Document of The World Bank FOR OFFICIAL USE ONLY Report No. P 7483- CO REPORT AND RECOMMENDATION OF THE PRESIDENT OF THE INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT TO THE EXECUTIVE DIRECTORS ON A PROPOSED STRUCTURAL FISCAL ADJUSTMENT LOAN IN THE AMOUNT OF US$400 MILLION FOR THE REPUBLIC OF COLOMBIA November 16, 2001 Colombia-Mexico-Venezuela Country Management Unit Latin America and the Caribbean Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. REPUBLIC OF COLOMBIA-FISCAL YEAR January 1-December 31 CURRENCY EOUIVALENTS (as of 25 October 2001) Currency Unit = Peso 2340.00 Pesos = US$1 1.12 Euros = US$1 WEIGHTS AND MEASURES Metric System SELECTED ABBREVIATIONS AND ACRONYMS AFORE Administradoras de Fondos para el IMF International Monetary Fund Retiro IPS Instituciones Prestadoras de Salud ARS Administradora del Regimen ISA Interconexi6n Electrica SA Subsidiado ISAGEN Interconexi6n Electrica SA CAS Country Assistance Strategy (Generadora) CAS-PR Country Assistance Strategy ISS Instituto del Seguro Social Progress Report LDP Letter of Development Policy CMU Country Management Unit NFPS Non-Financial Public Sector DANE Departamento Administrativo PAYG Pay-As-You-Go Nacional de Estadistica PBG Policy Based Guarantee DB Defined Benefit PFMP Public Financial Management Project DC Defined Compensation POS Plan Obligatorio de Salud DECPG Development Prospects Group PPP Purchasing Power Parity DNP Departamento Nacional de QAG Quality Assurance Group Planeaci6n SDR Special Drawing Right EPS Empresa Promotora de Salud SFAL Structural Fiscal Adjustment Loan EUR Euros SGP Sistema General de Participaciones FINDETER Financiera de Desarrollo SISBEN Sistema de Identificaci6n de Territorial Beneficiarios FOGAFfN Fondo de Garantias Financieras SNSSS Sistema Nacional de Seguridad Social FONPET Fondo de Pensiones Territorials en Salud FOPEP Fondo de Pensiones Publicas SOAT Seguro Obligatorio de Accidentes de FOSYGA Fondo de Solidaridady Garantia Trdnsito FSAL Financial Sector Adjustment Loan SSAL Social Sector Adjustment Loan GDP Gross Domestic Product TES Colombia Treasury bonds IBRD International Bank for UPC Unidad per Capita de Capitacion Reconstruction and Development URC Unified Registry of Contributors IDB Inter-American Development Bank WBG World Bank Group IFI International Financial Institutions WEO World Economic Outlook IBRD Vice President David de Ferranti Chief Economist: Guillermo Perry Director, LCSPR Ernesto May Country Director: Olivier Lafourcade Lead Economist/Manager: Marcelo Giugale Task Managers: Vicente Fretes-Cibils Marcelo Giugale Team Production Support: Michael Geller This operation was prepared by a World Bank team composed of Messrs/Mmes. Rojas (LCSPS); Webb (LCSPE); Giugale, Chal]a, Fretes-Cibils, Velez Bustillo, Cuevas, Partow, Geller (LCC] C); Escobar, Uribe (LCHH); von Gersdorff (LSCFR); Carvalhc, (LEGLA); del Valle, Zhang (FSD); de Henao, Lezaca (LCCCO); Fernandez, Gottret, and Panopoulou (consultants). The tean was led by Messrs. Fretes-Cibils and Giugale (LCC 1C), and worked under the general guidance of Mr. Olivier Lafourcade (Director, LCCIC). FOR OFFICIAL USE ONLY REPUBLIC OF COLOMBIA FISCAL STRENGTHENING PROGRAM-STRUCTURAL ADJUSTMENT LOAN TABLE OF CONTENTS LOAN AND PROGRAM SUMMARY .............................................................i I. BACKGROUND AND RATIONALE ............................................................1 II. THE STRUCTURAL CHALLENGES OF FISCAL STRENGTHENING ........................................ 5 Intergovernmental Fiscal Relations ..........................................................5 Education ............................................................6 Health .............................................................7 Pensions ............................................................9 Public Sector Reform ............................................................ 10 Public Sector Debt .............................................................11 III. THE ECONOMIC REFORM PROGRAM OF THE COLOMBIAN GOVERNMENT .. 11 A. Government Program Policies ............................................................ 12 B. The Program's Outcomes ........................................................... 17 IV. THE PROPOSED STRUCTURAL FISCAL ADJUSTMENT LOAN (SFAL) .. 20 A. Objectives ........................................................... 20 B. Content ........................................................... 21 C. Risks ............................................................ 23 D. Poverty Impact ........................................................... 25 E. Loan Amount, Conditions, Financial Safeguards, and Disbursements and Audits ........................................................... 25 V. THE MACROECONOMIC FRAMEWORK FOR THE PROPOSED SFAL .. 28 VI. BANK STRATEGY ............................................................ 28 A. The CAS and The Proposed SFAL ............................................................ 28 B. Complementarity with Other World Bank Operations and Activities ................... 30 C. Complementarity with Other Institutions' Operations .......................................... 31 VII. RECOMMENDATION .. 31 TEXT TABLES Table 1. Colombia-Public Sector Fiscal Balances Scenario with Reforms (2000-05) ...... 18 Table 2. Colombia-Matrix of Policy Actions for Tranche Releases of SFAL ...... 21, 22, 23 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. l FOR OFFICIAL USE ONLY TEXT FIGURES Figure 1. Colombia: Public Sector Balance, Real GDP Growth, and Poverty Rates (1978-99) ..........................................................3 Figure 2. Colombia: Public Sector Balances Under Alternative Reforms Scenarios (1999-2005) ......................................................... 19 Figure 3 Colombia: Net Public Debt Under Alternative Reforms Scenarios (1999-2005) ..................... .................................... l 9 TEXT BoxEs Box 1. Key Social Impacts of the Policies Supported by the SFAL ................................. 20 Box 2. The Colombian Constitutional Court ......................................................... 24 ANNEXES A. COUNTRY ASSISTANCE UPDATE NOTE ......................................................... 33 INDICATORS Al. Colombia at a Glance ......................................................... 41,42 A2. Key Economic and Program Indicators-Change from Last CAS ............................. 43 A3. Selected Indicators of Bank Portfolio Performance and Management ........................ 45 A4. IBRD/IDA Program Summary ......................................................... 47 A5. IFC and MIGA Program Summary ......................................................... 49 A6. Summary of Nonlending Services ......................................................... 51 A7. Colombia Social Indicators ......................................................... 53 A8. Key Economic Indicators ......................................................... 55-57 A9. Key Exposure Indicators ......................................................... 59 Al 0. Operations Portfolio (IBRD/IDA and Grants) ......................................................... 61 All. Statement of IFC's Held and Disbursed Portfolio ....................................................... 63 B. THE MACROECONOMIC FRAMEWORK .65 C. FISCAL SUSTAINABILITY, ECONOMIC GROWTH, AND POVERTY .73 D. PUBLIC SECTOR DEBT-STRUCTURE AND SUSTAINABILITY. . X..............91 E. LETTER OF DEVELOPMENT POLICY .97 F. LETTER OF PRESIDENT PASTRANA ............................................................. 109, 110 G. SPECIFICATIONS, DEFINITIONS, AND EVIDENCE FOR TRANCHE RELEASE CONDITIONS .11 H. MANAGEMENT OF PUBLIC FINANCES ........................................ 123 I. MAP This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OFFICIAL USE ONLY REPUBLIC OF COLOMBIA FISCAL STRENGTHENING PROGRAM-STRUCTURAL FISCAL ADJUSTMENT LOAN (SFAL) LOAN AND PROGRAM SUMMARY Borrower: Republic of Colombia Amount and Terms: US$400 million fixed-spread loan, with automatic interest rate fixing whenever disbursements reach an aggregate amount of US$20 million; final repayment maturity of 13.5 years with level repayments of principal, each equivalent to 10 percent of the loan amount. Commitment charge: 0.85 percent p.a. for the first four years, and 0.75 percent p.a. thereafter. Front-end fee: 1 percent of the Loan amount, to be paid by the Government out of its own finances. Implementing Agency: Ministry of Finance and Public Credit Description: The SFAL is the Bank's response to the government's demonstrated commitment to reaching an inflection point in Colombia's fiscal accounts path, an essential first step toward achieving full fiscal sustainability, economic growth, and poverty reduction. The program to be supported by the SFAL focuses on a core set of policies that can strengthen the country's fiscal outlook. This policy package involves (a) rationalizing the system of transfers to local governments, and imposing more market- driven and more binding budget constraints on their finances; (b) establishing mechanisms to arrest the exploding cost of inefficiency in the provision of public health services; (c) halting the accumulation of pension- related contingent liabilities; (d) advancing the reorganization of public agencies and their current expenditures; and (e) setting up a better system for managing public debt. The combination of measures to improve governmental fiscal relations, rationalize public sector expenditures, and improve financial management of the public sector's assets and liabilities will lead to a more sustainable fiscal path and to greater efficiency in allocation of public resources, including those allocated to the social sectors. Both are necessary to sustain economic growth and poverty- reduction efforts, and to improve the provision of key public services. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. i FOR OFFICIAL USE ONLY Risks and Benefits: The proceeds of the loan will help to implement fiscal reforms, provide a confidence-building signal to international markets, reduce the cost of borrowing, and lengthen debt maturity. These will, in turn, grant the government the fiscal space for protecting social sector expenditures and implementing difficult structural measures. In addition to their fiscal implications, the measures supported by the SFAL have social impacts that have been identified and evaluated, and which will provide the basis for additional policies to protect the most vulnerable. The SFAL carries both internal and external risks. The main internal risks include (a) the fragility of the overall political context. Prolonged social disruption and resistance from interest groups could derail the reform agenda, and the internal conflict could further escalate. The government is committed to continuing its dialogue with the guerrillas, and is working to mitigate political and social risks through building a consensus with a variety of interest groups. (b) Blocking or substantial alteration by the opposition-led Congress of the policy initiatives in the program. This risk is mitigated because the bulk of the reforms have already been discussed with Congress. The SFAL has also sought to include reforms that require Congressional approval up front in the program or in the floating tranche, and has stressed measures to be carried out by the Executive. (c) A reversion to fiscal mismanagement. The government's commitment to reform militates against this, as do hard budget constraints faced by subnational governments. The loan also requires a set of nonreversible actions prior to tranche disbursement. The most important external risk is that of a deeper downturn in the world economy. This could be transmitted to the Colombian economy through a further deterioration in terms of trade and the closure of access to international financial markets. In this regard, the government intends to continue its flexible exchange rate policy and retains access to IFI financing, particularly the IMF through continued fulfillment of conditions and the right to draw upon resources under its Extended Arrangement. Schedule of Disbursements: The loan is expected to be disbursed in three tranches, of US$160 million in December 2001, US$180 million expected in March 2002, and a US$60 million floating tranche to be disbursed upon fulfillment of its specific conditions. Project ID Number: CO-PE-P073572-LEN-BB This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. ii REPUBLIC OF COLOMBIA FISCAL STRENGTHENING PROGRAM STRUCTURAL FISCAL ADJUSTMENT LOAN (SFAL) President's Report I. BACKGROUND AND RATIONALE 1. Starting in the mid-1990s, Colombia entered an unsustainable path of overall fiscal imbalance-by the end of the decade, central government expenditures had grown by about 8 percentage points of GDP, while revenues, despite no fewer than eight tax reforms, increased by only about 2.5 points. As a result, the total gross stock of public sector debt as a percentage of GDP almost doubled (to about 50 percent by 2000). Such a fragile fiscal position made Colombia particularly vulnerable and ill prepared to respond to the economic recession that would unfold in 1998. This was triggered by a combination of a sharp deterioration in the country's terms of trade, the effect of the East Asia and Russia crises, and a weakening of investor confidence due to stepped-up guerrilla and illegal drug activities. The downturn, the first in 70 years, revealed critical structural weaknesses in the government accounts-notably, the encroachment of entitlements in favor of politically powerful sectors, unbridled subnational spending, the deterioration in quantity and quality of social services, and the substitution of current for capital expenditures and weaknesses in the banking sector (its mortgage industry all but collapsed, adding to the fiscal malaise). In the end, Colombia, a country previously reputed for its sound macroeconomic management, lost its sovereign investment grade and saw its access to international financing tightly restricted. 2. To deal with the immediate effects of the recession, the then incoming Pastrana Administration formulated a three-year stabilization program for the period through 2002 based on exchange rate flexibility (the peso was floated in September 1999), fiscal stabilization, and the implementation of structural reforms, including financial sector restructuring and privatization. The international financial community mobilized to support that program, and the IMF Board of Directors approved in December 1999 a three-year, SDR 1.96 billion Extended Arrangement Facility-the first of its kind for Colombia. This arrangement was entered into with a clear understanding on the part of the Colombian authorities that (a) it was precautionary in nature, (b) it was meant to send a confidence-building signal to international markets, and (c) it would not be drawn upon unless there was a pressing balance-of-payments need. The authorities have proceeded along the precautionary path in not drawing upon the arrangement, and despite the deteriorating external environment, to date this has proven to be the right policy decision (the spread of the country's sovereign borrowing rate continues to decline and is gradually converging to the level of investment-grade borrowers such as Mexico).' 3. The World Bank joined the stabilization efforts. Its Board approved in November 1999 an EUR 482 million Financial Sector Adjustment Loan, the second tranche of which would be 1. However, should an international crisis occur and the financial markets further tighten access for all emerging country borrowers-particularly following the tragic attacks of September 11, 2001 and subsequent events- Colombia's drawing upon the IMF arrangement would not be perceived as a negative signal about the country's economic management. I converted, in February 2001, to a Policy Based Guarantee (PBG) operation of up to about EUR 238 million, fiurther reinforcing credibility in intemational markets. Moreover, the government requested last year that the World Bank take the lead in identifying and supporting the implementation of key structural reforms to (a) complement the stabilization effort under the IMF program; and (b) help bring Colombia's fiscal accounts onto a sustainable path in the medium- and long-term. The operation proposed in this document is the Bank's response to suoh request, and reflects our extensive dialogue with the authorities regarding fiscal structural weaknesses. 4. The implementation of the stabilization program produced some positive results-key indicators, such as growth, inflation, and the fiscal balance have improved; the extemal position and financial system have been strengthened (see Annexes A and B); and compliance under the above-mentioned IMF and World Bank operations has been satisfactory. The 1998--99 recession had a devastating social impact and provided a crucial policy lesson in this regard. In Colombia, there has been a close positive correlation between growth and poverty reduction (see Figure 1 and Annex C). During periods of positive growth (and fiscal balance), Colombia enjoyed substantial declines in poverty (even though inequality increased). Specifically, fr2,m 1988 to 1995, as average GDP grew at about 3.5 percent per year, poverty rates declined frcm about 65 to 60 percent of the population. This reduction, and the government's efforts that helped achieved it, were completely wiped out by the 1998-99 recession. Poverty levels today are not substantially different from what they were in the late 1980s (and inequality has increased). More immediately, unemployment climbed to and has since remained at, almost 20 percent of the labor force. Put differently, the evidence suggests that sustainable growth has been Colombia's best social safety net.3'4 5. That link between growth and poverty in effect dictates Colombia's economic policy agenda over the next two years-the priority is to sustain first, and then consolidate, the observed, albeit timid, recovery of growth, especially in light of the impacts of the events of September 11. For several reasons, the key to sustaining and then consolidating econoiriic recovery is the strengthening of the fiscal accounts. First, in spite of important progress, the initial round of stabilization-related reforms did not attend to the structural, sector-specific 2. Since the approval of the IMF Extended Arrangement, two reviews were successfully completed in September 2000 and March 2001, and the third review is expected to be completed by January 2002. The borrowing facilitated by the World Bank's PBG has been used to meet the country's financing needs for 2001. While it provided market access at a critical time and contributed to the decline in Colombia's sovereign bond spread over the U.S.'s sovereign bond, that borrowing has been relatively expensive (probably a reflection of investors' concern over delayed structural reforms and potential contagion effects from other emerging markets in che region). 3. As part of the authorities' efforts to deal with the immediate, recession-driven unemployment problem, the World Bank approved a US$100 million Community Works Project (January 2000), and a US$150 million Huiran Capital Protection Project (March 2001). While these interventions mostly deal with recession-drixen unemployment, it is recognized that there are structural rigidities in the formal labor market including high payroll taxes, binding minimum wages, and inflexible contractual arrangements, causing structLiral unemployment. The government will tackle these rigidities to make the labor market more flexible through labor reforms supported by the forthcoming SSAL. 4. A recently completed draft World Bank report on the evaluation and causes of Colombia's poverty reconfirns the formidable, dominant role that economic growth plays among the determinants of the country's poverty levels. 2 imbalances that have weakened, and continue to weaken, the country's fiscal position. In the timid economic recovery path, these weaknesses will further push down economic growth. Second, the recovery has been smaller than previously anticipated and it will now take longer to achieve significant positive economic growth; thus, tax revenues have not met expectations, and will be lower than previously anticipated. Third, the period leading to next year's Congressional and presidential elections will exacerbate political pressures toward spending. Fourth, an intensification of the internal conflict, and the violence associated with illegal drug activities, could dilute investor confidence in Colombia, from an already low point. Finally, further deterioration in the external environment, contagion from financial crises in major markets and/or a major emerging economy, terms of trade shock caused by a global recession (especially the U.S. economic downturn, and the impact of oil and coffee price declines), or a combination of the above, could restrict Colombia's access to international capital markets even more (as explained later, the financing scenario for the rest of 2001 and for 2002 calls for significant foreign funding). In brief, in the absence of further structural fiscal strengthening, the probability of a sharp, recession-driven rise in poverty is high-a probability that has been heightened by the events since September 11. Figure 1. Colombia: Public Sector Balance, Real GDP Growth, and Poverty Rates (1978-99) 85- 6 ; 75- ; > ~~~~1990-1995 5 e 0 - 19852002< _5 4 3 ~ 5 1978 1988 ~~~~ ~ ~~1995 1999 .70 ,~65 - > ~~~~~~~~~~~~~~~~~~~~~~~~-2 o60 -3 55 ~~~~~~~~~~~~~~~~~~~~~~~-4 1978 1988 1995 1999 _ Average Real GDP Criowth Rate - Average NFPS Balne (Percentage of GDP) - Poverty Rate Source: National Accounts, IMF, and World Bank. 6. The Colombian authorities understand the urgent need for policy action, and, as indicated above, have requested the World Bank's technical and financial assistance in the design and implementation of a wide-reaching fiscal strengthening program, to be supported by the proposed US$400 million Structural Fiscal Adjustment Loan (SFAL). This President's Report (a) summarizes the diagnosis of the main sectoral imbalances that lie behind Colombia's fiscal weaknesses; (b) describes the policy package that the SFAL will support; and (c) explains the strategic fit, risks, and modalities of the operation. 3 7. At the outset, it is crucial to notice that, on the one hand, the sociopolitical situation in the country severely limits the government's space for reform (notably, through the dynamics of the ongoing civil conflict and of the upcoming elections) and, on the other, many of the required reforms are long-term undertakings that will take time to be fully implemented and show results. There is, however, strong government commitment for reform, and ownership of the refonn program is widespread-this is a coalition government and the program has been discussed with, and has the support of, the main political parties, ensuring both viability and continuity. The government should be able to fully implement its fiscal reform package within this administration's term, and leave the fiscal accounts on a more sustainable path for the next administration, smoothing both the political and, more important, the economic and social transitions. The program to be supported by the SFAL is, nevertheless, not intended to deal exhaustively and conclusively with all areas where action could improve Colombia's fiscal outlook. Rather, it is conceived as a first-step package that focuses on a core set of critical policies that can sufficiently strengthen and consolidate that outlook. In particular, the SFAL- supported policy package involves (a) rationalizing the system of transfers to local governments (especially education-related transfers), and imposing more market-driven and more binding budget constraints on their finances; (b) establishing mechanisms to arrest the exploding cost of inefficiency in the provision of public health services (particularly the cost of public hospitals and the health arn of the Social Security Institute); (c) halting the accumulation of pension- related contingent liabilities; (d) advancing the reorganization of public agencies and their current expenditures; and (e) setting up a better system for managing public debt. 8. The Country Assistance Strategy Progress Report (CAS-PR), discussed by the Banlk's Board in November 1999, recognized that the performance of the Colombian economy-most notably, economic growth and fiscal stability-had deteriorated sharply during the two years since the presentation of the last full CAS. As a result, restoring the fiscal balance Was determined as being crucial for the success of Colombia's efforts to recover growth and reduce poverty. The CAS-PR also recognized that the government had made reducing the fiscal deficit one of its key macroeconomic objectives, and that the worsening economic situation, including fiscal imbalances and financial instability, could lead to financing gaps and problems in accessing external financial markets on reasonable terms that, in turn, would prevent the achievement of the CAS objectives. In light of the deteriorating external environment, strengthening the fiscal account remains as crucial as before, if not more so, for sustaining and consolidating economic recovery, and preventing the deterioration of social conditions. The CAS-PR moved the assistance strategy to a high-case lending scenario, including adjustment operations, and justified this movement due to (a) the need to cushion the adverse social impact of the economic recession, financial instability and internal conflict on the poor and vulnerable groups, and (b) the government's commitment to undertake fiscal adjustments and implement structural reforms to be supported by an IMF program. As explained in the Country Assistance Update Note (Annex A), the reasons and triggers to operate in a high-case scenario not only remain current but have also been enhanced by recent developments-a deteriorating external macroeconomic environment that threatens growth sustainability, the continuing govermnent commitment to fiscal discipline, and the strength of the reform agenda. 9. The SFAL is the Bank's response to the government's demonstrated commitment to reaching an inflection point in the country's fiscal accounts path-an essential first step toward achieving full fiscal sustainability. The ambitious policy agenda that the operation will 4 support-as elaborated in the government's Letter of Development Policy (LDP), see Annex E-has been discussed and designed over a series of technical missions starting in early 2001.5 Moreover, given the likelihood of social impacts from these policies (see paragraph 54), the agenda envisions that the SFAL would be followed up and complemented by a Social Sector Adjustment Loan (SSAL) that would enhance the social sustainability of the reforms. II. THE STRUCTURAL CHALLENGES OF FISCAL STRENGTHENING 10. The Colombian government has maintained its commitment to a program of fiscal adjustment, despite the difficult economic, political, and security conditions. The authorities have made advances in implementing a number of key reforms, including strengthening the financial system and passage of legislation to (a) allocate funds for financing local government pensions; (b) strengthen the fiscal position of local governments; and (c) increase core taxes, including the presumptive income tax, the value added tax (from 15 percent to 16 percent), and the financial transactions tax (from 0.2 percent to 0.3 percent). However, along with these advances, there have been some setbacks, most notably (a) a court order in October 2000 requiring backward indexation of public sector wages; and (b) the failure to privatize two large enterprises in the electricity sector (ISA and ISAGEN). To ensure sustainability in the medium term, the government must urgently deepen the fiscal reforms by addressing some additional fundamental structural weaknesses that are discussed below. 11. INTERGOVERNMENTAL FISCAL RELATIONS. Colombia has been recognized as one of the leaders in decentralization in Latin America, and has led the region in explicitly acknowledging that subnational governments have different management capabilities. The constitutional reform of 1991 and ensuing legislation transferred sizeable revenues and responsibilities to subnational govermments.5 The transfers were established through formulas and include (a) general-purpose transfers for municipalities; and (b) earmarked transfers for education and health, primarily to the departments. In 1993, the central government began a gradual process to transfer education and health resources and responsibilities to those subnational governments that demonstrated sufficient capacity and received certification. In addition to promoting results orientation, the decentralization included a market-like set of rules and monitoring mechanisms to regulate subnational borrowing. The Ley de Semaforos (1996) introduced measures and market signals of subnational indebtedness capacity, with the intention of transferring the cost of risky lending to commercial credit institutions. 12. By the end of the 1990s, however, transfers to subnational governments had become a major source of the fiscal imbalances for the central government, primarily because the transfer of resources preceded the transfer of full responsibility. The resulting lack of accountability led to a decline in the quality of public services, notably education and health. In reality, the transfer of responsibilities has been limited because many expenditure decisions, such as the teacher payroll (including compensation and benefit levels) are still controlled by the central government. Since the central government intervenes and is too weak politically to resist demands of special interest groups (for example, teachers unions), local expenditures have 5. A collection of detailed policy notes underpinning the agenda was produced and is available on request. 6. The major revenue sources of the transfers are (a) the so-called Situado Fiscal, (b) the Participaciones Municipales, and (c) part of oil and mining royalties. 5 continually increased. The problem has been compounded because the central government carried out the certification process for decentralization of services under political pressure, meaning that the standards for certification were largely ignored and the central government approved the transfer of education resources and responsibilities to most departments, even to those without adequate capacity. With the fiscal problems of the subnational governmenl.s caused, to a large extent, by the interference of the central government, the latter has consequently been forced to provide an implicit guarantee of subnational debt. This has created perverse incentives and moral hazard behavior when commercial banks provide financing to the subnational entities. 13. The two major transfers (the Situado Fiscal and the Participaciones Municipales, which absorb over 40 percent of central government current revenues) have not sufficed to cover increases in the subnationals' current expenditures, particularly the growth of the teacher payroll. The central government has come to the rescue with additional temporary transfers, which then become permanent. However, these additional transfers have frequently been insufficient to pay for expenditure increases, and most subnationals have inadequate tax bases to make significant local revenue contributions. These elements have resulted in a soft budget constraint for the subnationals and, as a consequence, several departments and municipalities have overspent and are heavily indebted. Moreover, the national government has carried out significant "bailouts" of subnational governments and their creditors, despite its limited capacity to effectively monitor subnational fiscal efforts and the provision of services. 14. In summary, when the central government kept control of human resource policy and collective bargaining, it blurred the division of responsibilities (especially in education) and created perverse incentives, leading to fiscal deficits, demands for additional transfers, increased moral hazard, and bailouts. When resources were insufficient to guarantee coverage and national quality standards, or national policies restricted mobility of input factors (particularly labor), subnational governments had little choice but to spend funds beyond the allocations provided by the central government (as they did for teacher salaries, for example) and to run deficits that were financed by increased borrowing. This put pressure on the central government to bail out the subnational governrments. Bailouts took many different forms, both hidden (as in the case of Treasury credits in 1997-98 for a selected list of departments) and explicit (as in the case of Law 617 of 2000). At other times, the central government paid for subnational responsibilitics directly with resources from the Treasury, resorted to the creation of special funds (the compensation fund, and/or emergency funds), or transferred higher royalties. 15. EDUCATION. Public expenditure in education has expanded rapidly during the last decade-from about 3 percent of GDP in the early 1990s to about 6 percent by the end of the decade. This expansion has been determined primarily by the supply of inputs, particularly teachers, generating inefficiencies in the education system and neglecting significant parts of the demand for education. 16. Over the last decade, education expenditures increased at both the national, and more important, subnational, levels. The expansion has mainly resulted from the implementation of the 1991 Constitution and subsequent decentralization laws that mandated the decentralization of education services. The decentralization was to be financed mainly through two sources (a) the Situado Fiscal-transferring resources to departments, and (b) the Participaciones-transferring 6 resources to municipalities. Beginning in 1995, however, to help the transition toward a full decentralization of education services, a compensation fund increased transfers from the central to the subnational governments. This compensation fund has become permanent and has partly financed the emerging fiscal imbalances at the subnational levels. 17. As a result of the large allocation of resources to education and slow expansion of the school population, the unit cost of providing public education has increased over the long term, particularly in the last decade-the unit cost of public education doubled in real terms during the 1990s. Moreover, teacher salaries and wages represent about 80 percent of the total cost of public education, and the share has increased continuously since the early 1990s. The total supply of teachers-distributed unevenly geographically-and not the student educational needs, has driven the growth in education expenditure, resulting in a suboptimal allocation of resources per student.7 18. Despite expenditure increases, educational coverage is not universal, and inequality, in terms of access by the poor, remains an important issue. Improvements in educational achievements have also been slow-Colombia took about 35 years to significantly raise the average number of years of formal education for people over 18 years of age, from about 3 years in the mid-1960s to about 7 years in late 1990s-and, the urban-rural formal education gap has not changed over this period. The average level of education for adults is still below comparable countries in the region-for example, average formal schooling for adults is 8 years in Mexico, and 13 years in Argentina and Uruguay. Moreover, the public education system remains internally inefficient-there are high repetition and dropout rates at all grade levels, particularly in primary school-despite the improvements over the last decade. 19. HEALTH. The Social Security Institute (ISS) and the public system of hospitals are absorbing a ballooning amount of public resources and, unless tackled and reversed, that trend threatens fiscal stability. As a result of the law reforming the provision of health services (Law 100 of 1993), together with the decentralization law (Law 60 of 1993), public resources allocated to health have increased from about 1 percent of GDP in 1991 to about 4 percent in the late 1 990s. Health insurance coverage has increased from about 20 percent of the total population in 1992 to about 57 percent in 1997, while the quality of health services has improved. Despite these improvements, emerging financing imbalances in the ISS health services, the high cost of public hospitals, increasing evasion of contributions and supply subsidies-delaying the transformation of fiscal transfers to demand subsidies-are threatening the stability of public health services. They are, indeed, undermining both overall fiscal sustainability and the objectives of the health system reforms-namely, universal coverage of health services through a solidarity system, with managed competition of insurers and providers, and resulting improvements in equity and efficiency. 20. Health Reform. The health reform of 1993 (through Law 100) created a mandatory universal health insurance system. Individuals could choose and affiliate to any insurer (Empresa Promotora de Salud, EPS) that receives the individual's contribution (4 percent of 7. To overcome these problems, the special district of Bogota, with legal attributes of both a department and a municipality, has started to implement education reform using its own resources. This may become an example for other fiscally strong municipalities to take over and reform basic and secondary education. 7 salary from employees and 8 percent from employers) as an insurance premium. Independent workers would contribute 12 percent (over two minimum salaries), and the State would subsidize the contribution of poor individuals. In turn, individuals and their families would have acce<;s to a defined health benefits package (Plan Obligatorio de Salud, POS). A solidarity and guaraitee fund (Fondo de Solidaridad y Garantia. FOSYGA) would monitor and administer the contributions, paying each insurer the risk-adjusted premium for the affiliated individual (called Unidad Per Capita de Capitacion, UPC); the fund would also function as a redistribution mechanism, allocating one percentage point of the collected salary-based contributions to subsidizing insurance premium for the poor. The insurers would contract, on a competitive basis, with health care providers (Instituciones Prestadoras de Salud, IPS) to offer health care services to beneficiaries. 21. Users of the new system were divided into two groups (a) those with income capacity to pay the contribution ("contributory regime"); and (b) those without income capacity to pay ihe contribution (the poor), and who thus needed to be subsidized ("subsidized regime"). Under this system, municipalities have to identify the poor for affiliation to the subsidized regime, using a beneficiary identification system. Moreover, selected individuals can choose the insurer of the subsidized regime (Administradora del Regimen Subsidiado, ARS), which receives from municipalities a UPC per beneficiary with the corresponding cost for a basic health benefit package plus an administration fee. The ARS could enter into contracts with public and priv ate health care providers. 22. The ISS-Health Problems. In the context of overall health system reform, Law 100 also gave the ISS the legal form of a public industrial and commercial company. As such, it could manage retirement pensions, workers' compensation, health insurance (ISS-EPS), and health service delivery (ISS-IPS). In health, ISS-EPS and ISS-IPS business are in practice a simple financing operation, which remains inefficient and unable to compete with the private seclor. These problems arise primarily from (a) mixing the financing (EPS) and the delivery (IPS) of services, with large hidden cross-subsidies; (b) difficult labor issues, including large pension liabilities and labor rigidities; (c) high fixed costs relative to service provided, making both ISS- EPS and ISS-IPS not profitable and noncompetitive in the market; and (d) poor management, including inadequate planning and control, causing abrupt and costly decisions. In addition, poor quality of services and noncompliance with minimum health and financial indicators led the Superintendency of Health to prohibit registration of new affiliates by the ISS-Health (until, in the Superintendency's assessment, those minimum standards were fulfilled), compounding its financial and economic problems. 23. ISS health is both illiquid and insolvent. The consolidated ISS-EPS and ISS-IPS incoirme statement has registered losses over the last five years; cumulative losses over this period totaled about US$440 million, and losses for fiscal year 2000 amounted to about US$70 million. lTo cover these losses, in the past, the Treasury provided transfers or credits. Moreover, the consolidated ISS-EPS and ISS-IPS balance sheet shows negative equity starting in 1994, reaching a negative US$270 million as of 31 December 2000, and the accumulated overdue debt with private and public hospitals surpassed US$250 million as of the end of 2000. 24. The Public Hospital System's Problems. In the context of health reform, the public network of hospitals, health centers, and other health facilities was supposed to (a) compete with 8 and complement the private network of health providers, and (b) sell services to the affiliated population through the EPSs and ARSs. Simultaneously, since the National System of Social Security in Health (Sistema Nacional de Seguridad Social en Salud, SNSSS) did not cover the entire population, the public hospitals provided services to those not affiliated with an insurer. The implementation of health system reform is an ongoing process, and during the transition several problems have emerged. These include (a) public hospitals have not fully adjusted their cost structures, thus becoming less competitive than private health providers and less attractive for the EPSs and ARSs; (b) real labor costs have increased 40 percent since 1995 due to centralized salary negotiations, distorting cost structures even more and requiring large transfers from the Treasury; and (c) the transformation from supply subsidies to demand subsidies mandated by the reform has moved slowly (to about 50 percent by mid-2001 for the main sources of transfers), if not stalled, because of the pressure to continue financing public hospitals through supply subsidies based on historic budgets independently of output.8 25. Health Financing Problems. One of the key problems in the sources of health system financing, affecting both the contributory and the subsidized regimes, is evasion in the contributory regime-in 2000, total evasion was estimated to be about 36 percent of total potential contributions, representing about US$900 million. Evasion results from two main sources (a) under-declaration of income by individuals who already contribute in the system, and (b) lack of contribution payments by individuals. In addition, the cumbersome transfer system results in delays that translate into accumulation of cross debt at different levels-that is, from subnational governments to insurers and from insurers to both public and private providers. Moreover, subnational governments use most of their resources for payments to public hospitals, neglecting the possibility of increasing affiliation in the subsidized regime. As a result, the transformation from supply to demand subsidies (and hence to insurance affiliation) has also been slow. Finally, the new health care and insurance system was intended to cover a limited health benefit package. However, the Constitutional Court ruling of "right to life" expanded health care benefits, creating additional pressure to finance unexpected benefits, including the provision of health services abroad. 26. PENSIONS. The current condition of Colombia's pension system constitutes a major risk to fiscal sustainability over the medium and long term. The pension system is technically insolvent and the annual imbalances in the public sector worker regimes are increasing. The estimated net present value of the 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 problems include generous benefits, benefit guarantees, perverse incentives for reserve management, and recent Constitutional Court rulings expanding pension benefits. These causes are rooted in the slow reform transition (see below) and the large number of workers exempted from the reform. This has resulted in pension expenditures still growing at a fast rate, particularly the pension expenditures for public sector workers. 8. If public hospitals do not change their cost structures and continue to be financed through supply subsidies at the same pace as in the last six years, the total accumulated operating cost of public hospitals over the next 10 years could amount to about US$30 billion (about 35 percent of GDP in year 2000). 9 27. To confront the problems of a financially bankrupt and fragmented defined benefit 1 lDB) 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 DB and defined contribution (DC) pension system-workers could choose between the DB regime run by ISS or the DC regime run by 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 persion funds for public sector 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). The law, however, exempted the pension funds for Ecopetrol (6,000 members), teachers (300,000 members), the armed and police forces (200,000 members but only half with pension rights), and representatives of Congress. Furthermore, the law allowed for an extended transition to the new system, scheduled to be completed by 2014. 28. PUBLIC SECTOR REFORM. Colombia's current public sector structure and fiscal imbalances still reflect practices of a welfare state and state interventionism, which are not consistent with the market-oriented policies and decentralized framework adopted during the last two decades. To correct these inconsistencies and tackle the growing central government fiscal imbalances during the 1 990s, Colombia attempted to revamp the public sector structure through three important reform initiatives. First, during 1992-94, the government adopted a public sector modernization strategy that covered most central government agencies (and a few decentralized ones). The main purpose of the reform was to set up an organizational structure more consistent with the provision of decentralized public services. The reform, however, led only to a modest expenditure rationalization at the central level. Second, in 1998-99, a proposed reform was priniarily targeted to public sector rationalization through streamlining and downsizing of the central government administration, including civil servants. The scope and impact of the reform was significantly limited when the Constitutional Court ruled that some of the decrees enacting the reform violated the Constitution. And finally, in 2000, the government introduced a refo-rm package focusing mainly on fiscal rationalization. 29. The inertia of the public sector structure and its functioning, combined with the numerous attempts to adjust the provision of public services to new trends and country needs, seems to account for the eclectic nature of Colombian's public sector today. For example, agencies and civil servants provide services typically assumed by the private sector (including ancillary services, data processing, accounting, and external auditing), and they coexist with modern regulatory agencies, privatization processes, market-oriented organizations, and demand-drivmn public services. The culture of state enlargement and political patronage competes with the culture of streamlining and efficiency seeking in allocation and production of public services. This bureaucratic dualism has been compounded by political fragmentation, internal conflict, corrupt practices, and high crime indexes that seriously constrain good governance. In addition, legitimate concerns for the protection of human and property rights have often led to misguided state intervention, which has further restricted and complicated the design and implementation of economic policy. Unfortunately, this has distorted market signals and incentives, creating I 0 opportunities for moral hazard behavior, misallocation of public resources, and costly bailouts, hampering the efforts to correct fiscal imbalances. 30. PUBLIC SECTOR DEBT. The current level and structure of Colombia's public sector debt is risky and makes the overall fiscal position vulnerable. Should current trends continue in the absence of comprehensive fiscal reform, public sector debt is likely to become unsustainable. Moreover, Colombia's debt portfolio is characterized by high interest rate and foreign exchange exposures, and high refinancing risks significantly increase the vulnerability of government finances. Management of these exposures is difficult due to the absence of comprehensive debt portfolio and funding strategies. 31. Over the last six years, to finance fiscal imbalances at different levels of government, the authorities have issued large amounts of public debt. As a result, total gross public debt has been rising steadily in absolute terms and as a proportion of GDP (see Annex D). The total gross stock ofpublic debt increased from about 27 percent of GDP in 1994 to about 50 percent of GDP in 2000. A sizeable 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 2000.10 Most of the increase in external debt has been driven by central government borrowing; other levels of government currently face statutory and procedural restrictions on borrowing abroad. Public sector internal debt also rose from the mid-1990s, reaching (in gross terms) 27 percent of GDP by 2000, because both the central government and subnational entities sought financing domestically. As a consequence of the increase in the stock of public sector debt, total public debt interest service increased from about 2.2 percent of GDP in 1994 to about 4.4 percent of GDP in 2000. Interest payments on internal debt increased more than proportionately-almost threefold-from 1 percent of GDP in 1994 to about 2.7 percent of GDP in 2000. III. THE ECONOMIC REFORM PROGRAM OF THE COLOMBIAN GOVERNMENT 32. The Colombian government is well aware of the weaknesses in its fiscal accounts; the shortcomings in its intergovernmental fiscal relations; the inefficiencies in the provision of key public services, including education and health; the increasing fiscal pressure from the pension system and public sector debt; and their negative effects on long-term real economic growth and poverty reduction. The authorities therefore sought to design and implement, with support from the World Bank, a fiscal strengthening agenda that would put the non-financial public sector accounts on a more sustainable path while contributing to enhancing and improving the provision of social services. This section summarizes the rationale behind the various policies in that agenda and compares the projected evolution of Colombia's fiscal accounts under the reform and no-reform scenarios. 9. In net terms, the total stock of public debt represented about 42 percent of GDP in 2000. 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 reference only to gross aggregates because estimates of net debt are available only from 1999 onward. 10. The sources of external credit also changed considerably throughout the decade. 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, bonds gained considerable ground over this period, increasing from about 2 percent in 1990 to over 36 percent of the total external debt of the public sector by 1998. 11 (A) Government Program Policies 33. The policies that make up Colombia's economic reform program, including its overall strategic vision and the fiscal strengthening initiative, are described in the government's letter of development policy (LDP, see Annex E). The authorities' ownership of the program is strong and widespread, and so is their commitment to continue implementing reforms until the very end of the current administration (see letter of President Pastrana, Annex F).1' The program has also been discussed with, and has the support of, the main political parties, ensuring its viability and continuity beyond this administration. Moreover, at the government's suggestion, through a series of seminars the Bank has started a technical dialogue with the economic teams of the main political candidates contending for the Presidency, something that is intended to smooth the economic transition between administrations and facilitate the continuity of these reforms. While supporting the full reform program of the government, the Bank will explicitly requiire certain key commitments for the disbursement of tranches under the proposed SFAL operalion, as elaborated later on. 34. The specific targets of Colombia's fiscal strengthening effort are to (a) increase the primary fiscal surplus of the Non-Financial Public Sector (NFPS)-from 0.8 percent of GDP in 2000 to at least 1.3 percent of GDP in 2002, and to at least 2.5 percent of GDP in 2005; and simultaneously, (b) reduce the NFPS fiscal deficit-from about 3.6 percent of GDP in 2000 to 3.1 percent of GDP in 2002, and to 1.5 percent of GDP in 2005. This will prevent and reverse an otherwise unsustainable debt and debt-service accumulation and will be achieved in a marmer that protects social sector expenditures and vulnerable groups. To deliver those targets, a combination of policies-described below-will enhance revenues and rationalize spending.' 2 35. INTERGOVERNMENTAL FISCAL RELATIONS AND EDUCATION. The government program in intergovernmental fiscal relations includes reforms to (a) make subnational governments filly responsible for expenditures, (b) increase subnational tax revenues, (c) simplify and make more transparent the transfer system, and (d) reduce moral hazard from subnational borrowing ard reinstate hard budget constraints. Key components of the reforms are presented below-for expenditures, the presentation covers both general expenditure policy and the education sector. Given the magnitude of the problem and the characteristics of the health sector, while the general expenditure policy applies, the specific sectoral reforns are discussed separately (see below). 36. Expenditures. The authorities seek to control the cost of general administration of subnational levels and to allocate public resources more closely toward the original design of decentralization in the Constitution-which foresaw an allocation according to service needs by population, with adjustments for poverty, among other factors. To control general administration costs, subnational governments are getting structural adjustment loans to finance downsizing of 11. A strong demonstration of this commitment was evidenced by the fact that in early October 2001, three key Ministers of the Colombian government (Finance, Labor, and Planning) and their technical staffs had two fUll days of meetings in Washington at Bank headquarters just to discuss technical issues on pensions (including reform models and lessons in other parts of the world) in order to absorb and exchange information for finalizing the design of the country's pension reform. 12. These policies also preserve the autonomy and accountability of the local authorities to implement environmental policies, including the management and allocation of public resources resulting from the transfers of oil and mining royalties from the national to subnational governments, particularly municipalities. 12 personnel (severance pay, etc.) under the terms of Law 617 of 2000. In addition, the central government has recognized the need to amend Law 60 of 1993-which governs the sectoral allocation of territorial transfers under the decentralization framework-to ensure consistency with the Constitutional reform (Acto Legislativo). To achieve this, the government has recently submitted to Congress an amendment to Law 60, which includes, inter alia, (a) the automatic certification of municipalities (that are capitals of departments and/or have more than 100,000 inhabitants) to autonomously manage the provision of education services; and (b) specific education performnance goals included in a performance matrix (matriz de desempeno). Moreover, in the event that passage of the amendment to Law 60 is delayed, the government intends to proceed with certification (via performance agreements) of key municipalities for autonomous management of resources, particularly education-related resources. These new, more efficient practices embedded in performance agreements are consistent with the ones that are institutionalized in the proposed amendment to Law 60 of 1993. 37. More specifically, the automatic certification or perfornance agreements will imply that (a) the transfer of resources to municipalities will be calculated on the basis of the number of students who attend class (not the number of teachers on the payroll); (b) minimum coverage goals will be established with the objectives of increasing enrollment and reducing drop-out ratios; (c) certified municipalities will be authorized to reallocate resources from teacher vacancies-both existing and new-within the education sector and to eliminate teacher positions; (d) the certified municipalities will be able to reallocate teachers within the local jurisdiction; and (e) the Ministry of Education will provide technical assistance to certified municipalities, if requested. The central government will publish and otherwise disseminate quarterly performance monitoring and evaluation reports of the certified municipalities. 38. Revenues-Taxes. The government intends to strengthen the subnational tax system- which is in contrast to previous reforms focusing on the national tax system. In particular, it will revise the tax system of territorial entities (Estatuto de Ingresos Territoriales) to rationalize subnational tax power and strengthen enforcement. Specifically, the revision will include measures to regulate (a) the basic principles and rules for all departmental, district, and municipal revenues, including taxes, fees, and special contributions; (b) the distribution of tax revenues by level of government; (c) the tax base and tax rates (or delegation to subnational entities) for liquor, beer, and tobacco excises, vehicle taxation, property tax, and industry and commerce tax, valorization contribution, and participation in surplus value of real estate; (d) penalties for noncompliance by taxpayers or beneficiaries; and (e) tax procedures. In addition to benefiting subnationals, the proposed reforms will also benefit the central government because increases in subnational revenues would decrease pressure for more transfers and bailouts. 39. Revenues-Transfers. The Constitutional reforn (Acto Legislativo) recently approved by Congress, intends to de-link transfers from current revenues, and put the growth of transfers from the central government to the territorial entities on a more sustainable path, thus limiting the growth of central administration's operational expenditures (gastos generales).13 The reform combines the current three major sources of transfers financing-Situado Fiscal, 13. The limits on the growth of the central administration's operational expenditures will be equal to (a) 1.5 percent in real termns per year for 2002-08; and (b) thereafter, a rate of growth equal to the average growth of the central administration's current revenues during the previous four years. 13 Participaciones, and FEC-into the Sistema General de Participaciones (SGP), and includes an eight-year transition period (2002-08). Specifically, the reforms include that total transfers will (a) grow by 2 percent in real terms during 2002-05, and 2.5 percent during 2006-08; and (b) from 2009 increase at the average growth rate of the central administration's current revciiues during the four preceding years. In addition, if real GDP growth exceeds 4 percent in any year of the transition period, SGP resources would be increased proportionally, but only after adjusting for the transfers made during years where real GDP growth was less than 2 percent (2002-05;) or 2.5 percent (2006-08). It is expected that after the Acto Legislativo is implemented, the annual transfers would decline from about 5 percent of GDP in 2001 to about 4.0 percent of GDP in 2010, and would generate fiscal savings equivalent to a total accumulated amount of about 5 percent of GDP over the next 1 0-years. 40. As indicated above, the Constitutional reform of transfers is complemented by the amendment of Law 60 of 1993, which specifies the mechanism for distribution of resources among entities. Within this reform framework, the government seeks to (a) prohibit the distribution of any cash transfers to subnational governments except where the distribution is specified by formula or by law; (b) require that any national decision affecting subnational government spending be accompanied by a transfer that is sufficient to cover the cost; and (c) require the annual publishing of the subnational distribution of budgeted expenditures, along with the regulations and matching grant ratios of each cofinancing program. These reforms, over the long term, will help to (a) reduce earmarking of transfers so that they specify only the sector allocation (for example, education and health), not the economic category (for example, wages and investment); and (b) distribute all transfers either on a capitation basis to pay for social services (as equalization grants to entities with lower than average per capita tax bases) or on a matching fund basis (for functions with national externalities). 41. Subnational Borrowing. The government's reform strategy stresses a series of actions to correct perverse incentives to subnational governments and their lenders, and to begin to eliminate discretion on the part of the national government. In accordance with Law 617 of 2000, since July 1, 2001, the central government has stopped providing territorial entities with bailouts or guarantees on their debt.' It has also revised its regulations concerning debt and borrowing of territorial entities in order to start halting unsustainable borrowing, limiting bailouts with national budget resources, and eliminating discretionality in the treatment of debt. To this end, the authorities have issued a decree whereby the debt of territorial entities contracted after December 31, 2001 and held by commercial and development banks would be subject to a differential capital-risk-weighting grid based on a current published credit ratings performed by an internationally reputable credit-rating agency. 42. In addition to making the capital-risk-weighting for the debt of territorial entities helc by banks dependent on international credit ratings, new banking and securities regulations altow pension funds, insurance companies, and other regulated investment firms to hold territorial entity bonds only if they have been given investment-grade credit ratings by at least two international agencies. Further, the government has revised the regulations of the Ley de Sem6foros to (a) link the "color" of the Semtiforos (that is, the central government's 14. This law allowed for bailouts and national government guarantees of territorial debt up to June 30, 2001 under strict conditions to enter, implement, and perform under fiscal adjustment programs at the local level. 14 authorization to borrow) more clearly to the ratings of the international credit agencies; and (b) restrict commercial borrowing for all entities having loans subsidized under Law 617 of 2000, until they have complied with their respective adjustment program for at least four years. In addition, the government will continue with the practice that each request for subnational government foreign borrowing has to be approved by the National Congress. 43. HEALTH-The ISS-Health Service. The government's main objectives with respect to ISS-Health are to (a) improve the quality of services to affiliates; (b) eliminate inefficient supply subsidies to its ISS-IPS; and (c) limit fiscal costs to the national government. To achieve these objectives, both ISS-Health "stock" and "flow" financial problems need to be addressed-that is, the existing accumulated deficit and its underlying causes both need to be eliminated. To tackle the "stock" problems, the program includes (a) transferring resources to the ISS only after the renegotiation of the collective bargaining agreement with its labor unions has been successfully completed; (b) lifting the sanction applied by the Health Superintendency prohibiting new affiliations to the ISS after the underlying causes of the sanction have been resolved; and (c) withholding compensation transfers from FOSYGA to the ISS until more accurate information on the number of its affiliates has been provided. 44. To resolve the "flow" problem, the government has defined, approved, and initiated the implementation of a restructuring plan for ISS-Health (that is, ISS-EPS and ISS-IPS). This plan focuses on both reducing costs and increasing revenues. Measures for reducing costs include (a) restructuring of accumulated debt; (b) renegotiating the collective bargaining agreement in what refers to primas extralegales (related to vacations, seniority, and services), seniority-based severance, and pension benefits; (c) rationalizing clinics and ambulatory health services; (d) improving contracting mechanisms with own and external health care providers; and (e) effectively verifying beneficiary rights to services, based on improvements in the enrollee database. Measures for increasing revenues include (a) better collection of contributions from affiliates through reforms to the current system of self-estimated contributions (autoliquidaci6n); (b) increased collection of arrears from, among others, FOSYGA, SOAT, and the ISS-ARP; and (c) enrolling more affiliates after the sanction imposed by the Superintendency of Health is lifted. 45. Public Hospitals. As part of its public hospital network reform, the government is committed to (a) reducing the network's cost structure; (b) rationalizing the supply of health services and adapting to actual demand; and (c) making the public network effectively complementary of the private network, thus creating a consistent health system of both public and private facilities. To achieve these objectives, the government is tackling both sides of the equation-namely, supply and demand-simultaneously. On the supply side, last year the government started a pilot program to restructure 27 public hospitals (out of a total of about 170 medium- and high-complexity public facilities). As a result of this program, hospital costs have already been reduced by 5 percent in real terms in just one year. On the demand side, total public resource for health allocation has been partially reformed from supply to demand subsidies, and by mid-2001 about 50 percent of resources allocated to public hospitals were demand-driven. 46. Based on these results, the government will expand its public hospital network reform and implement further measures to reduce costs, including (a) reforming the personnel structure 15 of public hospitals; (b) resolving accumulated debts (to public hospital employees and providers); and, critically, (c) renegotiating collective bargaining agreements with hospital unions, especially relating to seniority-based severance. It will also implement actions to increase public hospital networks' productivity, such as (a) reaching agreements among na-ional, departmental, and municipal governments, and hospital directors to advance the long-lasting restructuring of public hospitals and to develop functional health service networks; and (b) identifying and analyzing existing information regarding supply and demand of health services in different regions in order to better define plans for managerial and technology improvements. In addition, the transformation of resources from supply to demand will continue. 47. Health Financing. To tackle health financing problems, and particularly evasion in the contributory regime, the government's program includes (a) registration and periodic updates of all affiliates of the system in the Health Superintendency's Internet database; and (b) creation of a unified reporting system, with authorizations for entities different from EPSs to collect contributions. In addition, the authorities intend to fully implement Decree 40 of 2000, which allows for payments of subnational debt, including that for the health system, directly from resources to be transferred from the Treasury to subnational governments (that is, departments and municipalities). A study to redefine the content of the health benefit package and the corresponding risk-adjusted premium (UPC) will be launched. Based on the study, the National Council of Social Security in Health (Consejo Nacional de Seguridad Social en Salud, CNSSS) would establish the cost-benefit relationship in the health package, and the level of the risk- adjusted premium. Coverage for health care services outside Colombia will be severely lirnited, and at a later stage, the health benefits package will be redefined by law. 48. PENSIONS. The major objective of the second-generation pension reform is to rmake Colombia's pension system more equitable and sustainable. Thus, members of new cc horts entering the pension system will all have the same rights and obligations and will not increase the current implicit pension debt (IPD). While the proposed reform maintains the culTent situation of concurrent defined benefit (DB) and defined contribution (DC) regimes, it deepens the reform envisioned in Law 100 of 1993 and strengthens the pension system by reforming the overall pension system and/or at least one of the exempted pension regimes. It also seeks to harmonize or introduce a single set of unified pension rights and obligations for all new entrants, while increasing coverage and improving services to all participants. 49. The proposed reform-which includes the reform of the overall pension system thrz ugh new legislation-also envisions, among other things, (a) a deadline for a final switch of current members of the system between the DB and DC regimes; (b) increased eligibility requirenlonts in terms of age and number of weeks of contribution for a pension; (c) centralizing the administration and control of transfers of public pension regimes into a single agency; (d) improving the data-bases required for the recognition and recertification of pensions, (e) pre- funding new accrual of rights to extraordinary benefits; and (f) creating a unified registry of contributors (URC). Moreover, the new legislation reforming the pension system would (a) continue to transform the DB scheme through better individual work records, which wo'uld constitute the basis for pensions at retirement in the ISS-managed scheme; (b) fund the minimum pension guarantees through an insurance scheme; and (c) close entry into any pension fand except for ISS and AFPs. Finally, the reform of the exempted regimes, if advanced---for example, the teacher's pension fund-would contain features such as (a) individual pension 16 saving accounts; (b) a close link between contributions and benefits; and (c) the application of the new regime to all new entrants. 50. PUBLIC SECTOR REFORM. The government's public sector reform effort is framed within the overall fiscal strengthening program described above. The authorities intend to (a) issue a decree for the rationalization of central government current expenditures, notably personnel, and procurement of goods and services, including temporary workers and ancillary services; (b) close redundant national government agencies, and redundant and nonessential decentralized agencies; and (c) build consensus with the Judiciary branch for reduction of the budget allocation to the Superior Judicial Council (Consejo Superior de la Judicatura). The government is also committed to additional public sector reforms with a view to strengthening governance, including deepening the dialogue with the Constitutional Court to enhance the effectiveness of economic policies. 51. PUBLIC SECTOR DEBT. To improve debt management and debt portfolio structure, the government is developing new portfolio and funding strategies, and corresponding benchmarks. These strategies-that integrate fiscal projections with analysis of portfolio risk exposure-will reflect the government' s assessment of an acceptable level of fiscal risk (that is, "budget-at-risk") relative to the expected reduction in the cost of funding, and thus, include the maximum amortization that the government can manage in any given year. The funding strategy will also allow for the use of active debt management tools (for example, debt exchanges, swaps, and buybacks) to enhance the structure of the portfolio. However, significant institutional strengthening will be required to carry out and implement these strategies. 52. To reduce central government debt exposure, a debt portfolio restructuring strategy has recently been defined. As part of this strategy and to tackle the bunching of domestic debt amortizations, in June 2001, the authorities swapped domestic debt due in 2001-05 with debt issues of longer maturities, extending the average maturity of domestic debt from 3.5 to 4.5 years and minimizing the need for rollovers in the next few years. With regard to foreign exchange exposure, the government continues to use the currency composition of international reserves to hedge against net liability positions. In addition, to facilitate the management of the explicit contingent liabilities of the public sector, the government's program includes the development and implementation of a plan to improve the registration and quantification of these types of liabilities. (B) The Program's Outcomes 53. The fiscal strengthening program described above puts the NFPS accounts on a more sustainable path. Each of the proposed measures carries fiscal implications, and they are singled out in the projections through 2005, shown in Table 1 and Figure 2. The combination of measures to improve governmental fiscal relations, rationalize and reallocate public sector expenditures, and improve financial management of assets and liabilities of the public sector brings about both a more sustainable fiscal path-a necessary condition for sustaining economic growth and poverty reduction-and greater efficiency in allocation of public resources in the social sectors-a necessary condition for improving the provision of key public services, particu- 17 Table 1. Colombia: Public Sector Fiscal Balances Scenario with Reforms (2000-05) Percent of GDP 2000 2001p 2002p 2003p 2004p 2005p Total Revenues 28.8 28.7 27.N 27.9 28.0 28.: Tax Revenues 179 10.8 19.3 19.5 197 1999 of which Revenues from Territorial Tax Reform' 00 0.0 0 2 0 2 0.2 0.: of which: Savings due to Evasion Reduction2 0.0 0.0 0 1 0 1 0.2 0o: Non-tax Revenues 10.9 9.9 8.5 8.4 8.3 h.' Property Income 1 7 1.5 1.3 1.1 1.1 I Operational Surplus of public enterprises s5o 4.2 3.9 3 8 3.7 3.7 Other 63 42 33 35 36 31 Total Expenditure and Net Lending 32.4 32.0 30.9 30.4 29.9 29., Current Expenditures 24 7 24.1 23.6 23.0 22A4 22 Wages and Salaries 7.5 7 3 70 6.8 6 7 f of which: Effect of Acto Legislativo3 0.0 0.0 -0.2 -0.3 -0.3 -0 of which: Closure of Agencies4 oo 0.0 -0.1 -01 -0.1 -0.8 Goods and Services and Other 3.5 3.1 3.0 2.9 2 8 2.8 of which : Savings from Public Hospital Restructuring5 00 0.0 -0.2 -.3 -0A4 -0 5 *(includes: Lower Transfers due to ISS health restructuring) Interest 4.4 4.4 4.4 4.2 4.0 4.0 of which: Debt Management' 00 0.0 -07 -0 2 -0.2 -0 3 External 17 20 I'S 1.6 7.S 1.5 Internal 2 7 24 2.6 2.6 2.5 25 Transfers to Private Sector 9.3 9.3 9.2 9.1 S.9 8 0 of which : Savings due to Pension Reform 0.0 0.0 0.0 -0.2 -0 3 -0 4 Capital Expenditure 7.9 7.9 7 3 7.4 7.5 7.5 of which: ISS Health Restructuring' 00 0 3 0.5 0.0 0.0 0.0 Fixed Capital Formation, cash basis 8.5 81 7 4 7.4 7.5 7.5 from floating debt -0.7 -0 3 -02 -0.7 -0.e -0. Transfers 0.1 0.1 0.l 0 X 0.1 0.7 Net Lending -03 0.0 0.0 0o0 0 0 0.0 Nonfinancial Public Sector Balance -3.6 -3.3 -3.1 -25 -2.e -tL Quasi-fiscal balance 0.5 0.5 0.3 0.5 03 03 Fogafin balance 0 0° 0) 0A. 0.7 07 Net cost of financial restructuring -0.4 -0 9 -0.6 -0 2 -0.2 -0.2 Overall Balance -3.5 -3.6 -3.1 -2.1 -1.6 -l.l Overall Financing 3.S 3.6 3.1 2.t 16 1.1 Memorandum item: NFPS Primary Balance, with reforms 08 1. 7.3 1.7 2.1 2.5 NFPS Primary Balance, no reforms 0.8 1.4 09 0.3 0.2 0o A'ote: p =projected 1. Increased annual revenues, ranging from US$150 million to US$200 million, are projected from strengthening of the territorial entities' tax system. Departmental and municipal tax revenues are projected to eventually increase by about 15 and 25 percent, respectively. R.-forms include broadening of tax bases and increase in tax rates at the territorial level. 2 Annual tax collection increases, on average, of about US$200 million during 2002-O5 as a result of increased tax complianve and enforcement. 3. Operating expenses of the public sector fall by about US$350 million annually during 2002-05 resulting from the Xcto Legislativo. The Acto Legislativo sets a 2-percent ceiling on the real growth rate of transfers from the national to subnational govemments. 7he Acto Legislativo also sets a 1.5-percent ceiling on the real rate of growth of national govemment operating expenses. 4. Average annual savings of US$130 million result from closure of nonessential and redundant public entities and agencies. 5. Annual savings averaging US$350 million during 2002-05 results from restructuring of public sector hospitals. 6. Annual savings on interest payments averaging about US$275 million are projected through 2005. Savings on interest payment,s results from a 250-to-400-basis-point fall in average interest rates (relative to a no-reforms scenario). It is projected that a 1-percentage point fall in the debt-to-GDP ratio leads to a spread reduction of 7 basis points in average interest rates. 7. Average annual savings of US$230 million are projected during 2002-05. Savings accrue as a result of strengthened application of reforms already introduced through Law 100 of 1993. These reforms include (a) completion of the transfer of administration and control o i public pension regimes into a single agency, (b) improvement of the databases required for recognition of pensions, and (c) creation of a unified registry of contributors. It is also assumed that there is an increase of I percentage point in ISS pension contributions, and an increase2 from 3 to 16.5 percentage points in contributions to the Fondo del Magisterio. These measures slow down the rate of depletion of pensionl fund assets and reduce the need for govemment transfers. 8. Hospital restructuring costs amount to US$200 million in 2001. This amount increases to about US$600 million in 2002 and falls tIl zero thereafter. Source: Ministry of Finance, IMF, and World Bank staff estimates. 18 larly education, health, and pensions. Alternatively, if no reforms were implemented or if these reforms were delayed until a new administration is in office, the NFPS would remain on an unsustainable path, and probably sink into deeper, and more explosive, deficits. Moreover, as a result of the reform package, the public sector's debt path will be modified-instead of entering into a rapidly deteriorating, unsustainable spiral, Colombia's debt-to-GDP ratio would be put on a more manageable path (Figure 3). Figure 2. Colombia: Public Sector Balances Under Alternative Reforms Scenarios (1999-2005) 0.0 -1.0 -2.0- -3.0 -4.0- -5.0- -6.0- -7.0 M O N m CC 0 0 O N C C C C ONFPS Balance (No Reforms) ENPFS Balance (With Reforms) Source: World Bank. Figure 3. Colombia: Net Public Debt Under Alternative Reforms Scenarios (1999-2005) 65 - 60 - 55- 50- ; 45- 40 - ON 0 0 0 ON C s C C C C - Net Public Debt (No Reforms) Net Public Debt (With Reforms) Source: World Bank staff estimates. 19 IV. THE PROPOSED STRUCTURAL FISCAL ADJUSTMENT LOAN (SFAL) (A) Objectives 54. The proposed SFAL is the instrument through which the World Bank will support the Colombian government's fiscal strengthening program described above. The program is seen as a critical component in the efforts to sustain the economic recovery, consolidate growth and, thus, reduce poverty. The proceeds of the loan will help reduce the cost of borrowing, lengthen debt maturity, and provide a confidence-building signal to international markets, all of which will, in turn, give the government fiscal space for protecting social sector expenditures and implementing difficult structural measures-such as refornming transfers and expenditure responsibilities, restructuring public hospitals, and redefining pension systems. In addition to their fiscal implications, the measures supported by the proposed SFAL have social impacts that have been identified, evaluated, and will provide the basis for additional policy decisions to protect the most vulnerable groups (to be supported by the forthcoming Social Sector Adjustment Loan). However, overall, the policy actions supported by the SFAL are not expected to hurt Colombia's poor (see Box 1). Box 1. Key Social Impacts of the Policies Supported by the SFAL The key policy actions of the SFAL program on pensions, health, ISS, and public sector reform will resulf in an adjustment in the relative real wages of public employees-for example, special benefit cuts in ISS and an equalization of public pension benefits with overall Colombian pensions. These two measures could negatively impact the poor by affecting their income-generation ability. However, the evidence shows that even under sizeable adjustments, adverse effects on the poor due to an adjustment of public sector wages or pensions are very tuilikely for two reasons (a) the poor disproportionately depend on the private sector for employment; and (b) the poor are isolated from the special pension benefits of the public sector. Relative to the private sector, Colombia's public sector employment is characterized by much higher wages and is concentrated among households in the upper levels of the income distribution. Even after controlling for skill level, public salaries exceed the private sector by about 20 percent. Public employment earnings mostly benefit households in the top 20 percent of the income distribution, which receive 80 percent of these earnings; while the poorest 50 percent receive less than 5 percent of the total public sector earnings. Furthermore, the population belonging to households who benefit from public labor earnings is barely present among the poor-they represent only 2 and 17 percent of such households among the poorest 20 and 50 percent, respectively, whereas the richest 20 percent represent almost 50 percent. Just as the poor do not depend on public sector wages, they also do not benefit much from current pension benefits. The poorest 50 percent of the population receive less than 10 percent of total pension benefits. Moreover, 65 percent of pension earnings go to the richest 20 percent of households. Although distinctions between private and public sector pensions are not possible with current data, the very regressive distribution of public employment leads to the conclusion that public sector pensions should be even more concentrated among the upper percentiles of the population The SFAL's direct benefits for the poor will come from improved efficiency in the delivery of social services, particularly education and health services. Public social expenditure (except pensions) has been progressive. For example, first, in education, coverage rates (in primary, secondary and tertiary education) have improved over the past years for all income quintiles. In addition, access to education services have become more progressive for secondary education, and for tertiary education at the margin. Second, in health and nutrition, child malnutritien and infant mortality, heavily concentrated in the poorest section of the population, have also been decreasing as a result of public sector interventions. Finally, in basic infrastructure, access to public sector water and sanitation has also improved, particularly for those people in the lowest quintile of the income distribution. Thus, given the progressiveness of public expenditures, the poor will benefit from policies that ensure transfer levels and autonomous management of public resources at the local level, increasing public expenditures effectiveness and efficiency. 20 (B) Content 55. The government's overall economic and fiscal strengthening program is laid out in the Letter of Development Policy (see Annex E), which provides the broader context-a strategic vision-and its dynamic implications. From that program, a core subset of policy actions supported by this operation and listed in Table 2 has been selected as the legally agreed actions that would trigger the disbursement of each of the two scheduled tranches and of the floating tranche (details on tranche release conditions are included in Annex G). 56. The timing of the tranches is calibrated to (a) support the government in its efforts to deliver an immediate confidence-building signal to the markets through major refonns that have recently been or are to be implemented before the end of the year 2001 (first tranche expected in December 2001); (b) allow time for the implementation of policy decisions that require the verification of the 2001 budget execution, the approval of the 2002 budget and its execution in the first months of 2002, and coordination with various stakeholders (second tranche in March 2002); and (c) accommodate for policy commitments whose fulfillment does not solely depend on the Executive but on Congressional approval (floating tranche). This timing fits well with Colombia's political calendar-Congressional elections in March 2002 and Presidential election in May 2002. 57. The tranche release commitments were selected for their centrality to the government's fiscal strengthening program. Other measures are also undeniably important for improving the efficiency and quality of public service, but none of them can be sustained unless the overall fiscal stance is put on a sustainable path. Table 2. Colombia-Matrix of Policy Actions for Tranche Releases of SFAL FIRST TRANCHE-Expected December 2001 A. General Condition of the SFAL The macroeconomic framework of the Republic of Colombia is consistent with the objectives of the program. B. Specific Conditions of the SFAL Intergovernmental Fiscal Relations 1. The constitutional amendment (Acto Legislativo No. 1, of July 30, 2001) reforming the transfer of resources from the central government to subnationals has become effective, limiting the rate of growth of fiscal transfers to subnational governments and the rate of growth of the central administration operational expenditures for 2002 and there-after. 2. The government has submitted to Congress a draft law reforming the tax system of territorial entities (Estatuto de Ingresos Territoriales) to increase subnational tax revenues by more than 20 percent in real terms in the first year of its application term compared to realized tax revenues in 1999. 3. The government has issued a decree to regulate debt and borrowing of territorial entities that will contribute to halting unsustainable borrowing, limiting bailouts, and eliminating discretionality in the treatment of debt. Since July 1, 2001, the territorial entities have received no bailouts or guarantees of their debt in accordance with Law 617/00. Health 4. The Ministry of Health, the Department of National Planning, and the Ministry of Finance have defined and approved a national public hospital restructuring policy and an implementation program for the next five years, and have begun the implementation of the first phase, including the selection of at least 10 departments. For the year 2002, the implementation program has covered more than 15 percent of the total public hospital budget included in the FYOI budget. 21 ISS 5. The government has initiated a restructuring plan for ISS Health, including negotiating with ISS unions about labor costs and benefits, and has implemented the first phase, including the completion of renegotiating 50 percent or more of ISS's total outstanding debt to health service providers and the reduction in at least 30 percent in the waiting list for elective surgeries (as of June 30, 2001). Pension 6. The government has created a new social security department in the Ministry of Finance for improving social security system control. Improved control has generated fiscal savings equivalent to more than US$100 million in 2001. Public Sector Reform 7. The government's approved budget for FY02 has included a reduction in total central government current expenditure (net of interest payments and transfers to subnational entities and to social security) of more than 4 percent in real terms in relation to the same expenditures in the approved budget for FYOI. 8. The government's approved budget for FY02 has included a reduction in central government's general expenditures (gastos generales) of more than 15 percent in real terms in relation to the same expenditures in the FY00 executed budget. SECOND TRANCHE-Expected March 2002 A. General Condition of the SFAL The macroeconomic framework of the Republic of Colombia is consistent with the objectives of the program. B. Specific Conditions of the SFAL Intergovernmental Fiscal Relations 1. The government is implementing the Constitutional amendment (Acto Legislativo No. 1, of July 30, 2001) in accordance with its terms, limiting the rate of growth of (a) total transfers to subnational goverrunents to last year inflation plus 2 percentage points; and (b) the central administration's operational expenditures to last year inflation plus 1.5 (percentage points)-as both were included in the FY02 budget, approved by Congress. 2. The decree governing debt and borrowing by territorial entities (referred to in condition 3 of the first tranche) is being implemented in accordance with its terms. Education 3. (a) Law 60, 1993 has been amended to provide for certification of municipalities that, together with districts, represent at least 40 percent of the country's school enrollment enabling such municipalities to autonomously manage the provision of education services (including teacher payroll) and to establish education performance monitoring mechanisms; or (b) the central government has certified municipalities that, together with districts, represent at least 14 percent of the country's school enrollment enabling such municipalities to autonomously manage the provision of education services (including teacher payroll) and intergovernmental performance agreements have been signed between the Central Government and such certified districts and municipalities. In either case, no extra budgetary transfers for education to certified districts or municipalities have taken place. Health 4. The government has signed contracts with at least three subnational governments to finance the restructuring of public hospitals, and the implementation thereof has started. These contracts have included specific annual targets for (a) productivity increases in inpatient and outpatient services and (b) labor cost reductions. The resulting fiscal savings are equivalent to at least 5 percent in real terms per month, on average among those participating subnational governments, compared to the previous year in a manner satisfactory to the Bank. 5. The government has transformed 60 percent or more of the 2001 budgeted health situado fiscal 'after legal mandatory deductions) to demand subsidies. ISS 6. The government and the ISS have reached agreement to generate annual savings in ISS' total costs over a 10-year period which will, in the opinion of the Bank, make ISS financially and economically viable. 22 Pension 7. The government is carrying out and controlling the transfers of public pension regimes into a single agency, reducing payments of ineligible pension claims. Ongoing controls generate fiscal savings of, on average, at least US$10 million per month. Public Sector Reform 8. The central government's total actual current expenditures (net of interest payments and transfers to subnational governments and to social security) that are being incurred in 2002 are generating average savings of more than 4 percent in real terms in relation to the same expenditures incurred during the same period in 2001. 9. The budget approved for FY02 has included general expenditures (gastos generales) that in real terms are not higher than those executed in the FY01 budget. FLOATING TRANCHE A. General Condition of the SFAL The macroeconomic framework of the Republic of Colombia is consistent with the objectives of the program. B. Specific Conditions of the SFAL Intergovernmental Fiscal Relations 1. Congress has approved the law referred to in condition 2 (first tranche), substantially in accordance with the proposal submitted by the government. Pension 2. Congress has approved a law to refonn the general pension system to put it on a sustainable path; or Congress has approved a law to reform at east one of the exempted pension regimes to bring it/them, in the opinion of the Bank, in line with the general pension regime. (C) Risks 58. The proposed SFAL carries high risks, both internal and external, but these risks are much lower than the risk of inaction. With regard to internal risks, the first is that the overall political context, including the ongoing internal conflict, is fragile, and the government may be unable to sustain prolonged social disruptions and resistance from major interest groups (such as teachers and health sector unions) that could derail the reform agenda and lead to more social and political instability. Moreover, the internal conflict and the violence associated with illegal drug activity could further escalate derailing both the stabilization and structural adjustment efforts. The government is aware of this deep downside risk and is committed to the peace process, continuing its dialogue with the guerrillas, and has also stepped up its war against illegal drug activities. It has worked to mitigate other political and social risks through reaching out to, and deepening its dialogue and building consensus with, different political parties, representatives of civil society, and economic and interest groups, including unions. In addition, members of opposition parties have privately endorsed the preparation of this project, and some of their senior officials have participated in important Bank meetings with the Colombian authorities. To help ensure sustainability, the Bank is preparing a collection of policy notes that will contribute to smoothing the economic and social transition from the current administration to the next. 59. Second, the Congress, currently controlled by an opposition party, could block or substantially alter one or more of the policy initiatives contained in the adjustment program. This risk is real, but is substantially mitigated because the bulk of the reforms have already been discussed with members of Congress, particularly with members of the majority party. 23 Moreover, to avoid delays in supporting the government's efforts, those reforms that rec uired Congressional approval are either included up front in the program or as conditions o( the floating tranche. In addition, several components of the reform put more discipline and accountability on the Executive (for example, the decision to rationalize central goverrnment current expenditures, notably personnel, and procurement of goods and services; and to close redundant national government agencies), something that well serves Congressional and subnational government interests. 60. Third, like any fiscal adjustment efforts in a country with a relatively recent reccr:d of weak fiscal policies, there is a risk of reverting to fiscal mismanagement, particularly in the year leading to both Congressional and presidential elections. However, this is unlikely in Colombia's case, for two reasons. On the one hand, the government is committed to imprz)ving its fiscal accounts (as prior actions, such as tax increases and expenditure cuts, have demonstrated), and subnational governments will now be under a harder budget constraint--their commercial bank lenders will face stricter lending regulations, and general bailouts by the government (as of July 2001) are no longer available. On the other hand, this loan will req-iire a set of nonreversible actions prior to disbursements of the tranches, limiting the possibil ly of fiscal indiscipline. 61. The final internal risk is the threat that future rulings and interpretations of the Constitutional Court may reverse key legal reforms, with negative fiscal impacts, as it has happened in the past. This risk is now reduced because recent changes (March 2001) in the composition of the Court have resulted in a body more accepting of pragmatic and ma:-ket- oriented policies, which has already reversed decisions of the previous Court such a, that requiring that adjustments of salaries of civil servants should be based on the consumer price index (see Box 2). Box 2. The Colombian Constitutional Court The Colombian Constitution of 1991 created a specialized constitutional tribunal-the Corte Constitzic-ional (Constitutional Court). The purpose of this Court was, among other things, to ensure protection of human rights, including social and economic rights (so-called "rights of the second- and third-generations"). The 1991 Constitution also introduced the Acci6n de Tutela as a means of simplifying and expediting the Court's protec ion of human rights whenever threatened or violated. Court Justices are appointed for eight years and are guarinteed independence and stability. Following a transition Court in 1992, the first Court (1993-2001) distinguished itself for pursuing the consolidation of human rights protection. The first Court's interpretation of labor and social security laws (including pcision laws), however, was broad and inclusive, creating unanticipated liabilities for the government and affecting fiscal sustainability. Moreover, without properly anticipating the economic impact of its decisions, the Court's rtA ings have also affected the housing market, threatening the financial stability of mortgage banks. The second Court (2001-09) has just taken office, and has had a different approach in the interpretation of lab r and social security laws, having already reversed some decisions of the previous Court. 62. There are also several external risks; most notably a further deterioration in the external macroeconomic environment led by a deeper slowdown and/or recession in the Nkorld economy-particularly as a result of the attacks of September 11 and subsequent events. The most important transmission channels are (a) a further deterioration in the terms of trade, particularly resulting from continued unfavorable coffee prices and volatile oil prices; and (b i the closure of access to international financial markets due to crises in major markets and/or 24 emerging market countries. In the event of (a), the government intends to continue its flexible exchange rate policy to encourage nontraditional exports and foreign direct investment. With regard to (b), the government has committed itself to sound fiscal management (as supported by this operation) and retains access to IFI financing, particularly the IMF through continued fulfillment of the conditions and the right to draw under the Extended Arrangement. 63. In spite of the above-mentioned mitigating factors, the SFAL remains a high-risk, high- return operation. Thus, it will be essential to maintain a close working relationship with the client, and to provide ample, ongoing dialogue and assistance during the implementation period. Because this operation will form the core basis of the Bank's interventions in Colombia, its successful implementation remains critical for future Bank assistance. To support timely implementation, a schedule of frequent project supervisions, and continuous contact and exchange of views with the government will be maintained, greatly facilitated by the presence of a well-equipped Country Office in Bogota. (D) Poverty Impact 64. The proposed SFAL will contribute to poverty reduction at the national, regional and local levels. First, it will help protect the macroeconomic framework of the country as a whole by reducing the likelihood of fiscal crises at different levels of government. As mentioned before, the Bank's analysis shows that negative macro shocks are the most important determinant of poverty increases in Colombia. Second, the various sectoral reforms, supported by the SFAL, will sustain economic recovery, consolidate economic growth in the medium- and long-term, and reduce poverty-as economic growth continues to be the major source of poverty reduction in Colombia (see Annex C). Third, the SFAL will also ensure that within the fiscal strengthening program, public resource allocations to social programs are protected, particularly those for education and health, facilitating the provision of public services for the poor. Fourth, the most important poverty-related activities under the responsibility of the subnational level are the provision of education and health services, and the SFAL support measures to improve the efficiency and coverage of these services, benefiting the poor directly. 65. Fifth, the reforms supported by the SFAL will also seek to increase equality of distribution of subnational resources in different parts of the system, and will accelerate their transformation from supply-driven to demand-driven subsidies. Such reforms will particularly benefit the poor as they are the main demanders of public services. Finally, bringing balance and transparency to departmental and municipal finances, together with improving the delivery of education services and strengthening the provision of health services in public hospitals, will ensure that essential public services will continue to be more efficiently available to those who could not otherwise procure them by their own means. (E) Loan Amount, Conditions, Financial Safeguards, and Disbursements and Audits 66. The proposed (US$400 million) SFAL in the form of a fixed spread, U.S.-dollar- denominated loan would be made to the Republic of Colombia. Disbursements under the proposed loan would be made to an account (Deposit Account) of the Republic of Colombia, established at the Banco de la RepuTblica for this purpose. The proposed loan would have two scheduled tranches and a floating tranche, which is de-linked from the other two tranches. These 25 tranche arrangements have been discussed, and designed with the government, ensuring their full ownership. They capture the government's request and the Bank's assessment of the necessary timing for undertaking the actions and implementing the measures, while simultaneously separating those actions that are fully under the administration's control and those that are shlared with other institutions (including Congress and labor unions) and involve negotiations. The first tranche would be US$160 million-scheduled to be disbursed in December 2001, the second tranche would be US$180 million-scheduled to be disbursed in March 2002; there would also be a floating tranche of US$60 million-to be disbursed when its specific conditions are fulfilled. 67. To meet the objectives elaborated above, a single loan (with two tranches plus a floating tranche, and an integrated set of conditions tackling cross-cutting fiscal weaknesses) was juclged to be more appropriate than other alternatives, including a single-tranche adjustment loan or a programmatic loan. It is expected that fulfilling the set of conditions of a single loan during the last year of the administration would convey strong messages to the market regarding the government's commitment to reform and fiscal stability. Moreover, the government would thus be able to fully implement its fiscal reform package within this administration's term, and [eave the fiscal accounts on a sustainable path for the next administration, smoothing both the political and, more important, the economic and social transitions. 68. To ensure compliance with financial safeguards, the government is committed to continue improving public expenditure management, including both procurement and financial management, and combating corruption (for details, see Annex H) Over the last few years, the Bank (at the authorities' request) has undertaken a substantial amount of analytical work to support the government's efforts for improving public expenditure management. This included Colombia-Country Financial Accountability Assessment (September 1998; a second Country Financial Accountability Assessment is scheduled to be carried out in FY03, and will pay particular attention to financial accountability arrangements at the sub-national and municipal levels); the Country Procurement Assessment Report (CPAR, March 2001); the Public Financial Management Project (PFMP-I, closed March 2001); and the on-going PFMP-II (February 2001) which follows and expands on the earlier project. In addition, the Bank has supportedl the government's anti-corruption program through the World Bank Institute as well as an Institutional Development Facility (IDF) grant completed earlier this year. 69. The government has developed and is implementing strategies based on the recommendations contained in the work mentioned above. More specifically, the governrment's strategy for improving public procurement includes the establishment of an institutional framework for an orderly and consistent development and dissemination of policies, regulaiions, and standard bidding and contract documentation. In addition, it gives priority to simplification and not to the introduction of new rules, control and punitive measures that could easily aggravate existing problems and provide additional opportunities for corruption. These measures have been developed in consultation with the major stakeholders. With regard to financial management, the Office of the Accountant General is being strengthened to allow it to become more effective, as are institutions (both public and private) involved in the professional practice of public accounting. As indicated above, these efforts have been supported through two specific Bank operations-the PFMP I and the follow-on PFMP II-both highly praised by QAG, and recently completed sector work on corruption. 26 70. Moreover, these two projects have also supported the government's measures to strengthen institutional capacity and analysis in a number of key areas of public expenditure, such as macro-programming, formulation and monitoring of the budget, budget execution- including the development and implementation of an Integrated Financial Management System (SIIF)- and evaluation of results of public expenditure-including the implementation of the National System for Evaluation of Results of Public Sector Performance (SINERGIA). Despite important progress in areas of public expenditure, further strengthening of financial management processes is needed. These deficiencies include (a) divergence between budgeted expenditures and the government's strategic development priorities; (b) lack of clarity regarding the roles and reporting requirements of government institutions; and, (c) the need to enhance public sector transparency and accountability by improving the dissemination of the objectives, performance evaluations and budgetary results of public entities. The on-going PFMP II is supporting reforms in these areas. 71. In addition, the IMF's safeguard assessments report on the audit mechanism for the Colombian central bank, completed in May 2001 under the transitional procedures, concluded that the current auditing arrangements are capable of providing high-quality audit services. However, while external audit of the central bank's financial statements is undertaken, it does not comply with the strict definition of an independent audit under the International Standards on Auditing (ISA). The principles governing central bank auditing mechanisms are enshrined in the Colombian Constitution of 1991, whereby the President is responsible for auditing the central bank, and is permitted to delegate this function to the Banking Superintendent and to an independent State agency, but not to a private firm. Based on this finding, and on the legal issues involved, the IMF and the Colombian authorities are looking for alternatives to overcome this matter. The government also intends to request the IMF to assist it in addressing issues in the area of fiscal transparency, and an IMF fiscal transparency mission is envisaged early next year. 72. For the purpose of SFAL's disbursement and audits, the government will open, prior to furnishing to the Bank the first request for withdrawal from the Loan Account, and thereafter maintain in its central bank, a deposit account in dollars on terms and conditions satisfactory to the Bank. All withdrawals from the Loan Account will be deposited by the Bank into the Deposit Account. The Bank will monitor and assess ex-post (through supervision missions with the government's permission) the use of the Deposit Account based on the Comptroller General Audit Report. The government undertakes that the proceeds of the Loan will not be used to finance ineligible expenditures. If the Bank determines at any time that any proceeds of the Loan have been used to make a payment for an ineligible expenditure, the government will, upon notice from the Bank, (a) deposit into the Deposit Account an amount equal to the amount of said payment, or (b) if the Bank has requested, refund such amount to the Bank. Amounts refunded to the Bank upon such request will be credited to the Loan Account for cancellation. Finally, the government will, at the Bank's request, (a) have the Deposit Account audited in accordance with appropriate auditing principles consistently applied, by independent auditors acceptable to the Bank; (b) furnish to the Bank, as soon as available, a certified copy of the report of such audits; and (c) provide to the Bank other information concerning the Deposit Account and the audit at the Bank's request. 27 V. THE MACROECONOMIC FRAMEWORK FOR THE PROPOSED SFAL 73. Sound macroeconomic management consisting of (a) good fiscal management, inc'uding the implementation of fiscal reforms; (b) an inflation-targeting monetary policy; and (c) a flexible foreign exchange regime, will provide the basis for sustaining the economic recovery and for a potentially respectable rate of economic growth, single-digit inflation, and a v-iable external balance (see Annex B for a fuller discussion of recent economic developmenTs and prospects). 74. Real economic growth for 2001 is now projected to be about 1.5 to 2.5 percent--down from about 3 percent in 2000. While this projected real economic growth is above the country's annual real growth trend of less than 1 percent observed over the past five years, it is still well below the 4.5 percent annual real economic growth of the early 1990s. The manufacturing industry and nontraditional exports have provided the main impetus to growth over the past year; improved performance in agriculture, commerce, and construction has also contributed to real economic growth. However, weakening of investment spending, disruption of oil exports, falling coffee export prices and extensive strikes in both the public and private sectors have negatively affected economic growth. Inflation continues its downward trend, and is expected to attain the Central Bank's year-end target of 8 percent for 2001. The balance of payments current account deficit (overall external deficit) is projected at 2.6 percent of GDP-up from the deficit of about 0.2 percent of GDP in 2000 but within manageable proportions. This increase is due to an increase in domestic demand for imports, and the simultaneous weakening of traditional exports, particularly oil and coffee. Gross international reserves (in months of goods and services) are projected at about six months for this year. 75. For the medium term (2002-03), average real economic growth is projected at about 2.5 to 3 percent per year, and for the long term (2004 and on) at about 3 to 3.5 percent annually, fueled mainly by investment expansion. As a result of the implementation of the forthcoming fiscal reforns and the continuation of sound macroeconomic management, the deficit of the NFPS would decline and reach about 1.5 percent of GDP in 2005. Inflation is projecTed to continue its downward trend, and to stabilize at about 4 percent in 2003 and beyond. The external current account deficit would reach about 2 to 3 percent of GDP, and gross international reserves are projected at about seven months of goods and services imports. Finally, in this macroeconomic framework, Colombia's net total public debt would stabilize at about 50 percent of GDP by 2005. Vl. BANK STRATEGY (A) The CAS and the Proposed SFAL 76. As mentioned before, the CAS-PR recognized that the worsening of the economic situation could lead to fiscal imbalances and financing gaps that would prevent the realization of the CAS objectives of poverty reduction, social development, and sustainable growth. Hence the CAS-PR proposed to move the assistance strategy to a high-case lending scenario, including adjustment operations. While a specific adjustment operation to support fiscal reforms was not explicitly envisioned in the CAS-PR, the strategy identified the need to support Colombia's efforts to strengthen its fiscal position, and the objectives of poverty reduction, social 28 development, and sustainable growth. These objectives still remain valid to date, and they together with the Bank strategic interventions in Colombia are also confirmed in the Country Assistance Note accompanying this Memorandum. 77. The SFAL will directly contribute to achieving the CAS-PR's objectives of poverty reduction and social development. The fiscal consolidation supported by the proposed operation would protect the country's macroeconomic framework by reducing the likelihood of fiscal crises and economic recession that would most impact the poor. By preventing major fiscal crises, the SFAL also ensures that essential public services would continue to be available to those who could not otherwise procure them-namely, the poor and most vulnerable. In addition, the fiscal reforms that bring about fiscal stability at the macro level also bring efficiency and equity at the sectoral levels, especially in the provision of public education and health that benefit the poor relatively more. The efficiency and the equality of distribution of subnational resources in different parts of the system and the transformation of resources from supply to demand subsidies, aiming at improving the provision of education and health services in public hospitals, not only help the poor, but also increase the balance and transparency of subnational government finances. Moreover, the forthcoming Social Sector Adjustment Loan (SSAL) will complement this operation as it will help (a) protect the poor from the potential effect of the fiscal adjustment; (b) ensure that key social services, including safety net programs, are targeted to the poor, and delivered more efficiently and effectively; and (c) start reforming existing labor regulations which have negatively impacted the employment of the poor, particularly young, female, and self-employed workers. 78. The proposed SFAL operation is fully consistent with the strategy proposed in the CAS and CAS-PR, not only in terms of policy content but also in terms of overall lending and exposure limits. The CAS-PR envisioned and triggered a high-case lending program of around US$450 million for FY98-01. If this program had been completely delivered, the lending program would have amounted to about US$1.8 billion. To date, the total amount committed is about US$1.4 billion-leaving space for the SFAL operation under the envelope proposed by the CAS and CAS-PR. More specifically, in FY98 and FY99, total commitments of about US$130 million and US$140 million were approved, respectively. In FY00, total commitments of about US$930 million were approved, including US$500 million to support financial sector reform and US$225 million to support the reconstruction efforts after the devastating earthquake of January 1999. In FY01 total commitments of US$190 million were approved. 79. Given its fiscal constraints, the government requested that the Bank stretch the lending program into FY02. As explained in the accompanying Country Strategy Note, the Bank expects to present six operations, including the SFAL and the Social Sector Adjustment Loan (for about US$150 million), in FY02. This would bring our proposed FY02 program to about US$630 million. Under this scenario, the lending program for FY98-02 would total about US$2 billion and average about US$400 million annually, which is below the high-case lending ceiling of US$450 million. All of the operations that are being proposed for FY02 were explicitly mentioned in the CAS-PR, with the exception of the SFAL. However, the CAS and CAS-PR clearly identified the need for Colombia to strengthen its program of fiscal reforms, which is the objective of the SFAL. 29 80. The overall external debt exposure of Colombia is relatively high, but manageable--total external debt as a percentage of GDP is about 42 percent, about three fifths of which is plublic sector; total external debt service represents about 51 percent of exports of goods and services, about two fifths of which corresponds to the public sector; and the ratio of total short-term external debt to reserves is about 90 percent. The World Bank is not a dominant credilor in C'olombia's balance sheet: IBRD debt outstanding as a proportion of public sector external debt is about 10 percent, and IBRD debt service as a proportion of exports of goods and services represents less than 3 percent. Moreover, over the last five years, on average, net disbursements from the World Bank to Colombia have been negative. The share of Colombian debt in the Bank's portfolio is modest, and the headroom for new lending is ample. (B) Complementarity with Other World Bank Operations and Activities 81. In order to help economic recovery and improve the financial system, the Financial Sector Adjustment Loan for US$500 million was approved by the Board in November 1999. This operation supported a comprehensive program of financial structural reforms that sought to promote the development of a healthy and sustainable financial system. In addition, part of its proceeds helped Colombia regain access to international capital markets through a Policy-Based Guarantee Operation. The SFAL's objectives will further contribute to these objectives by ensuring their sustainability and by ascertaining that national and subnational governments are better able to manage their finances and fiscal programs, including public sector debt management within a market-disciplined framework for subnational borrowing. 82. To alleviate the impact on the poor of the 1998/1999 economic crisis, the government identified segments of the population that had been hit hardest by the crisis, including the unemployed poor, young entrants into the labor market, female heads of households, children, the elderly, and those displaced by the ongoing conflict. In response to the problems of these vulnerable groups, the government decided to strengthen and expand its social safety net program. This program has significantly been helped by two World Bank operations (the Community Works and the Human Capital Protection Projects) for a total of US$250 million. These operations will contribute to shielding the most vulnerable groups by creating jobs and by protecting human capital through the provision of education and nutrition grants. The SFAL will ensure that, within the fiscal strengthening program, public resource allocations to social programs are preserved, particularly for education and health. 83. In addition two analytical studies, which are underway, complement the SFAL in the overall effort to reduce poverty in Colombia. The Colombia Poverty Report assesses and reviews (a) the evolution of poverty and inequality in urban and rural areas over the past two decades; (b) the incidence of crime and violence and related household responses; and (c) selected aspects of public assistance, including an overview of public social expenditures and a specific review of the subsidized health insurance system. The Social Safety Net Assessment evaluates reforms to Colombia's social safety net, both for addressing the needs of vulnerable groups during normal times and for developing a counter-cyclical strategy that could be implemented rapidly during future crises. The study (a) assesses vulnerability and risk management in Colombia; (b) evaluates how the three new emergency Red de Apoyo Social safety net prograrns fit into the broader social safety net over the long-term; (c) assesses the 30 efficiency, effectiveness and gaps in Colombia's existing social assistance programs; and (d) proposes technical, program and policy reforms aimed at improving the social safety net. (C) Complementarity with Other Institutions' Operations 84. The proposed SFAL follows the SDR 1.96 billion (about US$2.5 billion equivalent), three-year, Extended Arrangement Facility-the first of its kind for Colombia-that was approved by the IMF Board in December 1999, which helped stabilize the economy and provide a good macroeconomic framework for the implementation of the government's reform program. Within this framework, the Bank takes the lead, through the proposed SFAL, in supporting the authorities' effort at structurally strengthening the country's fiscal accounts and bringing them onto a more sustainable path. Division of labor and responsibilities, as well as continuous coordination between the IMF and the Bank takes place in the review of economic policy and the preparation of the respective interventions in Colombia. 85. In addition, the above-mentioned financial sector adjustment operation was prepared in close partnership with the IDB, which committed about US$1 billion in fast-disbursing loans through three operations-one in the electricity sector which was approved in December 1998 for US$350 million, another for public sector restructuring which was approved in January 1999 for US$500 million, and a third in the financial sector for US$100 million which was approved in October 1999. These operations have been complemented by another recently-approved loan focusing on territorial finances, for US$400 million. Assistance to the social safety net program was also designed in close collaboration with the IDB, which has committed US$250 million to this area. VII. RECOMMENDATION 86. I am satisfied that the proposed loan complies with the Articles of Agreement of the Bank, and I recommend that the Executive Directors approve it. James D. Wolfensohn President By Shengman Zhang Washington, D. C. November 16, 2001 31 32 Annex A THE REPUBLIC OF COLOMBIA Country Assistance Update Note The Current Strategic Framework Al. The last Country Assistance Strategy (CAS) for Colombia was discussed by the Board of Executive Directors in November 1997, and the Country Assistance Strategy Progress Report (CAS-PR) was discussed in November 1999. Under normal circumstances a new CAS would have been presented in the middle of 2002. However, it was decided that a new CAS would be better justified following the start of the new administration in early FY03 to coincide with Colombia's political cycle. In addition, we are currently preparing a collection of policy notes summarizing the major challenges in Colombia's social-economic development. The policy notes will serve as a bridge to the new administration, which will take office in August 2002. This work will be the analytical basis for the new CAS. The present note provides an update on the strategic framework under which the Bank proposes to operate until the Board of Executive Directors discusses the new CAS. A2. Following participatory consultations with government leaders and civil society the CAS identified, and the CAS-PR reconfirmed, the main objectives of poverty reduction, social development and sustainable economic growth. Anchored with a stable macro-economic environment and fiscal sustainability, these objectives were to be achieved by interventions in key areas (a) promoting peace and development; (b) promoting rural development; (c) developing human capital; (d) attaining public sector responsiveness and efficiency; (e) improving infrastructure services; and (f) ensuring sustainable development. These areas of intervention have not changed and continue to be supported by the Bank's lending program. Furthermore, the CAS-PR recognized that Colombia's economic performance, especially economic growth and the fiscal balance, had deteriorated sharply since the presentation of the CAS and recommended strengthening fiscal stability to restore growth and reduce poverty. The CAS-PR confirmed that a reduction of the fiscal deficit was a key macroeconomic objective of the government, and it recognized that without a correction of fiscal imbalances, Colombia could face serious financing gaps and problems in accessing external financial markets. Worsening economic conditions combined with heightened uncertainty as a result of the country's long-standing internal conflict, would hinder the achievement of the government's objectives as reflected in the CAS. In addition, the international economic and financial repercussions of the tragic events of September 11, 2001 have reinforced the importance of a return to fiscal sustainability as these repercussions may be severe and could further restrict capital flows to Colombia and augment the deterioration of the country's terms of trade. A3. The CAS presented a base-case IBRD lending program of around US$300 million per year for FY98-00. This relatively low lending ceiling was determined by Colombia's preference at the time to limit its IBRD borrowings. The CAS-PR moved the assistance strategy to a high- case lending program, with increased emphasis on adjustment lending, to around US$450 million annually. This movement was justified by (a) the need to cushion the adverse social impact of the economic recession, financial instability and internal conflict on the poor and vulnerable, and (b) the Government's commitment to undertake fiscal adjustments and implement structural 33 Annex A reforms to be supported by an IMF program. These triggers remain valid. Furthermore, the CAS-PR indicated that we would reassess the lending program and move to a base-case or low- case scenario if (a) the fiscal stabilization program supported by the IMF program went off track; (b) project implementation problems arose; or (c) increased violence and social unr-est significantly reduced our ability to work effectively in Colombia. These risks have not materialized. Recent Developments A4. Peace Process. Colombia has been ravaged by intense conflict for the past 40 years, financed in part by the illegal drug trade. About 40,000 people have died in clashes during the last decade alone. President Andres Pastrana, who has one year left in his four-year term, has made a highly ambitious peace initiative the focus of his administration. However, the progress achieved so far has been disappointing and many Colombians are increasingly skeptical that the peace process as currently conceived will achieve tangible results. In light of the attacks of September 11, 2001, it is likely that there will be international pressure on the government to take a stronger stance against the insurgent groups. The two main rebel groups in Colombia, the FARC and ELN, as well as the paramilitary group, the AUC, are on the United States' official list of foreign terrorist organizations. On October 7, 2001, Mr. Pastrana decided to continue the peace negotiations by extending the demilitarized zone for three months, but with firmer conditions, most notably the negotiation of a truce and agreements to end kidnappings and attacks on the civilian population. The refocused peace discussions will center on a ceasefire and on obtaining commitments from guerrillas to respect international humanitarian law. Colombia will continue to need the assistance of the international community to fight terrorism and to address the socio-economic determinants of violence in order to construct the underpinnings for peace and development. A5. The Economy. Economic growth for 2001 is now projected to be about 1.5-2.5 percent-down from last year's 3 percent. While this projected real economic growth is above the country's annual average real growth trend of less than 1 percent observed over the past five years, it is still well below the 4.5 percent annual real economic growth of the early 1 990s. TFhe manufacturing industry and nontraditional exports have provided the main impetus to growth over the past year; improved performance in agriculture, commerce, and construction have also contributed to the expansion of output. However, weakening of investment spending, disruption of oil exports, falling coffee export prices and strikes in both public and private sectors have negatively affected economic growth. Inflation continues its downward trend and is expected to attain the Central Bank's year-end target of 8 percent for 2001. The balance of payments' current account deficit is projected at 2.6 percent of GDP-up from the deficit of about 0.2 percent of GDP in 2000. This is due to an increase in domestic demand for imports, and the simultaneous weakening of traditional exports, particularly oil and coffee. Nevertheless, gross international reserves (in months of goods and services) are projected at around six months for the rest of 2001. While these projections take into account the initial impact of the tragic events of September 11, 2001, they will have to be further assessed. Nonetheless, it is certain that the impact on global financial markets and economic growth will be substantial and that as a result, Colombia's economy will also be negatively affected, including by restricted access to capital markets and further deterioration of terms of trade (particularly coffee and oil). Colombia's total 34 Annex A external financing needs are projected at about US$ 8 billion per year for the 2001-2003 period. Total public external debt has risen steadily, both in absolute terms and as a proportion of GDP. In gross terms, it has almost doubled from 27 percent of GDP in 1994 to 50 percent in 2000. A6. The government's macroeconomic program has three pillars (a) sound fiscal management, including the implementation of fiscal reforms; (b) an inflation-targeting monetary policy; and (c) a flexible foreign exchange regime. If this macro-economic policy mix is maintained, and given a stable economic environment, average real economic growth is projected for the medium term (2002-2003) at about 2.5-3 percent per year, and for the long-term (2004 and on) at about 3-3.5 percent annually, fueled mainly by investment expansion. As a result of the implementation of the forthcoming fiscal reforms and the continuation of sound macroeconomic management, the deficit of the Non Financial Public Sector would decline and reach about 1.5 percent of GDP in 2005. Inflation is projected to continue its downward trend, and to stabilize at about 4 percent in 2003 and beyond. The external current account deficit would reach about 2-3 percent of GDP, and gross international reserves are projected at about seven months of goods and services imports. Finally, in this macroeconomic framework, Colombia's net total public debt would stabilize at about 50 percent of GDP by 2005. A7. IMF Program. The government's stabilization program is being supported by an IMF Extended Facility, the first of its kind for Colombia, which was approved in December 1999 and extends through the end of 2002. The government entered into an Extended Facility Program in order to send a positive signal to international markets; it was understood that the arrangement was precautionary and that it would not be drawn upon unless there was a pressing balance-of payments need.' The government requested that the Fund focus on stabilization and balance of payment issues and that the Bank take the lead on key structural issues. Therefore, the Fund's program concentrates on stabilization of the fiscal accounts within the duration of the program, and the Bank has taken the lead in identifying and supporting key structural reforms that will contribute to bringing Colombia's fiscal accounts onto a sustainable path in the long term. Colombia has so far met, and at times exceeded, the performance targets under the IMF program; two program reviews have been successfully completed, and the third is expected to be presented to the Fund's Board in early 2002. All performance indicators are expected to be met with the exception of the original program targets for the fiscal deficit. The weakness of the economic recovery this year, combined with external variables such as low commodity prices and the events of September 2001, make it likely that deficit targets for both 2001 and 2002 will be modified in the current program review-the targeted fiscal deficit is expected to shift from 2.8 to about 3.3 percent of GDP for 2001 and this adjusted target will be met. 1 We support this strategy for several reasons, including: (a) the signal to international markets is working as can been seen by Colombian spreads, particularly in light of the recent events of September 11; (b) the fact that the Central Bank is bound by strict constitutional restrictions against lending to finance the deficit-an exemplary practice; and, (c) more importantly, the division of labor and responsibilities between the Fund (stabilization and balance of payments) and the Bank (structural issues). 35 Annex A Proposed Bank's Assistance Program Through the Next CAS A8. Portfolio Performance. At the presentation of the CAS-PR, it was recognized that the quality of the portfolio had improved substantially since the presentation of the CAS; this trend has continued. Portfolio performance in Colombia remains satisfactory. The portfolio currently contains 23 projects, with net commitments of about US$1.3 billion and an undisbursed balance of about US$663 million. Only one project is currently rated unsatisfactory. The disbursement ratio was 24.6 percent in FYOI, up from 20.2 percent in FY95. The proactivity index is lOO percent. Colombia has surpassed all of the key portfolio performance indicators included in the CAS and CAS-PR. A9. Country Procurement Assessment Report (CPAR). A CPAR was completed in March 2001. Its findings will be broadly discussed and its recommendations incorporated into the next CAS. The report found that for Bank-assisted projects procurement is carried out in accordance with Bank policies and procedures, and procurement performance under those projects is generally satisfactory and free of major problems. Two procurement audits and four agency procurement capacity assessments carried out in the last two years have found no major problems and determined that the procurement risk in Bank projects is lower than average. However, the CPAR found that public procurement and contract management had deficiencies that had the potential to result in inefficient use of public funds and could be a source of corruption. The findings and recommendations of the CPAR to strengthen public procurement were discussed with the Colombian authorities. Following recommendations of the CPAR, the government is implementing with its own resources a Public Procurement Project. In addition, the second Public Financial Management Project (PFMP), approved by the Bank Board on March 22, 2001 includes a component to strengthen public procurement. AIO. Country Financial Accountability Assessment (CFAA). A CFAA was carried out in September 1998. As a result of the recommendations in the 1998 CFAA report, the Office of the Accountant General and public accounting institutions have been strengthened. These efforts have been supported by recently completed sector work on corruption and by the Bank-financed PFMP I and II. These projects have supported the government's measures to strengthen institutional capacity and analysis in a number of key areas of public expenditure: macro- programming, formulation and monitoring of the budget, budget execution, including the development and implementation of an Integrated Financial Management System (SIIF), and evaluation of results of public expenditure, including the implementation of the National System for Evaluation of Results of Public Sector Performance (SINERGIA). Despite important progress in public expenditure management, further strengthening of financial management processes is needed. These weaknesses include (a) divergence between budgeted expenditures and the government's strategic development priorities; (b) lack of clarity regarding the roles and reporting requirements of government institutions; and (c) the need to enhance public sector transparency and accountability by improving the dissemination of the objectives, performance evaluations and budgetary results of public entities. The on-going PFMP II is supporting reforms in these areas. The next CFAA will be completed prior to the CAS and will have a special focus on sub-national finances. Its recommendations will be incorporated into the strategy and program of the next CAS. 36 Annex A Proposed Bank Lending and Analytical Program Through FY02 All. Lending Program for FY02. As mentioned above, the CAS-PR moved the assistance strategy to a high-case lending program, including adjustment lending, to around US$450 million for the lending program of FY98-01. Including FY02, the proposed lending program would total about 2 billion and average about US$400 million per year. In FY98 and FY99, we committed US$132 million and US$142 million, respectively. In FY00, six operations were approved for a total of US$935 million (including the US$500 million Financial Sector Adjustment Loan and the exceptional US$225 million Earthquake Recovery Loan). In FY01 we committed US$185 million. In FY02, we expect to present six operations, including the SFAL currently under consideration and the Social Sector Adjustment Loan (for an amount of about US$150 million). This would bring our proposed FY02 program to about US$630 million. With the exception of the SFAL, all of the operations that we propose for FY02 were specifically included in the CAS- PR. Although a fiscal adjustment operation was not explicitly envisioned in the CAS-PR, the strategy clearly identified the need to support Colombia's program of fiscal and other structural reforms. A12. The Bank's FY02 program supports the short-term and longer-term efforts of the Government of Colombia to protect and promote the welfare of the poor through a series of interrelated lending operations, as well as analytical work. * Magdalena Medio II. The Bank has supported this program through a recently completed LIL (Magdalena Medio Regional Development Project) and will continue its support through a second LIL approved in September, 2001. The first project was very effective in creating a longer term vision of a better future for the poor in this region in the midst of conflict. It aims to strengthen human and social capital, and start a community-led process of improving basic services and investment. The second project will gradually shifting the focus of decision-making and leadership of the program to local and regional actors, with communities playing a lead role. * Water Sector Reform Assistance Project. The Bank approved a $40 million project in October, 2001 to support Colombia's initiative to reform municipal water utilities in underserved rural and urban areas in the country's Caribbean coastal region. The project will provide technical assistance to guide municipal leaders as they incorporate the private sector in the management and operations of their water utilities and expand water and sewerage services to more people, especially the poor. * Productive Partnership Project. This project aims to establish a successful program of productive partnerships between agribusiness and small farmer producer organizations. It will do so by providing an integrated package of incentives and assistance for the development and implementation of demand-driven, sustainable subprojects that generate income and employment opportunities for rural communities. The project would contribute to (i) reactivating the rural economy; (ii) improving the welfare of poor rural inhabitants; and (iii) generating an enabling environment for the peace process in- the country. 37 Annex A * Judicial Reform. This project constitutes the first phase of a long-term initiative to improve judicial conflict resolution services in the civil, labor and family law jurisdictions. The proposed project is expected to have three key impacts 'i) improvement of the judiciary's social legitimacy and ability to respond to society's justice needs; (ii) increased incentives for the resolution of labor, family and civil conflicts through peaceful means, thus improving the prevalence of the rule of law and peaceful coexistence; and (iii) more effective enforcement of contracts at lower cost. * Structural Fiscal Adjustment Loan (SFAL). The proposed SFAL would support Colombia's efforts to achieve macroeconomic stability and consolidate growth by undertaking structural reforms that would put the country's fiscal accounts on a sustainable path. The Bank's strategy would assist the govenment to (i) ensure that the fiscal decentralization process is economically sustainable through the reform of intergovernmental fiscal relations, including transfers, and the imposition of hard budget constraints on subnational governments; (ii) rationalize central government expenditures to sustain fiscal discipline; and (iii) support structural reforms in sectors where both national and subnational governments share responsibilities, including education, health and pensions. The SFAL will ensure that within the fiscal strengthening program, public resource allocations to social programs are protected, particularly those for health and education. By consolidating growth and fiscal sustainability, it will support poverty reduction. History has shown that Colombia has enjoyed significant declines in poverty during periods of positive growth and fiscal balance. For example, from 1988 to 1995, as average GDP grew at rates of about 3.5 percent annually and poverty rates declined from about 65 to 60 percent of the population. Furthermore, by protecting macro-economic stability and reducing the likelihood of fiscal crises at the national and subnational levels, the SFAL will help to prevent the social and political instability that negative macro- economic shocks could engender. * Social Sector Adjustment Loan (SSAL). In order to alleviate the impact of the economic crisis on the poor, the government asked the Bank and the IDB to support programs targeted to segments of the population most impacted by the crisis. In this context, the Bank is currently financing the Community Works and Employment project which provides temporary employment to poor unemployed workers in urban areas through small community projects. The Bank is also financing the Human Capital Protection project that funds a conditional cash transfer program to protect children's health, education, and nutrition. To ensure that the impact on poverty and the sustainability of the government's interventions in this area are maximized, the government has asked the Bank to prepare a Social Sector Adjustment Loan to support structural reforms needed in the social sectors. The project would assist Colombia in establishing a social protection system that reacts quickly in response to economic crises and that is better targeted, more transparent and properly evaluated. The project would also assist in the reform of labor regulations that have hindered the employment of the poor, particularly young, female and self-employed workers A13. Key Analytical Work for FY02. To support the proposed program the Bank is undertaking the following analytical work: 38 Annex A * The Colombia Policy Notes will summarize the major challenges in Colombia's social- economic development and will serve as a bridge to the new administration. It will include thematic notes on peace, growth, governance, natural resource management and poverty and inequality. It addition, it will summarize essential sector issues for development in about 33 areas. * The Colombia Poverty Report assesses and reviews (i) the evolution of poverty and inequality in urban and rural areas over the past two decades; (ii) the incidence of crime and violence and related household responses; and (iii) selected aspects of public assistance, including an overview of public social expenditures and a specific review of the subsidized health insurance system. * The Social Safety Net Assessment evaluates reforms to Colombia's social safety net, both for addressing the needs of vulnerable groups during normal times and for developing a counter-cyclical strategy that could be implemented rapidly during future crises. The study (i) assesses vulnerability and risk management in Colombia; (ii) evaluates how the three new emergency Red de Apoyo Social safety net programs fit into the broader social safety net over the long-term; (iii) assesses the efficiency, effectiveness and gaps in Colombia's existing social assistance programs; and (iv) proposes technical, program and policy reforms aimed at improving the social safety net. Risks and Opportunities of the Proposed Program A14. There are serious risks associated with the proposed program. If security conditions were to significantly deteriorate, implementation of the Bank program would be negatively affected. In addition, the current broad-based support for the reform process could weaken, particularly during the upcoming national elections. However, the Pastrana administration, and leading opposition candidates, continue to support peace and economic reform. President Pastrana has expressed his commitment to use the final year of his current administration to support the implementation of vital reforms. Moreover, the program has been discussed with the main political parties (which would like to have painful but necessary reforms out of the way before they take office), helping to increase its viability and continuity beyond this administration. There are also several external risks that have been augmented by the events of September 11, namely a further deterioration in the terms of trade and closure of access to international markets. However, the implementation of the fiscal reform program will mitigate these risks by sending positive signals to the market that Colombia is committed to maintaining sound economic management. A15. The risks of not proceeding with the program are even higher. First, the possibility of a deterioration in the fiscal situation and growing fiscal deficits, jeopardizing macro-economic stability and growth prospects, which could in turn lead to deepening political and social conflict in Colombia. Second, a missed opportunity to alleviate the impact of economic crises on the most vulnerable citizens and to ensure that key social services are targeted to the poor and delivered efficiently and effectively. A16. Although this is a high-risk program, its returns could be substantial. Present political and economic circumstances in Colombia, as well as recent performance of the Bank portfolio, 39 Annex A do not warrant any significant departure from the strategy outlined in the CAS and reaffirmed in the CAS-PR. Indeed, they confirm the need to focus urgent attention on several important aspects, such as promoting key social and fiscal structural reforms. Under these circumstances, it is management's view that there is no reason to deviate from the high-case scenario for Bank lending at the rate of approximately $450 million per year. 40 Annex Al Colombia at a glance 9/7/01 Latin Lower- POVERTY and SOCIAL America middle- r Colombia & Carib. income Development diamond* 2000 Population, mid-year (millions) 42.3 516 2,046 Life expectancy GNI per capita (Atlas method, USS) 2,020 3,680 1,140 GNI (Atlas method, US$ billions) 85.3 1,895 2,327 Average annual growth, 1994-00 T Population (%) 1.9 1.6 1.0 Labor force (%) 2.7 2.3 1.3 GNI Gross per - primary Most recent estimate (latest year available, 1994-00) capita enrollment Poverty (1% of population below national poverty line) 55 Urban population (% of total population) 74 75 42 Life expectancy at birth (years) 70 70 69 infant mortality (per 1000 live births) 23 30 32 Child malnutrition (% of children under 5) 8 9 11 Access to improved water source Access to an improved water source (% ofpopulation) 91 85 80 liliteracy (% of population age 15+) 8 12 15 Gross primary enrollment ( aofschool-age population) 113 113 114 --Colombia Male 113 .. 116 Lower-middle-income group Female 112 .. 114 KEY ECONOMIC RATIOS and LONG-TERM TRENDS 1980 1990 1999 2000 Economic ratiosa GDP (US$ billions) 33.4 40.3 84.8 81.3 Gross domestic investment/GDP 19.1 18.5 12.5 12.2 Exportsofgoods and services/GDP 16.2 20.6 18.6 21.9 Trade Gross domestic savings/GDP 19.7 24.2 12.5 13.7 Gross national savings/GDP 19.6 21.6 13.4 14.9 Current account balance/GDP 0.3 1.3 -0.2 -0.2 Domestic Investment Interest payments/GDP 0.9 3.9 2.8 2.5 savings Total debt/GDP 20.8 42.8 41.0 41.9 Total debtservice/exports 15.1 40.9 51.3 51.1 Present value of debt/GDP .. .. 39.7 Present value of debt/exports .. .. 221.8 Indebtedness 1980-90 1990-00 1999 2000 2000-04 (average annual growth) GDP 3.6 3.0 -4.3 2.8 2,8 -Colombia GDP per capita 1.5 1.1 -5.7 1.0 0.9 - Lower-middle-income group Exports ofgoods and services 7.5 5.3 4.7 5.3 4.4 STRUCTURE of the ECONOMY 1980 1990 1999 2000 Growth of investrent and GDP (%) (Yo of GDP) 40 Agriculture 19.9 16.7 13.8 13.8 ;0 Industry 32.5 37.9 28.1 30.5 0 Manufacturing 23.9 20.6 13.7 13.8 Services 47.6 45.4 58.1 55.7 -20 -40 Private consumption 70.2 66.4 64.4 68.1 -s6 Generalgovermmentconsumption 10.1 9.4 23.1 18.2 _GDl 0 GDP Imports ofgoods and services 15.6 14.8 18.6 20.4 1980-90 1990-00 1999 2000 Growth of exports and imports (%) (average annual growth) Agriculture 2.9 -2.2 0.2 5.2 30 Industry 5.0 1.7 -9.0 3.8 5 Manufacturing 3.5 -2.3 -10.2 9.33 o Services 3.1 4.3 -1.1 1.9 Private consumption 2.6 2.6 -4.5 6.5 General government consumption 4.2 9.5 3.9 -14.5 -30 Gross domestic investment E.4 2.1 -38.3 15.2 -Exports 0 Imports Imports of goods and services 0.4 9.4 -22.3 5.8 1 Note: 2000 data are preliminary estimates. * The diamonds show four key indicators in the country (in bold) compared with its income-group average. If data are missing, the diamond will be incomplete. 4] Annex A I Colombia PRICES and GOVERNMENT FINANCE 1980 1990 1999 2000 Inflation (%) Domestic prices (% change) so - Consumer prices 26.4 32.4 9.2 8.8 40 Implicit GDP deflator 27.6 28.6 10.2 10.7 3 20 _ Government finance 12 - _C- (% of GDP, includes current grants) ,o Current revenue 15.8 12.7 13.2 95 96 97 98 99 0o Current budget balance . 6 8 -2.3 -2.9 --GDP deflator OCPI Overall surplus/deficit 3.9 -7.5 -6.1 TRADE 'US$ millions) 1980 1990 1999 2000 Export and import levels (USS mill.) Total exports (fob) 4,255 7,139 11,569 13,115 20.000 Coffee 2,361 1,415 1,324 1,069 Petroleum 101 1,951 3,757 4,569 15000 Manufactures 973 1,759 3,947 5,190 Total imports (cif) 4,663 5,589 10,659 11,538 10,000 Food 296 245 1,426 1,428 5OD Fuel and energy 564 330 252 234 Capital goods 1,616 2,034 3,651 3,414 o Export price index (1995=100) 7 61 194 243 94 95 96 97 98 99 oo Import price index (1995=100) 5 56 161 201 U Exports * Imports Termsoftrade(1995=100) 126 110 121 121 BALANCE of PAYMENTS (US$ millions) 1980 1990 1999 2000 Current account balance to GDP Exports of goods and services 5,747 8,687 13,866 15,608 0 Imports of goods and services 5,597 7,089 13,408 14,301 l 99 co Resource balance 150 1,597 457 1,307 Net income -211 -2,080 -1,422 -2,429 |2 Net current transfers 166 1,026 785 990 t i3 r-0 Currentaccountbalance 105 544 -180 -132 |4 Financing items (net) -430 53 502 -737 5t Changes in net reserves 325 -597 -322 869 | I Memo: Reserves including gold (US$ millions) .. .. 8,103 9,006 Conversion rate (DEC, local/USS) 47.3 502.3 1,756.6 2,087.6 EXTERNAL DEBT and RESOURCE FLOWS 1980 1990 1999 2000 (US$ millions) Composition of 1999 debt (USS mill.) Total debt outstanding and disbursed 6,940 17,222 34,678 34,081 IBRD 991 3,859 1,960 1,920 IDA 21 15 8 7 3965 1960 a Total debt service 951 3,889 7,117 7,974 52 I1ARD 144 751 390 370 IDA 0 1 1 1 1470 Composition of net resource flows Official grants 8 35 103 Official creditors 280 -40 970 275 Private creditors 531 -155 2,502 -415 Foreign direct investment 157 500 1,109 2,247 Portfolio equity 0 0 25 17 22883 World Bank program Commitments 351 536 591 350 A- IBRD E - Bilateral Disbursements 218 213 511 265 8-IDA D- Other multilateral F - Private Principal repayments 66 435 271 244 C - IMF G - Short-term Net flows 152 -221 241 21 1_1 Interest payments 79 318 121 127 Net transfers 73 -539 120 -105 Development Economics 9/7/01 42 Annex A2 Annex A2. Key Economic & Program Indicators-Change from Last CAS As Of Date 11/12/2001 Forecast in Last CAS Actual Current Forecast Economy (CY) 1997 11998b 999b 2000 1999c 2000c 2001 2002 2003 2004 Growth rates (%) GDP 2.4 4.2 4.0 3.2 -4.3 2.8 2.0 2.5 3.0 3.5 Exports 4.7 12.4 10.6 1.1 7.2 3.0 2.9 0 4.7 3.3 Imports 5.9 3.0 4.2 1.7 -25.4 8.0 12.2 3.5 3.8 8.2 Inflation (%) 18.5 18.0 17.0 16.0 9.2 8.8 8.0 6.0 4.0 4.0 National accounts (% GDP) Current account balance -4.8 -4.2 -3.7 -4.1 -0.2 -0.2 -2.7 -2.6 -2.9 -2.8 Gross investment 20.5 19.6 18.8 19.2 13.9 14.5 14.0 16.5 18.3 19.1 Public finance (% GDP) Central Administration balance -3.6 -2.4 -1.7 -1.4 -6.7 -5.9 -4.0 -3.9 -3.7 -2.8 International reserves 6.1 5.2 4.7 4.4 6.8 6.9 6.6 6.8 6.8 7.1 (as months of imports) Program (Bank's FY) FY97a FY98 b FY99 F FFyoo Y99c FY0Oc FY01 FY02 FY3b FY04' Lending ($ million) 345.0 375.0 550.0 140 930 190 630 665 Gross disbursements 265.0 372.0 404.0 238 471 183 490 450 ($ million) __= = a. Estimated year b. Projected year c. Actual outcome 43 44 Annex A3 Annex A3. Selected Indicators* of Bank Portfolio Performance and Management As Of Date 11/12/2001 Indicator 1999 2000 2001 2002 Portfolio Assessment Number of Projects Under Implementation a 20 26 22 23 Average Implementation Period (years) b 3.3 3.4 3.2 3.1 Percent of Problem Projects by Number a, c 0 7.7 4.5 4.3 Percent of Problem Projects by Amount a, c 0 6.9 5.5 5.7 Percent of Projects at Risk by Number a, d 0 7.7 4.5 4.3 Percent of Projects at Risk by Amount a, d 0 6.9 5.5 5.7 Disbursement Ratio (%) e 33.3 36.6 24.6 6.5 Portfolio Management CPPR during the year (yes/no) YES YES NO YES Supervision Resources (total US$) 1,033 1,245 1,240 1,594 Average Supervision (US$/project) 51.65 47.88 59.04 83.89 Memorandum Item Since FY 80 Last Five FYs Proj Eval by OED by Number 86 7 Proj Eval by OED by Amt (US$ millions) 6176 698.1 % of OED Projects Rated U or HU by Number 25.9 28.6 % of OED Projects Rated U or HU by Amt 23.7 20 a. As shown in the Annual Report on Portfolio Performance (except for current FY). b. Average age of projects in the Bank's country portfolio. c. Percent of projects rated U or HU on development objectives (DO) and/or implementation progress (IP). d. As defined under the Portfolio Improvement Program. e. Ratio of disbursements during the year to the undisbursed balance of the Bank's portfolio at the beginning of the year: Investment projects only. * All indicators are for projects active in the Portfolio, with the exception of Disbursement Ratio, which includes all active projects as well as projects which exited during the fiscal year. 45 46 Annex A4 Annex A4. IBRD/IDA Program Summary As Of Date 11/12/2001 Proposed IBRD/IDA Base-Case Lending Program a Strategic Rewards b Implementation b Fiscal year Proj ID US$(M) (H/M/L) Risks (H/M/L) 2002 Magdalena Medio 11 5.0 H H Water Sector Reform Assistance Project 40.0 H M Judicial Reform 5.0 H H Structural Fiscal Adjustmcnt Loan 400.0 H H Social Sector Adjustment 150.0 H H Productive Partnerships 31.5 M M 2003 Agricultural Technology Development 35.0 M M Labor/Tax SAL 400.0 H H Higher Education Financing Reform 150.0 H H Colombia-Bogota Urban Services 80.0 M M a. This table presents the proposed program for the next three fiscal years. b. For each project, indicate whether the strategic rewards and implementation risks are expected to be high (H), moderate (M), or low (L). 47 48 Annex A5 Annex A5. IFC & MIGA Program Summary FY 1999-2002 1999 2000 2001 2002 IFC approvals (US$m) 55.10 21.60 176.38 Sector (%) COLLECTIVE INV VEHICLE 11 FINANCE & INSURANCE 65 79 NONMETALLIC MINERAL 10 OIL, GAS AND MINING 35 Total 100 0 100 0 Investment instrument(%) Loans 45 4 Equity 9 33 Quasi-Equity 45 6 Other 57 Total 99 0 100 0 MIGA guarantees (US$m) 155.42 155.42 0.00 49 50 Annex A6 Annex A6. Summary of Nonlending Services As Of Date 11112/2001 Product Completion FY Cost (US$000) Audience ' Objective b Social Sector CEM FY02 87.5 G, B KG, PD Agro-Ecology FY02 60.0 G, B KG, PS Improv Pub Expend EF FY02 45.0 G, B KG, PS Colombia Rural Finance FY02 100.0 G, B KG, PD, PS Poverty Selected Topics FY02 100.0 G, B, PD KG, PD, PS Social Sector Review FY02 150.0 G, B, PD KG, PD, PS Environment, Land & Poverty FY02 120.0 G, B KG, PS Policy Notes FY02 100.0 G, B PD Labor Reform Agenda FY03 80.0 G, B, PD KG a. Government, donor, Bank, public dissemination. b. Knowledge generation, public debate, problem-solving. 51 52 Annex A7 Colombia Social Indicators Latest single year Same region/income group Latin Lower- America middle- 1970-75 1980-85 1994-00 & Carlb. income POPULATION Total population, mid-year (millions) 25.4 31.7 42.2 515.6 2,046.0 Growth rate (% annual average for period) 2.4 2.1 1.8 1.6 1.0 Urban population (% of population) 60.7 67.0 74.9 75.3 42.4 Total fertility rate (births per woman) .. .. .. 2.6 2.1 POVERTY (% of population) National headcount index .. .. 21.0 Urban headcount index .. .. Rural headcount index .. .. INCOME GNI per capita (US$) 560 1,150 2,460 3,680 1,140 Consumerprceindex(1995=1O0) 1 11 215 152 149 Food price index (1995=10O) .. .. INCOMEICONSUMPTION DISTRIBUTION Gini index Lowest quintile (% of income or consumption) 3.9 .. 3.1 Highest quintile (% of income or consumption) 58.8 .. 61.5 SOCIAL INDICATORS Public expenditure Health (% of GDP) . .. .. 3.4 2.7 Education (% of GNI) .. .. .. 3.6 4.7 Social security and welfare (% of GDP) .. .. .. 7.4 Net primary school enrollment rate (% of age group) Total .. .. .. 91 99 Male .. .. .. .. 100 Female .. .. .. .. 99 Access to an improved water source (% of population) Total 64 91 75 85 80 Urban 86 100 90 93 95 Rural 33 76 32 62 69 Immunization rate (% under 12 months) Measles .. .. .. 90 87 DPT .. .. .. 87 87 Child malnutrition (% under 5 years) .. 17 8 9 11 Life expectancy at birth (years) Total 62 67 70 70 69 Male .. .. .. 67 67 Female .. .. .. 73 72 Mortality Infant (per thousand live births) 62 37 24 30 32 Under 5 (per thousand live births) .. .. .. 38 40 Adult (15-59) Male (per 1,000 population) .. .. .. 207 192 Female (per 1,000 population) .. .. .. 122 133 Matemal (per 100,000 live births) .. .. Births attended by skilled health staff (/.) . .. 10105101 Note: 0 or 0.0 means zero or less than half the unit shown. Net enrollment ratios exceeding 100 indicate discrepancies between the estimates of school-age population and reported enrollment data. 53 54 Annex A8 Key Economic Indicators National accounts (as % GDP at current market prices) Gross domestic product 100.0 100.0 100.0 100.0 100.0 100.0 1000 100.0 Agriculture' 13.7 14.2 13.8 13.8 16.0 15.9 15.9 15.8 Industyy' 29.4 28.3 28.1 30.5 28.3 28.9 29.1 29.2 Services 56.9 57.5 58.1 55.7 55.7 55.2 55.0 55.0 Total Consumption 85.0 86.2 87.5 86.3 85.4 85.5 85.8 86.0 Grossdomesticfixed 21.3 19.6 139 14.5 14.0 16.5 18.3 19.1 investment Public investment .. 8.0 8.7 7.5 7.0 8.2 8.3 8.3 Private investment .. 11.6 5.2 7.0 7.0 8.3 10.0 10.8 Exports (GNFS)5 14.8 15.1 18.6 21.9 18.9 19.9 19.2 18.7 Imports (GNFS) 20.8 21.0 18.6 20.4 20.8 20.8 20.6 19.7 Gross domestic savings 15.0 13.8 12.5 13.7 14.6 14.5 14.2 14.0 Gross national savingse 15.6 14.2 13.4 14.9 15.0 15.5 15.4 15.2 Memorandum items Gross domestic product 106719 98806 84845 81290 82285 85151 89150 94697 (US$ million at current prices) Gross national product per 3450.0 2600.0 2200.0 2020.0 1870.0 1900.0 1960.0 2030.0 capita (US$, Atlas method) Real annual growth rates (%, calculated from 1994 prices) Gross domestic productat 3.4 0.6 -4.3 2.8 2.0 2.5 3.0 3.5 market prices Gross Domestic Income 3.6 -1.1 -3.8 4.2 1.8 3.9 2.6 4.2 Real annual per capita growth rates (%, calculated from 1994 prices) Gross domestic product at 1.5 -1.3 -5.7 1.0 0.3 0.9 1.4 1.2 market prices Total consumption 3.3 -1.7 -4.2 -0.8 2.6 0.4 0.4 0.3 Private consumption 0.5 -2.7 -6.2 4.6 3.4 0.5 0.3 0.4 55 Annex A8 Balance of Payments (USSm) Exports(GNFS) 14195.0 13560.0 13865.5 15608.0 155118 16204.2 164736 172554 MerchandiseFOB 115293 10944.3 11556.1 13619.7 13722.3 13520.0 13655.2 13928.3 lmports(GNFS) 184109 17539.5 13408.3 14301.4 16515.5 16978.1 17657.2 181123 Merchandise FOB 14371.4 13691.9 9899.5 11076.9 12162.4 12537.9 13039.5 13561.0 Resource Balance -4216.0 -3979.5 457.2 1306.6 -1003.7 -773.9 -1183.6 -856.9 Net current transfers 613.6 444.0 785.4 989.8 382.2 887.9 1056.6 1105.0 (including official current transfers) Current account balance -5952.9 -5270.3 -179.8 -132.4 -2665.0 -2524.0 -2514.0 -2602.0 (after official capital grants) Netprivateforeign direct 4829.6 2432.5 1011.6 1181 1218 1233 1798 1810 investment Long-term loans (net) 2199.8 932.1 504.7 -294.0 607.0 1383.0 986.0 1386.3 Official -452.1 103.6 969.5 274.6 -73.5 -618.7 -928.3 -589.0 Change in reserves' 277.4 -1390.1 -321.8 869.0 296.0 278.0 234.0 310.0 Memorandum items Resource balance (% of -4.0 -4.0 0.5 1.6 -1.2 -0.9 -1.3 -0.9 GDP at current market prices) Real annual growth rates YR94 prices) Merchandise exports 5.7 9.1 7.2 3.0 2.9 0.0 4.7 3.3 (FOB) Primary -3.7 3.2 24.3 -0.6 -0.6 -8.3 4.3 1.5 Manufactures 4.2 19.3 1.1 5.4 5.2 5.9 4.3 4.3 Merchandise imports -4.5 0.3 -25.4 8.0 12.2 3.5 3.8 8.2 (CIF) Public finance (as % of GDP at current market prices)' Current revenues 12.6 11.9 12.7 13.2 14.9 14.7 14.1 14.0 Current expenditures 12.4 13.9 15.0 16.1 16.2 15.9 15.0 14.4 56 Annex A8 Actual Estimate P:ctd 7 0i ator - 1997 1998 1999 2000 200t1 2002 .203 2004 Current balance, surplus (+) 0.2 -2.0 -2.3 -2.9 -1.3 -1.2 -0.9 -0.4 or deficit (-) Monetary indkators M2/GDP (at current market 36.0 34.4 36.0 33.1 34.7 34.9 35.3 35 5 prices) GrowthofM2(%) 25.8 10.9 10.5 47 12.0 6.0 4.0 4.0 Private sector credit growth! 79.2 76.5 -492 20.1 66.2 78.0 110.7 195.3 total credit growth (%) Price indices( YR94 =100) Merchandise export price 1084 935 92.9 102.2 97.7 100.1 98.1 97.5 index Merchandise import price 86.0 81.5 79.6 115.9 78.8 81.6 83.0 84.8 index Merchandiseterrnsoftrade 126.1 114.7 116.6 88.2 124.0 125.7 122.3 126.5 index Real exchange rate 119.1 1136 102.7 98.9 96.2 93.4 93.4 93.4 (USsLCU)J Consumerprice index 17.7% 16.7% 9.2% 8.8% 8.0% 6.0% 4.0% 4.0% (%/o growth rate) GDP deflator 16.8% 15 2% 10 2% 10.7% 11.1% 7.5% 6.2% 5.7% (% growth rate) a. If GDP components are estimated at factor cost, a footnoote indicating t'Data-YR' fact should be added. b. "GNFS" denotes "goods and nonfactor services." c. Includes net unrequited transfers excluding official capital grants. d. Includes use of IMF resources. e. Data refer to the Central Administration f "LCU" denotes "local currency units." An increase in US$/LCU denotes appreciation. 57 58 Annex A9 Key Exposure Indicators 2g ~~~~~~tt '-.,. Estimate PrIo . 99,~~a. 13. 19" 1999 2000 2001 2002 X2003--,- Z Total debt outstanding and 31800 33338 34678 34081 35958 38403 41990 46307 disbursed (TDO) (US$m)a Total debt service (TDS) 6960 6584 7117 7974 7947 8986 9112 10280 (US$m)a Debt and debt service indicators (%) TDO/XGSb 224.0 246.0 250.0 218.0 232.0 237.0 255.0 268.0 TDO/GDP 29.8 33.7 41.0 41.9 43.7 45.1 47.1 48.9 TDS/XGS 49.0 48.6 51.3 51.1 51.2 55.5 55.3 59.6 IBRD exposure indicators (%) IBRD DS/public DS 21.0 11.9 8.2 10.4 9.2 9.5 9.2 8.4 Preferred creditor DS/public 42.1 24.7 21.1 25.7 26.200 24.8 22.3 21.1 DS (%)c IBRD DS/XGS 4.0 2.6 2.8 2.4 2.3 2.3 2.4 2.2 IBRD TDO (US$m) 1723 1740 1961 1920 2090 2340 2350 2220 IDA TDO (US$m) 10 9 8 7 6 5 4 3 a. Includes public and publicly guaranteed debt, private nonguaranteed, use of IMF credits and net short- term capital. b. "XGS" denotes exports of goods and services, including workers' remittances. c. Preferred creditors are defined as IBRD, IDA, the regional multilateral development banks, the IMF, and the Bank for International Settlements. 59 60 Annex A 10 Annex AIO. Operations Portfolio (IBRD/LDA and Grants) A. Of D0e. II/12620 CIsd Pn,J.. 141 IORD/MA . Trs1 Dunb.d (Aov-) 616 50 ofnt.nh h6. brcn n:rd 35.16 T-1I Dhsobnnd (Clo-d) 7,567.30 .f vhich hr. bonn np.Wd 7,646 40 To1l D)bunud (AO-n I Cl1-n 81,83 S2 of th.h h. been rnpid 7,681 56 o.b1 UWnIi,b-nnd (A-On) 67s56 Tobl Und6b.e.d (CI.Md) D W T.Wl UOnd.b.n-d (Afiv0 Cl-d) 678 56 A,ti, Peeio.b niffer.ne- BeOt-ten L-t PSR E.Wpted and Aeto.l Su-prrvbirn Ratig Orid,md An-int hn US$ Miio- D.b ......_b _ P.jeni ID Pnjeet N..,c mn-an,M 1a.0lmn ta"tPn i F..lI Ye- IBlRD GRANT Cranel. Undl,b. 0Ori. Fr R evd P057692 2nd Magdalena Medio Prmject 6 2002 5 5 P006880 AGRICULTURETECHNOLO S HS 1995 51 11.8 118 -5.7 P044140 CARTAGENAWATERSUPPLY &SEWERAGE ENVIRO S S 2000 85 68.8 239 P040109 CO PUBLIC FiNANC MANAGEMENT PROJECTII S S 2001 3535 35 2 -0.2 P006861 CO URBAN INFRASTRUCTURE U U 1990 75 573 29.5 P006891 CO- ANTIOQUIA EDUCATION S S 1998 40 26 183 16 P068762 CO- COMMUNITY WORKS (MANOS A LA OBRA) S S 2000 100 96 28 P069964 CO- H.man Capit Patn- Cash Trmsfer S S 2001 150 150 0 P046112 CO- PASTOEDUCAT7ON S S 1998 72 36 29 03 P050578 CO- RURAL EDUCATION S S 2000 20 I0.0 42 P050576 CO- YOUTH DEVELOPMENT S S 1999 5 3 3 P063317 COLOMBtAHIGHANDES S S 2001 154 14.7 0.7 P065263 EARTHQUAKE RECOVERY HS HS 2000 225 104.2 55 9 POl6S93 ENERGY TECHNICAL ASSISTANCE S S 1995 11 0.5 0.5 P006884 FINANCIAL MARKETS DEVELOPMENT S S 1997 15 5 5 3 2 8.7 P054125 ODS-PHASE OUT S S 1999 0.7 8 3 0 P053243 PEASANTENTERPRISEZ S S 199S 5 27 2.4 P006887 POWER MARKET DEVELOPMENT & ENERGY (TA) HS S 1996 2493 1 1 1 P040102 REG REFTA HS S 1997 12.5 5.3 53 P006894 SANTAFE I (Wat.r/S.pply) S S 1996 145 39.9 39 P057326 SIERRA NEVADA SUSTAINABLE DEVELOPMENT S S 2000 5 4 4 -0.6 P039291 URBANENVIRONMENTTA S S 1996 20 1.9 1 9 -04 P065937 WATER SECTOR REFORM ASSISTANCE PROJECT 6 # 2002 40 40 Overall result Result 13015 24 55 7015 236.1 -3 2 a Intended d,sb.rsements to date mmuns aenasl dsbursement. to dale as pojected at appraisal 62 Annex Al I Annex All. Statement of IFC's Held and Disbursed Portfolio As of 7/31/2001 (In $US Millions) Held Disbursed FY Approval Company Loan Equity Quasi Partic Loan Equity Quasi Partic 1969/85/88/93/95 CF del Valle 0 0 0 0 0 0 0 0 2001 CHMC 0 0.21 0 0 0 0.21 0 0 1974/01 Cementos Caribe 4.05 10 0 12.95 4.05 10 0 12.95 1963/90 Coltejer 6.02 0 0 0 6.02 0 0 0 1995/99 Corfinsura 25 0 25 0 0 0 25 0 1987 PRODESAL 0 0.59 0 0 0 0.59 0 0 1977/89/92/94/96 Promigas 6.25 0 0 12.5 6.25 0 0 12.5 1994/95 Promisan 0 0.2 0 0 0 0.2 0 0 1996 Proyectos 0 5 0 0 0 5 0 0 1997 Suleasing 9.1 0 0 0 3.8 0 0 0 1999 Surenting 0 5.1 0 0 0 2.5 0 0 2001 Tolcemento 3.33 0 0 10.67 0 0 0 0 Total Portfolio: 53.75 21.1 25 36.12 20.12 18.5 25 25.45 Approvals Pending Commitment Loan Equity Quasi Partic 2001 CHMC 100000 0 0 0 2001 CCGF 0 19000 0 0 Total Pending Commitment: 100000 19000 0 0 63 64 Annex B ANNEX 1B REPUBLIC OF COLOMBIA FISCAL STRENGTHENING PROGRAM-STRUCTURAL FISCAL ADJUSTMENT LOAN (SFAL) THE MACROECONOMIC FRAMEWORK RECENT ECONOMIC DEVELOPMENTS B1. A moderate recovery took place in the Colombian economy in 2000 and this recovery has continued, although at a slower pace, in 2001. The recovery comes on the heels of the 1999 recession, when real GDP contracted by 4.3 percent (Table A.B. 1), and which was the deepest economic crisis Colombia has experienced in the last 70 years. The recession had its roots in large imbalances in both the fiscal and external accounts, which were exacerbated by the Asian and Russian crises and drops in oil and coffee prices in the late 1990s. Fragile and deteriorating Colombian domestic security conditions further alarmed investors, both domestic and foreign. B2. Profound changes in the early 1990s conspired against fiscal and macroeconomic stability in Colombia. Changes in the Constitution in 1991 generated significant new financial obligations for the government, and set the stage for what should have been an important shift in the role of government. The most important changes were the consolidation of the decentralization process, reforms in social security, and a renewed emphasis on the role of the state in the provision of social services, justice, and security. The fiscal implications of the new rules were large. Transfers from the central to local goverments nearly doubled from 22 percent of current revenues in 1990 to over 40 percent in 2000, social security expenditures at the central level grew from 2.9 percent to 5.3 percent of GDP between 1990 and 1997, and defense spending increased from 1.4 percent to 3 percent of GDP over the same period. Table A.B.1. Colombia: Selected Macroeconomic Indicators, 1990-2000 In Percent of GDP (Unless Otherwise Indicated) 1990-94 1996 1997 1998 1999 2000 Real GDP Growth (%) 4.5 2.0 3.1 0.6 -4.3 2.8 NFPS Balance -0.7 -3.0 -3.9 -4.6 -6.4 -3.6 Central Government Balance -1.3 -3.7 -3.7 -4.9 -7.5 -6.1 Current Account Balance -3.7 -4.9 -5.5 -5.3 -0.1 -0.2 Inflation Rate (%/o) 24.8 21.6 17.6 16.7 9.2 8.8 Investment 18.4 22.2 21.3 19.6 13.9 14.5 Unemployment Rate (%) 9.5 11.2 12.4 15.2 19.4 20.2 B3. The emphasis on decentralization was not, however, compensated by a reduction in the size of the central government. Central government expenditures, other than transfers to local 65 Annex B governments and to social security, continued to rise during the 1990s, growing by 50 percent, from 4.2 percent to 6.3 percent of GDP during the decade. The mounting pressures led to continuous attempts to raise revenues through no fewer than eight tax bills, with little empnasis until recently on trimming spending. While central government tax revenues increased by 1.9 percentage points, expenditures did so by 3.3 percentage points during the decade. Fromn a situation of a nearly balanced budget in the early 1990s, Colombia thus ended 1999 with a Non- Financial Public Sector (NFPS) deficit of 6.4 percent of GDP. B4. In addition, a rapid real appreciation of the peso through the early and mid-I 990s, partly as a result of large capital inflows due to new oil discoveries, increased borrowing from abroad as a result of the country's new investment-grade rating, and the repatriation of drug profits, undermined trade liberalization efforts. Capital inflows pushed up asset prices, particularly real estate. A strong peso was supported by a newly independent Central Bank with a Constitutional mandate to control inflation. The appreciating currency was particularly detrimental to the agricultural sector and to textile manufactures. In addition, while capital and imports were relatively cheap, labor became increasingly costly. The combination of these factors led to growing fiscal and external deficits, pushing Colombia first toward slower growth and rising unemployment from 1996 onward, and later into a full-blown recession. The 1998-99 Recession B5. The current account deficit had reached worrisome levels by 1997-98, standing at above 5 percent of GDP. The deterioration in the current account began in 1993 as a result of rapid growth in private investment and borrowing spurred by a boom in the real estate sector. By 1997, however, the external deficit reflected the imbalance in the fiscal accounts, spurred by growing public expenditures and rising interest payments. In addition, export revenues fell in 1997 and 1998 as a result of the drop in oil and coffee prices, Colombia's two largest exports, further aggravating the problem. B6. Colombia's deteriorating fumdamentals, combined with a trade shock estimated at :.5 percent of GDP, and with events in international credit markets, resulted in the virtual closing of financial markets to the country. This forced the Colombian authorities to turn to the domestic market to fulfill their large financing needs, putting increasing pressure on domestic interest rates. Interest rates were driven even higher by continued attacks on the peso in 1998 and 1999, confirming market concerns regarding the sustainability of Colombia's fiscal and external imbalances. The failed attempts by the Central Bank to defend the existing exchange rate band led to a cumulative loss of US$1.8 billion in reserves. The benchmark deposit rate had riser to almost 20 percent in real terms, while the annual real lending rate rose to 26 percent in late 1998. The exchange rate band was finally abandoned in September 1999 and the peso was allowed to float freely., B7. The economy, already suffering from the earlier demise of the construction sector in 1995-96, was thus hit by the collapse of productive activity as a result of galloping interest rates. Private investment fell by over 60 percent and private consumption dropped 6 percent. 1 lie 1. The peso has depreciated by about 15 percent in nominal terms since the collapse of the band, and the behav ior of the exchange market has remained largely orderly. 66 Annex B sectors that suffered the most marked contractions were construction (-20 percent), manufacturing (-15 percent), and commerce and financial services (each -7 percent). With the severe recession that hit the country, however, imports plummeted as domestic demand collapsed, while exports benefited from the high international price of oil and the devaluation of the peso. The current account in 1999 and 2000 was therefore in near balance. B8. The condition of the financial system, already under strain, worsened during 1998 and 1999 due to the sharp deterioration of loan portfolios and a large volume of nonproductive assets. High interest rates and negative economic growth stretched the payment capacity of most borrowers, and arrears in the loan portfolios grew to almost 15 percent in 1999. To head off a systemic crisis in the financial system, the government put in place programs to assist financial institutions and mortgage holders. Measures were also put in place to recapitalize viable private banks and to address the problems of public banks.2 B9. Economic woes have been accompanied by high unemployment levels of about 20 percent, which have not reacted to the moderate resumption of growth over 2000-01. Shrinking demand has also been partially responsible for the drop in inflation from nearly 18 percent in 1997 to under 9 percent in 2000. Macroeconomic Adjustment BlO. In response to the severity of the recession, the Colombian government embarked on a program of stabilization, supported by a three-year (2000-02) Extended Fund Facility from the IMF. The government program aims at increasing revenues and reducing expenditures; this is to be complemented by a rationalization and definition of the roles and competencies of central and local government. B 11. A number of measures, mainly directed to enhancing revenues, have been taken. These include (a) passage of sound budgets for 2000 and 2001; (b) widening of the value added tax base; (c) improving tax collection and fighting evasion and smuggling; (d) changes in royalty allocation to attract private investment, particularly to the oil sector; and (e) recent passage of a tax bill expected to yield additional revenues on the order of 1.8 percent of GDP. These revenue measures are being complemented with some structural measures in a number of areas, including expenditure reduction. A financial system reform law was passed in 1999 raising minimum capital requirements and strengthening the faculties of the Banking Superintendency and Financial Guarantee Fund (FOGAFIN). A constitutional amendment was passed that would save part of the transfers to local government in order to finance local pension funds. Moreover, another constitutional amendment that de-links territorial transfers from the central government's current revenues for a period of eight years has also been approved by Congress. In addition, legislation to limit the operational expenditures of government at the local level was also approved. Further, a number of critical reforms are either currently under consideration by Congress or are expected to be presented to legislative sessions in 2001 and/or in the first session 2. The govermment estimates the cost of cleaning up the balance sheets of private banks at about 4.5 percent of GDP, nearly half of which would come out of public coffers through FOGAFIN loans, with an additional 4 percent of GDP for the restructuring of public banks. 67 Annex B of 2002. These include a reform of the resources and competencies of local government (Law 60/93), and a territorial tax reform. Economic Recovery B12. The government's efforts have yielded encouraging results. Real GDP grew by 2.8 percent in 2000 following the 4.3 percent drop in output in 1999. The main engine of economriic growth during 2000 was the recovery of exports, both traditional and nontraditional. Manufacturing output expanded by about 9 percent, and inflation ended the year at 8.8 percetnt, well below the Central Bank's 10 percent inflation target. Interest rates have come down from their historic highs to about 4 percent in real terms. B13. Colombia's fiscal performance was helped by higher than projected oil revenues and lower capital spending during 2000. Notwithstanding a number of setbacks, including the Constitutional Court decision of October 2000 to retroactively raise public sector wages, and the financial deterioration in the public pension and health systems, the target Non-Financial Public Sector (NFPS) deficit of 3.6 percent of GDP was achieved in 2000. B14. The recovery is nevertheless fragile, because external conditions have deteriorated substantially in 2001, with a sharp slowdown in world economic growth and international trade, accompanied by drastic declines in oil and coffee prices.3 Internal demand growth has been sluggish since 2000. Prospects for a significant reactivation in the construction and real estate sectors looked reasonable until recently, but credit from the financial system has stalled in real terrns. The private sector has been paying off its debt and appears cautious about borrowing significantly for accelerating investments. The financial system, although stronger as a whole, remains fragile. The oil sector, Colombia's main export earner, will see some declines in production as a result of the aging of oil fields and guerrilla attacks on pipelines, combined xNith volatile international prices. Unemployment has remained at about 20 percent since the beginning of 2000, and in response much of the government's current investment program focuses on strengthening the social safety net, targeted to mitigate the negative impact of The economic contraction on the poor. B15. The shortfall in privatization revenues during 2000 forced the government to rely heavily on both extemal and domestic credit. Public sector external debt at the end of 2000 stood at about 26 percent of GDP, up from 19.5 percent and 24 percent in 1998 and 1999, respectively; total net public debt was about 42 percent of GDP. Domestically, there was a significant increase in the holding of government securities by the private sector. At the same time. a significant shift to more liquid monetary instruments took place, with broad money growing by only 2.7 percent, while narrow money increased by nearly 14 percent. This was the result of the combined effect of the financial transactions tax and the problems of the financial system. Financial system credit to the private sector stagnated during the year, and the econom-ic recovery was financed largely by the productive sectors through the use of own capital and installed capacity. 3. The slowdown in world growth and international trade has intensified following unprecedented terrorist attacks against the United States on September 11, 2001, and subsequent events. 68 Annex B ECONOMIC PROSPECTS B16. GDP growth of 1.5 to 2.5 percent is projected for 2001, with a bias toward the lower end of the range. The slowdown from 2.8 percent growth in 2000 is due to uncertainties regarding capital markets and a weakening of investment spending, deteriorating terms of trade and disruption of oil exports, and extensive strikes in the public and private sectors in the first half of 2001 (see Table A.B.2). Under a medium-term scenario that assumes full implementation of structural reforms to strengthen the fiscal accounts, including the reforms to intergovernmental fiscal relations and education expenditures, and the restructuring of the ISS and public hospitals, an increase in the rate of GDP growth to levels approaching 2.5 to 3 percent is projected for the 2002-03 period, and to 3 to 3.5 percent for 2004-05. Table A.B.2. Colombia: Selected Macroeconomic Projections, 2000-05 In Percent of GDP (unless otherwise indicated) 2000 2001 2002 2003 2004 2005 GDP Growth (%) 2.8 2.0 2.5 3.0 3.5 3.5 NFPS Balance -3.6 -3.3 -3.1 -2.5 -2.0 -1.5 Central Government Balance -6.1 -5.7 -4.6 -3.6 -2.6 -2.3 Current Account Balance -0.2 -2.6 -2.6 -2.9 -2.8 -2.7 Inflation (%) 8.8 8.0 6.0 4.0 4.0 4.0 Investment 14.5 15.7 16.5 18.3 19.1 19.2 Unemployment Rate (%) 20.2 18.0 17.5 15.7 15.3 14.8 Source: World Bank Staff projections B17. It is expected that in 2001 the NFPS fiscal deficit will be about 3.3 percent of GDP (higher than the 2.8 percent of GDP target originally agreed under the IMF program although it is likely that the program target will be modified). The expected shortfall is due to weaker than programmed tax collections, mainly as a result of weakening economic activity, which will offset lower-than-projected investment spending and bank restructuring costs. The objective of the government was to end its four-year term in 2002 with a fiscal deficit close to 2.5 percent of GDP. Achieving this goal, and maintaining sustainable deficits into the future, requires the passage of structural reforms outlined above that are designed to improve control over public sector expenditures. A reform of the public pension system is assumed-this reform, however, would not significantly affect the fiscal balances within the horizon of the projections included below. Thus, the full impact of the reform would fall outside the projection period, and most of the savings in the short term arise from the improved administration of the ongoing transition, an important component of the overall pension reform efforts. Under such a scenario, the NFPS deficit would decline from 3.6 percent of GDP in 2000 to 3.3 percent in 2001, to about 3.1 percent of GDP the following year, and would steadily approach 1.5 percent of GDP by 2005. The accruing benefits of structural reforms would make this fiscal stabilization possible despite expected declines in oil production over the next five years. Under this scenario, Colombia's total public debt would stabilize at about 50 percent of GDP in the long term. 69 Annex B BI 8. Attaining the envisioned level of economic growth would also be supported by increases in public savings, thus helping to raise national savings and maintain a current account deficit in the 2 to 3 percent-of-GDP range over the medium term. The current account deficit is projected to increase by about 2.5 percent of GDP in 2001 as traditional exports, particularly oil and coffee, weaken sharply both in volume and price, while growing domestic demand sustains imports. The decline in the current account balance is expected to be moderated by a devaluation rate of 10 percent this year, helping the growth of nontraditional exports, despite the risk of weakening demand in the United States in particular, and increasing the cost of imports. The capital account is expected to strengthen in 2001 as a result of further public seclor borrowing and reduced net capital outflows from the private sector. International reserves are projected at about seven months of imports in the coming years, with small yearly increases resulting largely from interest earnings on the reserves. B19. A reduction of inflation from a targeted level of 8 percent in 2001 to 6 percent and 4 percent in 2002 and 2003, respectively, is forecast as the Central Bank moves toward inflation targeting in conducting its monetary policy. While some inflationary pressures can be expected as the economic recovery proceeds, continued high levels of unemployment, slackness in external and domestic demand, and productivity gains should enable the attainment of inflation targets. Also forecast is the resumption in the growth of broad money aggregates and financial system credit to the private sector following the decline experienced in 1999 and 2000. B20. A higher national savings rate and a sustainable current account deficit would create the required space for expansion in private investment that is critical to achieving an economic growth of 3.5 percent or above. Both private and public investment dropped significantly during the 1998-99 recession, and have not rebounded to levels sufficient for sustaining long-term growth at the 3.5 percent level envisioned-total investment in Colombia stands at about 14 percent of GDP. Central government investment has been reduced to about 1.5 percent of GD]', half its average level during 1990-97, and represents about half the resources currently spent on interest payments by central government debt. Private investment has undergone a decline since 1994, and now stands at only 7 percent of GDP. B21. The key ingredient for improving the incentives for investment is the reduction in uncertainty, both political and economic. In the political front, peace negotiations have been the main priority of the Pastrana Administration; progress, however, has been slow, and a comprehensive agreement is unlikely to be achieved in the near future. While realizing Colombia's full growth potential will not happen under conditions of conflict, much can nevertheless be done on the economic front to improve the investment climate in the country. Bringing the fiscal accounts onto a sustainable path is therefore doubly important, and would help to significantly reduce the country risk associated with Colombia. The reduction of the deficit, and therefore of the need to tap domestic markets to finance the deficit, would not only open needed space for the private sector, but would also enhance the credibility of the stabilization program and the prospects for the future. Fiscal stabilization would also create the space for public sector investment in areas complementary to private sector activity, effectiveyv ''crowding in" private investment. B22. The prospects for a consolidation of the economic recovery in 2002-03 and beyond thus depend on the success of the government in garnering support for structural reforms to 70 Annex B strengthen the fiscal position and on improvements in the climate for investment in the country. The rationalization of decentralization is a particularly critical reform, and is central to the solution of the fiscal challenges facing Colombia in the coming years. The realignment of revenues and responsibilities at the different levels of government is fundamental to achieving control over Colombia's fiscal balances, not to mention its importance in improving the delivery of education and health services, and in the strengthening of human capital in the country. External Financing Requirements B23. The government plans to obtain the bulk of its deficit financing from external sources and has already placed about US$2.7 billion in international bonds to cover its privately provided financing for the year, including US$1 billion backed by a World Bank policy-based guarantee operation. The remainder of its gross financing needs, approximately US$1.8 billion, is expected to come from multilateral institutions. As such, although Colombia's spreads would be vulnerable to a deterioration in the external environment toward the Region, the impact on access to financing from international capital markets is likely to remain limited, at least during 2001. The government's strategy of decreasing its reliance on domestic financing is expected to be maintained over the coming years in order to decrease the crowding out of the private sector and to set the stage for sustained investment and growth. Negative net domestic financing is envisioned from 2002 onward, with limited recourse to privatization.4 B24. Projections of Colombia's external financing requirements and sources of finance are presented in Table A.B.3. Total public and private sector gross external financing requirements are estimated at US$8.1 billion, US$7.8 billion, and US$8.4 billion for 2001, 2002, and 2003, respectively. These external financing requirements are expected to be financed largely through public and private debt from both multilateral and private sources, including bond placements. A smaller portion, roughly 15 to 20 percent, of these financing needs will be covered by private investment flows. Private investment is initially projected to remain at levels close to their current levels of around US$1.2 billion. A small accumulation of reserves is forecast in the next three years, mostly as a result of interest income. B25. The public sector will continue a strategy of refinancing external public debt to improve its cost and maturity structure if international market conditions allow. Maintaining adequate access to longer-term financing as provided by the World Bank and other international financial organizations will continue to be essential for improving the structure of Colombia's external public debt. 4. Domestic debt restructuring has already improved the government's domestic debt profile. In early June the authorities swapped US$2.4 billion of domestic debt due in 2001-05 with paper of longer maturity. As a result of this swap, the average maturity of domestic debt increased from 3.5 to 4.5 years; projected domestic debt redemptions fell 22 percent in 2001, 34 percent in 2002, and 40 percent in 2003. 71 Annex B Table A.B.3. Colombia: External Financing Requirements, 200-03 (US$ Millions) 2000 2001 2002 2003 Financing Requirements (1+2+3) 5,285 8,118 7,799 8,400 1. Current Account Deficit 132 2,665 2,524 2,514 2. External Debt Amortization 4,284 5,157 4,997 5,652 a. Public and Publicly Guaranteed 1,464 2,401 2,717 3,602 b. Private 2,820 2,756 2,280 2,050 3. Change in Reserves 869 296 278 234 Sources of Financing (5+6) 5,170 8,118 7,799 8,400 5. External Debt 3,989 6,900 6,566 6,602 a. Public Sector 3,395 4,875 4,503 4,377 Multilateral and Bilateral 1,703 1,950 1,850 1,702 Gross Disbursements Private Creditors 1,692 2,925 2,653 2,675 b. Private Non-Guaranteed 594 2,025 2,063 2,225 6. Net Private Investment 1,181 1,218 1,233 1,798 Source: World Bank Staff projections 72 Annex C ANNEX C REPUBLIC OF COLOMBIA FISCAL STRENGTHENING PROGRAM-STRUCTURAL FISCAL ADJUSTMENT LOAN (SFAL) FISCAL SUSTAINABILITY, ECONOMIC GROWTH, AND POVERTY BACKGROUND Cl. The proposal for fiscal reforms to ensure fiscal sustainability is a necessary condition to recover high positive GDP per capita growth in Colombia and to return to the poverty reduction path that the Colombian economy enjoyed until the recent recession. Such a reform would reduce poverty by 2010 to the prerecession levels of 1995. However, without fiscal sustainability, low growth and high unemployment will continue and poverty will remain almost unchanged. C2. This annex assesses the consequences of two alternative policy actions for the poor in Colombia (a) implementing a fiscal reform package to recover positive real GDP per capita growth; and (b) letting the fiscal situation deteriorate toward unsustainable levels. Given two scenarios of fiscal balance and growth during 2001-10, microdata simulations of poverty levels under these two alternative policy paths are projected. To calibrate the model, four urban household surveys from the last two decades are used, measuring the impact of GDP growth on income generation and poverty levels-directly and mediated via employment generation connected to poverty profiles. C3. Two main findings are presented here. First, under the status quo, or fiscal instability, low growth and high unemployment would continue and poverty would remain almost unchanged relative to the current level. Fiscal sustainability would, however, deteriorate even further if the GDP growth rate fell below the population growth rate. Second, if the proposal for fiscal reform is implemented, high positive growth of GDP per capita is recovered. As a result, Colombia would experience substantial reductions in poverty during the next decade, recovering the losses suffered during the recent recession. C4. This annex (a) shows why fiscal sustainability is necessary to recover economic growth; (b) examines the experience of growth and poverty in Colombia; and (c) shows why growth has been the key factor in poverty reduction. It measures both the elasticity of poverty to growth and the links between the probability of being poor and growth via employment generation. This annex also estimates the potential gains (or losses) in poverty reduction if fiscal reforms were implemented, and if they were not implemented; shows whether the proposed fiscal adjustment has any adverse effects on the poor via the delivery of social services, public salaries, or pension equalization; and presents a summary and conclusions. 73 Annex C FISCAL SUSTAINABILITY IS NECESSARY TO RECOVER ECONOMIC GROWTH C5. Sustainable fiscal policy is an essential component for strong macroecononic performance. Fiscal deficits can lead to slow growth rates since they often lead to rising interest rates and financial repression, as well as an increase in the current account deficit and, in turn, real exchange rate appreciation. Likewise, deficits financed through seignorage indisputably lead to higher inflation, under which uncertainty increases, relative prices are distorted, and private investment falls. On the other hand, low and stable fiscal deficits can increase prospects for growth, which in turn allow for further sound fiscal management, thus creating a virtuous circle. I C6. Until the early 1990s, prudent management of the Colombian economy allowed for low goxvernment debt levels which, together with low inflation rates by Latin American standards. led to steady-although moderate-growth rates. However, Colombia's current fiscal path is leading to an unsustainable increase in deficits and debt-to-GDP ratio over the medium and long term, with dim prospects for GDP growth if reform is not undertaken. C7. The government has recognized the need to strengthen fiscal accounts and set a sclid foundation for further reforms. If all reforms of the proposed comprehensive fiscal strengthening program (including Acto Legislativo, pension reforms, territorial tax reform) are approved before 2001, it is expected that Colombia will be able to undertake a high, sustainable growth path for the next 10 years. Projections cited in Table A.C.1 are set up such that the main effect of fiscal reform is on debt accumulation, with growth rates converging to 4.5 percent and 2 percent in :he respective scenarios. Table A.C.I. Growth Projections for Colombia: High Growth versus Low Growth 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 High Growth 2.5 3.0 4.0 4.5 4.5 4.5 4.5 4.5 4.5 4.5 Lowh Growth 1.5 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 GROWTH HAS BEEN THE KEY FACTOR IN POVERTY REDUCTION IN COLOMBIA C8. Poverty reduction can be achieved in one of two ways: increasing the income of the poor 2 through economic growth, or distributing income from the nonpoor to the poor. If poverty alleviation is achieved through economic growth, the result will likely be more wellare enhancing than a mere redistribution of income. Yet some argue that the benefits of economic growth for the poor are undetermined due to the sharp increases in inequality that often accompany growth. 1. Easterly, W., C. Rodriguez, and K. Schmidt-Hebbel. 1994. Public Sector Deficits and Macroecono nic Performance. New York: Oxford University Press. 2. Provided that poor and nonpoor positions are not reversed; that is, the principle of transfers holds. 74 Annex C C9. Nevertheless, international evidence reveals the essential role of economic frowth for poverty reduction. A recent cross-county study using data over the last four decades finds that the average income of the poor (the bottom fifth of the income distribution) rises proportionately with average income; thus, the poor do reap the benefits of growth. The study also finds that growth-enhancing policies, such as reducing government consumption and stabilizing inflation, not only raise overall income, but also appear to have an additional positive effect on the distribution of income, further increasing the income of the poor. CI0. In this respect Colombia's experience follows international patterns. Until the recent economic recession Colombia had seen a reduction in poverty rates through economic growth, even though the income distribution simultaneously deteriorated since 1988. Poverty and Growth: The Colombian Experience During the Last Two Decades C I1. Using data from the Colombian urban household survey, Table A.C.2 provides poverty, growth, and inequality indicators for 1978, 1988, 1995, and 19994 for seven cities in Colombia- Barranquilla, Bogota, Bucaramanga, Cali, Manizales, Medellin, and Pasto. Extreme poverty is defined as a situation in which a person lacks the income needed to purchase a very basic basket of food products.5 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 allow cross-country comparability with World Bank statistics, the US$2 per day poverty rate is also calculated.6 C12. From 1978 to 1995 poverty reduction was substantial during positive growth periods. During the periods of growth (1978-88 and 1988-95), Colombia experienced significant declines in the poverty rate, even though inequality deteriorated (see Table A.C.2). The headcount decreased at a rate of about 2 percent per year between 1978 and 1995, going from 70 percent in 1978 to 48 percent in 1995. In the following four years, the recession resulted in an increase in the poverty rate back to 1988 levels, along with further income inequality. During this period, poverty increased at a rate of 4 percent per year. 3. Dollar, D., and A. Kraay. 2001. "Growth is Good for the Poor." 4. The analysis assesses the elasticity of poverty to growth to provide a forecast of potential future poverty paths under the two economic scenarios-high or low-depending on whether fiscal policy is strengthened. 5. That is, if his or her household per capita income falls below the extreme poverty line defined by DANE, based on the 1985 Encuesta de Ingresos y Gastos. 6. Adjusting for Purchasing Power Parity (PPP). 75 Annex C Table A.C.2. Colombia: Income Inequality and Poverty Indicators (1978-1999) 1978 1988 1995 1999 Poverty Poverty rate 70.2% 55.0/o 47.9% 55.4% Poverty Gap 35.1% 23.2% 18.90/o 26.0% P(2) 21.4% 12.7% 9.8% 15.4% Extreme poverty rate 27.5% 17.4% 9.8% 14.5% US$ 2 per day poverty 34.00/o 4.7% 2.5% 5.4% Mean household income per capita2 157,080 235,163 294,522 277,469 Average annual growth rate NA 5.00/o 3.6% 1.4%t GDP growth rate 4.1%3 4.1% 5.2% -4.3% Income inequality (Gini) 47.3% 48.6% 52.2% 54.5% Source: Colombia Poverty Report 2001. 1. Based on Purchasing Power Parity Converters from WDI database. 2. 1999 pesos. 3. 1980 figure * 1978-1988. ** 1988-1995. t 1995-1999. C l 3. Between 1978 and 1995, Colombian GDP growth rates averaged between 4 and 5 percent per year, producing proportional welfare gains for the poor and more than proportional gains for the extreme poor. Figure A.C. 1 shows the cumulative distribution functions for mean house hold income per capita in 1978, 1988, and 1995 urban Colombia. Welfare levels improved continuously and unambiguously for every percentile of the population between 1978 and 1995. The proportion of the population below the 1995 poverty line (at C$166,204 or US$2647) decreased from over 70 percent to under 50 percent during that interval. During 1978-88, improvements in earnings reached 47 percent for nearly all percentiles, except at both extremes of the distribution, where the improvements were slightly higher. Those at the bottom of the income distribution thus experienced income increases that were actually more than proportional, unlike Dollar and Kraay's findings that, on average, the income of the poor rises one for one with average income.8 This is also confirmed by the fact that extreme poverty rates declined by 37 percent during 1978-95, whereas poverty rates fell by only 22 percent. In the next interval, between 1988 and 1995, the extreme poor fared even better when compared with the poor-their headcount declined by 44 percent, whereas it declined by only 12 percent for the poor (Table A.C.2). Large declines in the poverty gap and the P(2) are further evidence of the fact that those at the bottom of the income distribution did benefit from economic growth. 7. The Purchasing Power Parity (PPP) adjusted exchange rate was I US$ = 629 C$ in 1999. 8. Dollar and Kraay's study (2001) is based on the average income of the bottom 20 percent of the distribution. They point out that outcomes would differ if they used poverty headcounts in their study, but that the elasticity of the poverty headcount with respect to average income varies widely across countries. 76 Annex C Figure A.C.1. Colombia: Cumulative Income Distribution 1978, 1988 and 1995. 0.9 0.8 -1995 Poverty Line: $1l66,204 - 0.8-* ,S 0.7- l7 *K) 0.7 0.6 Pe19 ia9otl5hueod noe 19 eo .~0.5 (U 0.4 30.3- 0.2 0.1- 0 1 0 50000 100000 150000 200000 250000 300000 350000 Per capita monthly household Income, 1999 pesos C14. During 1995-99 economic recession hit the extreme poor harder and partially erased the substantial gains obtained during 1978-95. Although extreme poverty had sharplv decreased with income growth, half of these gains were erased during the economic slowdown. Just as the extreme poor benefited substantially during periods of growth, Colombia's economic recession seems to have more severely hurt the very poorest segments of the population. Both extreme poverty rates and the US$2 per day rate increased disproportionately-by around 50 percent, compared to 16 percent for the poverty rate-during the recession years.'0 The behavior of the P(2) measure provides one more confirmation: it rose by 22 percent between 1995 and 1999, indicating both deeper poverty and a higher level of income inequality among the poor. It is worth noting that all these results are very robust to the poverty line chosen. ' I C15. That poverty has deepened during the economic slowdown is not surprising, since the current crisis tends to disproportionately affect the poor. Using Fedesarrollo's Encuesta Social, Gaviria (2000)12 found a positive relationship between poverty and vulnerability-the latter defined as a household's risk of facing a significant loss in income-during Colombia's recent economic recession. He cites the fact that the poor tend to be employed in more volatile sectors, 9. Dollar and Kraay (2001) find no evidence that periods of negative growth affect the poor disproportionately. However, our findings for Colombia's recent economic recession show otherwise. 10. Nevertheless, the extreme poverty rate is still 2.9 percentage points below the 1988 rate. 11. We computed poverty and extreme poverty rates for the US$1 and US$2 per day lines, as well as for a wide range of hypothetical poverty lines, with no change to the above-described behaviors. 12. Gaviria, Alejandro. 2000. "%Sobre quien ha recaido el peso de la crisis? Un analisis preliminar usando la Encuesta Social de Fedesarrollo." October. 77 Annex C such as construction, and that less-skilled labor tends to be more inconie elastic, implying that low-skilled workers experience disproportionate losses during times of crisis. The poor are also more likely to experience welfare losses due to loss of income, since they disproportionately lack savings or access to credit. Thus, overall welfare losses during periods of negative growth are even higher than just increases in the poverty headcount. C1 6. In 1999, following a drop in mean household income per capita, urban welfare unambiguously deteriorated compared to 1995. Figure A.C.2. shows the cumulative distribution functions for 1988, 1995, and 1999. The cumulative distribution of income for 1999 lies to the left of the curve for 1995 for all monthly per capita household incomes below C$350,001) (or about US$556). The percentage of poor people, as measured by the 1999 poverty line (C$175,831 or about US$280) increased by about 5 percentage points between 1995 and 1999. Figure A.C.2. Colombia: Cumulative Income Distribution 1988, 1995 and 1999. 0.9 - _ 7 _ X _ __ *. 0.8 - - - _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ 1999 Poverty Line: $175,831 0.7 -- Li__ _ _ _ _ _ _ 0 .6 Xa 198g C=0.51--~-& 1t995 0 CO _ ______ ~0.6 0.4 0.3 0 0.2- - _ 0.1 0 50000 100000 150000 200000 250000 300000 350000 Per capita monthly household income, 1999 pesos Cl7. During the recession, the poor suffered more than the rich. Both the 5th and 10th percentiles lost 25 percent of income, while the median income decreased by 11 percent, and the 90th percentile income rose by 9 percent and the 95th percentile by 1 percent. Since first-order stochastic dominance does not hold, we turn to the generalized Lorenz curves of those two years in Figure A.C.3, which provide additional information on second-order stochastic dominance. The 1995 curve lies above the 1999 curve, implying that income was more equally distributed in 1995; welfare in 1999 is thus unambiguously worse than in 1995. 78 Annex C Figure A.C.3. Colombia: Generalized Lorenz Curves 1988 and 1995 (1999 Pesos) 350000 __ __ _ _- 300000 - ___ _ ___ __ __ _ _ V l-1999$ E 250000 ------------ _ _ o 200000 - v . 150000 -- _ 10(0000 5000 _ 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Cumulative percentage of population C18. Welfare losses due to economic recession are so massive that poverty levels rise above those of 1988. The Generalized Lorenz curves, displayed in Figure A.C.4, intersect around the 60th percentile. This implies that social welfare measures with substantial weights on the poor tend to favor the 1988 distribution. The comparison of poverty measures in Table A.C.2 illustrates this point: the poverty count was marginally higher in 1999 (55.0 percent in 1988 compared to 55.4 percent in 1999), but both the poverty gap (at 26 percent) and the P2 index (at 15.4 percent) were clearly higher in 1999. C19. In summary, poverty rates are countercyclical, and the extreme poverty rate appears more sensitive to growth than the "normal" poverty rate. Thus, periods of economic growth produced substantial welfare gains for the poor, and even more for the extreme poor, in Colombia, whereas the economic recession sharply reversed these gains. After high periods of growth, under any distributional weights, 1995 social welfare in urban Colombia was better than in any previous year. However, due to the economic downturn, a similar claim cannot be made about 1999: although that year is still unambiguously better than 1978, when social weights concentrate on the poor, it is clearly dominated by 1995 and even by 1988. Economic recession thus sharply reversed the improving welfare dynamics that had been observable until 1995. 79 Annex C Figure A.C.4. Colombia: Generalized Lorenz Curves 1988 and 1999 (1999 Pesos) 300000 250000 E E . 200000 -1999 1988 o 150000 E E 100000 50000 o - . . . . . . . . . . 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Cumulative percentage of people The Elasticity of Poverty to Economic Growth in Colombia C20. Income per capita growth explains most of the gains and losses in urban poverty firom 1978 to 1999. Changes in the headcount of the poor over time can be decomposed into three components (a) economic growth (proxied by mean household income per capita growth), (b) changes in the real value of the poverty line (relative prices); and (c) changes in inequality (as measured by the share of each household's income distribution percentile in total income).13 Results are displayed in Table A.C.3. Actual poverty reduction was in the magnitude of 15 percentage points between 1978 and 1988, and 7 points between 1988 and 1995. Growth was the only contributor to poverty reduction during 1978-95 (29 percentage points), accounting for 112 and 164 percent of the fall in poverty during 1978-88 and 1988-95, respectively. 13. For basic methodology, see Datt, G., and M. Ravallion, 1992, "Growth and Redistribution Components of Changes in Poverty Measures: A Decomposition with Applications to Brazil and India in the 1980s," Journal of Development Economics 38 (2) :275-95. Here, the same methodology is used with some modification to control for changes in the value of the poverty line. Formally, poverty rates in dates I and 2 can be expressed as functions of the level of the Poverty Line or price index of the poor's basket (p), mean income (u), and the level of inequality in those dates (q), thus: PI = P(pl, p,u ql) and P2 = P(P2, u2, q2) and our decomposition computes these three components (1) Prices: P (P2,P2, q2)-P(Pl-,2,q2) (2) Growth: P (P2, P2 q2)-P(P2, P1, q2) (3) Inequality: P (P2 P2 q2) - P(P2-42, q) 80 Annex C Meanwhile, higher inequality counterbalanced the positive growth effect by nearly 7 percentage points (2.5 percentage points first, and 4.2 points subsequently). Likewise, more expensive food items raised the poverty line and increased poverty by 2.0 percent during 1978-95. Nevertheless, despite the negative consequences of a deterioration of the income distribution and a rise in prices, poverty was still reduced, due to the dominant, positive effect imposed by economic growth. Table A.C.3. Colombia: Decomposition of Poverty Changes by Growth and Inequality Actual Poverty- change in Poverty> Income income poverty Growth Distribution Line Residual change elasticity 1978-1988 -15.2 -17.0 2.5 1.7 -2.3 50% -0.343 Contribution 100% 112% -16% -11% 15% 1988-1995 -7.1 -11.7 4.2 0.3 0.1 25% -0.462 Contribution 100% 164% -59% -4% -1% 1995-1999 7.5 3.3 2.2 3.3 -1.3 -6% -0.572 Contribution 100% 44% 29% 44% -18% Source: Colombia Poverty Report 2001. C21. During the recession, poverty increased due to the combined effect of growth loss, persistent inequality, and high relative prices of food items. Poverty increased by 7.5 percentage points during the crisis period, wiping out the 1988-95 gain. This resulted from contributions of all three components in similar proportions: high prices and recession contributed 44 percent, while inequality contributed an additional 29 percent. C22. The elasticity of poverty to mean household income has increased over time, reaching its highest level, -0.57, during the recession period. This implies that proportional effects on poverty rates have been larger during the economic crisis than during the previous years. Thus, a 1 percent decrease in GDP is now associated with a larger change (increase) in poverty than the change (decrease) in poverty that a 1 percent rise in GDP once produced. A boost in the elasticity of poverty to growth means more poverty alleviation can be achieved through growth, but the consequences can be more severe if economic slowdown is revisited. If the poor are disproportionately vulnerable during times of crisis, as noted above, an economic slowdown also causes further deepening of poverty. How MucH DOES EMPLOYMENT REDUCE POVERTY? The Links of Poverty to Economic Growth via Unemployment C23. The relationship between employment and poverty and employment and aggregate demand is an alternative way of showing the way in which economic growth affects poverty levels. Employment levels are positively associated with aggregate demand. Since low 81 Annex C employment rates have been shown to lead to higher poverty rates, an increase in aggregate demand has positive consequences for poverty alleviation, partly through this mechanism. In addition, earnings per worker tend to move procyclically, especially for self-employed workers. C24. Households with high poverty risks have lower unemployment levels of spouses and other household members of working age. Table A.C.4 reveals that the proportion of poor heads employed is only marginally above the nonpoor's: from 2 to 4 percentage points during booms and down to 1 percent during the recession year 1999. However, employment of other household members (spouse and other adults) is consistently lower among poor households, by 11 to 13 percentage points during 1978-99. In fact, the employment rate of poor households fell more than that of nonpoor households during the recent crisis. Furthermore, household employment rates are associated with poverty: a household with an employment level between only 0 and 20 percent faces a 26 percent increase in poverty risk, whereas a household with an employment level between 40 and 50 percent is protected from poverty, by 25 percent 'Table A.C.5). It is also important to note that unemployment also disproportionately affects the poor: members of the first quintile experience unemployment rates over 4.5 times as high as those of the fifth quintile (Table A.C.6). Table A.C.4. Colombia: Household Employment The Poor vs. the Non-Poor (1978-1999). 1978 1988 1995 1999 Household employment Poor Non-poor Poor NVon-poor Poor Non-poor Poor ANon-poor Household head 83.7% 80.7% 80.3% 78.3% 81.3% 77.3% 74.5% 73.4% Other adults 17 & over 56.8% 68.0% 55.3% 68.8% 59.1% 71.1% 56.2% 68.9% Source: Colombia Poverty Report 2001. Table A.C.5. Colombia: Relative Risk of Being Poor by Household Employment Rate (e)' (1978-1999) Share of total Relative risk population in Emplooyment rate 1978 1988 1995 1999 1999 zero -0.8% 13.3% 4.5% 55.0% 7.6% O0.75 -31.1% -39.7% -39.1% -36.7% 18.3% Source: Colombia Poverty Report 2001. 1. e=# of employed/# of persons older than 12 years of age 82 Annex C Table A.C.6. Colombia: Urban Unemployment for Various Demographic Groups. Quintile 1978 1988 1995 1999 1 12.2% 20.8% 18.0% 36.1% 2 12.4% 15.2% 11.90/% 26.4% 3 7.7% 11.1% 8.8% 20.0%/o 4 6.4% 7.4% 5.9% 14.6% 5 3.8% 3.6% 3.6% 7.8% * Quintiles are taken on household per capita income. Source: Colombia Poverty Report 2001. C25. Higher household employment substantially reduces the probability of falling into poverty. A Multinomial Logit regression model on the probability of being poor reveals that a 2 percentage point increase in the employment level of household members was associated with a 1 percent reduction in poverty risk in 1999. This is also equivalent to a 15 percent decrease in poverty risk if one additional household member becomes employed. 14 (The marginal effects are shown in Table A.C.7). It is important to note that of all other household characteristics examined in the Logit model (including demographic, skill endowment, and labor market attachment variables), those associated with employment variables had the largest impact on the probability of being poor, with only the marginal effect from having a college education following closely. Table A.C.7. Colombia: Logit Results Marginal Effect of Household Employment Rate on the Probability of Being Poor (1978-1999) 1978 1988 1995 1999 Marginal Marginal Marginal Marginal coeff coeff coeff coeff Household employment -35% -55% -73% -51% Source: Colombia Poverty Report 2001. * Significant at 10% level or less ** Significant at 5% level or less Marginal Logit coefficients evaluated at means. Elasticity of Total Employment to GDP Growth C26. Employment clearly responds to aggregate demand fluctuations, with an elasticity close to 80 percent. The country's GDP per capita grew at an average annual rate of 1.8 percent, with 14. With the average number of working-age members per household being 3, and the average household employment rate of 54 percent, one more household member working translates into a 33-percentage-point increase in the average household employment rate. 83 Annex C very low instability.'5 Between 1978 and 1988, GDP grew at an average. rate of 3.3 percent per year, compared with 4.3 percent per year between 1988 and 1995. As growth rates declined, so did employment's response: its growth fell from 5.2 to 3.2 percent for the corresponding subperiods. This yields an estimate of the elasticity of employment with respect to growth of 0.75 (Velez and others 2001). 16 Similarly, a study by L6pez (2001) finds an elasticity of 0.81 during 1991-98. 17 POTENTIAL GAINS IN URBAN POVERTY REDUCTION: FISCAL REFORM COMPARED TO NONREFORM C27. If fiscal reform is not implemented, income per capita growth would be negligible. Depending on whether fiscal reform is adopted in Colombia, GDP growth rates could reach sustained levels of 4.5 percent (similar to those between 1978 and 1995), or remain at on]y 2.0 percent. If the population growth rate continues at its current level of 1.8 percent per year, the growth rate in GDP per capita under fiscal reform would be about 0.7 percent in the first year and about 2.5 percent on average in the nine years following, compared to about -0.3 percent in the first year, and about 0.2 percent on average in the following nine years if fiscal policy :is not strengthened. 18 C28. With no fiscal reform and low growth, poverty would remain almost unchanged. Figure A.C.5 depicts potential poverty reduction in Colombia under the above low- and high-growth Figure A.C.5. Colombia: The Poverty Rate Under High and Low Growth Scenarios (1978-2010) 6% Bars represent average annual income per capita growth. 75% 1978-1988 Line represents poverty rate. X 5% 70% 4 4 65% c l .t ; 1995-1999 ,1995 0' 3% 60% 1 999-201 0 -o 0.~~~~~~~~~~~~~~~~~~~~~ C~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C CD2% ~ 5% C tu 0% ~~~~~~~~~~~~~~~~~~~~~~~~~~~45%V 1978 1988 199519920 ID -1% 40% -2% 35% 15. A standard deviation of 0.02 percent. Moreover, a cross-country study (Rodrik, D. "Why is There So Much Economic Insecurity in Latin America?" World Bank mimeo, October 1999) shows that the probability of the Colombian economy entering episodes of "high volatility" in the last 30 years was close to zero. 16. Velez and others. 2001. "The Reversal of Inequality Trends in Colombia, 1978-1995." IDB and World Bank. 17. L6pez Castanlo, H. 2001. "Caracteristicas y determinantes de la oferta laboral colombiana y su relaci6n con la dinamica del desempleo: Consideraciones te6ricas y de politica." 18. L6pez (2001) estimates that population growth might fall to 1.7 percent during 2001-os. Hence, per capita growth estimates should be adjusted marginally. 84 Annex C scenarios. Without reform, sluggish growth would decrease poverty rates to 54.4 percent by 2010-this is only 1 percentage point less than today's poverty rate, and still 6.5 percentage points higher than the rate in 1995 and similar to 1988 levels-two decades earlier! In short, by 2010, Colombia will not be able to recover the poverty losses it endured during the recession unless it reaches high sustainable growth rates. C29. Fiscal reforms and positive income per capita growth would reduce poverty to prerecession levels, or that observed in 1995. Fiscal reform, through its positive effect on GDP per capita growth rates, could result in a decrease in the poverty rate-currently at 55.4 percent-or less thanl percent in the first year, and about 1.5 percent on average per year thereafter.'9 At this rate, the headcount in urban Colombia could fall back below prerecession levels (48 percent) by 2010. C30. High GDP growth would raise the employment rate by almost 8 percentage points during the next decade. As mentioned, employment rates are highly sensitive to economic cycles. Figure A.C.6 shows the two possible paths the Colombian employment rate could undertake under low- and high-growth scenarios. If comprehensive fiscal reform is adopted, one could expect average employment rates to increase slightly in the first year, and by about 1.5 percent per year for the following nine years20 (provided that the elasticity of employment to growth does not fall throughout time and that the working-age population grows at a rate of 2.2 Figure A.C.6. Colombia: The Employment Rate Under High and Low Growth Scenarios (1999-2010) 61% 59% :j7% -0-High growth -4--Low growth t 55% E 53% - 51% 49% - 47% - 45%- 1998 2000 2002 2004 2006 2008 2010 2012 19. Calculated with a poverty-income elasticity of 0.57 (see above). Although we do not have elasticities calculated for extreme poverty rates, past experience shows us that this rate is more sensitive to growth than the poverty rate, since extreme poverty endured larger fluctuations during periods of growth and economic decline. 20. Using an employment-growth elasticity of 0.75, percent change in employment = 0.75 percent change in GDP/capita - 0.25 percent change in labor force. 85 Annex C percent 2). If Colombia follows this pattern, starting with an employment rate of 51 percent in 1999, it could reach a rate of about 58 percent by 2010. By 2004, the employment rate could already surpass the 1995 rate (53 percent). C3 1. Higher employment rates can have clear positive effects on poverty reduction, as they have been shown to play a significant role in combating poverty risk. If average household employment rates rise by 1 percentage point (a probable scenario with an employment rate that is growing at 1.5 percent per year), poverty risk will decline by about 0.5 percentage poinits on average. C32. In the absence of fiscal reform and the high growth rates it would bring, employment rates would fall-by about 0.3 to 0.4 percent per year in the next nine years. At tha: rate, Colombia's employment rate would stagnate, falling only 1 percentage point-from its current level of 51 percent-in the next decade. Household employment, on average, would fall by about 0.2 percentage points, increasing average poverty risk by 0.1 percentage points. C'learly, the prospects for poverty alleviation through an increase in employment are dismal if. high growth rates are not achieved. C33. In summary, positive growth of per capita income can bring reductions in poverty during the next decade and recover the losses suffered during the recent recession. In adclition, predicted poverty levels under alternative growth scenarios are similar if we use direct effects on income distribution of per capita income or indirect effects via employment levels. Potential discrepancies between the two come from additional effects of growth on individual earnings that would reduce poverty levels even further. THE EFFECT OF FISCAL ADJUSTMENT ON THE POOR C34. The proposed fiscal adjustment does not affect the delivery of social services to the poor, nor does it reduce their main sources of earnings. Colombian public social expenditure is currently well targeted toward the poor: 43 percent of total public subsidies go to the poorest 40 percent of the population. In addition, public subsidies via social services provide substantial welfare improvements for the poor. Subsidies are equivalent to a third of household incoiTie for the poorest 20 percent of the population. The largest and most progressive categories of public expenditure are childcare, basic education, and health services, with the poorest 40 percent receiving 59, 54, and 50 percent of the expenditures on these services, respectively. C35. Some of the components of the proposed fiscal reform tend to improve the efficiency of the delivery of social services. Given that current public social expenditure is progressive, it is evident that the poor would benefit indirectly. By contrast, if fiscal sustainability were not recovered, well-targeted public spending in social services might become unfeasible. For example, the budget for social service delivery could easily be strangled if Colombia's country- specific risk premium increases and public debt payments rise. C36. The proposed fiscal reform includes an adjustment in relative wages of public employees-for example, special benefit cuts in ISS-and an equalization of public pension 21. DANE's estimate for 2000-05. 86 Annex C benefits with overall Colombian pensions. The evidence shows that even under sizeable adjustments, adverse effects on the poor due to an adjustment of public sector wages or pensions are very unlikely. C37. Relative to the private sector, Colombia's public sector public employment is characterized by much higher wages and is concentrated among the upper levels of income distribution. Comparing the distributions of labor earnings in the public and private sectors (Figure A.C.7) shows remarkable differences. While only 25 percent of private sector employees receive wages above C$540,000 per month (equivalent to about US$858 in 1999), 75 percent of the workers in the public sector receive wages above that amount. Figure A.C.7. Colombia: Labor Earning Density Functions Public vs. Private Sector (1999) 0.9- 8Publi inm 0.8 0.7 Private income 0.6- 0.5 0.4- 0.3 0.2- 0.1 _ _ 0 10,000 100,000 540,000 1,000,000 10,000,000 Labor earnings, 1999 pesos (log scale) C38. Only a minor fraction of the poor depend on the public sector for employment-the welfare of the poor depends almost completely on private sector employment. Figure A.C.8 depicts concentration curves and concentration coefficients for total labor earnings, and the The CAS-PR envisioned and triggered a high-case lending program of around US$450 million for the lending program of FY98-01 within which the proposed operation was prepared. The operation is fully consistent with the strategy proposed in the CAS and CAS-PR, including the policy content and the overall lending and exposure limits. Indeed, if this program had been completely delivered, the lending program would have amounted to about US$1.8 billion. To date, the total amount committed is about US$1.4 billion, leaving space for this operation under the envelope proposed by the CAS and CAS-PR. More specifically, in FY98 and FY99, total commitments of about US$130 million and US$140 million were approved, respectively. In FY00, total commitments of about US$930 million were approved, including US$500 million to support the financial sector reform program and US$225 million to support the reconstruction efforts after 87 Annex C the devastating earthquake of January 1999. In FY01, total commitments of US$190 million were approved. Figure A.C.8. Colombia: The Distribution of Private & Public Labor Earnings by Household Income Percentile-Concentration Curves (1999) 100%.. . - 90% 80% Concentration Coefficients: 70% Public sector: 0.736 Private sector: 0.492 60% Total labor: 0.531 50% 40% -D{ Private 30% 20% - 10% 0% 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Mean household income per capita percentile C39. Given the fiscal constraints, the government requested the Bank to stretch the lending program into FY02, and Bank support to Colombia will continue during this fiscal year as proposed by the Country Strategy Note. Specifically, we expect to present six operations, including the SFAL and the Social Sector Adjustment Loan (totaling about US$150 million) in 1FY02. This would bring our proposed FY02 program to about US$630 million. Under this scenario, the lending program for FY98-02 would total about US$2 billion and average about US$400 million annually, which is below the high-case lending ceiling of US$450 million. All of the operations that we propose for FY02 were explicitly mentioned in the CAS-PR, with the exception of the SFAL. However, the CAS and CAS-PR clearly identified the need for Colombia to strengthen its program of fiscal reforms, which is the objective of the SFAL. C40. Only a minor fraction of the poor depend on the public sector for employment--the welfare of the poor depends almost completely on private sector employment. Figure A.C.8 depicts concentration curves and concentration coefficients for total labor earnings, and their public and private components in 1999 urban Colombia.22 Public employment eamings mostly benefit households in the top 20 percent of the income distribution, which receive 80 percent of these earnings. (Consequently, only 20 percent of public sector labor earnings go to the poorest 80 percent of the population.) While 30 percent of total labor earnings go to the poorest 65 percent of households, their share of public sector labor earnings is only 10 percent. Moreover, 22. For further emphasis, Figure A.C.7 depicts the shares of public and private sector income for mean household income percentiles. 88 Annex C the poorest 50 percent receive less that 5 percent of the total public sector earnings. Finally, the population belonging to households which benefit from public labor earnings is barely present among the poor: the poorest 20 and 50 percent represent only 2 and 17 percent of such households, respectively, whereas the top 20 percent represent almost 50 percent. C41. Just as the poor do not depend on public sector wages, they also do not benefit much from current pension benefits. Figure A.C.9 shows concentration curves and concentration coefficients for total earnings and their pension and nonpension components, again for 1999 urban Colombia.23 The poorest 50 percent of the population receive less than 10 percent of total pension benefits. Moreover, 65 percent of earnings via pensions go to the richest 20 percent of households. The concentration coefficient for pensions-a measure of inequality comparable to the Gini Coefficient-is 0.624, much higher than the Gini Coefficient for income in 1999, which was 0.543 .24 One caveat, however, is that the pension data do not distinguish between private and public sector pensions. Nevertheless, one can suspect that public sector pensions are even more concentrated among the upper percentiles of the population. This follows from the fact (as seen above) that public employment is very regressively distributed. Figure A.C.9. Colombia: The Distribution of Pension and Non-Pension Income by Household Income-Concentration Curves (1999) 100% 90% Concentration Coefficients: 80% Pension: 0.624 Non-pension: 0.539 70% Total: 0.543 60% 50% - _ Pensions -C- Non-pensions 40%- - 30%- 20% 10% 0%. . . .. . . . . . 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Mean household income per capita percentile 23 Once again, we provide further emphasis by depicting the shares of pension and nonpension income for mean household income percentiles in Figure A.C.10. 24 Even by Latin American standards this seems a very large level of inequality. Comparing with Wodon (2000), in Colombia the Gini Elasticity for pensions (ratio of concentration coefficient to Gini) is higher than all other countries except Brazil. See Wodon, Q., 2000, "Poverty and Policy in Latin America and The Caribbean," World Bank Technical Paper No. 467, Washington, D.C. 89 Annex C C42. In summary, the analysis of this section does not find significant effects of the proposed fiscal reform on public expenditure for the poor either via reforms on delivery of public services or public employment and/or public salaries for the poor. Fiscal stability guarantees the sustainable provision of progressive social expenditure to the poor. Likewise, the proposed public employment reforms are in effect targeted toward the middle class and the rich and not the bottom 60 percent of the population. Furthermore, our analysis of public sector employ ment further emphasizes the role that fiscal stability can play in alleviating poverty: given that the poor disproportionately depend on the private sector for employment, sustained growth that stimLulates private sector investment is increasingly important for raising the incomes of the bottcrn 55 percent of Colombia's population. CONCLUSION C43. Fiscal sustainability is necessary to sustain economic recovery and achieve high positive GDP per capita growth in Colombia, and to retake a poverty reduction path. During the last two decades Colombia has replicated international patterns of growth and poverty. Positive growth has been the key instrument for poverty reduction in Colombia: each additional point of GDP'-per capita growth reduces inequality by approximately 0.6 percentage points. At the same time, the recession in the second half of the 1990s resulted in high costs in terms of poverty increase- back to 1988 levels. C44. In addition, evidence shows that the extreme poor in Colombia receive amplified shlocks from macroeconomic cycles. That is, both gains and losses are more than proportional for them. Therefore, from the poor's perspective, positive, stable, and sustained growth is preferalb[e to unstable economic performance. In summary, poverty rates are countercyclical, and the extreme poverty rate appears more sensitive to growth than the "normal" poverty rate. F iscal sustainability reform is a necessary condition to recover positive income per capita growtl, and to alleviate unemployment for the poor. That, in turn, would reduce poverty by 2010 to the prerecession levels of 1995, as shown by microdata-based simulations. However, without f iscal sustainability, low growth and high unemployment would follow and poverty would renain almost unchanged at the current level-similar to that of 1988. C'45. Finally, negative effects are not anticipated for the poor from the fiscal reforms associated with changes in public employment or the equalization of wage benefits and pension rights of public employees with the rest of the population. If applied, they would affect mostly the upper quintile of the income distribution. 90 Annex D ANNEX D REPUBLIC OF COLOMBIA FISCAL STRENGTHENING PROGRAM-STRUCTURAL FISCAL ADJUSTMENT LOAN (SFAL) PUBLIC SECTOR DEBT-STRUCTURE AND SUSTAINABILITY EVOLUTION OF PUBLIC SECTOR DEBT D1. Over the last six years the level of Colombia's total gross public debt has been rising steadily in absolute terms and as a proportion of GDP. Public debt servicing costs have risen substantially and there is pressure on the government to strengthen its primary balance. D2. The total gross stock of public debt,' internal and external, increased from about 27 2 percent of GDP in 1994 to about 50 percent of GDP in 2000. This resulted primarily from rising fiscal imbalances at different levels of government throughout the decade (see Figure A.D.1). The steady growth in fiscal imbalances first became visible in the central government accounts, but this was initially masked by surpluses in other parts of the public sector. The central government accounts started the decade with a small surplus. However, starting in 1992 the central government posted a series of increasing deficits, which peaked in 1999 with a deficit of 7.4 percent of GDP. Figure A.D.1. Colombia: Public Sector Balances (1991-2000) 2.0 - 0.0 cp -2.0 -4.0 -6.0 -8.0 - m~i ct m- *IN r 00 c> O O ON ON ON ON ON ON ON ON o 0 NFPS * Central Government Source: IMF D3. Behind the deterioration in the central government accounts are some of the institutional changes brought about by the 1991 Constitutional reforrn, which transferred to subnational levels 1. There are several types of public sector liabilities other than "debt." For more details on the classification of public sector liabilities, see Allen Schick, Targetingfor Fiscal Risks, Mimeo, World Bank, September 1999. 2. In net terms, the total stock of public debt represented about 42 percent of GDP in 2000. 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 reference only to gross aggregates because estimates of net debt in Colombia are available from only 1999 onward. 91 Annex D of government discretionary spending powers, which the central govemrnent has been forced to finance through fiscal transfers. In addition, other policy reforms put pressures on revenues (for example, trade liberalization led to a decline in tariff revenues and several tax reforn packages aimed at strengthening tax revenues were only partially successful). D4. By the mid-1990s it became evident that the accounts of the public sector were :,-apidly deteriorating. The public sector deficit rose steadily-basically tripling from just over 2 percent of GDP in 1995 to 6.4 percent of GDP in 1999. Large central government deficits, growing deficits at the subnational level, dwindling social security surpluses, and volatile oil revenues, all contributed to the deterioration in the public sector balance. This is evidenced by the continued erosion of primary balances 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 2.6 percent of GDP in 1999. D5. The weakness in public sector finances is also due in part to increasing debt-servicing costs. Total public debt interest service increased from about 2.2 percent of GDP in 1994 to about 4.4 percent of GDP in 2000. Total interest service on internal debt increased more than proportionately-almost threefold-from I percent of GDP in 1994 to about 2.7 percent of GDP in 2000 (see Figure A.D.2). Figure AD.2. Colombia: Interest Payments of the Public Sector (1994-2000) 5.0 3 4.0 - 09. 3.0-_ r 2.0 v 1.0 0.0 "t In \O r- O x ON oo C7 a o ol o\ CD ON O ON N O N O * External U Internal Source: IMF D6. During the last decade the structure of public sector debt also changed. External debt fell sharply from about 32 percent of GDP in 1991 to about 14 percent of GDP in 1996 (thanks mostly to the surpluses generated in the decentralized public entities, for example, through oil revenues). Internal debt rose throughout this period, but not enough to offset the sharp decline in external debt. However, as the public sector accounts began to deteriorate in the mid-1 990s (particularly central government accounts), debt levels began increasing again. Throughout this period, a sizeable 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. Most of the increase in the external debt after 1996 has been driven by central government borrowing, because other levels of government currently face statutory restrictions on borrowing abroad. However, throughout this period gross public sector internal debt continued to rise steadily, reaching 27 percenit of GDP by 2000, as both the central government and subnational entities sought financing 92 Annex D domestically (see Figures A.D.3 and A.D.4). Figure A.D.3. Colombia: Gross Figure A.D.4. Colombia: Gross Public Public Sector Debt (1991-2000) Debt by Debtor (1991-2000) 60 60 50~~~~~~~~~~~~~~~~~~5 4h 10 0 40 30 3 c 20 20 - _ In \_ N 00 _ - ON C S e - : - -C- 00 ON 0 ON s sa a soON OCC 4 ~0 ~ ~O U External a Internal * Central Government * Sab-National Governmi Sources: IMF, MoF, Contraloria 0 Other Public Sector Sources: IMF, MoF, Contraloria D7. The structure of domestic financing changed throughout the 1990s as well. First, implementation of the Constitutional changes of the early 1990s led to elimination of Central Bank financing of the fiscal deficit, implying that domestic financing would now have to be obtained at market terms by the central government.3 The domestic debt of the central government has nevertheless continued to rise through a large expansion in issuance of Treasury securities. D8. In contrast to the case of central government debt, the stock of subnational government 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 are often met through fiscal transfers from the central government. However, as a result of the restrictions on foreign borrowing by subnational entities, their domestic debt has increased (see Figure A.D.5). The main source of this increase is domestic commercial bank lending. Financing has been readily available mainly because of an expectation that there is an implicit central government guarantee on the liabilities of subnational public entities; and because of failure in the implementation of the "traffic lights" law, which aimed at regulating subnational government borrowing. 3. 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. However, the Central Bank balance still continues to be an important contributor to public sector accounts (the balance of the Central Bank amounted to about 0.8 percent of GDP in 2000). 93 Annex D Figure A.D.5. Colombia: Debt Structure of Sub-National Entities 8 -6 4 CM 0 ' t 0C CD ON Oa m ON O ON O ON O 0 * External Internal Source: Contraloria D9. Recent adjustment efforts have gone some way toward restoring balance in the fiscal accounts. In 2000, the public sector deficit was almost halved, to -3.6 percent of GDP; moreover, the primary balance showed a surplus (albeit small) of around 0.8 percent of Gl)P- for the first time in several years. However, there are very large and growing implicil. and contingent public sector liabilities (arising mainly from the social security pension system and subnational levels of government), which are not yet being fully reflected in the fiscal accounts but will add pressure to them in the future. DEBT SUSTAINABILITY ANALYSIS D1O. To analyze the sustainability of public sector debt, we calculate the primary balance necessary to keep public sector debt on a sustainable path,4 using a model that includes (a) the present value of all net public sector liabilities (including implicit and contingent liabilities from social security), and (b) key economic parameters and projections in a base-case scenario. DI 11. According to our calculations, to avert an explosion of the debt-to-GDP ratio, the public sector has to generate a permanent primary surplus of about 2.5 to 3 percent of GDP.- 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. 4. This is the level of the primary balance expressed as a ratio to GDP that if generated "permanently" by the government would keep the debt-to-GDP ratio sustainable. Moreover, a steady-state debt-to-GDP ratio is achieved in the long term. 5. Primary balances are defined here with respect to the consolidated public sector, including social security and subnational government entities. For consistency, we are assuming that seignorage revenues accrue ic the government "above the line" (through the reported Central Bank balance). 94 Annex D Table A.D.I. Colombia: Key Macroeconomic Assumptions and Projections 2001 2002 2003 2004 2005 Real GDP Growth (p.a.) 2.0% 2.5% 3.0% 3.5% 3.5% Inflation (p.a.) 8.0% 6.0% 4.0% 4.0% 4.0% Nominal Exchange Rate Depreciation 8.0% 6.0% 4.0% 3.0% 2.5% Total Public Sector Debt (% of GDP) 44% 45% 47% 49% 50%0 Domestic 19% 19% 20% 20% 20% Extemal 24% 26% 27% 29% 30% Nominal Interest Rate (domestic) 15.9% 14.4% 12.3% 11.2% 10.6% Nominal Interest Rate (foreign) 8.5% 8.5% 8.5% 8.6% 8.6% D12. It follows that in the absence of reforms, keeping debt on a stable path would require a sustained fiscal adjustment in the primary balance of 2 to 2.5 percent of GDP (to bring the primary surplus to about 3 percent of GDP, from about 0.8 percent in 2000). If the primary surplus were kept permanently at 2.5 to 3 percent of GDP, it is estimated that total net debt of the public sector would stabilize at about 50 percent of GDP by 2005 (see Table A.D.1). Figure A.D.6. Debt-GDP Ratio Figure A.D.7. Debt-GDP Ratio Primary Balance Constant at 2000 Level Primary Balnce Set to Keep Debt-GDP Ratio Constant 10l 70 7.0 50 60 6 " LiC.0 Prinery Balance DebtlG Ratio40 50 50 0 30 en~~~~~~~4 0o nO 40940i 3 0 E Q 20~ 10 10 I 0 o 0 0 0a - 4 Pj Baane ebtGDRato -Primary Balance Debt/GDP Ratio (2000) Source World Bank estimates Source: World Bank estimates D13. The vulnerability embedded in public finances in the absence of comprehensive reforms can 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 2000 surplus level (just 6 over 0.8 percent of GDP). As can be seen in Figure A.D.6, total net debt would rise from about 42 percent of GDP in 2000, to over 65 percent of GDP in 2005 (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 4.5 percent per year, a small deceleration from the 1994-2000 level (when the debt-to-GDP ratio increased at an average annual rate of about 6 percent). D14. Figure A.D.7 shows the path of the primary balance required to stabilize the debt-to-GDP 6. If the 2000 primary balance was adjusted to filter out cyclical or noise components, the resulting "structural" primary balance would in fact be lower, since unusually high oil revenues 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 2.3 percent of GDP. 95 Annex D ratio at the 2000 level (about 42 percent of GDP). As can be seen, the primary surplus would have to be maintained in the range of 4 to 5 percent of GDP in the next five years to keep the debt-to-GDP ratio constant (higher than the 3 percent of GDP required just to keep debt on a more sustainable path). D15. Table A.D.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 about 4.5 percent of GDP (instead of 3 percent as before). The sensitivity of these calculations to the inclusion of liabilities arising from social security (mainly pensions) in a scenario of limited structural reforms is also shown in Table A.D.2. For example, if GDP is assumed to grow at a rate of 3.5 percent per year and social security liabilities are excluded from our calculations, a permanent primary surplus of 0.5 percent of GDP would be sufficient to k eep the debt-to-GDP ratio on a sustainable path. This should be contrasted with the 3 pe-cent primary balance-to-GDP ratio required once Social Security liabilities are included in these calculations.7 Table A.D.2. Permanent Primary Surplus: Sensitivity Analysis GDP Growth Rate 1.5% 2.0% 2.5% 3.0% 3.5% 4.0% Including Social Security 5.7% 5.2% 4.6% 3.9% 3.1% 2.1% Excluding Social Security 1.1% 0.9% 0.8% 0.6% 0.5% 0.3% D 16. In sum, our analysis shows that in the base-case macroeconomic scenario and withoul full reforms, a minimum of about 2 percent of GDP adjustment in the primary surplus from the 2000 level is required to bring about a more sustainable fiscal path in Colombia. If this adjustment in the primary balance is carried out, the debt-to-GDP ratio would stabilize at about 50 percent of GDP by 2005. 7. Should comprehensive social security and other reforms take place, the present value of contingent and implicit government liabilities would fall sharply, implying that the permanent primary surplus required to keep the debt- to-GDP ratio on a sustainable path would be smaller than 3 percent of GDP. 96 Annex E November 14t^, 2001 Mr. James D. Wolfensohn President The World Bank Group Washington, D.C. 20331 Dcar Mr. Wolfenson: 1. This le(Ler intends to briefly rcview the progress achieved under our economic program (supported by both the IMF and the World Bank), describe the reccnt economic and social dcvelopments, including the economic recovery, inllation, and the fiscal and extemal positions. We would also like to outline the main challenges that remain as Colombia enters the last year of its three-ycar economic recovery program, and to present the policy changes that arc being introduced to achieve the objectives of the govenmuent's program, including strengthening and consolidating the fiscal accounts, for the next year and for the medium term. I. Recent Developments 2. Economic conditions improved in 2000 as real GDP recovered from a drop of 4 percent in 1999 to a growth of an estimated 2.8 percent, based on a recovery of exports and some reboui in momesu,c aernanU'; J11W,u1cr. L [lLL UULtA %e I _t A , J^ 1.a 10 percent. inflation ended the year at 8.8 pcrcent, wcll within the 10 percent target and has declined to about 8 percent in October 2001. fn 2001 economic growth has slowed down reflecting the weakening of external and internal demand. In pa rticular, the slower growth in cxternal markets has affected both coffec and oil exports and, most recerntly, the export of manufactures. Colombia has also suffered supply shocks: oil exports were disrupted by sabotage in the first half of the year and strikes affected economic activity. The unemployment rate has remained high, around 18 percent. 3. Policy implemcntation under our economic program has progressed well-all quantitative perfornance cnrtcria agreed with the IME have been observcd up to September 2000, and a nurnber of significant reforrns have been implemented. 4. In 2000 thc combincd public scctor dcficit l'cil to 3.5 perccnt of GDP, from 5.5 percent of GDP in 1999, achieving the target set in lhc IMF program. In October of last 97 Annex E 0 0 7 8 6 year, the constitutional court handed down a dccision that revoked the wage freeze that was being applied most public employees, requiring that they instead bc awarded increases corresponding to the previous year's intlation (9.2 percent, payablc before the end of 2000), and that future public wage increases must, at a minimum, compensate for inflation in the preceding year. This raised the NFPS wage bill by about 0.5 perccnt of GDP in 2000 and absorbed somc of the good perfonnance dtat the governmcnt had planned to achieve relative to the fiscal deficit target. 5. Fiscal performnance remained on tar-et through the first half of 2001, despite a shortfall in revenue collections induced by the cyclical downturn, the disrupfion of exports, and some initial delays in impleTncnting the tax package that was approved last year. The revenue shortfall was compensated by a reduction in spending, lower than projected bank restnrcturing costs, and largcr than projected quasi-fiscal profits of the central bank. After taking into account the effect of the cyclical downturn and the temporary reduction in oil exporls, the deficit for 2001 is expected to be 3.3 percent of GDP, 04 perceat higher than the 2.9 percent of GDP indicative target that had been set in the last rcview of the IMF program. Recently, the Constitutional Court revoked thc decision mentioncd abovc. partially reducing backward indexation of public wages. While this will not represent significant savings in 2001, it does represent an important improvcment for fiscal policy in the future. 6. Monetary developmcnts over the past year lhave been characterized by slow growth of deposits and credit to thc private sector. Whilc the pnrvate sector's financial portfolio is projected to grow by 8 percent in 2001, broad money would increase by 4 percent, up from 1.3 percent in 2000. Outstanding credit to the private sector is projected to rise by 5.5 percent this year when adjusted for loan writeoffs and mortgage debt relief, r:ainly due to strong growth of credit by private commercial banks. Interest rates have declined since the bcginning of the year, by mid-Novrember lending rates were lower by 377 bps and deposit rates 175 bps. The central bank has reduced its interest rates in six occasions in 2001 (March, July, August, September, October and November ) for a total of 3.25 percentagc points in light of the continued favorablc outlook for inflation in 2001 and 2002, the economic slowdown, and declining international interest rates. 7. The recovery of thc financial system has continued ovcr the past year-the average capital adequacy ratio increased from 11 percent in December 1999 to 13 percent in December 2000, and to 12.7 pcrcent in August 2001; at the same time rionpcrfonning loans fell to 10.9 percent of total loans, reflecting lower delinquencics for consumer and conmuercial credit. Mortgage portfolios, however, have deteriorated sincc May 2000 (following an inmprovement in the preceding m eonths that reflected debt rclief programs 9S 2 Annex E 00 07;8 6 z r k .7 - W,, Mier .ffiX -~0 7 introduced at the beginning of the year), with nonperforning loans hovering around 20 percent of total loans this year. 8. The foreign exchange market has remained orderly since the peso was floated in September 1999, and the ccntral bank's foreign exchange auction system is tansparent and has becn functioning well. After experiencing a depreciating trend since April 2000 tha. reflected political events and increasing difficulties in executiiig the govermrent's pTivatization program, the peso was relatively stable in the last few months of 2000 and during 2001. 9. The external current account is estimated to have posted a surplus of 0.38 percent of GDP in 2000, aftcr a surplus of 0.2 percent of GDP in 1999; the trade supltus rose to 3.1 percent of GDP from 2.1 percent of G7TDP a year earlier. Boosted by higher oil prices and a more competitive exchangc rate, Colombian exports grew by nearly 13 percent (in U.S. dollar terms) in 2000, with petroleum exports rising by 22 percent and nontraditional cxports by nearly 17 percent, while coffee exports fell. In 2001, non traditional exports will grow by around 8 percent, while imports are expected to grow by 12 perccnt, causing the trade balance to fall to a deficit of USSI billion from a surplus of US$1.3 billion in 2000. The capital account was in near balance in 2000 as net repayments of debt by the financial sector and the non-financial private sector were compensatcd by net loan inflows to the non-financial public sector. Nct international reserves rose by more than US$800 million, thus remraining at more than six months of imports of goods aTd services. In 2001 a similar accumulation of reserves is expected. bringing the stock to UIS$9.6 billion. 10. As a result of the strong implementation to date of the governent's 3 year rccovery program, the external vulnerability of the Colombian economy is relatively low (the ratio of NIR to 2002 external debt-service payments excecds 100 percent). Tp to September 11 evcats in the United States and despite the turbulence in the international capital markets and the financial difficuJtics faced by some countries in the region, Colombia cnjoyed access to external financing with falling bond spreads; more recently, in November 14, markets reopened for Colombia when we were able to place US$0.5 billion in bonds towards the financing of next year's needs. The floating of the exchange rate and the fiscal consolidation since 1999 havc strengthened the confidence in Colombia's economic nanagemcnt and given the authorities room to mancuver in case of extemal shocks. The timely fmancing of this year's fiscal deficit and the progress in financing next year's requirements provide additional assurances, especially in view of the temporary closure ofexternal markets for all emcrging countries since September. 99 3 Annex E 00 0 7 8 6 II. The Economic Program A. Macroeconomic Policies Iit Undcr the economic program for 2001, the expected real GDP growth was initially set at 3.8 percent. However, the deteriorating extemal environment, and continuing depression of coffec prices and the volatility of oil pnrces conspired against these projections and GDP growth is now projected at about 2 percent for 2001. The GDP growth projection for 2002 has been reduced to 3 percent from 4.5 percent, in line with the revisions that have becn made for growth of the world economy. The target for consumer price inflation is 6 percent and for NIR accumulation USS326 rmlIion. 13. As noted above, the combined public sector deficit in 2001 wvill be 3.3 percent of GDP. On this basis, the NFPS primary surplus will continue to improve in 2001, to 1.1 percent of GDP from 0.9 percent in 2000. The strengthening of the primary fiscal balance, when the oil sector and the operations of the public health (TSS) and pension systems are excluded, illustrates the fiscal consolidation effort to date under Colombia's three year recovery program. This balance improved by 3.3 percent of GDP in 2000 and will be strengthened by another 1.9 percent of GDP in 2001. 14. Jn order to secure adequatc credit resources for the private sector to sustain the economic rccovery and prevent any 4;t#;;cbivc upvcd priaouro on dom..a.tic int!roct rnta the governmcnt has made every effoTt to limit the access to domestic financial savings by the combined public sector during 2001. To this cad, the govcmmtnt has secured foreign financing, with the equivalent of US$2.7 million in bond issues betwecn November 2000 and September 2001, including thc arrangement of the World Bank's Policy Based Guarantec Operation for a leveraged bond placement of about US$1 billion successfally carried out in early 2001. Most of the remaining financing necd will be covered by disbursements from multilateral institutions, including the proceeds fromn the first tranche World Bank's Structural Fiscal Adjustment Loan (SFAL). With this extemal financing plan, the program is providing for Lhe net use of domestic credit by the combined public sector of 0.2 percent of GDP in 2001, a sharp reduction compared with 2000, when the comparable number was 1.5 percent of GDP. In June this year the governnent swapped domestic debl amounting to COLS5.6 trillion (US$2.4 h:llion) falling duc in 2001-2005 with securities of longer maturities in order to smooth the amortization profile. This voluntary opcration extended the avcrage maturity of the dlomestic debt from 3.5 to 4.5 years and redutcedi amortization payments by 22 percenit in 2001 and 34 perccnt in 2002, at a marginal increasc in costs. 100 4 Annex E 000786 J M, .,4A. 15. Based on current and additional fiscal policies, supportcd mainly by thc World Bank's Structural Fiscal Adjustnent Operation, the NFPS deficit for 2002 is projected to be 2.5 percent of GDP given current forecasts for economic growth, while the balance for thc combined public sector will amount to 2.6 percent of GDP. The expected deficit takes into account the effect on revenue of 0.4 percent of GDP of two consecutive years of lower than projected growth of the Colombian economy. In addition, Colombia will have to spend some 0.3 percent of GDP to promote thc ongoing peace process in the context of implementing the "Plan Colombia" which sets out the govcmment's strategy to settle the intemnal conflict, curb the production of and trafficking of illicit drugs, and deal with the social consequences of these problems. The additional spcnding reflects the ,ovcrnment's intcnsified efforts in these areas. Nevertheless, the targets for 2002 reflect the continuation of the adjustment effort that will strengthen the primary balance by morc than I percentage point of GDP- The improvement of this balance over the three years of the government's recovery program supported by the IMF and the World Bank, would amount to 3.8 percenl of GDP. Thc primary deficit excluding oil and social secunty accounts will improve by another 1.1 percent of GDP in 2002, for a total adjustment of 6.3 percent of GDP between 1999 and 2002. 16. The NFPS deficit in 2002 is projected to be financed mainly through medium and long-term external resources from nmultilaterals and private financial markets and, to a lesser extent, with domestic resources. To this end the central government has already made considerable progress by pre-financing its 2002 needs in amount of US$1.5 billion of a total of US$2.8 billion. It should be noted that US$0.5 billion were obtained in November, signaling the rc-opening of the marklets for Colombia. The resources that are expected from the multilateral institutions add up USS845 millions. 17. Last year the central bank continued to take steps to make inflation targeting the norminal anchor of its monetary policy. Consequently, Colombia's monctary policy fiamework for 2001 is being based on a scheme for inflation targeting. The new framework is transparent and provides a clear signal to the markets of the central bank's determination to continue to achieve the program's inflation objective. The authorities have developed performance criteria for inflation, jointly with the rMF staff. These include consultations with the Fund about inflation control policies if current inflation werc to dcviate significantly from the projected quarterly path. B. The Fiscal Strengthening Program and Other Structlhral Reforms 101 5 Annex E st,$ z DO4t, 0 3 00 786 18. Tn July 2000, President Pastranuappointed a member ofthe Liberal party to the ministry of finance in an effort to broaden the political support for the economic program of the government. This has strengthened govemnability and has been decisive in advancing refomis through Congress. At the same bme, the govenmuent initiated roundtable discussions with the political opposition, the terTitorial govemmments, the business community. the unions, and others on the most important structural reforms. This broad dialogue on issues that affect lar;e scgments of the population has been usefiul, and the views expressed in these fora arc being reflected in the structural refonn proposals that have been presented to Congress. 19. The govntunent's stctiral reform agenda includcs ambitious measures to increase economic efTiciency. stimulate economic growth, and foster social progress. To this end. a large numbcr of rerorms have been introduced so far under the three year recovery program, starting with the floating of the peso at the inception of the program in 1999. Other important rcforms havc been introduced through the territonral fiscal adjust-mmnt law (Law 617); the adoption of an inflation targeting framcwork for monetary policy; the restrucring of the financial sector; two revenuc enhaincing tax reforrns; the establishment of finds to support temritorial govenunent pensions; privatization/liquidation of public entcrprises and banks; and actions to deepen domestic financial markets and develop longer-term finazncial instruments for the private sector. 20. Tn order to ensure that medium-term fiscal sustainability is achieved a tax packagc was approved in December 2000 that will yield about 1.9 percent of GDP annual)y. The approved tax package is based on an increase in the VAT rate from 15 to 16 percent; an increase in the presumptivc incomc tax ratc from 5 percent to 6 percent of net assets; and an increase in the rate of the financial transactions tax to 0.3 perccnt from 0.2 percent. In addition, the tax package establishes temporary tax incenlives for the repatriation of capital. 21. Congress recently approved tle Constitutional Reform (Acto Legislativo), changing the system of revenue lransfers (revenue sharing) to the tenritorial govermments. The reform of the system of revenue transfcrs to the territorial governments that was adopted through a constitutional amendment in June will be complementcd by the modification of Law 60. The proposed changes in this law, submitted to congress in Scptember, will improve the cflicicncy in the use of revenue sharing funds and strengthen the control over the use of these resources. In the area of education, the proposed reform establishes clear definitions of the responsibilities bctween regional and MITunicipal governimenits for providing education services; distributes resources based on the nurbcr of students rather 102 G Annex E CO Q 7 8 6 e4f/44~ than teachers; incorporates performance indicators for leaming; and establishes limits on current spending. A special regulatory commission will be created to monitor the performance indicators and determine the technical paramcters of the new system, such as the average cost per student, optimal student to teacher ratio, etc. In the area of health, the roles of each jurisdiction are being clarificd; a number of intcrmnediary institutions will be eliminated; and performance-based indicators also will be established. Whilc this law is niot expected to generatc fiscal savings, it will contribute to significant efficiency gains that will allow for a widening of coverage in the provision of these services. In the health seclor, witl IDB and World Bank support, the govemment is also implementing a restructtuing plan for public hospitals, which is deepened under the Structural Fiscal Adjustmcnt Operation. 22. In addition, considerable progress has been made over the past few years to streamline the opcrations and reduce the size of territonral governments under the diirection of the ministry of finance (Direcci6n de Apoyo Fiscal-DAF). Additional actions were taken last year to introduce systems designed to help control public spceding and strengthen the public financos at the territorial level. Law 617 ("Ley dc Sancamiento Fiscal de las Entidades Temritoriales" or territorial fiscal responsibility law), passed in October 2000, strengthens the control of spending at the territorial government level and limits territorial govemment current spending and introduces restrictions on the creation of new municipalitics, allowing only those capable of generating sufficient own revenues. It also provides for a debt-restructuxing scheme that involves support from the central govenmcnt for thosc local govemments that enter into fiscal adjustment programs monitored by the DAF. In addition, the govemnment will modify regulations governing subnational borrowing, reducing the scope of the central government in approving access to credit to local govemments and introducing incentives for a more active participation of international reputed credit rating agencies and market determined mechanisms for credit access. 23. For the past two years, and again in the budget for 2002, the government has been taking strict measures to reduce current expenditures in the central administration. The paymcnts that are under the control of the government, that is current expenditures cxcluding interest payments and transfers to local governments and social security, have decrcased by 4 percent in real terms from 2001 to 2002, while general cxpcnditures (supplies, utilities, etc-) will not increase in real terms. Both Ley 617 and the constitutional amendment recently approved, includc rules for limiting cenlral governmnt expenditures. The govemment also submitted to Congress a draft law that grants powers to the executive to merge and liquidate public institutionis, which is 103 Annex E 000786 * # t a¢. -~ , WE% gBt, expected to be discussed and hopefully approved during the first half of next year. The final element of the strategy to strengthen the control over public spending is the presenitation to congess in the first half of 2002 of a fiscal responsibility law that complements the actions taken under "Ley 617" and thc "Acto Lcgislativo." 24. The govemnment supports thc principles underlying thc Code of F.scal Transparency, and will begin to assess Colombia's compliance with the codc. In addition, public financial management has been strengthencd through the implementation of- the SIIF (Tntegrated financial itformation system), which registcrs the exccutior or govemment expenditures of 46 institutions of the central admuiistration in real time. This system increases the efficiency and transparency of the cxpenditure process. 25. Regarding the plan to provide territorial governments with increased fiscal autonomy, the government has drafted a law, which was presented to Congress in the second half of 2001. In addition to strengthening the territorial tax administrations, the main provisions of the proposed law are to simplify the regulations and procedures for regional taxes, such as those on liquor, tobacco, property, and the turnover tax ("industria y comercio"), providing the regional administrations with a transparcnt fianework for applying and collecting cxisting taxes. An advancc in this direction was the congress:onal approval of a law in September 2000, that increases the proceeds from lotteries and helps strngthen the finances of the health sector. 26. The authorities have initiated the reform the public health system unlder the social securitv institute (ISS), which is facing a difficult financial position as a result of rapidly rising costs and stagnating revenues. Operational aencits or arunu v..; put t;z ur GDP appeared in 1999 and the system's deficit was projected to widen further in 2001 and 2002 Action has been taklen in recent months to rcverse the financial deterioration of the ISS-health. To this end, a wide-ranging restructuring plan has been adopted and is being iniplemented. The main elements of the plan include an unprecedented renegotiation of the collective bargaining agrcement with the health worker unions which significantly reduces wage and pension bcnefits; a set of measurcs to strcamline the operations of the ISS; and the restructuring of ISS's debt. It has been decided to rcduce rss costs further through a more equitable distribution of the financial burden related to high-cost illncsses among health-service providcrs. The superintendency of hcalth has lifted the sanction that prevented the ISS from incrcasing affiliation. This will allow the ISS to generate additional income that will contribute to the reduction of the deficit. These reforms involve significant savinigs that will reduce the health system's cash deficit by 0.2 perccnt orGDP in 2002, and will climinatc the deficit by 2005. 104 Annex E w 0 786 27. Colombia has initiated a second-generation reform of its pension system with the establishment of funds to help finance territorial government pension liabilities, and the authorities are currently in the process of introducing a fundamental and broad based pension reform. During the past months the government has been discussing a pension reform proposal that aims to bolster the financial position, which would othcrwise require significant fiscal support over the next decade, and to reduce its long-term actuarial dcficit. Ample discussions have been held betwecn the technical teams of the ministry of finance, the national planning department tind the labor ministry, with the technical advice of World Bank experts. The government's draft pension reform has been submitted to the comrnission that was called by the President last May, integrated by representatives of the industrial associations. labor unions, congress and the presidcntial candidatcs. This commission will study the proposal and suggest changes before presenting it to congress by December 14. The reform of the special pension regimes (military, teachers and those included in labor bargaining agreements) will be discussed bv a special commission that has been entrusted to present recommendations for their reforTn by March 2002 which would be added to the reform of the general pension system. The reform proposal seeks to achieve the financial equilibrium for new cntrants into the system, finance the contingent liabilities generated by the minimumn pension guarantce, as well as reduce the fiscal effort required to finance pension pavments in the mcdium term. The main elements of the reform are: a) an increase of up to 2 percentage points in payroll contributions (1 point when the reform is implemented and then 0.25 points yearly froTn 2005 to 2008); b) increased retirement ages (from 60 to 63 years for meni and from 55 to 58 for women in 2014 and then to 65-60, respectively in 2020); c) an iincrease in the minimum number of weeks of contributions required for a pension (from 1000 in 2014 to 1250 in 2020, increasing 35 weeks pcr year); d) making higher-income pensions subject to income taxes; e) establishing a cap for new pensions equivaleat to 20 minimunm wages; f) new public employees must enroll in the public pension system. The government expects that the final approval of the reformn will be made in the first half of 2002. The fiscal savings of the pension reform amount to 2.2 percent of GDP to 2010, And the net present value of the pension liabilities to 2050 wrill be reduced from 206 percent of GDP to 159 percent of GDP as a result of the proposed reforn. 28. Tn dealing with a resurgence of the difficulties faced by the savings and loans -s.sociqIinn% (WAVs). the authorities recognize thc need to strengthen further their capital base and assure their efficient rcstructUring in ordcr to faisilitetz the rocovery of the housing industry. lTe strategy adoptcd for the CAVs consists of introdlucing rigorous capitalization and restructuring plans to be monitored by the banking superintenldcncy with clear targets and a credible set of actions to hc taken by the authorities should those 105 9 Annex fE .0 7 8 6 targets not be met. Simultaneously, FOGAFTN, the securities and banking superinendencies, the ccntral bank, and the ministry of finance are developing a full- fledged package of mcasures to facilitate the restructuring process and the develo-Dment of capital market instrumcnts for long-term rnortgagc financing. The annual fiscal cost of the additional assistance to the mortgage nnance sector nas Deen CsLunaLLCU LU VIC :Iuau and has been incorporated within Lhe fiscal program 29. Recent developments suggest that the private banks which have receivea support from FOGAFTN have irnproved their performance, although some of them remain w eak. Conditional on the implementation of a restructuring plan, the authoritics stand ready to assist potentially viablc institutions that may nced support. The remrraiirmg statc bank-s are being prepared for privatization or a process of gradual asset shrinking and final closire. as noted above. The clean-up of their portfolios and additional capitalization in preparation for their sale or liquidation by mid-2002 is progressing, with the additional costs having been incorporated in the fiscal program. 30. Togethcr with the IME staff and the technical assistance of Bank staff on structural issues supported by the SFAL, the authorities have revicwed the medium-term outlook for Colombia's public finances, external debt, and the balance of payments. The projections reflect a conservative scenario in wlhich productiVity gains. unemploymnent, the investment rate, and 1GDP growth return gradually to their historical levels. As a -esult of the extensive fiscal adjustment mcasures and the structural reforms that have 'een implemented so far, the growth of public sector debt is projected to moderate significantly firom the rapid increascs in the late 1 990s, when the debt rosc from 26.7 percent of GDP in 1997 to 42.4 percent of GDP in 2000 Al the same timc, the external current account deficit would be reduced from 5.4 percent of GOP on average during 1997-98, to 2.9 percent of GDP on average in the period to 2010. The authorities recognize that the fiscal adjustment and the structural reforms being pursued in 2001 and 2002 will not be sufficient to stabilize thc public debt at 44-46 percent, as is desirable. To achieve this, it will bc necessary to continue the Colombia's fiscal consolidation cfforts beyond the three years of the economic recovery program and put in place a combination of measures that permanently reduce the fiscal deficit by around I percent of GDP. Howevcr, it should be noted that this medium-term scenario does not take into consideration the impact on the economy that additional progress in the peace process would bring about or the fact that the pcnsion refomr will havc its most significant impact on public debt beyond the projection period. Thcsc two elements do improve the performance of the debt oYver a longer pcriod of time. 106 10 Annex E qRr4"M UU 0 7 8 6 31. This brief roview of Colombia's economic performance and our efforts towards fiscal consolidation shows the govermment's continued cornitxnment to the achievement of macroeconomic stability for the recovery of growth. The support of the World Bank throughout these past years has been a decisive factors in the succcss of our efforts. Sincre yours, - i a±ulSantos ster of Finance and Public Credit 107 ll Annex E 108 Annex F Bogota, D.C., 8 de agosto de 2001 Doctor OLIVIER LAFOURCADE Director Subregional Colombia, Mexico y Venezuela Banco Mundial Washington,D.C.,U.S.A. Estimado doctor Lafourcade: Recibf con agrado su comunicaci6n en la que hace referencia a su reciente visita a Colombia en la que pudo conocer mas de cerca la situaci6n social y economica del pais. Para mi es verdaderamente gratificante recibir esta manifestaci6n de apoyo de parte suya, mas aun cuando afirma que usted y su equipo pudieron apreciar en su visita el constante esfuerzo del Gobierno Nacional por mejorar la calidad de vida de los colombianos. En su mensaje tambien hace alusi6n a las multiples oportunidades que tuvo de hablar con personas que estan verdaderamente agradecidas con este esfuerzo. Son esas personas las que nos animan para continuar con esta importante labor que, sin lugar a dudas, esta contribuyendo eficientemente at desarrollo del pais. 109 Annex F Le reitero nuevamente mi agradecimiento y le confirmo una vez mas mi firme decisibn de continuar con las reformas que se han iniciado, pues estamos logrando, como usted mismo pudo constatar, la transicion de una sociedad castigada por la inequidad a una sociedad en vfas de mejorar la distribuci6n de su ingreso y de dar mejores oportunidades a su gente. Reciba un cordial saludo, , ,C /-t--- 110 Annex G ANNEX G REPUBLIC OF COLOMBIA FISCAL STRENGTHENING PROGRAM-STRIICTURAL FISCAL ADJIJS1 MENT LOAN (SFAL) SPECIFICATIONS, DEFINITIONS, AND EVIDENCE FOR TRANCCHE RELEASE CONDITIONS First Tranche (Board Presentation) A. General Condition of the SFAL The macroeconomic framework of the Republic of Colombia is consistent with the objectives of the program. Specificity and Definition. The macroeconomic framework of the Republic of Colombia is consistent with the objectives of the program (as described in the government's Letter of Development Policy sent to the Bank) means that the macroeconomic program supported by the IMF's three-year extended arrangement facility (approved in December 1999) is on track. The macroeconomic program is considered to be on track if the quarterly performance criteria- namely, the overall fiscal deficit of the combined public sector, the inflation targets, the net international reserves, the increase in medium- and long-term debt, and the issuance of guarantees by the public sector for private external debt-were observed; and the Bank staff is satisfied with the overall progress of the program as defined in the Letter of Development Policy. Evidence. The Bank staff, in consultation with the IMF and the government, will exchange views on the macroeconomic performance, and will assess progress in the program's implementation through exchange of information via e-mail and periodic (for example, every two months) missions to Colombia. B. Specific Conditions of the SFAL Intergovernmental Fiscal Relations 1. The constitutional amendment (Acto Legislativo No. 1, of July 30, 2001) reforming the transfer of resources from the central government to subnationals has become effective, limiting the rate of growth of fiscal transfers to subnational governments and the rate of growth of the central administration operational expenditures for 2002 and there-after. Specificity and Definition. Congress has approved a Constitutional amendment (Acto Legislativo) limiting the rate of growth of the transfer of resources from the central government to subnationals and combining different kinds of transfers into a single transfer system (Sistema General de Participaciones). According to the Constitutional amendment, the rate of growth of fiscal transfers to subnational governments will be limited to the inflation rate plus 2 percent for years 2002 through 2005, and inflation rate plus 2.5 percent for years 2006 through 2008. The lII Annex G limits on the growth of the central administration operational expenditures will be equal to inflation plus 1.5 percent for 2002-08. Evidence. Official copy of Acto Legislativo No. 01 of 2001, as published in the Diario Of, the following principles: (a) links between contributions and benefits, and (b) no new entry into the unreformed exempted regime. Lvidence: Copy of the law. 122 Annex H ANNEX H REPUBLIC OF COLOMBIA FISCAL STRENGTHENING PROGRAM-STRUCTURAL FISCAL ADJUSTMENT LOAN (SFAL) MANAGEMENT OF PUBLIC FINANCES HI. This note reviews the background and advances that have taken place in Colombia in the area of accountability and effectiveness in the management of public finances, and outlines some outstanding issues as well as the government's agenda, and Bank support, in this area. The note draws on the substantial amount of work already undertaken in the context of the Colombia Country Financial Management Assessment (CFMA, September 1998; a second CFMA is scheduled to be carried out in FY03), the Country Procurement Assessment Report (CPAR, March 2001), the Public Financial Management Project (PFMP, closed March 2001) and the on- going PFMP-II which follows and expands on the earlier project. In addition, the Bank has supported the government's anti-corruption program through the World Bank Institute as well as an Institutional Development Facility (IDF) grant completed earlier this year. H2. The proposed operation, in tackling issues of fiscal control and sustainability, will help to fulfill the necessary conditions for improved performance in financial management. At a more specific level, however, the PMFP-II, following earlier activities under the first PFMP, explicitly addresses weaknesses in financial management and procurement. The project includes a component on public expenditures as well as a sub-component on procurement, the latter reflecting many of the recommendations of the CPAR. In addition to a number of quantitative objectives' outlined in the text below and largely related to the expansion of integrated information systems, the PFMP-II, which was approved in 2001 and with an expected completion date in 2006, is supporting activities that will lead to: (i) enhanced accountability for achievement of performance objectives, spending of public resources, and procurement of goods and services; (ii) increased value for money in public spending; (iii) increased transparency; and (iv) enhanced civil society capacity to monitor public sector performance. At the operational level, the project contributes to clarifying inter-institutional roles and equipping decision-makers with the tools, information systems and skills needed to make well-informed decisions about allocation, use, control and evaluation of public expenditures. Many of these qualitative improvements are also expected to be accompanied by financial gains to the government as a result of savings in expenditures. These financial gains are, however, difficult to quantify ex- ante. Other issues related to financial management are being addressed by the IMF and the IDB, and are also discussed below. H3. Looking forward, the CFMA scheduled for FY03 will add to the considerable existing knowledge base in the fiduciary area in Colombia by focusing specifically on issues of financial management in sub-national entities. The recommendations of the CFMA, as well as those of the 1. Briefly, some specific quantitative objectives of the PFMP-11 include an increase in the coverage of the central budget by the Integrated Financial Management System (SIIF) to at least 90 percent, the system's implementation in 145 regional offices of central government entities, and the development of a modified version of the system for decentralized entities and its implementation in 1 I such entities that spend about 12 percent of the national budget. 123 Annex H recent CPAR and other on-going activities, will in turn constitute an important input into the Colombia CAS which will be prepared with the new administration taking office in Colombia in August 2002. I. Background H4. During the last decade, various steps have been taken to strengthen public expen(liture management in Colombia. The Constitution of 1991 created the office of the Accountant General of the Nation (Contaduria General de la Naci6n, CGN) charged with the responsibility of consolidating the accounts of all levels of government, changed the role of the Comptroller CGeneral of the Republic (Contraloria General de la Repuiblica, CGR) from ex-ante control to ex- post audit of government expenditures, and emphasized the establishment of a modern financial administration based on efficiency and effectiveness in public spending. Reforms have been introduced to strengthen fiscal stability and macroeconomic coordination, implement a new planning framework based on increased civil society participation, provide autonomy in budget execution to operative agencies, and introduce performance management in the public sector. H5. The Bank, through its first PFMP project has supported the government's measures in these areas, focusing on strengthening institutional capacity and analysis in a number of key areas of public expenditure: in macro-programming, formulation and monitoring of the budget; in budget execution; and * in the evaluation of results of public expenditure. H6. In the area of macro-programming. formulation and monitoring of the budget, the PFMP financed the work of a multi-party Commission for the Rationalization of Public Expenditures which carried out a detailed review of public expenditure management issues, and whose recommendations have been an important input into policy reforms undertaken by the current government in various areas, such as pensions, territorial transfers and privatization. The project also assisted in the development of statistical accounting methodologies to strengthen the capacity of the MHCP and the DNP in macro-economic analysis as well as information systems for monitoring public finances and public sector cash flow. In the area of budget formulation the development of a system for formulation of the investment budget has been initiated, and a system to provide online access to budgeting norms has been implemented. W[7. In the field of budget execution, the development and implementation of an Integrated Financial Management System (SIIF) has been a major accomplishment supported by the PF MP. The system has been in operation since January 2000 in 46 central government entities and covers about 80 percent of the central government budget. The SIIF has automated the entire process of budget execution by line entities and has already resulted in a reduction in idle cash balances by about US$ 45 million per month on average. Its database is available to the CGR for real time audit of central government accounts. HE. The project also supported the establishment of the CGN and the training of its staff in financial audits. The CGN has since developed and implemented new accounting norms, a new chart of accounts and trained officials at all the three levels of government. The project also supported the CGR's efforts to implement a system of social control, by involving civil society in oversight and audit of public expenditures. The project has also assisted in the development of an investment planning module for the Treasury. In the area of public credit, the PFMP has 124 Annex H provided support in developing methodologies for risk management and in the identification of options for the replacement of the existing debt management system. H9. In the area of public expenditure results evaluation, Colombia has pioneered a comprehensive National System for Evaluation of Results of Public Sector Performance (SINERGIA) to institutionalize systematic evaluation of government programs, projects and entities; this results-oriented approach to management was assisted by the Bank through the PFMP. SINERGIA has been in operation since 1997 and has been implemented in all 16 sector ministries and 170 public entities. Under this system annual Indicative Plans are established for each sector and entity, outlining strategic objectives, specific indicators of results to be achieved, along with performance targets. Based on the SINERGIA, the National Planning Department (DNP) and Ministry of Finance and Public Credit (MHCP) have also started the practice of signing Efficiency Agreements with specific ministries and public entities. In exchange for commitments regarding availability of funds made by the DNP and MHCP, the ministry or public entity concerned commits to deliver specific outputs. II. Current Agenda: Addressing Outstanding Issues H1O. As can be appreciated from the above, Colombia has undertaken a number of fundamental reforms to its system of public expenditure management in recent years. These have included the creation and restructuring of institutions, the development of information systems and accounting methodologies, and the introduction of results evaluation systems for the public sector. Despite this progress, weaknesses continue to reduce the effectiveness of financial management and procurement processes in some areas. These outstanding issues were identified during the implementation of the Bank's PFMP and some were also highlighted in the Country Financial Assessment completed in September 1998. They now constitute the basis of reforms being supported by PFMP-II, as is outlined below. A. Financial Management HI 1. In the area of budget formulation and monitoring, the government, with Bank support through the PFMP-II, is addressing three key issues: • Ensuring consistency between the government's strategic development priorities and budgeted expenditures; * Increasing clarity in the roles of different levels of government and different institutions within the central government; and * Reducing the proliferation of norms as well as of reporting requirements. The background to these issues and government response are expanded on below. H12. Ensuring consistency between the government's strategic development priorities and budgeted expenditures. A number of constraints affect the degree to which the budget reflects the government's development concerns. Foremost among these is the fact that earmarking limits the degrees of freedom available to the central government to implement its policy priorities through changes in expenditure patterns. There is also a tendency to overestimate revenues and underestimate costs. The consequent relaxation of the budget constraint allows low priority programs to be included in the budget. The inadequacy of systems for evaluating programs and projects, the lack of good information on unit costs of inputs and outputs, and weak capacity at the sector ministries to carry out evaluations make project prioritization more difficult and affect 125 Annex H the ability to evaluate budget requests. The government's ability to match public expendit ires to strategic priorities is further affected by poor budget estimates compounded recently by irmacro- economic shocks. This has meant that the government is unable to execute a significant portion of budgeted expenditures. Budget turn-outs, relative to originally budgeted amounts, var) year to year, often as a function of revenue projections and as a result of spending requests made after the budget is approved. The latter have in the past been accommodated by postponing their execution and payment to the subsequent year. Fiscal arrears grew from 1.0 percent of CrDP in 1990 to 2.6 percent in 1999 although they declined in 2000 as a result of the government's fiscal adjustment program. Arrears are fundamentally the result of fiscal imbalances rather than financial management, although the two are closely related (i.e., fiscal imbalances lead to inefficient methods being used to control expenditures). The budget law of 2000 begins to address this issue by establishing that commitments carried over to the next year be subtracted from the budget appropriations of the concerned entity for that year. Additionally, the government, in an effort to improve fiscal management, began in 2000 to measure the fiscal deficit on a commitments basis (i.e., by accounting for commitments in the period in which they are incurred). This is expected to strengthen the Ministry of Finance in dealing with spending requests that come after budget formulation. 1113. The PFMP-II comprises specific actions to improve the responsiveness of the budget to the government's development plans. These include: (i) better quantification of earmarked expenditures and the development of options for reforms to reduce earmarking; (ii) enhancing information and methods for revenue projection and for evaluating expenditure and revenue policies; (iii) improvements in macro-programming through developing, enhancing and implementing models for medium- and long- term macro-economic projections and simulations; (iv) perfecting the methodology for calculating unit costs of inputs and outputs in a number of areas; and (v) upgrading the tools for formulating and monitoring the central government's investment budget. Hi4. Increasing clarity in the roles of different levels of government and different institttions within the central government. A lack of clarity regarding the respective roles of national, regional and local governments has at times resulted in institutions at the central level continuing to perform, and budget for, functions that have been devolved to local authorities. Within the central level the need also exists to clarify the roles and responsibilities of entities that participate in different stages of expenditure management. The division of expenditures responsibilities between the MHCP and the DNP has led to problems in budget integration. MHCP is responsible for the preparation of the recurrent budget, while DNP is charged with the formulation of National Development Plans and the budget for capital expenditures. Recently, efforts have been made to improve coordination between these two entities as well as with the CGR through an inter-institutional agreement. The PFMP-II is providing assistance in clarifying and adjustin,:g the roles and responsibilities of the various entities that participate in the programming, budgeting, execution and evaluation of public expenditures. 1115. Reducing the proliferation of norms as well as of reporting requirements. In Colombia, agencies monitoring public spending impose multiple reporting requirements on line agencies. In addition to increasing the compliance overload of the agencies, this leads at times to a proliferation of mutually inconsistent information that affects the quality of programming and budgeting decisions. The norms for formulation and execution of the budget need i.0 be consolidated and streamlined. The PFMP-II is addressing this issue by implementing existing 126 Annex H inter-institutional agreements between the MHCP, DNP and CGR to reduce the costs of monitoring of budgetary expenditures. H16. In the area of budget execution and accounting the government's activities concentrate on addressing the following issues: * Improving the quality and coverage of information systems; * Enhancing the quality of accounting data; and * Upgrading institutional capacity. 1117. As outlined above, the development and implementation of the SIIF responds to the need to integrate all accounting, budgeting and payments of the public sector at all levels. While major progress has been made, the system is still in its initial phase. The SIIF will also enable line ministries implementing investment projects to have immediate access to external and internal funds, beginning in January 2002. The PFMP-II is providing assistance to expand the coverage of the central budget by the SIIF to at least 90 percent, and to extend the system to 145 regional offices of central government entities; a modified version of the system for decentralized entities is also being developed and will be implemented in 11 such entities which spend about 12 percent of the national budget. Strengthening the information systems of the CGN to capture and integrate accounting infornation at the national, decentralized, and territorial levels is another key objective of the project. H18. Improvement in the quality of accounting information that is generated by public entities, especially those not covered by the SIIF, is another area receiving attention as part of the PFMP- I1. This is addressed through mechanisms to enhance the registration and reconciliation of accounting records; to fortify institutional capacity, the project is supporting activities in the areas of dissemination, advisory services and training for the improvement of accounting practices. More specific actions by the project in the area of management systems include improvements to the Treasury's investment portfolio management system to improve control of transactions and optimize returns on investment, and an improvement in the debt management system of the Directorate of Public Credit. While the latter system has adequate information on external debt, it suffers from various deficiencies, including an inability to connect with the SIIF or to allow debt service and cash flow projections. H19. In order to effect further improvements in the area of public expenditure results evaluation, the government, with the support of the PFMP-II, will: Expand SINERGIA, the system of public sector results evaluation; and . increase transparency through dissemination strategies to inform the public on performance targets and results achieved. H20. As background, the general budget (Presupuesto General de la Naci6n) covers all central government expenditures. Territorial and decentralized entities are governed by separate budgets. All are subject to control and audit by the CGR. The CGR is charged with carrying out audits of all public entities and individuals who manage public funds. The resulting audit reports are made available to the audited entity and to the general public. Each year, a General Auditing Program (Plan General de Auditoria) is established whose coverage extends, at a minimum, to 80 percent of the total budget. This year the CGR has also decided to audit 127 Annex H territorial transfers in the areas of health and education, as well as royalties transferred to twelve departments. The reports are timely and of generally good quality. H21. Besides these "micro-audits" the CGR is also responsible for macro and fiscal control of public finances and for an assessment of the accomplishment of development objectives. In this context the CGR prepares a number of reports that evaluate public finances and policies, including the National Development Plan, some of which are mandated by the Constitution. These reports are also are publicly available on the CGR's website2 and are presented to and discussed by Congress. In the economic area, the CGR's reports include: an audit of the general balances of the consolidated public sector, a report on the general budget and the treasury, an annual report on the state of public finances, and an annual report on public debt. H22. In addition to the audit and evaluation work undertaken by the CGR, there is an institutionalized process in Colombia for public sector results evaluation, as mandated bv the Constitution and by law 152 of 1994. The DNP is charged with implementing this system (SINERGIA) through its Department of Management and Results Evaluation (Direcci6o, de Evaluaci6n de Gesti6n y Resultados, DEGR). Annual sector targets are set jointly by DNP and line ministries, and are approved by the CONPES (Consejo Nacional de Politica Econ6mica y Social). It is important to note that SINERGIA is conceived as a system to facilitate the implementation of internal controls by entities of the public sector and one that enables the gradual achievement of improvements in accountability, transparency and service delivery. 'It is not seen as a system that is purely concerned with quantitative outputs (that falls under the purview of the CGN and CGR). Two general types of results agreements are envisioned: institutional goals (metas), and specific commitments (compromisos). General sector goals are those contained in the Indicative Plans (IPs) for each sector; these do not depend exclusively on the entity in question, but are the result of actions by agents in numerous sectors and of variables not entirely under the control of the public sector. It is therefore very difficult to measure the extent of the deviation from IPs. Commitments, on the other hand, are those results that can be measured and where a direct line of responsibility can be established. The latest results evaluation for the central government in 2000 has been completed, and is centered on a comparison of sector performance indicators between 2000 and 1999, their contribution to the achievement of development objectives, and the resources used to achieve said objectives. The various sector commitments and results are obviously far too numerous to mention here, but a summary can be found in the DNP website3. This process is carried out annually. H23. While much progress in the area of results evaluation has been made through the introduction of SINERGIA, advances have been constrained by: (i) the uncertainty of budgeted funds which remains as an impediment in enforcing compliance with agreed performan-ce targets; (ii) weak links between performance and incentives or penalties; (iii) the monitoring of performance only with respect to investment expenditures, not recurrent expenditures; (iv) the need to extend the evaluation system beyond the central level to public enterprises and territorial entities; and (v) limited systematic analysis of the incentives faced by public sector managers in executing the budget. H24. In this regard, measures to improve the quality of budget formulation, and thus to improve the ability of entities to comply with performance targets, have been addressed above. 2. www.contraloriagen.gov.co 3. www.dnp.gov.co 128 Annex H To address the remaining issues related to results assessment, the government, with support from the PFMP-I1, is expanding SINERGIA's project evaluation module to include recurrent expenditures in the evaluation process. New modules are being developed for strategic perfonnance evaluations of different sectors and for expanding performance evaluation to territorial and decentralized entities. Additionally, dissemination strategies are being designed to provide information to the public on performance targets for different policies, programs, projects and public sector entities, as well as on evaluation of the results achieved and the amount of money spent in the process, with a view to increase the transparency of public sector operations. H25. Other initiatives. Other initiatives in progress in the area of financial management include an agreement between the Bank and the CGN that is to be used as a model for other countries in the LAC region of partnership in the implementation and follow-up of fiduciary Bank policies and procedures. This agreement is building on PREM's work over several years with the Ministry of Finance. A Memorandum of Technical Understanding has been drafted that aims to complement the work of CGN to enhance the World Bank's fiduciary responsibility through financial control over cooperation and international development projects. In particular, the objective is to provide support and guidance to those responsible for the accounting and financial aspects of projects and programs in a number of areas: on fiduciary issues, by making available to the Bank the instruments and tools of financial information needed to enforce contractual commitments, policies and procedures established both in individual loan contracts and in the general rules governing their operation; in the production of integrated financial data on projects, through the SIIF or another compatible system that is easy to develop and can be quickly and reliably fed into the SIIF; in ensuring that financial information from borrowing and executing entities in World Bank-financed programs and projects are prepared in accordance with CGN accounting principles, are consistently applied and acceptable to the World Bank, and submitted to professional audit by the managers of those entities; and ensuring that in programs and projects, financial statements convey the existence of transparent fiscal mechanisms, a properly functioning financial management system, and compliance with the accounting, financial, administrative and managerial clauses of the respective World Bank loan contracts, in accordance with CGN regulations. H26. Financial management issues at the sub-national level are addressed in a number of the initiatives outlined above, namely: increasing the clarity of functions at different levels of government, strengthening the information systems of the CGN to capture and integrate accounting information at the decentralized level, and extending the results evaluation system to territorial entities. Nevertheless, in view of the importance of this issue for fiscal outcomes in the country as a whole, the upcoming Country Financial Management Assessment, scheduled for FY03, will pay particular attention to addressing financial management issues at the departmental and municipal level in a comprehensive manner. H27. IMF. Related activities are being conducted by the IMF. The safeguard assessments report on the audit mechanism for the Colombian central Bank, completed in May 2001 under the transitional procedures, concluded that while audit of the central bank's financial statements is undertaken, it does not comply strictly with the definition of an independent external audit under the International Standards on Auditing (ISA). The principles governing central bank auditing mechanisms are enshrined in the Colombian Constitution of 1991, whereby the President is responsible for auditing the central bank, and is permitted to delegate this function to 129 Annex H the Banking Superintendent and to an independent State agency, but not to a private firm. The central bank's financial statements are thus subject to an audit by an especially-assigned external Auditor General who is appointed directly by the President of the Republic. The Office's independence from the central bank was, in 2000, validated by the Colombian Constitutional Court, which also validated the legal basis of the auditing arrangements. In practice, the President names an auditor who is assigned an independent staff to assist him or her, using a methodology developed with the technical advice of Price Waterhouse. H28. In the Fund's assessment, the current auditing arrangements are capable of providing high-quality audit services. However, as indicated above, the existing mechanism falls short of ISA requirements regarding independence. Moreover, the IMF report concludes that the public perception of independence may be compromised by the fact that the costs of the Audit Office are paid by the central bank, and by appointments to staff positions in the Audit Office being made in accordance with central bank procedures. The assessment also noted that the Auditor General's main responsibilities are functions typically associated with an internal audit department, and that under international best practice, this would preclude the Audit Office from also performing the external review. The Fund report recommends the publication of the audit opinion together with the central banks's annual financial statement; the latter are made public through the journal of the bank, in major Colombian newspapers and on the bank's website. Based on this finding, and on the legal issues involved, the Fund and the authorities are exchanging views on the appropriate means to overcome this matter. The government also intends to request the Fund to assist it in addressing issues in the area of fiscal transparency. A fiscal transparency mission is envisaged early next year. B. Government Procurement and Contracting H29. One of the areas that has a major impact on the results of public expenditure is government procurement and contracting. The recently completed CPAR highlights the strengths and weaknesses of the Colombian legal and administrative public procurement framework and makes specific suggestions to improve it. H30. Law 80 of 1993, together with some 80 regulatory decrees and other laws provide the legal framework for public procurement in Colombia. Law 80 was intended to be a law of principles, which is positive. Some of the strengths of the law are the full decentralization of the procurement function to national and territorial entities, the prohibition of excessive formalities, and the use of alternative dispute resolution procedures. In terms of procedures and practices, the bidding and contract award process is relatively expeditious. Additionally, the President has issued a directive in 2000 (Agenda de Conectividad) which aims at establishing E-Government in Colombia; some institutions have begun to establish information systems for a wider dissemination of their procurement activities. Law 598 of 2000 established, within the CGR, a monitoring system for State contracting. The directive and the law are first steps towards the objective of establishing a procurement information system. H31. In terms of Bank-assisted projects, the CPAR found that agencies responsible for project implementation carry out procurement in accordance with Bank policies and procedures, using its standard bidding documents. Procurement performance in Bank projects was therefore concluded to be satisfactory and free of major problems. Two procurement audits and four agency procurement capacity assessments carried out in the past two years found no major problems and determined that procurement risk in Bank projects is lower than average. 130 Annex H H32. There is, however, a consensus that procurement and contracting activities are affected by serious deficiencies that have led to the inefficient use of public funds and that have become a source of corruption. Measures are needed to improve institutional aspects and to increase efficiency and transparency. The legal framework lacks clarity, over-regulates public procurement, and contains loopholes. Over-regulation and the high concern with corruptiqn has in turn led to an environment of distrust between the government and those doing business with it. Weak institutional leadership in the area makes it difficult to develop and improve the public procurement framework on an ongoing basis. There is also room to improve the monitoring and evaluation of procurement practices. H33. The government recognizes these problems and has taken steps to institute changes, as outlined below. Additionally, it is beginning the process of developing capacity for disseminating procurement opportunities, bidding documents and contract awards to the public via the internet, in an effort to carry out procurement transactions on-line. More generally, the government is currently working on an amendment to the Procurement and Contracting Law (law 80). The DNP has, in this context, decided to submit to the Consejo Nacional de Politica Economic y Social (CONPES), the Government's highest policy-making body in social and economic matters, a policy paper on public procurement which should draw on the findings and recommendations of the CPAR. H34. Drawing on the action plan outlined in the CPAR, the PFMP-II includes a procurement component that is providing assistance to: (a) develop legislation and regulations to support modifications to Law 80 with a view to improve the legal framework for public procurement; (b) design and implement institutional arrangements, including the possibility of establishing a national body, for developing and regulating public procurement and contracting policy, standardizing and disseminating procedures, bidding documents and contracts, and assisting central and territorial entities in procurement and contracting matters on an ongoing basis; (c) develop and implement a system for monitoring and evaluation of procurement and contracting practices in the public sector, piloting the system in two central level entities, one decentralized entity and two territorial entities; (d) design and implement an internet-based public procurement information system aimed at disseminating all information regarding procurement notices, bidding documents and contract awards to interested parties and the public, thereby setting the basis for online public procurement; and (e) carrying out a survey to determine procurement training needs and developing a pedagogical system aimed at providing public sector personnel with learning tools relating to the conceptual, technical, legal and procedural aspects of public procurement and contracting. H35. The government is also receiving assistance from the Inter-American Development Bank in the area of procurement, in particular for the modernization of control and supervision mechanisms in the public contracting information system (Sistema de Informaci6n de la Contratacion Estatal - SICE) implemented by the CGR. C. Anti-Corruption Initiatives H36. Colombia possesses an extensive legal framework to combat corruption. The Criminal Code, Law 190 (Estatuto anti-corrupci6n) and Law 200 (Estatuto Disciplinario de los Funcionarios Ptublicos) have provisions defining corrupt practices and the duties of public officials. Various entities monitor how the public sector carries out its functions in general and procurement in particular. Nevertheless, the perception in both the public and private sectors is 131 Annex H that corruption is common. The government's approach in the past has focused on introducing additional regulations, controsl and punitive actions-a strategy that has not fully resohled the problem. As a result, several new initiatives to combat corruption are presently being carried out both by the Presidency and Vice-Presidency of the Republic. H37. Following a request to the Bank made by President Pastrana, a joint initiative with the Bank, headed by the Vice-Presidency of the Republic, was launched in association with a variety of public and private institutions and civil society organizations engaged in anti-corruption activities. The objective is the construction of a National Strategy for Integrity and Governance (Estartegia Nacional de Integridad y Gobernabilidad). A central pillar for constructing such a strategy was the implementation of a national survey to establish perceptions regarding public sector institutional integrity and the incidence of corruption in the country. The survey, developed and implemented with Bank support, was conducted in seven of Colombia's main cities and covered over 8000 respondents from the public sector (central and local levels), the private sector, and households. The survey's results were summarized in a recent report. The report concluded that corruption is still perceived as a problem in Colombian societv, with negative perceptions regarding the legislative branch in particular. With respect to public sector institutions (166 of these, at the central and local levels, were surveyed), public sector officials were generally satisfied with their entities' performance and commitment to corporate objectives. Weaknesses did emerge regarding the coincidence between corporate objectives and citizens' needs. Governance and transparency are perceived to be weaker at the local level than at the central level, reflecting weaker controls in the former. The report also concludes that the incidence of corruption in the public sector-including state contracting and diversion of public resources - while being a matter of concern, does not present a serious enough situation to justify the high level of importance accorded to the topic at all social levels. Social concern with the issue appears to be the result of weaknesses in governance and a perception that the political system, particularly the legislative branch, lacks integrity, rather than being due to classic issues of corruption (i.e. rent-seeking activities). Taking into consideration these results, the government is designing a comprehensive anti-corruption policy. H38. Another initiative, an IDF grant for Colombia's Anti-Corruption Program, was begun in 1998. The grant was implemented by the Departamento Administrativo de la Funci6n Publica (DAFP) with the assistance of the UNDP, and was completed in December 2000. The main objectives of the grant were to improve understanding of the nature and causes of corruption in the Colombian public sector, to review the efficacy of measures taken so far, and to develop strategies to strengthen integrity, transparency and accountability. The grant carried out an in- depth institutional analysis of two public sector entities to identify the causes and patterns of corruption in the public sector and develop counter-measures to deal with them. While these case studies were intended to help prevent and control corruption within the entities covered, they were also expected to lead to general lessons and tools that could be applied to other public sector entities. Based on the case studies, a comprehensive report was prepared providing insights into corruption risks, the possible consequences of these risks, their relative severity and strategies to mitigate them. The report, Riesgos de Corrupci6n en la Administraci6n Publica, was published in book form, and 1500 copies of the book were distributed to public entities and other stakeholders. The book is also available in the Web-page of the DAFP. A gen1eric methodology for the preparation of corruption risk maps (Mapas de Riesgos) was developed. This tool enables any public sector entity to carry out a self-analysis of its vulnerability to corruption and to prepare a realistic strategy to combat corruption. In order to disseminate the 132 Annex H results of the activities supported by the IDF Grant, 38 workshops were held, and were attended by 1300 representatives of a wide range of ministries and departments. To help institutionalize the outputs of the IDF, the DAFP included anti-corruption as one of the major dimensions to be considered by sector ministries in the development of their Administrative Development Plans. These plans are required to be submitted to the DAFP under the Estatuto Bdsico de Administraci6n Publica. The DAFP is providing advice and guidance, on a continuing basis, to public agencies who have chosen to use the methodology. In parallel, considerable effort was made to integrate the activities of the IDF Grant with the broader Presidential Program for Combating Corruption, led by the Office of the Vice President. 133 Annex H 134 IBRD 26153 COLOMBIA - - . it b 0 *+Etx / t 7 -'( 0 cqf / sA' ~~~~~~VaN dupssr '9 EAI / C~~~~~~~~~~~~~~~~OLIV~A Sincle j a PANA A =Ate RGENTINA Sb- The boundaries, colors, denominations ['4' O~~~~~~~~~~nd any other informnation showsn on SAN ~~~Aj-i,-~~~'- 0 A~ics mop do oat ieply, on tie part of Sn -~~~~~~~~~~~ N. CM~~~~~~~~~~~~~~~~~ ~The World Bank Group, any judgment -8 1' P h) . ( >:> ; Si - Quto on the legal status of any territory, or any endorsement or acceptance 6t N/ > i , r n such boundaries t t-, a ~N T\ C; G ,., :td.A,^ .;;. V E N E Z U E L A ( 9 i eMdrilin S wf>R>q 1 A 9 / 12;. > Jose de Gusw a * \ . t <, t .1 > iR >J ASRan e n A ~~~ ~~~\ } Crenvi~~Plrec,o o; \ N >.t A R I h# aj7 :tV t*" \\MILES 0 ~~~~~~~~~~~~~~~~ c U ~~~~~~~~~~~~~~~~~~~~~~0501011 C-,C- ~ ~ ~ ~ ~ ~ ~ ~~~~~~ 0 ~~~~5011 0 150 200 2,50 8~~~~~~~~~~~~~~~~~~ R A Z I L ELEVATION ABOVE 1000 METERS . | * NATIONAL CAPITAL \ J Jl DEPARTMENT CAPITALS , E R U I RIVERS / I DEPARTMENT BOUNDARIES QI. -. - INTERNATIONAL BOUNDARIES .4,,i C / / 7r- 74- AUGUST 199A