Econ6mico Carmimia Moreno Consuelo Onofre Piedad Rodriguez Superintendencia de Servicios Puiblicos Domiciliarios Blanca Doris Garcia Alcaldia Mayor de Bogota Carmenza Saldias Empresa de Acueducto y Alcantarillado de Bogoti Diego Fernindez Expertos privados Ana Maria Ferreira Fabio Giraldo Isaza 12 Transport This Chapter was written by Mauricio Cuellar. I. Background At the end of the 1980s and the beginning of the 1990s Colombia carried out a series of policy and institutional reforms aimed at modernizing the State.' These reforms took place within the broader strategies of opening the economy to international competition and the reduction of protectionism of the industry and local services There was an effort to stimulate the participation of the private sector in activities previously developed by the public sector, and the decentralization of service provi- sion to subnational entities was promoted, as established by the 1991 Constitution. At the end of the 1990s, the changes in the economic and social environment, due particularly to the armed conflict, led the government to focus on solving the armed conflict, which resulted in a slowing down of the implementation of the reforms, gen- erating institutional, financial, and investment uncertainties across sectors. At the beginning of the 1990s, when the opening of the economy began, it was evident that the country did not have the necessary infrastructure and transport sys- tem to support that opening For example, (a) the railroad system was in disrepair and undergoing a major overhaul of its operations, (b) conditions for navigation on the Magdalena River had deteriorated badly, (c) there were considerable supply restrictions due to protection policies regarding local air and sea transport enter- prises, (d) the efficiency levels of the port terminals were among the lowest in the region and their tariffs were the most expensive, and (e) the road system was very poor in terms of condition and coverage. I This Chapter was prepared with the guidance of Aurelio Menendez, Senior Transport Economist, LAC Transport Cluster German Ospina (Consultant) assisted with data collection 363 364 COLOMBLA. THE ECONOMIC FOUNDATION OF PEACE During the same period, investments in the sector aimed at reaching the follow- ing key development objectives: (a) enhanced road accessibility to and from impov- erished areas and those with high levels of violence, (b) promotion of competition through provision and improvement of the transportation corridors of foreign trade, and (c) employment generation through the construction and maintenance of road infrastructure.2 The government strategy for the improvement and maintenance of transporta- tion corridors focused on the promotion of private sector participation under con- cession contracts. For the road system, the railroad system, maritime terminals, and the main airports, it was to be implemented through concession contracts with the private sector. An aggressive plan of private participation was then initiated. Transport is essential for development in Colombia. Transport-related activities represented nearly 5 percent of the country's gross domestic product in 2000, and generated approximately 8 percent of total employment. Adequate transportation is essential to promote foreign trade and internal demand, and to move products from production centers to ports and consumption centers. From 1994 until 1996, load mobilization in terms of tons per kilometer decreased, while the gross domestic product increased during the same period. From 1996 to 1999 the performance of both indicators has been similar. Figure 1 shows this relationship Figure 1. Transportation and Economy, GNP, Tons per Kilometer 80,000,000- 70,000,000 - - --- - - 60,000,000 -_ _ _ _ _ _ _ - 50,000,000 -__ _ _ 40,000,000 __ _ _ Z 30,000,000- _ -= ~~~~GNP .... River 20,000,000 - -.--- . Road - - _ Total 10,000,000 - - - Railways 0 1 - - I I 1994 1995 1996 1997 1998 1999 2000 2 Periodic maintenance contracts have generated 4,400 monthly employments (the average contract duration is five months) In addition, the Instituto Nacional de Vias (National Highway Institute, INVIAS) has carried out routine maintenance through microenter- prises, generating 3,700 permanent )obs (1999 figure). Regarding maintenance of rural roads, approximately 400,000 wages have been produced in the last two years. TRANSI'ORT 365 In 1999, in Colombia, approximately 220 million tons (includinig cargo trans- ported by pipelines and multiducts) were transported, represenrting a 20 percent increase over 1995. Between 1995 and 1999 there was an increase-from 48 to 60 percent-in the volume of cargo transported by pipelines and railroads. This was due to the increase in the volume of oil and coal that tised these modes of transport. Regarding the volime of cargo moved (not including cargo moved through pipelines and multiducts), 70 percent was done by road, 25 percent by railroad (mainly coal), 4 percent by river, and 1 percent by air. The majority of general cargo and containers in Colombia is moved over roads. Despite the reduction in volume in recent years, mostly due to the armed conflict and financial problems, this mode has had higher reliability levels, timeliness, and security Table I shows thc historical trend of cargo movement in Colombia by mode of transport From 1995 to 2000, Colombian exports increased from 42.7 million tons to 62 9 million tons, almost a 48 percent increase Most of the exports were made by sea (98 percent), particularly through specialized private terminals (coal, oil, and banana) In regional port terminals, the volume of exports was stable at approxi- mately 3.5 million tons, and consisted mostly of general cargo and containers 3 Air exports were approximately 250,000 tons, and road exports to neighborinig coun- tries were approximately I million tons Imports were stable at approximately 12.5 million tons. The higher movement was made by sea (90 percent), but in contrast to exports, the movement of special- ized ports represents only 10 percent of the total. The terminals of regional port companies mobilized 50 percent of imports, and there was a 30 percent reduction of air imports compared to 1995. There was also a reduction in road mobilization of imports originating in the Andean Community. With respect to the transportation of passengers, in 1999, 97 percent of total pas- sengers transported used roads, followed by air (8 percent), and river (5 percent). Railroad transport is almost null. The trend in recent years has been a ieduction in Table 1. Cargo Mobilization at National Level (thousand of tons per year) Railroad Sub- Pipe- Year Road Internal Paz Rio Coal River Air total lines Ducts Total 1995 86,742 882 2,438 17,054 2,634 140 109,890 48,859 23,405 186,154 1998 84,350 779 2,332 21,842 3,049 119 112,471 71,453 24,747 212,680 1999 77,674 367 2,076 27.478 3,735 134 111,464 80.875 22,822 216,546 Sources Milnistry of Tratisport (MT) 2001, MTI' Cargo Surveys 3 T he handling of chis export cargo has been linder the responsibility of the ports formerly managed by Colpuerros (a national government agency Barranquilla, Carragena, and Santa Marta, in the Atlantic Ocean, and Buenaventura and 'lTumaco in the Pacific Ocean 366 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE the number of passengers moved by all modes of transportation, possibly because of the security situation and the current economic crisis. Because of the importance of the road subsector within the transport system in terms of cargo and passenger mobilization, and because of its great importance in giving access to poor, isolated populations, this Chapter will emphasize this subsector. ROAD SECTOR. Colombia has approximately 146,000 kilometers of roads, of which 16,500 (11 percent) are national roads managed by the Instituto Nacional de Vias (National Highway Institute, INVIAS),4 100,000 kilometers are managed by the Departments (68 percent), 15,000 kilometers are managed by the Fondo de Caminos Vecinales (10 percent), and the remaining 14,000 kilometers have undefined owner- ship. Seventy-seven percent of the roads managed by INVIAS are paved (12,705 kilometers). Each network is defined by its ownership/management instead of by its functionality, which often generates a duplication of functions and difficulties in the preparation of the road inventory. The extension of the road system in Colombia and its condition are much lower than in comparable countries in the region, and substantially lower than those of more-developed countries For example, the num- ber of kilometers of paved road per million inhabitants in 1995 was 1,340 for Venezuela, 820 for Mexico, 347 for Peru, and 309 for Colombia. Accessibility of roads for the population differs considerably in the various regions. While there are Departments with very good road and transport cover- age, such as Atlhntico, Caldas, and Quindio, there are a large number of areas where inhabitants must travel for several days (such as Choc6, Caqueta, Putu- mayo, the Pacific Coast region among others) to gain access to basic services and to the capitals of the Departments. This isolation is linked to the lack of basic services provided by the State, facilitating the presence of guerrilla and delin- quent groups. Due to the lack of funds allocated to maintenance, the road system deteriorates quickly. Table 2 shows the deterioration percentages of the paved road system man- aged by INVIAS. Table 2. Deterioration of Paved Road System Managed by INVIAS (percent) Condition 1998 2001 Good 78 70 Regular 17 23 Bad 5 7 4 Regarding the national network, 63 percent is paved, 27 percent is unpaved, and 10 per- cent is to be constructed TRANsi'oawr 367 The situation is even more dramatic regarding unpaved roads-those that pro- vide access to poorer areas. According to the condition assessment of this network, and comparing 2000 and 2001, Table 3 shows the following. Table 3. Condition of Unpaved Roads (percent) Condition 2000 2001 Good 45 34 Regular 39 42 Bad 16 24 There is no information on the rest of the road system, however, due to the lim- itation of resources directed to the maintenance of roads, and according to what the transporters express, the deterioration of the secondary and tertiary networks has also increased. The main causes for this deterioration are the absence of sufficient resources to finance maintenanice activities, and the lack of adequate axle-weight controls of cargo vehiicles. Regarding the latter, it is worth noting that after discussions with freight transport operators, the allowed weight per axle was recently increased. Con- sequently, the damage to the road system is likely to increase.5 There is no informa- tion yet about the measures taken or the costs to upgrade the pavements to support the new axle weights. The traffic volume on roads managed by INVIAS remains low, with a national average below 2,000 average daily traffic (ADT). Volumes over 10,000 ADT are shown on only 375 kilometers, corresponding mainly to access roads to the major cities. Thirty percent of the INVIAS network has an ADT below 500. In the depart- mental network, an average of 150 ADT is estimated, and in municipal and rural networks, the average is approximate 25 ADT. Regarding the road traffic composi- tion, there is an average of 70 percent of light vehicles, 10 percent of buses, and 20 percent of trucks. However, in the main road corridors, the truck percentage may reach 60 percent, as is the case of La Lfnea in the Bogota-Buenaventura corridor. RECENT ACHIEVEMENTS. The following are the achievements reached in the trans- poit sector since the reforms initiated at the beginning of the 1990s * The government has carried out institutionial reforms that attempt to clarify the roles of each sector entity in planning, regulation, control, and execution. In addition to the planning activities, the Ministry of Transport is responsible for policymaking, the Regulatory Transportation Commission (Comlsz6n Reg- 5 In operations carried out in recent years, it was found that 10 percent of the trucks are overloaded 368 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE uladora de Transporte) is responsible for its regulation, the Superintendency of Transport and Ports is responsible for control and enforcement, and the implementing agencies are in charge of each mode of transportation. Due to its nature, the air mode is regulated by another entity. o The sector has directed its resources to support economic growth through investments in the main network, which links the large production and con- sumption centers to each other, and these with maritime ports, mainly through private participation. The sector has also supported employment- generation programs and contributed to the peace process, to the extent that through roads, the sector gives access to depressed and high-violence-level zones. * The participation of the private sector through concessions has extended to all modes of transport and administrative levels (central government, depart- mental, and municipal entities), from the development of infrastructure to operation and maintenance. Table 4 shows the activities taken over by the pri- vate sector in the last decade. The road concession program has represented an important and valid alternative to deal with the problems related to invest- ment slowness in road infrastructure, although it presents difficulties that hin- der its continued development and that generate a financial drag for sector resources. The port reform has shown positive results due to increased pro- ductivity and reduction of costs and, therefore, rates; reduction of service time to ships; and investments made in new technologies. Also, the high corrup- tion levels suffered by the old State-owned port agency have been eliminated. This increase in port productivity does not anticipate the need to build new port premises in the medium term. Regarding the air sector, three main air- ports and the second runway of the Bogota Airport have been given under concession contracts, with results still uncertain. Regarding the railroad sec- tor, two networks were given under concession contracts (Atlantic and Pacific), and their results have not been evaluated yet. o The actions taken by the government to reduce accidents and road hijackings have yielded positive results. Regarding measures taken against hijacking in particular, modern tracking systems have been implemented, leading to important reductions in land hijacking. The Ministry of Transport has adapted transit monitoring of vehicles through the use of location devices (the Global Positioning System, GPS) and the implementation of reaction pro- grams by the road police o The development in Bogota of a clear and consistent policy of urban trans- port has generated a new image of the city, reproduced in other cities within the country and in the region. The Transmdlenzo system, costing only 10 per- cent of what a heavy metro system costs, has opened the possibility for other cities to consider solving their public transport problems through a modern, cost-effective system with a high level of service, thus avoiding considerable and unsustainable investments. Table 4. Private Participation in the Transport Sector Type of Infrastructure! Type of Networkl Responsible Present Service Installation Entity Name of Project Scope Condition Roads National INVIAS 13 road concessions of first and second generation 2,418 km C, 0 National INVIAS 4 toll concessions in national roads 55 toll stations 0 National INVIAS/ Departments 2 national road concessions managed by Valle and Magdalena 122 km 0 Departmental Departments 3 road concessions in Cundinamarca and I in Valle 0 Department de Antioquia Jose Maria C6rdoba Airport - Las Palmas Credit/toll 0 Airports National Unidad Second runway - El Dorado Concession 0 Administrativa Especial de Aeronautica Civil Cartagena, Barranquilla, and Cals terminals Concession 0 Ports National Superintenden- Barranquldla, Buenaventura, Cartagena, and Santa Marta, cia Portuaria Sociedades Portuarias Regionales Concession 0 Railroad National Ferrovias Pacific and Atlantic networks Concession C Transportation Municipal Office of Mayor, Bogota License Servicios de Trinsito y Transporte Project, plates Concession 0 Office of Mayor, Cali Management of vehicle registries and driver licensing Concession Being assembled C = Construction O = Operation 370 COLOMBIA T HE ECONOMIC FOUNDATION OF PEACE II. Main Sector Issues and Strategies Notwithstanding the above-mentioned achievements, there still remain aspects to be developed and reviewed. Within the national context, the armed conflict has an important effect on the sector development, and the sector has become part of the national strategy to help reduce the armed conflict by seeking access to isolated areas where the guerrillas operate a) Armed Conflict Negatively Affects the Functioning of the Sector Like other sectors of the economy, the transport sector has suffered the effects of the armed conflict experienced throughout the country The increase in road hi)ackings, illegal armed posts (retenes), road closings due to actions of subversive groups where users are massively kidnapped (pescas milagrosas), and blockades caused by peasant mobilizations have generated great losses for the nation's economy. The situation has affected both the development and operation of transportation, with the following consequences: * The factors that cause most road closings are public disturbances, landslides, and anticipated work (for repairs). In 2000, of the total hours of roads closed (25,890 hours), 30 percent were caused by landslides, 14 percent were antic- ipated, and 56 percent were caused by disruptions.6 Road hijacking has increased considerably in recent years, leading to large losses and to insurance companies rejecting cargo insurance. Obviously, this situation affects the development of multimodal transportation due to the risk of assuming responsibilities in cargo management. In the case of passenger transportation and vehicle transit, the lack of secure roads and the pescas milagrosas have caused a reduction in the traffic volume and closing of roads at night. Equally important has been the negative effect in the financing of road concessions, particularly those of first generation where the State guaranteed minimum traffic. The obvious result of the security situation is that foreign investment in the sector has been discouraged. * Destruction of the transportation infrastructure by blowing up railroad, bridges, and toll stations further creates costs for the economy. The railroad system has also been affected by blowups of the track that links coal exploita- tion sites in the Cesar Department with the port. Also, the control by armed groups of the basins of navigable rivers has discouraged use of the river mode of transportation. * The displacements to major cities caused by the conflict are creating strong pressures on service provision, transportation being one of them Generally, 6. Twenty-three percent of these closures due to disruptions were caused by hijacking and theft, 37 percent by guerrilla actions, and 40 percent by community blockades. TRANSPORI 371 these people move to the outskirts of the cities where the availability of trans- portation services is reduced and its provision more expensive. b) Isolation ofAreas Due to Lack of Transportation Infrastructure One of the main causes of the social and economic deterioration of the various rural zones is their isolation due to lack of a transportation infrastructure. The lack of physical access to basic social services such as education and health, and to produc- tion and consumption centers, is an exclusioll factor There appears to exist a high correlation between the zones of higher sociopolitical conflict and the deficient road development of the various modes of transportation. This aspect was identified in the development of the Plan Nacional de Rehabilitactin at the end of the 1 980s and in the recent Road Plan for Peace (Plan de Vfas para la Paz) as part of Plan Colom- bia Currently, the construction of a series of roads to integrate wide regions into the rest of the country is being made through Plan Colombia (see Annex Table A. 1 for a list of the roads included in the Plan de Vias para la Paz) These investments seek to benefit the poorest rural popLilations in critical zones of the conflict through the following actions. (a) improvemenit of the internmodal transportation infrastructure, (b) generation of temporary employment through the intensive use of the local labor force, and (c) promotion of civil participation in the execution and sustainable maintenance of works. In addition, conservation and reconditioning programs of rtiral roads are being developed, as are the construction and improvement of river port premises INVIAS is responsible for the V/as para la Paz works which, upon completion, will be devolved to the Departments for maintenanice. c) The Unfinished Development of the Institutional and Regulatory Framework In 1 992, a reform of the transport sector was carried out with the goal of adapting its institutional nature to the new development strategies. To further the decen- tralization process, the road system to be managed by the central government was defined, and the remaining system was transferred to Departments and Munici- palities. At the beginning of 2000, a new institutional reform was created (Decree 101) This reform sought to cover the institutional gaps left by the 1992 reform The Ministry of Transport was given the lead role in transport policies at the national level, and its activity was focused on the formulation and adoption of poli- cies, general plans, programs, and projects in the matter of transit, transportation, and infrastructure, transferrinig the regulatory responsibilities to a Transportation Rcgulation Commission (Comitsin de Regulacidn de Transporte), similar to those existing in other public service sectors.7 Vigilance and control activities regarding transit, transportation, and infrastructure were transferred to the Superintendency 7 Energy and Gas, TelecomMLnillcationis, and Water, Sewerage, and Waste collection Regu- lation Commissions 372 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE of Transport (Superintendencia de Transporte),8 and the specialized entities contin- ued to be responsible for the execution of works.9 Implementation of this new organization began in 2001, and there are still some gaps and duplication of func- tions that need to be addressed.10 Annex Table A.2 shows the functions of the sec- tor entities The following addresses the reform aspects (including the ongoing decentraliza- tion process) and regulatory subjects where the progress initiated at the beginninig of the 1990s needs to be consolidated or resolved. i. Institutional * The purpose of having a Transport Ministry focused exclusively on plan- ning and policymaking was truncated, since it continues to be responsible for the execution of river works and dredging access channels to maritime ports. Continuing its role as an execution entity, biases its fundamental activity as policymaker due to budgetary, political, regional, and contract- ing pressures. * The institutional framework seeks to define the roles and specializations of each entity of the sector at the national level. However, this framework is not so evident at the regional level. For example, the jurisdictional scope of the Regulatory Commission or of the Superintendency is not clear about the stan- dards that must be applied in transport operations at the Departmental and Municipal levels, including concessions."1 * The Transportation Regulatory Commission (Comzs,dn Reguladora de Trans- porte) remained within the Ministry of Transport as a unit with neither legal status nor its own budget. This makes the Commission very vulnerable to political interests and eliminates the independence that, by nature, a regula- tory entity must have. This situation makes the Ministry, in fact, the deposi- tory of the regulatory role. 8. Created from the old Port Superintendency, established in 1991. 9 Except control and vigilance of the air sector, which remained under Aeroci vil. 10 Entities at the national, departmental, and municipal levels are contracting out road interventions through concessions without coordination In principle, the Regulatory Commission is in charge of those activities for the whole sector. However, in pracrice, departmental and municipal transit authorities are issuing regulations without consulting the Regulatory Commission 11 For the execution of major works, some Departments, such as Antioquia, have estab- lished special tolls, and others, such as Cundinamarca and Valle, have awarded regional concessions The lack of regulatory clarity and coordination of concessions has created problems in the various entities, particularly regarding the location of toll stations. For example, in the geographic area of the Sabana de Bogota, there are over 20 toll stations, and recently the municipal administration proposed the installation of six new access sta- tions to the capital. FIRANSI'ORI 373 ii. Regulatory * Regulation plays an important role in structuring and enforcing free compe- tition and tariff control. Transportation tariffs, which until 1997 were defined by the market, are now regulated by the Ministry. At present, the Ministry manages a maximum and minimum tariff band where, theoreti- cally, operators are positioned according to the operating costs In general, the increase of tariffs between 2000 and 2002 has been 19 percent. In prac- tice, tariffs are negotiated directly between the owners of the cargo and the operators. However, tariff regulation and negotiations have become major issues in the relation between the government and the sector. On the other hand, the profitability of freight operations has been affected by an oversup- ply of transport services (of approximately 40 percent), a simultaneous reduction in the demand caused by the economic crisis, and a considerable increase in operating costs. * Trade with the neighboring countries of the Andean Group has remained sta- ble in spite of the existing difficulties in border crossings. Notwithstanding existing agreements, border crossing between Colombia and Venezuela and Colombia and Ecuador continues to require truck transshipment. * Although reduced, accident levels remain high, with approximately 6,500 fatal accidents during 2000. Although the Ministry is working on the pre- vention of accidents, higher priority is given to delinquency problems * Regarding urban transportation, the regulatory treatment of municipalities is the same across all types and sizes. There is no national policy regarding urban transportation. * The Ministry is responsible for the information records of the various ele- ments and parties involved in transit and surface transportationi.2 At present, the technological development of territorial transit entities is deficient and does not allow for complete and reliable information. This leads to deficien- cies in the support of decisions taken at the local and national levels, poor service to users, insecurity and deficiency in the collection of fees, lack of con- trol and high inefficiency in processing of paperwork and, as a consequence, corrupt practices in the issuance of licenises. iii. Road Decentralization Process * The road decentralizationi process was accompanied by the diminishment of the role of the Fondo Nacional de Caminos Vecinales The fund still exists, and it appears it will continue to exist for the foreseeable future. However, its role is being limited to fulfilling political and regional requirements not necessar- ily responding to any program, except for those interventions included under the Plan de Viaspara la Paz. Some investments in the tertiary (rural) road net- 12 This information includes, among other things, registry of vehicles and motorcycles, national registry of drivers and offenders, penalties, and taxes. 374 COLOMBIA THE ECONOMIC FOUNDATION OF PEACE work have been made using cofinancing mechanisms with the Municipalities. The presence of the Fondo in the Departments is very limited. Since 1993, Departments and Municipalities have become responsible for the management and maintenance of the secondary and tertiary road systems. The Departments took charge of several roads transferred from the national government (INVIAS and Office of Caminos Vecinales) but its present main- tenance status is unknown. The transfer of road segments to the Departments has been suspended (there still remain 15,012 kilometers to be transferred) and the continuation of the process iS uncertain. This situation could gener- ate the reversion of the process with the possibility that the national govern- ment again may become responsible for the secondary and tertiary roads. * The foregoing is due to the lack of clear financial mechanisms (free of politi- cal pressures, as was the case of the Fondo de Cofinanczaci6n Vial) to support the decentralization of the secondary and tertiary road networks to the Departments. Given the lack of resources, the regional authorities did not find any type of incentive to assume responsibilities over those roads and, besides, the enhancement of the technical capabilities of the departments has not been sufficient. Therefore, the ability and readiness of the subnational governments to develop the technical and administrative capacity to effec- tively manage the road system within their jurisdictions and to secure the nec- essary resources to finance the required maintenance activities remain of high priority to advance the decentralization agenda in the road sector. d) Ensuring Adequate Resources to Address the Needs of the Transport Sector Investments in the transport sector have been limited by the financial restrictions of the entities that depend on the national budget (INVIAS, Caminos Vecinales, Ferrovias, Minstry of Transport). These restrictions have been more dramatic in recent years, reaching their lowest level in 2000. In 2001, appropriations for the sector increased as a result of the investments in the Plan de Vias para la Paz. Sector investment in 2000 was similar to that of 1991 (COP700,000 in 2001 pesos, approximately US$310 mil- lion). Between 1991 and 2001, however, investment grew considerably, reaching a maximum in 1996 (COP2.1 billion in 2001 pesos, approximately, US$930 million). The dependence on treasury resources, together with the existing complex budgetary processes, make investments more difficult each time. Figure 2 shows budgetary appropriations for investment in the transport sector in the last 12 years. Note that in 2001 (year 12) a small investment recovery was experienced. Public investment in the transport sector has been focused mainly in the road, airport, and railroad subsectors. In 2000, 71 percent of the total budget was allo- cated to roads. Financing came from the national budget and own resources. Table 5 shows budget allocations during 1998-2001. In the few years following its creation in 1993, INVIAS became an effective entity, capable of preparing multiannual investment plans, and with in-house tech- TRANSPORT 375 Figure 2. Transport Sector Investment 2,500,000 - 2,000,000 -__._ ___ __ _ __ u 1,500,000- v-___ 0 0 1,000,000- --- -_7 v 500,000 - _ 0 I 1 I I I I I I I I I I 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 nical capacity to confront the challenges of modernizing the road system. However, its high dependency on the national budget generated great vulnerability in the sus- tainable allocation of its resources. When the economic crisis of the end of the 1990s arrived, INVIAS's budget was reduced considerably. Its technical capacity was also negatively affected In addition, INVIAS's management was continuously reshuf- fled, and with that came a decrease in its efficiency in the management of contracts. INVIAS's own resources are limited to toll collection. Other financing sources are valorization contributions and tax surcharges on diesel, though the latter is not significant. In 2000, gross toll collections amounted to COP230,000 million, rep- resenting approximately 40 percent of the entity's budget (commitments) However, the trend of this source of resources is decreasing due to decreasing traffic volumes and the award of new road concessions to which toll collections in the respective roads are transferred. Twenty-three percent of the 2000 budget was directed to investments in new construction, 27 percent to concession projects,'3 30 percent to improvement and maintenance of the system, 10 percent to the microenterprise program for maintenance, and 10 percent to payment of recurring expenses.'4 Among these percentages can be noted the impact of concession guarantees on INVIAS's budget. The calculation of resources estimated only for the maintenance of the road sys- tem under INVIAS (80 percent in good condition, 15 percent in regular condition, 13 Corresponds to the guarantee payments of the concession contracts 14 Corresponds to the purchase of land, study programs, payment of court decisions, main- tenance of equipment, and so forth 376 COLOMBIA. THE ECONOMIC FOUNDATION OF PEACE Table 5. Transport Budget 1998-2001 (millions of pesos) Investment Item 1998 1999 2000 2001 INVIAS 949,612 961,178 561,866 1,318,029 Caminos Vecinales Funds 30,387 18,240 63,836 165,274 Roads Cofinancing Fund 22,753 0 0 0 Rivers 7,374 11,445 12,017 27,634 Railroads 24,031 119,112 127,616 187,563 Maritime ports and access chainels 12,904 5,900 5,873 8,524 Airports and air transportation 92,662 109,200 98,000 119,887 Superintendency of Transport 0 0 18,207 7,588 Roads operation 1,290 594 1,902 0 Total 1,141,013 1,225,669 889,317 1,834,498 1998 1999 2000 2001 Investment Item (percent) (percent) (percent) (percent) INVIAS 83.2 78 4 63 2 718 Caminos Vecinales Fund 2 7 1.5 7 2 9 0 Roads Cofinancing Fund 2.0 0.0 0 0 0 0 Rivers 0 6 0.9 1 4 1.5 Railroads 2.1 9.7 14.3 10 2 Maritime ports and access channels 1.1 0.5 0.7 0.5 Airports and air transportation 8.1 8.9 11 0 6.5 Superintendency of Transport 0 0 0.0 2 0 0.4 Roads operation 0.1 0.0 0 2 0.0 Total 100.0 100.0 100.0 100.0 and 5 percent in bad condition) is COP650,000 million per year (approximately US$282 million). The available resources to serve the national road system in 2001 were COP250,000 million (US$108 million), representing 40 percent of the total required. Departmental and Municipal resources allocated to interventions on roads come from fuel and dsesel tax surcharges and the vehicle tax. The tax surcharge is 5 per- cent over the fuel cost and 3 percent over the diesel cost.15 Also, the national gov- ernment assigns the proceeds from vehicle taxes to the Departments and Munici- palities for their use in any type of investments. Departments also have access to the Royalties Fund (Fondo Nacional de Regalias).1I Departments may also collect tolls on 15 There is a Subsidy Fund of Fuel tax Surcharge to subsidize Departments with low con- sumption levels. 16. This fund supports road, river, port, and airport projects. The largest portion of these resources is allocated to investments in roads In 1999, the approved resources reached approximately COP105,000 million, of which COP98,000 corresponded to the road sector. TRANSI'ORT 377 their roads and establish valorization contributions.'7 Notwithstanding these financ- ing sources, there are no clear regulations on the collection and use of these resources. The air subsector is financed with own resources. These are the Airport Tax (Tasa Aeroportuaria) and other duties paid by users and resources derived from conces- sions.'8 In financial terms, this subsector is self-sustaining because airports with income resources higher than costs subsidize airports with deficits. Thus, a financial equilibrium of the airport system has been achieved without major difficulties It is important to maintain this scheme since, in their structure, the private participation processes must take into account surplus margins to allow the operation of loss-mak- ing airports, but that are paramount for the country's integration e) Private Participation in the Development of the Sectors Infrastructure As mentioned, in the 1990s, the participation of the private sector was at the heart of the strategy for the development of transport projects. Through concessions the government sought the allocation of private sector resources to execute works in the transport sector. The participation of the private sector through concessions has extended to all modes of transportation and administrative lcvcls (National, Dcpart- mental, and Municipal), not only regarding infrastructure, but also in the operation and provision of services. The participation of the private sector has even extended to activities reserved for the State, such as the issuance of driver's license manage- ment and vehicle plates. The participation of the private sector has been essential for the development of the transport sector. Nevertheless, there has not been a participation strategy integrated into the trans- portation system. Thus, in the road sector as in others, development has been iso- lated, generating in some cases competition among the various modes of trans- portation. The participation of the private sector in concessions in the railroad and river systems could be financially uncertain due to the cargo volumes mobilized by these modes (except for coal, which is transported in large volumes by the railroad), and has caused a financial drain on the national government in those cases in which the contracts included government guarantees for minimum revenues. With the exception of oil and coal, which have their own specialized means of transportation, cargo volumes transported within the country could be covered by the existing sup- ply of road services. On the other hand, there is no development in dry docks or intermodal installations to facilitate multimodal transportation, especially in the 17 Thus, departments such as Antioquila and Cundinamarca collect tolls on some of their roads and impose contributions from valorization (or property revaluations) for works executed 18 In 2000, total collection reached COP251,000 million, of which approximately 57 percent was assigned for operation, 41 percent for investment, and 2 percent foi debt payment "**==>This document did not complete OCR process. <==**"