Report No. 63369-CM Cameroon The Path to Fiscal Decentralization Opportunities and Challenges September, 2012 Poverty Reduction and Economic Management Africa Region Document of the World Bank Currency Equivalents Exchange Rate Effective date December 31, 2011 Currency Unit 507 XAF = US$1 Fiscal Year January 1 – December 31 Abbreviations and Acronyms CA Commune d’arrondissement CAC Centimes additionnels communaux CBFP Congo Basin Forest Partnership CEFAM Centre de Formation de l’Administration Municipale CFAF CFA Franc CIFOR Center for International Forestry Research CIME Centre des impôts des moyennes entreprises CIT Corporate Income Tax COFOG Classification of the Functions of Government DGE Direction des Grandes Entreprises DGI Direction Générale des Impôts DGT Direction Générale du Trésor DS Droit de Sortie FCU Fiscal Coordination Unit FEICOM Fonds spécial d’équipement et d’intervention intercommunale FOB Free on board GDP Gross Domestic Product GFMIS Government Financial Management Information System GFS Government Finance Statistics IMF International Monetary Fund IS Impôts sur les Sociétés LGs Local Governments MDG Millennium Development Goal MINATD Ministry of Territorial Administration and Decentralization MINEF Ministry of Forests and Environment MINEP Ministry of Environment and Nature Protection MINFI Ministry of Finance i MINFOF Ministry of Forests and Fauna NPFD Non-Permanent Forest Domain PA Précompte sur achats PFD Permanent Forest Domain PFM Public Finance Management PIT Personal Income Tax PNDP Programme National de Développement Participatif PRSP Poverty Reduction Strategy Paper PSRF Programme de sécurisation des recettes forestières RCM Revenue sur Capitaux Mobiliers RFA Redevance forestière annuelle SE Surtaxe à l’Exportation SSV Sales of Standing Volume TA Taxe d’Abattage TEU Taxe d’Entrée en Usine TRP Timber Recovery Permit TS Taxe Salariale TSA Treasury Single Account UFA Forest Management Units UFA Unités Forestières d'Aménagement VAT Value-Added Tax VC Ventes de Coupe ZBA Zero-Balance Account Vice President : Makhtar Diop Sector Director : Marcelo Giugale Sector Manager : Mark Roland Thomas Task Manager(s) : Raju Jan Singh ii CAMEROON The Path to Fiscal Decentralization – Opportunities and Challenges Table of Contents ACKNOWLEDGEMENT .............................................................................................................................. 1 EXECUTIVE SUMMARY AND RECOMMENDATIONS .................................................................................. 3 A. LEGAL FRAMEWORK – NEED FOR CLARIFICATION ......................................................................................... 5 B. SPENDING ASSIGNMENTS—NEXT STEPS ..................................................................................................... 5 C. A SHORT TO MEDIUM-TERM STRATEGY FOR PFM REFORMS IN CAMEROON ...................................................... 6 D. AN AGENDA FOR TRANSFER REFORM ......................................................................................................... 9 E. A NEW DESIGN FOR ASSIGNMENT AND SHARING OF FORESTRY REVENUES ........................................................ 10 CHAPTER 1: OBJECTIVES AND POLICY FRAMEWORK FOR DECENTRALIZATION IN CAMEROON........... 12 A. INTRODUCTION ................................................................................................................................... 12 B. REFLECTIONS ON METHOD AND INTERNATIONAL EXPERIENCE ....................................................................... 12 C. WHAT DO WE SEE IN CAMEROON? THE MACROECONOMIC CONTEXT............................................................. 16 D. OUTCOMES OF THE DECENTRALIZATION PROCESS ....................................................................................... 17 CHAPTER 2: THE LEGAL AND INSTITUTIONAL FRAMEWORK ............................................................... 22 A. INTRODUCTION ................................................................................................................................... 22 B. ANTECEDENTS .................................................................................................................................... 25 C. INSTITUTIONAL FRAMEWORK ................................................................................................................. 25 D. NEXT STEPS ........................................................................................................................................ 30 CHAPTER 3: ASSIGNMENT OF RESPONSIBILITIES: WHO DOES WHAT? ................................................ 31 A. INTRODUCTION ................................................................................................................................... 31 B. THE UNDERLYING RESPONSIBILITY FRAMEWORK ......................................................................................... 31 C. HOW TO OPERATIONALIZE RESPONSIBILITIES? ........................................................................................... 33 D. SPENDING ASSIGNMENTS—NEXT STEPS ................................................................................................... 35 CHAPTER 4: BUDGETING, SOFT BUDGET CONSTRAINTS AND ACCOUNTABILITY ................................. 37 A. INTRODUCTION ................................................................................................................................... 37 B. BUDGETING ....................................................................................................................................... 38 C. DEBT MANAGEMENT, ARREARS AND IMBALANCES IN THE INTERGOVERNMENTAL SYSTEM ................................... 47 D. TREASURY AND CASH MANAGEMENT ....................................................................................................... 49 E. INSTITUTIONAL ARRANGEMENTS............................................................................................................. 51 F. A SHORT TO MEDIUM-TERM STRATEGY FOR PFM REFORMS IN CAMEROON .................................................... 52 CHAPTER 5: SUBNATIONAL REVENUES AND ADMINISTRATION .......................................................... 54 A. INTRODUCTION ................................................................................................................................... 54 B. WHAT LOCAL TAX DESIGN SHOULD LOOK LIKE ............................................................................................ 54 C. CURRENT PRACTICES ............................................................................................................................ 55 D. ASSESSMENT ...................................................................................................................................... 60 E. TAX ADMINISTRATION .......................................................................................................................... 65 F. A RENEWED SUBNATIONAL TAX REFORM AGENDA .................................................................................... 67 CHAPTER 6: TRANSFER DESIGN........................................................................................................... 71 A. INTRODUCTION ................................................................................................................................... 71 B. CURRENT TRANSFERS ........................................................................................................................... 71 C. CAPITAL TRANSFERS ............................................................................................................................. 72 D. OVERALL EFFECTS ................................................................................................................................ 72 E. SPECIAL PURPOSE TRANSFERS OR DIRECT PROVISION ................................................................................... 75 iii F. SOME ALTERNATIVE EQUALIZATION TRANSFER DESIGNS ............................................................................... 75 G. FORMULA-BASED GENERAL PURPOSE TRANSFERS ...................................................................................... 76 H. AN AGENDA FOR TRANSFER REFORM ....................................................................................................... 78 CHAPTER 7: FORESTRY, DECENTRALIZATION, DEFORESTATION AND RENT-SEEKING .......................... 80 A. INTRODUCTION ................................................................................................................................... 80 B. BACKGROUND .................................................................................................................................... 80 C. A DESCRIPTION OF FORESTRY TAXATION ................................................................................................... 82 D. COLLECTED TAXES................................................................................................................................ 83 E. DESIGN AND MANAGEMENT OF THE RFA ................................................................................................ 86 F. A NEW DESIGN FOR ASSIGNMENT AND SHARING OF FORESTRY REVENUES ........................................................ 95 REFERENCES ........................................................................................................................................... 98 APPENDIX ............................................................................................................................................ 100 ANNEXE: MAJOR PROBLEMS FROM THE 56 COMPTES ADMINISTRATIVES ........................................... 122 List of Charts Chart 1:1: Modified Subsidiarity Principles .................................................................................. 14 Chart 1:2: Relative Living Standards and the Wellbeing Indicator (average rank)....................... 18 Chart 1:3: Changes in Regional Living Standards (in percent) ...................................................... 20 Chart 2:1: Institutional Framework: Decentralization versus Deconcentration .......................... 26 Chart 2:2: Composition of the Regional Council .......................................................................... 27 Chart 2:3: Organization of Urban Metropolitan Areas ................................................................. 28 Chart 4:1: Local Governments’ Budgets and Execution, 2007 (CFAF millions) ............................ 41 Chart 4:2: CAC Element in Municipalities’ Budgets (in percent) .................................................. 42 Chart 4:3: Budget Execution (in percent) ..................................................................................... 43 Chart 4:4: Municipalities’ Current Budget Execution, Utilities (in percent) ................................. 44 Chart 4:5 : Municipalities Current Budget Execution, Pensions (in percent) ................................ 44 Chart 4:6 : Municipalities Current Budget Execution, Wages (in percent) .................................. 45 Chart 4:7: Debt Payments (CFAF millions).................................................................................... 46 Chart 4:8: Execution, FEICOM Grants Component (CFAF millions) .............................................. 46 Chart 4:9: Spending Circuit with a TSA ......................................................................................... 50 Chart 5:1: CAC Allocations, 2010 (CFAF per capita) ..................................................................... 61 Chart 5:2: RFA and CAC, 2010 (CFAF per capita) .......................................................................... 64 Chart 5:3: Revenue Circuit with a TSA .......................................................................................... 67 Chart 6:1: CAC and FEICOM Grants, 2010 (CFAF per capita)........................................................ 73 Chart 6:2: Overall Local Resource Availability, 2010 (CFAF per capita) ....................................... 73 Chart 6:3: Central Social Spending at Subnational Level, 2011 (CFAF per capita) ....................... 74 Chart 7:1: Specific and Export-Related Taxes, 1998-2010 (CFAF billions).................................... 86 List of Tables Table 1-1: Cameroon Fiscal Indicators (in percent of GDP) ......................................................... 16 Table 2-1: Cameroon – Timeline of Legislation Relating to Decentralization .............................. 24 Table 3-1: Subnational Spending Assignments, Cameroon .......................................................... 34 Table 3-2: China: Local Government Spending (in percent of total) ............................................ 36 Table 4-1: Content of Local governments’ Budgets ..................................................................... 40 Table 4-2: Budget Classification for Decentralized Governments ............................................... 40 Table 4-3: Eligibility and Conditions for Loans from FEICOM ....................................................... 48 iv Table 5-1: Administrative Accounts – Revenues (54 sample accounts, excl. Yaoundé and Douala) .................................................................................................................................. 57 Table 5-2: Administrative Accounts from Douala VI and Yaoundé VI, Revenues ........................ 58 Table 7-1: Forestry Taxes in Cameroon ........................................................................................ 84 Table 7-2: RFA and Municipalities’ Budgets ................................................................................. 91 Appendix Charts Chart A 1: International Comparisons ........................................................................................ 100 Appendix Tables Table A 1: Cameroon: Spatial dimensions of Poverty ................................................................ 103 Table A 2: Summary Table of Indicators ..................................................................................... 104 Table A 3: Comparison of Competences between Deconcentrated Regions/Departments and Decentralized Regions ............................................................................................ 107 Table A 4: Comparison of Competencies between Administrative (Deconcentrated) Departments and Municipalities ............................................................................ 110 Table A 5: Comparison of Competencies between Decentralized Regions and Municipalities . 114 Table A 6 : Municipal Taxes (Law 2009/019) .............................................................................. 118 Table A 7: Municipal Taxes, Fees, and User Charges (Law 2009/019) ...................................... 119 Table A 8: Tax Assignments between Urban Metropolitan Areas and City Councils ................. 121 v Acknowledgement This report was prepared by a Team led by Raju Jan Singh (AFTP3). The Team included David Abouem a Tchoyi (Consultant), Ehtisham Ahmad (Consultant, University of Bonn and London School of Economics), Bjorn Dahlin van Wees (Consultant), Faustin-Ange Koyassé (AFTP3), and Sylvie Ndze (AFMCM), as well as Paolo Omar Cerutti and Guillaume Lescuyer (Consultants, both CIFOR) and Bernard Dafflon and Thierry Madiès (Consultants, both University of Fribourg). Peer reviewers are Giorgio Brosio (Turin University), Jean-Jacques Dethier (DECRS), and Lili Liu (PRMED). Mary Barton-Dock (Country Director, AFCC1), Greg Binkert (Country Director, AFCC1), Kathryn Hollifield (Acting Country Director, AFCC1), and Jan Walliser (Sector Manager, AFTP3), Mark Roland Thomas ( Sector Manager, AFTP3) provided guidance and advice. The team would like to thank a number of colleagues who provided insights, comments and support in the course of the report preparation. They include James Acworth (AFTEN), Abel Bove (AFTPR), Simon Davies (AFTP3), Jan Drodz (AFTUW), Serge Emeran Menang Evouna (AFTEN), Kai Kaiser (PREMED), Meike van Ginneken (AFTUW), Dan Murphy (AFCC1), Amadou Nchare (AFTAR), Nneoma Nwogu (LEGAF), Manievel Sene (AFTAR), Cia Sjetnan (AFCC1), Peter Taniform (AFTR), and Serdar Yilmaz (AFTPR). The Team is extremely grateful for the cooperation of the Cameroonian authorities and for the invaluable contributions of senior officials and governmental agencies. We acknowledge in particular the following institutions: the Ministry of Finance, the Ministry of Economy, Planning and Regional Development, the Ministry of Territorial Administration and Decentralization, FEICOM, the National Institute of Statistics, and the PNDP. In particular, the Team would like to thank Mr. Akoa (Director General of FEICOM) and his staff, Mr. Bagueka (Directror General of Taxation), Mr. Edoa (Director General of Budget), Mr. Owona Owona (Director of Territorial Administration and Decentralization) and his staff, Mr. Penda, Permanent Secretary of the National Council of Decentralization, Late Mr. Sambo (Director General of Treasury, Financial and Monetary Cooperation), and Mr. Tasong (Secretary General of MINEPAT). Our deep recognition and thanks go to Mrs. Nga, PNDP National Coordinator and her staff, namely Mrs. Dang and Mr. Bekak, Sab and Essoh. Without them, our field visits and meetings with municipal representatives and officials of about a hundred municipalities from the Center, East, Littoral, and North West regions would not have been as efficient and productive. In this regard, the Team is deeply grateful for the extensive discussions with Government Delegates, Mayors, and Municipal officials in Bamenda, Bertoua, Douala and Yaoundé and their surrounding municipalities. The Team also gained a lot in discussions with representatives of civil society and Cameroon’s development partners. In particular, it wishes to thank representatives of Dynamique Citoyenne, PWYP, Center for Environment and Development (CED), Association Citoyenne de Defense des Intérêts Collectifs (ACDIC), Centre Régional Africain pour le Développement Endogène (CRADEC) and the Public Sector Labor Union, as well as Gerard 1 Daraspe (KfW), Sandrine Ebakisse (GIZ), Anja Heuft (GIZ), Michel Lazare (IMF), Frédérique Same Ekobo (EU), Zephirin Emini (UNDP), Ludger Kaup (GIZ), Leopold Nana (GIZ), and Marion Sejourné (French Foreign Affairs). 2 Executive Summary and Recommendations Decentralization: goals and aspirations Decentralization in Cameroon is an instrument to meet multiple objectives. The 1972 Constitution and supporting legal framework envisage that decentralization will: x ensure effective public service delivery; realize a better matching of local preferences and the services delivered; x provide for enhanced accountability of public officials; x reduce regional tensions, especially in the natural resource rich regions, by providing a share of revenues, particularly for the exploitation of forestry revenues; x effectively provide for minority and extremely vulnerable groups, such as the Baka; and x lead to a better environmental management and conservation of the world’s rain forest resources. The 2009 PRSP 1, also postulated that the renewed decentralization process will better mobilize Cameroon’s growth potential, including through the development of regional growth hubs in Douala and Yaoundé. It also expected that the decentralization would lead to better revenue generation. A further objective was to encourage gender equality. The World Bank and other multilateral and bilateral donors have supported these objectives. This report takes stock of Cameroon’s experience with decentralization. It examines whether the design has been coherently formulated and preconditions for decentralization have been put in place. It also suggests short to medium-term proposals to make the process work better, within the constraints of significant data limitations. Approaches to decentralization The design of the decentralization process in Cameroon reflects the “normativeâ€? approaches to decentralization—assuming benign officials and policy makers, and stipulating a normative allocation of responsibilities. In reality, a political economy story is at play, as suggested by developments in the “positiveâ€? literature that drop the assumption of disinterested officials. The focus of the political economy literature is mainly on “incentivesâ€? for local officials to provide services efficiently, or to cheat. The building blocks of both normative and positive approaches to decentralization are similar—spending and taxation assignments, design of transfers, debt management and information flows and instruments for implementation. However, the sequencing and mix of the instruments might vary, as we discuss below. The 2009 PRSP candidly acknowledges problems with the decentralization process, but renews the government’s commitment to its explicit goals. The process received a fillip by 1 See Government of Cameroon, 2009, Growth and Employment Strategy Paper, 2010-2020 (PRSP). 3 the 2009 decentralization legislation reiterating it as a priority of the administration. Consequently, the expectations of mayors and the wider public have been raised concerning greater responsibilities, adequate finances as well as expectation of better service delivery on the part of the electorate. A failure to meet these legitimate expectations could generate a political backlash. Service delivery indicators in Cameroon trail those of countries at similar income levels, including for child mortality and life expectancy at birth, which are worse than the average for sub-Saharan Africa. There is a wide dispersion in service delivery across regions in Cameroon, and the situation in some of the less developed regions has worsened in the past decade or so. There is an increase in poverty in the North, Far North, and East, as well as declining access to clean toilets, drinking water, as well as increase in child mortality. The PRSP assumes that renewed emphasis on decentralization would address both the inequalities in service delivery and low levels of provision. The management and sharing of forestry revenues have a major role to play in diffusing tensions in sensitive border regions. If incentives are properly designed, decentralization could reduce illegal logging and subsequent deforestation. However, local officials may not have the right incentives to provide for the marginalized and minority groups within the region (such as the Baka). Assessment, Risks and Recommendations Despite the good intentions in the Constitution and subsequent legislation, Cameroon lacks an effective strategy and an operational plan for decentralization. Indeed, line ministries appear to regard decentralization as a threat to their control over resources and influence. Periodic changes are introduced with limited consultation with the stakeholders or notice to local governments to adjust their plans and budgets. This occurred with the economic crisis in 2008, 2 and a decree of MINATD in July 20113 in the middle of the financial year adjusted the allocation of the shares in general revenues (CAC) between the metropolitan areas and their associated municipalities. The decree was not widely publicized, and there is insufficient information at present to simulate its effects. However, this change is unlikely to affect the overall assessment made at the time of the mission (at the beginning of March 2011). The authorities face a number of polar options; including to move: (1) with a “big bangâ€? implementation; or (2) implement decentralization in a measured manner, through certification linked to capacity enhancements. Preconditions for the big bang strategy would include (a) transfer of full functional responsibilities consistent with “subsidiarityâ€? principles; (b) clear own-source revenues; (c) a system of equalization transfers; (d) better information on who spends what and the results of the spending; as well as (e) hard budget constraints to minimize additional liabilities for 2 Ad hoc changes in the sharing arrangements for key revenue items, such as a “temporaryâ€? 50 percent reduction in the sharing of forestry revenues, the RFA, were introduced at this time. 3 Decree No 2011/1731/PM. 4 the center. A more gradual decentralization differs from the rapid strategy for item (a), as the full functions (or sub-sub functions) are devolved on the basis of certification as regions and municipalities meet specified criteria. The argument that local governments lack the capacity to carry out certain functions cannot be made strongly, as the transfer of human resources and financing can easily address any shortcomings that might be apparent under the present circumstances. Asymmetric arrangements may result with both rapid and more gradual strategies, providing greater responsibilities and financing options, e.g., for natural resource rich and border regions. Often asymmetric options are needed in order to suppress separatist tendencies, and may also play a role in the management of deforestation and addressing climate change. A. LEGAL FRAMEWORK – NEED FOR CLARIFICATION The legal framework relating to decentralization is overlapping, cumbersome and contradictory, and in many respects open to different interpretations. The main difficulty is that decentralized functions are ill-defined and not distinct from “deconcentratedâ€? operations of the central government. The tension between subsidiarity and concurrent competencies is one such example. There are also some provisions, such as the bailout clause for local governments that are extremely damaging in terms of the effects on local government behavior and incentives. A first step would be to “clean upâ€? the legal framework to remove obvious contradictions and ambiguities. It is important to address subnational accountability—including overlapping responsibilities for the same function, and bailout provisions that eliminate hard budget constraints on local governments. Equally important would be the promulgation of implementing regulations, in order to make the legislative intent more comprehensible for local officials. This can only be done if there is greater clarity on the extent and speed of the decentralization process. B. SPENDING ASSIGNMENTS—NEXT STEPS In the short to medium-term, it would be useful to consider which function or sub- function might be assigned to lower levels of government, consistent with the chosen strategy. The assignment of administrative arrangements at the discretion of central ministries will not result in effective decentralization, as electorates cannot hold mayors or local officials responsible for the effective implementation of the devolved function. While primary education is a typical decentralized function in many countries, in Cameroon local governments have limited effective responsibilities in this area. Under the recent changes, local governments can only build additional classrooms at the direction of the central Ministry of Basic Education. Clarity in the functions devolved has to be accompanied with better and timelier information on what is spent and the outcomes of the spending. This information is needed for macroeconomic coordination by the central government, and is also a critical element in making the local governments more accountable to their electorates. It should help with 5 “yardstick competitionâ€? as electorates benchmark the performance of local governments in relation to neighboring jurisdictions, and countries. Even as more information is needed on “decentralizedâ€? operations, there should also be better accounting for “deconcentratedâ€? operations on behalf of the central government. The same standards as for “decentralizedâ€? operations should apply, with full information on what was spent and the results of the spending by the prefects and the sub-prefects. They too should be subject to “benchmarkingâ€? and if they were unable to perform with a minimum level of efficiency, there would be a strong case for “devolvingâ€? the functions to subnational jurisdictions that might have better incentives to operate efficiently. C. A SHORT TO MEDIUM-TERM STRATEGY FOR PFM REFORMS IN CAMEROON The availability, timeliness, and reliability of information on subnational operations are very poor. As of July 2011, only around 16 percent of the municipalities had provided certified administrative accounts for 2009 to the Supreme Court’s Audit Bench. 4 The lack of timely and accurate information has led to serious allegations of misuse of resources and corruption in better off localities, such as Douala, as well as in less developed forest regions, like Yokadouma. The problems with the public finance management (PFM) system identified at the local levels in Cameroon, including the buildup of arrears, are linked to deeper and more pervasive issues, including at the center (IMF, 2011). However, a solution at one level of government cannot be found in isolation of adverse incentives and lacunae at the other. Several enhancements are needed to strengthen PFM at the subnational levels. These include: creating a common platform of fully developed Government Financial Management Information Systems (GFMIS) at each level of government; establishing linkages to other levels and using a common platform, and particularly a common chart of accounts and fully specified budget classification system (based on GFS2001 and COFOG); and initiating standardized monitoring of arrears and buildup of liabilities at the subnational level. The latter should also be a function of the central debt management department. There are serious difficulties in the synchronization of the budget cycles of the local and central governments. The local budgets must be prepared before the central budget is known in late November, and often the allocated resources do not materialize for several months. Further, changes in the sharing arrangements, including for forestry—critically important for many municipalities—were made without informing them, e.g., reducing allocations by half in 2009. This forced the municipalities to assume the stabilization function—which they were ill equipped to handle. The resulting arrears, particularly contributions for social insurance and utilities that must be covered by the center, are therefore not surprising. The local governments thus engage in retaliatory game-play. 4 The Supreme Court’s Audit Bench only managed to receive 16.5 percent of municipalities 2009 administrative accounts: http://www.cameroon-tribune.cm/national/1-politique/65530-gestion-des-fonds-publics-le-rapport-2009-de- la-chambre-des-comptes-disponible. 6 There have been advances in the design and operation of the Treasury Single Account (TSA). While this has the potential to improve the monitoring and management of cash, delays in payments for local governments and absence of information on their resources can damage confidence in what should be a positive reform for both the central and local governments. A system of correspondent accounts within the TSA should be created for local governments. In the absence of a uniform government financial information system, the TSA should permit the local governments to operate zero-balance accounts against their correspondent accounts, so that no payment delays ensue. This is critical in getting acceptance for the TSA and greater transparency in the management of public monies at the local level. Complete, standardized, and timely information on the budget process, using both the economic and functional classifications, is essential for accountability. This should be mandatory for the decentralized organs as well as the deconcentrated parts of government so that a full picture of who spends what and where is available to the governments at each level, as well as local electorates. Another critical reform will be to eliminate the bailout clauses from the legal framework, in order to generate appropriate incentives for accountable local operations. Realistically, performance budgeting is a distant goal for countries like Cameroon. However, as functions or sub-functions are identified for devolution, simple criteria for measurement of outcomes should also be identified—such as literacy or morbidity rates. These could be tracked separately and used to inform electorates, and the central government, as to how effective subnational governments are in the provision of public services and meeting assigned MDGs. These could also be used for performance-based transfers in cases where the infrastructure for service delivery is deficient. These criteria are also needed for the evaluation and measurement of deconcentrated operations of government at the local level. Thus, if local prefects or administrators are unable to meet established criteria, actions should be taken to change the administrators or devolve the functions. Institutional clarity is also needed—including on: x macro-fiscal coordination, leading to closer and better integration of the budget processes at each level of government; and x separation of the debt/lending and grant allocation functions. It is important that MINFI’s financial and fiscal responsibilities should be clearly defined. The administrative functions should be managed separately from these and should build on the initiatives under the MINFI and not attempt to contradict or duplicate them. MINATD should continue to play a major role in the areas of institutional and functional responsibilities, as well as certifying the readiness of local governments to assume new functions. It would also be important to clarify the roles of bodies that straddle areas—such as FEICOM (which should not duplicate DGI functions in relation to taxes and DGT functions in relation to revenue-sharing). A decision is needed as to whether FEICOM should be either a Grants Commission or a Development Bank (but not both). 7 Similarly, while the coordination functions of the newly operational Comité National des Finances Locales (CONAFIL) are welcome,5 it is unlikely to succeed unless the data are generated on a standardized basis, consistent with the norms of MINFI. If the current confusion is maintained, with separate formats generated by FEICOM as at present, it is unlikely that CONAFIL will be able to make any significant progress with the current availability of information. CONAFIL’s main functions are to monitor and optimize local governments’ revenue mobilization and efficient management of local finances—these are typical areas where MINFI standards are applicable, although it technically falls under MINATD. Enhancing local own-source revenues Financing local spending is the main objective for the revenue-sharing arrangements, particularly for the CAC and the share of forestry revenues (RFA). Some of the user fees introduced recently meet the benefit principle, and are steps in the right direction— provided that the responsibility for the services is clearly identified. Revenue-sharing arrangements address vertical imbalances, but are disequalizing, i.e., exacerbate horizontal imbalances. This may be partly a policy choice to create the growth centers in Douala and the capital—but the weak position of the municipalities, in these and other metropolitan areas, further reduces local government resources and accountability (although this point may have been partially addressed as the July 2011 decree is implemented). The additional resources allocated to the largest metropolitan areas have not led to a higher growth trajectory, and there are also doubts as to the enhancement in the quality and level of public services. A key element in the political economy story is to permit local governments to control their own taxes at the margin to meet local priorities. Some of the taxes that should be clearly local, given international experience, such as the property and vehicle taxes, are presently shared with other levels of government or FEICOM. This sharing dilutes incentives forging accountability. Thus, the property tax should become entirely “localâ€? with setting of the relevant rate (say within a band that could be legislated nationally, as in several unitary states). Other aspects of the administration, such as the cadastre or audit could be carried out by an independent or national agency without affecting the “local controlâ€? of the tax. Work needs to be initiated to establish a cadastre (this could be maintained at the central or regional level), as well as valuation mechanisms and a system to track property transactions. For the more advanced municipalities, such as Douala and Yaoundé, consideration could be given to applying the techniques of modern tax administration—using self-assessment and applying sanctions for misdeclaration (i.e., purchasing the property at above declared values). It is important to avoid taxes on internal trade. These distort incentives and induce price cascading. It is better to rely on the VAT, and share the revenues, as is currently the 5 CONAFIL was established by the 2009 legislation (Law 2009/011, art. 109) and the operational regulations were issued on July 18, 2011 (Decree No 2011/1732/PM). 8 practice. However, the administrative charge of 1 percent is high, especially as the tax has to be collected regardless of the sharing arrangements. A full assessment of the myriad of local taxes and fees should be conducted. It has been difficult for the mission to obtain verified accounts of all municipal operations for any of the recent years, but it appears that many of the numerous local tax heads (apart from the taxe globale) generate little or no revenues. Some of the nuisance fees and charges could be eliminated. Alternative, non-distorting subnational tax handles could be sought. A piggy-back on the income taxes could be assigned to the metropolitan areas, or regions. The piggy-back or surcharge could be applied to the central base, perhaps within a specified range and does not require local administration. The local return could be made directly to local governments, even if the audit and verification functions remain the responsibility of central tax administration, DGI. It is, however, likely that much of the additional “piggy-backedâ€? income tax would benefit Yaoundé and Douala and their municipalities—facilitating an employment and growth strategy, but also putting greater emphasis on the better design and implementation of an equalization framework for better overall service delivery. Once the basis for the new tax assignments has been thought through, it would be appropriate to revise the legislation on the tax assignments and also to promulgate the necessary implementation regulations. There have been attempts to streamline and modernize tax administration, including using a single return for the large taxpayers. This is a positive step, given the emphasis of the government to improve the business climate, and reduce burdens on taxpayers. However, this reform has changed the way the taxes are collected and the information available to local governments (including significant liabilities for local governments, particularly pertaining to the shared taxes). For local fears to be allayed, the reform needs to be supplemented with a timely flow of information (on a daily, weekly and monthly basis) to the local governments concerning resources that have been collected on their account and credited to their correspondent account in the TSA. Also, there should be no delay in permitting local governments’ access to their taxes and shared resources, as described above, in relation to the TSA. D. AN AGENDA FOR TRANSFER REFORM Deficit-filling transfers are specified in the law and subvert any incentives for local efficiency or accountability. Additional transfers for Yaoundé and Douala to promote growth centers have exacerbated inequalities without improvements in growth or living standards in the metropolitan regions, although the July 2011 decree should begin to redress this problem. Other than in Yaoundé and Douala, current transfers are made more or less on an equal per capita basis. The difficulty is that the “populationâ€? basis for the allocation is quite out of date, and has been “politicizedâ€?, with resistance to updating the numbers that might affect the shared revenues. Current transfers to finance local responsibilities should be based on assessments of need and relative cost of public service delivery, as well as the relative revenue-raising capacities of the local governments. Thus, a properly designed system would facilitate the provision of 9 similar levels of service delivery at similar levels of revenue effort, and also improve incentives for effective service delivery. While population remains an important element influencing cost, it would be effectively “depoliticizedâ€?, as more accurate bases for the allocation would be used. Initially, relatively simple factors should be used, given the paucity of accurate information in Cameroon. Gap-filling transfers need to be stopped, as they adversely affect incentives for efficient subnational operations, as well as accountability. The legal framework should be adjusted to eliminate the automatic financing of deficits. The case for using transfers to generate growth should be based on properly designed capital transfers. However, the effectiveness of these transfers needs to be carefully evaluated, and juxtaposed against an “equalizationâ€? strategy for current transfers. Special- purpose transfers may be needed in cases where the local politicians may not have the interests of marginalized groups and minorities as a priority. Work is needed to design and implement performance-based transfer mechanisms in which future allocations are based on monitorable outcomes. E. A NEW DESIGN FOR ASSIGNMENT AND SHARING OF FORESTRY REVENUES An efficient design of the assignments and sharing of forestry revenues will have a major impact on the poorest groups in Cameroon, generating development in some of the most backward areas of the country. It will also affect incentives to limit or enhance deforestation, and thus affect climate change. In this regard, what happens in Cameroon will also be relevant for other countries facing difficulties with the management of tropical forests. And with some modifications to address non-renewable resources, the underlying logic and models could become relevant for the treatment of other natural resources in Cameroon, and elsewhere. The RFA bestows asymmetric revenues on the forestry regions, and the shared revenues fluctuate with activity levels and international price variations. It is important to protect localities from downward swings in international prices and activity levels, and any “smoothingâ€? can generally be better performed by the central government, given its reserves and cheaper access to credit. However, the impact on the forestry region (both deconcentrated and decentralized) was exacerbated in 2009, as the proportion of RFA accruing to them was reduced by half. The loss in RFA revenues was to a large extent compensated by trade related taxes that are based on logging volumes, and reduced restrictions on “protectedâ€? exports. These, however, do not accrue to the local or regional governments. In addition, there is a shortfall in relation to the expected allocation of funds from FEICOM Thus, the perception is reinforced that income fluctuations were forced on poor or marginal communities, while the interests of the state were protected by the shift towards taxes on a more stable volume of logging that are not shared. This is bound to generate a negative impact that might offset the nation-building aspects of revenue-sharing. It might also encourage illegal forms of logging, whose rents are not shared with the center, paralleling the game play with arrears on utility payments. This can undermine the elaborate structure and incentives to move from informal to formal arrangements for managing forestry 10 resources. The temptation to divert, or misuse the resources in this region has been quite strong and predates the ad hoc reductions in the RFA which have only increased the incentives to divert resources (we have noted the recent jailing of the mayor and treasurer of Yokadouma for embezzlement). Aerial/satellite monitoring of logging is now technologically feasible (as in Indonesia). This can lead to greater accountability of the local officials, both deconcentrated and decentralized. While this is a subject that requires additional research, some possible components of the analysis are described below. In this report, we examine the building blocks of the decentralization framework—starting in Chapter I with a description of the status quo and the methods to be used. In Chapter II we discuss the unclear legal framework in Cameroon. The fuzzy spending responsibilities and lack of effective strategy to address the extent and scope of decentralization are examined in Chapter III. The weaknesses in information flows and public financial management (PFM) issues, including in the budgeting, treasury and debt management systems are addressed in Chapter IV. We examine in Chapter V the absence of effective own-source revenues that limits the accountability of local governments. This accentuates a general absence of hard budget constraints at the subnational level. In Chapter VI we review some options for redesigning the transfer system, and some institutional arrangements. Finally, in Chapter VII we address the special issues related to forestry. 11 Chapter 1: OBJECTIVES AND POLICY FRAMEWORK FOR DECENTRALIZATION IN CAMEROON A. INTRODUCTION 1.1 Decentralization in Cameroon is focused on improving local service delivery, better accountability, generating more equitable distribution of resources leading to a reduction in regional tensions, facilitating a more efficient use of natural resources, and more effective provision for the most vulnerable and marginalized groups. Other objectives include national unity, as well as equality and fostering of growth and development.6 A diagnostic assessment covers the design and implementation of policies and institutions to meet these objectives.7 The report draws on the standard arguments for decentralization as well as the constraints and incentives for local officials to provide public services efficiently. 1.2 The standard objectives for decentralized operations include better representation of local preferences and ability to tailor service delivery to local conditions, and if the process is managed correctly, should also facilitate the development of democratic institutions and oversight. These should generate better governance and elimination of corruption. Improved local public service delivery is also supported by multilateral agencies (World Bank, 2003) and bilateral donors. However, the incentives for local officials to meet these objectives may not be effectively in place, and form a main thrust of this report.8 B. REFLECTIONS ON METHOD AND INTERNATIONAL EXPERIENCE 1.3 The standard (or normative) approaches to fiscal decentralization assume that governments at all levels are benevolent, and that devolving functions would enhance efficient service delivery, and result in higher growth and poverty reduction. Typically, there is a separation of functions, and the center retains only responsibility for functions that affect all lower levels of government, such as defense and monetary policy, with lower levels responsible for all other functions especially related to service delivery. The principles of subsidiarity determine which functions should be assigned to each level of government. 1.4 Experiences outside the United States, particularly in the European Union and especially in developing countries, have led to a questioning of the normative approach, spawning a surge in the “political economyâ€? literature. The main difference between the approaches is that the assumption of “benevolentâ€? government is dropped, and incentives facing politicians and bureaucrats become important, as do the role of institutions and 6 See Government of Cameroon, 2009, Growth and Employment Strategy Paper, 2010-2020 (PRSP). 7 This report is the result of a team work led by Raju Jan Singh (Lead Economist), including David Abouem a Tchoyi (Consultant, Yaoundé), Ehtisham Ahmad (Consultant, LSE and University of Bonn), Paulo Cerutti (Consultant, CIFOR), Bernard Dafflon (Consultant, Fribourg University), Bjorn Dahlin van Wees (Consultant, Yaoundé), Faustin Koyassé (Senior Economist), Guillaume Lescuyer (Consultant, CIFOR) and Thierry Madiès (Consultant, Fribourg University). 8 For a review of the recent literature on the case for and preconditions for effective decentralization, and whether or not this leads to enhanced local public service delivery, see Ahmad and Brosio (2006, 2009). 12 information flows. If the process is partial in design or implementation, expected benefits in terms of improved service delivery and poverty reduction may not materialize.9 With inequalities in income distribution, and power, there may be “captureâ€? by vested interests, as well as incentives for “clientelismâ€? that provide a fig-leaf for continued centralized operations.10 1.5 There are dangers that decentralization may not result in greater preference matching, improved public service delivery or growth—unless the preconditions are effectively met. Indeed, the empirical links between decentralization and service delivery are at best tenuous, and require assiduous work on proper design, incentives and correct sequencing. 11 1.6 EU countries ascribe to subsidiarity principles (see Chart 1.1), and functions are assigned to the lowest level at which they can be efficiently implemented. This incorporates both normative as well as political economy considerations. National or supranational levels are primarily concerned with establishing rules and common policies, as well as financing and redistribution if there is a vertical fiscal gap. Typically, decentralization could generate effective local service delivery and accountability if there is adequate inter-jurisdictional “competition.â€? 1.7 Centralizing trends come about through the need for equalization or inter-personal equity, congestion, spillovers, as well as economies of scale. There is thus a tension between centralizing trends, as well as decentralization objectives. The assignments and institutional arrangements affect incentives for local governments to provide public services effectively. 9 See Devarajan, Khemani and Shah (2009). 10 See Bardhan (2006). 11 See Ahmad and Brosio (2009). 13 Chart 1:1: Modified Subsidiarity Principles Centralization Supranational (EU) Decentralization needs x Rules & x Devolution x Cooperation regulations x Participatory x Coordination x Common policies democracy x Control x Harmonization National/federal x Accountability and effective Factors x Policy provision x Financing x Equalization Political Economy x Economies of scale x Yardstick Provincial x Spillovers competition x Congestion x Financing at margin x Accountability Political Economy Local x Equalization x Interpersonal x Financing at margin Equity x Accountability Source: Adapted from Dafflon, in Ahmad and Brosio, 2006, Handbook of Fiscal Federalism. 1.8 While a decentralization process may take several years to implement fully, it is common for some functions to be decentralized, while others might be retained by the center and administered by officials responsible to the center—under a deconcentrated arrangement. Thus, primary education may be decentralized, while higher education remains deconcentrated or under direct provision by the center. The whole thrust of a decentralization program is that local preferences may be better represented, and the local officials are held responsible for the outcomes. With deconcentration, it is harder although not impossible for local preferences to be accommodated, and the spending may reflect the choices of a central planner, and accountability is effectively transferred to the central government. 1.9 Problems arise when there are decentralized and deconcentrated operations for the same function. Overlapping functions, say for primary education, would occur when one level of government is responsible for hiring and financing teachers, and another (usually the lower level) is only responsible for operations and maintenance or building classrooms. This complicates matters for the central government, which may continue to view primary education as a “deconcentrated function,â€? and for local electorates or assemblies that face only “partialâ€? decentralization. In this case, neither level has full accountability; hence service delivery failures are likely. 14 1.10 In order to ensure accountability of local politicians and officials, it is important that there be clarity in terms of the functions or sub-functions that they are to be held responsible for. In addition, to establish accountability it is essential to ensure that local governments face hard budget constraints. This involves four main elements, described below. 1.11 The first is that there should be an own-revenue source over which local governments have control at the margin—particularly in relation to setting of rates. This ensures that local spending priorities can be met out of local resources, helping to align incentives for local officials and their electorates. 1.12 The second element in establishing hard budget constraints is in ensuring that debt is kept to limits that can be financed out of local resources, including assured transfers. This also presupposes that local governments have own-source revenues, without which the threat of no-bailout is not credible. 1.13 The third element is that the intergovernmental transfer system should not automatically cover local deficits—if it did, there would be no incentive to raise own-source revenues. Such a provision would negate the credibility of hard budget constraints. Thus, accountability involves an interaction between spending, own-source revenues, transfer design and debt limits. 1.14 Finally, a key element in establishing hard budget constraints and adequate incentives for local governments is that there should be full information on local government operations that is available in a timely manner to local electorates, as well as to the central government. This puts considerable weight on standardized and timely public financial management operations at the subnational level, and coordination with MINFI. 1.15 The building blocks of both normative and positive traditions are similar—spending and taxation assignments, design of transfers, debt management and information flows and instruments for implementation. However, the sequencing and mix of the instruments might vary, as we discuss below. 1.16 The decentralization strategy would govern how these “building blocksâ€? are put together, and would link decentralization to growth and development targets. To some extent this has been attempted in the recent Cameroonian PRSP, but at a fairly high level. Indeed, the PRSP links decentralization with the generation of productive employment opportunities for the 75 percent of the labour force that is underemployed, and engaged largely in informal activities. Thus, decentralization is to be linked to a regional strategy, and allocation of resources to poorer regions would be offset to some extent by the development of “growth centersâ€?. These centers are to be established in the industrial zones in and around Douala, and Yaoundé, and are meant to provide productive formal sector employment opportunities in order to reduce underemployment to less than 50 percent by 2020, and the poverty rate from current levels of around 40 percent to less than 30 percent during the same period. 15 1.17 Perhaps the greatest danger at present is that the emphasis on the decentralization process has led to heightened expectations on the part of local politicians and the public at large that there will be substantive improvements in their public services and standards of living. However, these are unlikely to be realized without a better and more coherent implementation strategy. 1.18 In this report, we examine the current state of decentralization in Cameroon, as well as prospects for improving service delivery outcomes. We first examine the patterns of service delivery and offer some stylized facts about local governance in Cameroon in order to set the stage for a discussion of decentralization. In what follows, we describe the “building blocksâ€? of the decentralization and governance process. These include: ƒ the overarching legal framework specifying objectives, as well as strategy and implementation rules; ƒ the assignment of spending responsibilities; ƒ budgeting and governance issues; ƒ revenue assignments, including developing effective own-source revenues; ƒ shared revenues as well as transfer design; ƒ tax administration; ƒ management of debt and local liabilities including arrears; ƒ macroeconomic coordination and stabilization functions. 1.19 We illustrate the process from the perspective of the management of the forestry resources of the country. This will touch on various objectives, from ensuring civil order in border regions, to ensuring that producing regions have adequate financing for development and poverty reduction. C. WHAT DO WE SEE IN CAMEROON? THE MACROECONOMIC CONTEXT 1.20 Although Cameroon is an oil-producing country, with significant mineral and forest resources, total resource mobilization is not particularly high—at around 20 percent of GDP. Oil revenues were 7.4 percent of GDP in 2008, but are expected to decline to around 2.5 percent of GDP by 2020. Non-oil revenues, at around 12 percent of GDP, are among the lowest in sub-Saharan Africa and a source of considerable concern (see Table 1.1). This is why the medium-term plan to 2020 (PRSP) 12 puts considerable stress on the ability of local governments to raise additional revenues from as yet unutilized tax handles. Consequently, in aggregate, local governments rely heavily on a 9 percent share of the main revenue bases, the VAT and the income taxes (see chapters to follow) and own-source revenues are negligible. Table 1-1: Cameroon Fiscal Indicators (in percent of GDP) 2008 2010 2020 Total revenues 19.4 15.2 15.5 Oil revenues 7.4 2.8 2.5 12 See Government of Cameroon, 2009, Growth and Employment Strategy Paper, 2010-2020 (PRSP). 16 Non-oil revenues 12.0 12.4 12.9 Tax revenues 11.3 11.0 11.8 Non_tax revenues 0.7 1.4 1.2 Overall balance 2.0 -5.2 Primary non-oil -5.5 -5.9 balance* Sources : Cameroonian authorities; *IMF, 2011 Article IV Consultation 1.21 Unfortunately, data on subnational operations is patchy and incomplete, and not even available on a consistent basis to the Ministry of Finance, or bodies dealing with local government. Indeed, only 16.5 percent of municipalities’ 2009 accounts had been submitted to the Supreme Court’s Audit Bench as of August 2011. 13 While there are reports of irregularities in the finances of municipalities in the East, 14 with relatively less well- developed public financial management institutions and processes, it is worrying that even the most advanced local governments, such as Douala, are subject to irregularities.15 1.22 Consequently, even reports by the international agencies, including the IMF’s Article IV consultation, rely solely on fiscal data for the central government. Cameroon is also one of the few countries in the world not providing GFS data to the IMF. 1.23 Back of the envelope estimates put the size of subnational own-spending out of shared revenues and the limited bases available to the local governments at under 1 percent of GDP. Thus, despite the objectives, the decentralization process in Cameroon is in a nascent stage. 1.24 The impetus provided to the decentralization process in 2010/11 appears to have been largely in deconcentrated mode, with line ministries spinning off administrative “sub- areasâ€? to lower levels, often without the requisite human resources or financial means. Local officials commented that the Ministry of Basic Education does better than others. However, even in this case, the responsibility transferred is that of building additional classrooms in schools—in locations determined by the Ministry. These are generally not the priorities of the local authorities, even with respect to primary education—they may have a need for toilets, qualified teachers or books, and perhaps not additional classrooms. D. OUTCOMES OF THE DECENTRALIZATION PROCESS 1.25 It is hard to link observable service delivery outcomes to decentralization—given the paucity of information available. However, in order to build a picture of relative “wellbeingâ€? in Cameroon, we take a set of commonly used criteria for needs, including: (1) poverty (percent of population living on less than 738CFAF/day); (2) primary education 13 See http://www.cameroon-tribune.cm/national/1-politique/65530-gestion-des-fonds-publics-le-rapport- 2009-de-la-chambre-des-comptes-disponible. 14 Former mayor and treasurer of Yokadouma sentenced to 30 years in prison for embezzlement of public funds (over CFAF 1bn) between 2002 and 2007: http://www.cameroon-tribune.cm/national/7-regions/65913-yokadouma-30-ans-de-prison-pour-lancien- maire-et-son-receveur. 15 http://fr.allafrica.com/stories/201107260816.html. 17 attendance (percent of population aged 6-14 years); (3) Access to drinking water (percent of population); (4) access to electricity (percent of population); (5) literacy (in percent; ages 15 and older); (6) child mortality (deaths per 1000 live births; children under 5 years); (7) underemployment (ILO definition; percent of population in relevant age group); (8) access to clean toilets (percent of population); and (9) births assisted by qualified staff (percent of births). A rank order for each is determined (1= the best score); and the composite rank for various regions is shown in Chart 1.2. The ranks are not weighted and the examples below are for illustrative purposes only. A more detailed analysis using different weights for various factors could be utilized. This could be attempted by the Cameroonian authorities, as they would have a better perspective on what the relative weights should be, at least in accordance with their preference functions. In what follows, we use the simplifying assumption of equal weights. 1.26 As expected, Douala and Yaoundé score best, and the Far North is consistently the lowest. While the index does not imply that Douala and Yaoundé have no needs, in relative terms the Far North, North and East regions fare the worst. 1.27 It is useful also to examine the actual indicators as well (and not just the relative ranks)—these are shown in the Appendix Tables A1-A2. Again, while Douala and Yaoundé appear to have the best outcome indicators, given the additional resources that are provided to these areas, this is not surprising. What is interesting is that despite the additional funding and attempts to create “growth centersâ€?, underemployment in Douala and Yaoundé is still very high (above 70 percent), and these areas have the highest dissatisfaction in relation to the provision of electricity, as well as a strong perception of “corruptionâ€? and rent-seeking. 1.28 It is even more striking that the quality of key services has declined in Douala and Yaoundé. This has lead to worsening performance in terms of access to drinking water, sanitation as well as health care outcomes (see Chart 1.3). Chart 1:2: Relative Living Standards and the Wellbeing Indicator (average rank) 12 10 8 6 4 2 0 Sources: ECAM 3, National MDG Report 2008 and Staff calculations Note: For definitions and components of the wellbeing indicator, see text 18 1.29 Moreover, within Cameroon, there is a wide dispersion in standards across regions, and the situation in some of the more backward regions has worsened in the past decade or so. Particularly striking is the increase in poverty in the North, Far North, and East, as well as declining access to clean toilets, drinking water, as well as increase in child mortality (Chart 1.3). 1.30 It is particularly surprising that despite the emphasis on primary education, there was a decline in attendance in the East. Also, in the richest regions there were declining indicators—especially in Yaoundé for access to drinking water, clean toilets, and increased infant mortality. There was a declining access to drinking water in Douala (as mentioned above). Thus, the disparity in living standards adds to the very striking needs throughout the country. 19 Chart 1:3: Changes in Regional Living Standards (in percent) Poverty, 2001-2007 Access to Improved Sanitation, 2001-2007 40 20 20 0 0 -20 -20 -40 -40 -60 -60 -80 Littoral Adamaoua Center Yaoundé East Douala South West North South West Far North North West Littoral Adamaoua Yaoundé Center East Douala South West North South West Far North North West Sources: National MDG Report 2008 and Staff Sources: ECAM 2&3 and Staff calculations calculations Access to Drinking Water, Births Assisted by Qualified 2001-2007 Staff, 2004-2006 60 30 40 10 20 0 -10 -20 -30 -40 -60 -50 Adamaoua Littoral Center Yaoundé East South Douala North West Adamaoua Littoral Far North South West Yaoundé Center North West East Douala South North West South West Far North North West Sources: National MDG Report 2008 and Staff Sources: ECAM 2&3 and Staff calculations calculations 20 Child Mortality, 1998-2004 Primary Education Attendance, 2001-2007 60 20 40 15 20 10 0 5 -20 0 -40 -5 Douala West North Littoral Adamaoua Yaoundé East Far North Center South North West South West Douala North West Adamaoua Littoral Yaoundé Far North Center South East North West South West Sources: National MDG Report 2008 and Staff calculations Sources: ECAM 2&3 and Staff calculations Literacy, 2001-2007 Access to Electricity, 2001-2007 25 100 20 80 15 60 10 40 5 20 0 0 -5 -20 -10 -40 Littoral Adamaoua Yaoundé Center South East Douala North West Far North South West North West Douala North West Adamaoua Littoral Yaoundé Far North Center South East North West South West Sources: ECAM 2&3 and Staff calculations Sources: ECAM 2&3 and Staff calculations 1.31 It is also interesting that for each of the indicators of wellbeing, Cameroon falls some way below the standards of lower to middle-income countries. Indeed, for indicators such as child mortality and life expectancy at birth, Cameroon does worse than the average for sub-Saharan Africa. So in general terms, much improvement is possible in relation to living standards in an international context (See Appendix Chart A1). 1.32 There is clearly a need for better service delivery at the local level. But that has not happened, despite the attempted decentralization process. The PRSP is remarkably frank in 21 its assessment of the difficulties faced in the decentralization process. The key lacunae identified relate to implementation difficulties: (i) late operationalization of monitoring bodies; (ii) partial implementation of FEICOM, and CEFAM, the training centre for local government officials; (iii) incomplete legal and tax framework; and (iv) incomplete transfer of powers. 1.33 While the PRSP is undoubtedly correct in its evaluation of the difficulties with decentralization, we shall argue that there are also problems with the design of the decentralization process. In particular, there is inadequate attention to accountability and incentives facing various actors in the process. 1.34 Key difficulties derive from divided responsibilities for specific functions that lead to a confusion of responsibilities between deconcentrated agents of the center and decentralized local governments. The difficulties are reinforced by errors in the design of own-source revenues, explicit negation of hard budget constraints and incentives to run arrears and deficits, inappropriately designed transfer mechanisms, as well as the absence of comparable and timely information needed to evaluate the decentralization process and its outcomes. Chapter 2: THE LEGAL AND INSTITUTIONAL FRAMEWORK A. INTRODUCTION 2.1 The 1972 Cameroonian Constitution sets out “normativeâ€? objectives for decentralization. A 1996 revision of the Constitution created a unitary decentralized state,16 and also referred to decentralizing functional responsibilities for education, health, social and cultural activities 17 (see Table 2.1 for a timeline of legal changes). Keeping up with the trends in the EU, in 2004, the legal framework was changed again to introduce the principles of subsidiarity; progressiveness and complementarity. 18 However, these concepts were not defined; nor made operational in the Cameroonian context—and are in fact overridden by concurrent competencies of the center. The legal framework recognized deconcentration as a step towards devolution, 19 and envisaged that there would be a transfer of resources in parallel with competences. 20 However, deconcentration and decentralization for the same functions would severely constrain the exercise of decentralized operations at the local level, particularly in a situation of unclear lines of authority, and variable competence. 2.2 The institutional framework is based on central and deconcentrated authority exercised at the regional level, and elected mayors in rural and urban municipalities. In 16 Law 96/06 of January 18, 1996. 17 Law 96/06 (art. 55, 56); see also Law 1974/023. 18 Law 2004/17 (which only partially repealed 1974/023). 19 Law 2004/17 (art 24). 20 Law 2004/17 (art. 25, 31,and 80). 22 addition, traditional chiefs continue to play an important role, although the precise lines of authority and competence are vague. 2.3 Overall, the legal framework in relation to decentralization is overlapping, cumbersome and contradictory in many respects, and open to different interpretations. The tension between subsidiarity and concurrent competencies is one such example. An explicit bailout clause for local governments destroys hard budget constraints and introduces adverse incentives on local government behavior. Deficit financing removes incentives to manage spending efficiently or utilize own-tax bases, and encourages irresponsible behavior that could lead to economic crises. 2.4 Despite the good intentions in the Constitution and basic legal framework, and the very clear ideas of the Minister of State for Territorial Administration and Decentralization, the mission was unable to ascertain a strategy and an operational plan of action to achieve the desired results. Indeed, most of the line ministries appear to regard decentralization as a threat to their control over resources and influence. A description of the legal framework follows. 23 Table 2-1: Cameroon – Timeline of Legislation Relating to Decentralization Year Number of act Date Title 1972 2 June 1972 Constitution 1974 Law 74/023 8 December 1974 Municipal organization and various subsequent modifications. 1977 Decree 77/091 25 Mars 1977 Determining the powers of supervision over municipalities, municipal associations and unions, and municipal institutions, as well as subsequent modifications. 1987 Law 87/015 15 July 1987 Creation of Urban Metropolitan Areas 1992 Law 92/002 14 August 1992 Conditions for the elections of municipal councillors. 1996 Law 96/06 18 January 1996 Revision of the constitution of 2 June 1972. Decree N°77/85 7 May 1996 Modifying certain measures of decree N°77/85 of 22 March 1977 establishing the functioning and management of FEICOM, and its subsequent modifications. 2004 Law 2004/017 22 July 2004 Orientation of decentralization. Law 2004/018 22 July 2004 Rules applicable to municipalities. Law 2004/019 22 July 2004 Rules applicable to regions. 2006 Decree 2006/18 31 May 2006 Reorganization of FEICOM and modifying decree 2000/365 of 11 December 2000 pertaining to the same subject. Law 2006/04 14 July Rules for election of regional councillors. Law 2006/05 14 July Rules for election of senators. 2007 Decree 2007- 3 September 2007 The process of issuing, collecting, centralizing, distributing 1139-PM and transferring additional council taxes. Code 16 November 2007 Code of intervention of FEICOM 2008 Decree 2008/013 17 January 2008 Organization and functioning of the National Decentralization Council. Decree 24 April 2008 Specifying certain procedures pertaining to the organization 2008/0752/PM and functioning of deliberative and executive bodies of municipalities, urban metropolitan areas, and municipal associations. Decree 2008/376 12 November 2008 Administrative organization of the Republic of Cameroon. Decree 2008/377 12 November 2008 Duties of heads of administrative divisions and the organization and functioning of their services. 2009 Law 2009/019 15 December 2009 On the local tax system. Law 2009/011 10 July 2009 Financial regime of decentralized territorial entities. Decree 2009/248 5 August 2009 Assessment and distribution procedures for the allocation of central funds to decentralized entities. 2010 Various decrees 26 February Setting the terms for the exercise of certain powers 2010/239 to transferred by the state: drinking water, rural roads, women 247/PM and family, agricultural products and rural development, social aid to the poor, fish and pastoral production, culture, public health, basis education. Decree 23 February 2010 Concerning the general funds for decentralization in the 2010 2010/0165/PM budget. Decree 1 June 2010 Establishing the sectoral accounting plan for decentralized 2010/1734/PM entities. Decree 1 June 2010 Fixing the budget nomenclature of decentralized entities. 2010/1735/PM 24 B. ANTECEDENTS 2.5 The Cameroonian legislation concerning decentralization is characterized by a wide variety of overlapping legislative texts of different levels, ranging from the 1972 Constitution, its amendment in 1996, and the further legal changes in 2004. In addition, there are various Presidential decrees and ministerial directives. The overlapping arises since it is difficult to know if successive texts covering the same topic replace earlier ones or complement them. The legal texts that mention in fine that “the current law or (decree) revokes all opposing previous provisionsâ€? gives rise to ambiguity as to what exactly has been repealed. It would have been clearer to explicitly mention the items that no longer apply. Some articles only partially repeal provisions of previous laws, leaving others intact. For instance, article 88 of Law 2004/017 on the scope of decentralization states that the previous Laws “are repealed and replace corresponding provisions of Law 1974/023 relating to local municipalities.â€? Hence, Law 1974/023 stands, but certain articles are repealed. C. INSTITUTIONAL FRAMEWORK 2.6 The co-existence within the same geographical area of both a “deconcentratedâ€? region run by a governor who reports to the minister in charge of local entities (MINATD), and a “decentralizedâ€? entity (see Chart 2.1), which may be a partly elected body, generates ambiguity between the competences (functions) that are deconcentrated and delegated and those which are effectively decentralized or devolved to regions or local governments. As emphasized in the previous chapter, the difficulty is that the same function may be decentralized and yet subject to deconcentrated control. Functional responsibility should be strictly and explicitly defined. 2.7 Article 26, Law 2004/017, explicitly specifies that deconcentration may be a first step before devolution. The same article provides that “material and human resourcesâ€? may be put at the disposal of decentralized regions and municipalities by deconcentrated authorities in order to exert their “new competences.â€? In this case the deconcentrated authorities receive “a share of the resources listed in article 25,â€? i.e., those resources that are to be passed on to decentralized local governments (LGs) in parallel with the transfer of competences. 21 2.8 Chart 2.2 illustrates the design of regional councils. Chart 2.3 describes the urban metropolitan areas operated under deconcentrated authority. 21 See also Law 2004/017, articles 31 and 80, and the discussion below as well as Chart 2.1. 25 Chart 2:1: Institutional Framework: Decentralization versus Deconcentration Decentralization Deconcentration = urban and rural municipalities = administrative divisions MINATD 10 Regions Governor 10 Regions (Provinces: art. 61Cst) 374 territorial entities of which: x 14 urban metropolitan areas Prefect 58 Departments x 45 urban municipalities x 315 rural municipalities Source: Law 96/06 of January 18 1996 revising the Constitution of June 2, 1972: the Cst Sub-Prefect Sub-divisions sets the number of regions. Municipalities are recognized by the Cst, but are determined (Head of district) by law (Cst, 1, art. 55). Law 2004/017, art. 3, of July 22, 2004, on the orientation of decentralization. Cst, Art. 61. The President of the Republic can: Cst, art. 58: a delegate is appointed by the a- rename a local authority or change its geographic boundaries; President of the Republic (the Governor Decree b- create a new local authority. In this case, he would provide it 2008/377, art. 12) to represent the State in the with a name and geographic boundaries. Traditional regions and ensures administrative control. Also art. 6 Law 2004/017 and art. 2 Law 2004/019 chiefdoms The Prefect ensures administrative control over the municipalities and municipal parastatals Sources: decree 2008/377 of 12 (Decree 2008/377, art. 41) November 2008, art. 32 and 47 (Also Law 2004/017, art. 66 and 67) 26 Chart 2:2: Composition of the Regional Council Parliamentarians The regional council from the Region must be presided over Regional council have an advisory by an “indigenousâ€? = capacity in the personality Delegates of Representatives of traditional Regional Council (Cst, art. 57) departments + rulers (Cst, art. 58; Law 1996/06) Municipal councillors Traditional rulers = electoral college of departments st nd rd 1 , 2 , 3 degree (law 2006/04, art. 6, 13 and 17) = electoral college (regional or departmental?) 27 Chart 2:3: Organization of Urban Metropolitan Areas Executive: Delegate and Urban Metropolis deputy delegate nominated by the CA CA CA = Municipal President. Council CA CA Legislature of the urban metropolis CA CA CA CA CA Mayor + 5 councillors 2.9 The institutional framework also explicitly refers to the role of the traditional rulers, which introduces a complex triangular relationship between the different stakeholders at the regional level. This affects the legitimacy of the locally elected officials, who are stranded between the legitimacy of the traditional rulers and the deconcentrated authority of the central government. 2.10 Many of the provisions of the different decentralization laws relate to the explicit and implicit supervisory power of the State over regional and local authorities. In particular, a delegate appointed by the President of the Republic “exercises the supervisory authority of the State over the Region.â€? 22 Thus, the governor, appointed by the center, could exercise powers over the elected President of the Regional council. 2.11 In addition, the President of the Republic exercises considerable powers over electoral boundaries, giving him influence over the elected members. He has unrestrained powers to create, or redraw by decree, the geographical boundaries of the Regions (Constitution, art. 61). This gives him influence over members of the Senate, elected by the Regions. The President can also create new municipalities, merge or split them.23 While the President must seek the views of a commission, which includes the representatives of the elected municipalities, he is not obliged to follow their advice. The Senate 2.12 The Senate in principle represents the decentralized entities,24 which include the regions and the municipalities. 25 Thereafter the Constitution prescribes that each region is 22 Law1996/06, #58. 23 Law 2004/018, art. 2, 4 and 6 respectively. 24 Constitution, art. 20(1). 25 Constitution, art. 55(1). 28 represented in the Senate “by 10 senators, 7 of these are elected by indirect universal suffrage on a regional basis and 3 appointed by the President of the Republic.â€? 26 2.13 There is a lack of clarity about representation in the Constitution. According to art. 20(1) the Senate represents the regions and the municipalities (the territorial entities defined in art. 55(1); but art. 20(2) specifies that only the regions are represented in the Senate, not local governments. Indeed, only the regions constitute an electorate (Law 2006/005, art. 3 and 6). Besides, senators are elected according to a list system (the lists are compiled by the political parties) without the possibility of preferential vote and without the possibility of combining candidates from different lists.27 Since the lists are compiled by political parties, it may well be that the candidates reflect party preferences and interests and not specifically those of the region or the local governments. Local Governments: Regions and Municipalities 2.14 In the colonial period, municipalities enjoyed considerable autonomy. The 1972 Constitution only dealt with the regions (art. 55-62). Municipalities operated as deconcentrated organs, and Law 1996/06 clarified that decentralized local governments comprise regions and municipalities. However, Law 2004/018 began to further delineate the responsibilities and organization of the municipalities. The Electorates 2.15 There are many cases where neither executive nor legislative bodies are elected through direct universal suffrage. The head of the executive body of the “urban metropolitan areaâ€? is a delegate of the central government appointed by the President of the Republic (Law 2004/018, art. 115)—this is particularly important for Yaoundé and Douala. Thus, 14 of Cameroon’s largest and economically and politically most important cities – covering at least one-third of Cameroon’s population – are under the authority of an appointed delegate who wields varying but significant influence over the municipalities, considerably watering down the decentralization process. Moreover, regional councillors are elected through indirect suffrage by municipal councillors and representatives of the traditional rulers. 28 However, mayors of the municipalities within an urban metropolitan area (the “communes d’arrondissementâ€?, CAs hereafter) are ex officio members of the metropolitan council. Besides the mayor, the municipal council of each CA designates five councillors for the metropolitan council (see Chart 2.3). 29 The Regional Council 2.16 The regional council is established through indirect suffrage (Law 2006/04), leading to questions concerning the autonomy of its members. These include delegates of the “departments,â€? which are administrative deconcentrated bodies. However, because municipal councillors form the electoral college of the “departmentâ€?, the link with local governments is not broken. The regional council includes representatives of traditional rulers elected by their peers. 26 Constitution, art. 20(2); also Law 1996/06, art. 20, and Law 2006/05. 27 Law 2006/05, art. 7. 28 Law 2006/04 art. 3 and 17, and Law 2004/019, art. 26. 29 Law 2004/018, art. 121. 29 2.17 Members of the State Parliament (MPs) sit “in an advisory capacityâ€? in the regional council of their regions. They can voice their opinions, but their votes do not count.30 This proviso may not create problems if it enhances coordination between the regional and national levels. However, if the MPs voice only the opinion of the majority party, and influence others, this could constitute interference by the center in the workings of the regional assembly. Urban Metropolitan Areas 2.18 The creation of “urban metropolitan areasâ€? (see Chart 2.3 above) depends on the President of the Republic. 31 The law does not require a participatory dialogue with the various stakeholders. 2.19 The delegate, or executive officer, is appointed by the center, as described above, and he sits above the elected municipal mayors. However, metropolitan council is composed of both the mayor and five councillors of each municipality (art. 121). The legal framework engenders tensions within metropolitan areas: the delegate reports directly to the central supervisory authority 32 but the municipal councils, made up of elected members, must be consulted. D. NEXT STEPS 2.20 A first step would be to “clean upâ€? the legal framework to remove obvious contradictions and ambiguities. Especially important are provisions that erode accountability in the decentralized operations—including overlapping responsibilities for the same function, as well as provisions that eliminate hard budget constraints on local governments. 2.21 Equally important would be the promulgation of implementing regulations, in order to make them more comprehensible for local officials. This can only be done if there is greater clarity on the extent and speed of decentralization envisaged. 30 Law1996/06, art. 57. 31 Law 2004/018, art. 109 and 126. 32 Law 2004/018, art. 116, 117 and 123. 30 Chapter 3: ASSIGNMENT OF RESPONSIBILITIES: WHO DOES WHAT? A. INTRODUCTION 3.1 Subnational responsibilities in Cameroon are unclear. The legal framework reflects the lack of clarity as to what decentralization involves and how to implement it. The recent attempts to make the process operational also reflect a top-down approach that is more consistent with a deconcentrated structure, and does little to enhance effective decentralization and responsibility for specified functions. The 2010 directive to strengthen decentralization encourages ministries dealing with education, healthcare, agriculture, culture, and women’s issues to identify programs to be assigned to local governments, with the transfer of competencies and resources. However, most of the central departments “allocatedâ€? relatively minor programs, whether or not these were local priorities. For instance, the Ministry of Basic Education decided to “devolveâ€? the building of additional classrooms to municipalities, at the direction of the Ministry. 3.2 This devolution had not pleased any of the municipalities that the mission was able to interview—as local priorities were not accurately gauged, or elicited. Many mayors protested that they did not need extra classrooms in existing schools. In some cases there was a need for toilet facilities and teaching materials, and in others the local priority was for paved streets and water and sanitation. 3.3 In the mission’s field visits to municipalities across Cameroon, it was hard to pin down exactly what functions local governments are currently held responsible for. Most official bodies were carrying out projects on behalf of higher levels. Moreover, mayors felt that the transfer of competences was at the discretion of the line ministries and that requisite financing was generally not made available. In the absence of adequate own- source revenues, the transferred functions become unfunded mandates, and it is hard to hold mayors accountable. And the fact that mayors do not receive a salary generates undesirable disincentives. B. THE UNDERLYING RESPONSIBILITY FRAMEWORK 3.4 The Constitution states in a very general manner that the competences devolved to local governments shall be to promote the economic development, and social, health, educational, cultural and sports capacities. This was reiterated in the 1996 revision as well as the 2004 Law, 33 which defines the assignment of responsibilities between the State and local governments to meet the criteria of (1) subsidiarity, (2) progressiveness, and (3) complementarity, ostensibly following developments in the EU. 34 However, these criteria are not defined, which paves the way to various interpretations as to what they might mean in the Cameroonian context. It is also worth noting that Law 2004/017, art. 20, refers to “devolved powers,â€? hence the reality of the centralized unitary state ceding responsibilities is recognized. 33 Constitution, art. 55 and 56, Law 1996/06, and reiterated in Law 2004/017 art. 4. 34 Law 2004/017, art. 9. 31 3.5 Since subsidiarity and complementarity are not defined in the legal framework, they are not operational. There is no indication as to how these criteria might be applied in the Cameroonian context. In fact, both principles are inoperable since the law ensures that competences of the different layers of governments will remain concurrent (see Tables A3- 5) and not exclusive. 35 Indeed Law 2004/017, art. 15, clearly specifies that the State, regions and municipalities do not have exclusive competences but rather carry out the same competences concurrently. Thus, the principle of subsidiarity does not apply. Laws 2004/018 relating to municipalities, and 2004/019 relating to decentralized regions, both promulgated on July 22, 2004, enumerate the respective competences of both regions and municipalities. However, several competences overlap each other (see Appendix, Tables A3-5), which raise a genuine problem of vertical coordination between the two levels of governments. 3.6 The words used to describe the functions that are deconcentrated to the administrative regions and those functions that are devolved to the decentralized regions are not precise and very often intersect with each other. The same happens between the administrative “departmentsâ€? and decentralized municipalities. Finally, we find the same functions assigned to elected regions and municipalities. There is thus a lack of clarity between deconcentration and devolution; between the deconcentrated region and the decentralized region, and also between departments and municipalities. 3.7 Similarly, there is lack of clarity in the responsibilities to be assigned to elected regions and municipalities. The vertical coordination between regions and municipalities is also missing in the legislation: x Sometimes the designation of jurisdiction uses exactly the same words for the regions and municipalities: therefore it is not clear who is responsible for what function. x Often the same wording is used to describe the competences of both tiers, but with an additional qualification of “regionalâ€? or “localâ€? interest. It is not clear who defines, and according to which criteria, the frontier between regional and local (not an easy matter, for example, for ground water). x Occasionally, the competence is classified as regional with the “participationâ€? of the municipalities. It is not clear whether participation means consultation, co-decision, co-financing, or a combination of these. x Finally, some competences are devolved to local governments, but remain under the constraint of “conformity to a national planâ€? or “the authorization of the supervising institution.â€? It is not clear how far the supervision extends: only as far as the issues of policies or national standards, or also to aspects of its material content. 3.8 "Progressiveness" has several possible meanings and dimensions that relate to: x The contents of the devolution, by establishing successive sets of tasks; x The time path, according to a calendar for each set of tasks; 35 Law 2004/007, art. 15. 32 x The way to proceed: This could be big bang, or gradual. 3.9 The “big bangâ€? approach would devolve all the functions identified in the Constitution to the appropriate level, given principles of subsidiarity, together with adequate resources. These would typically involve some shared revenues, and a critical element of own-revenues, at the margin, to ensure accountability. Another element in the “packageâ€? should be an equalization element to permit local governments to provide similar levels of service at roughly similar levels of own-tax effort. 3.10 A gradual strategy could follow two typical routes: x The first approach would be to deconcentrate some competences to the administrative regions, and then transfer (devolve) these competences and the correspondent means (financial, material and human) to the decentralized entities. The difficulty is that the deconcentration postpones the “accountabilityâ€? for the service, without necessarily addressing capacity constraints. After all, if resources are provided adequately, skills can be hired or obtained. x An alternate approach would be through a process of certification, under which “certifiedâ€? local governments could apply for additional functions, together with resources, if there were adequate progress on capacity and institutional underpinnings—including budgeting, accounting and reporting infrastructure. 3.11 Decentralizing a function or sub-function (such as primary education) and yet keeping it under deconcentrated control, as seems to have happened in Cameroon, is likely to be quite problematic—leading to split responsibilities and little or no accountability. These problems would be exacerbated by weaknesses in the public financial management system, leaving public funds open to “captureâ€?, “rent-seekingâ€?, or clientelism. C. HOW TO OPERATIONALIZE RESPONSIBILITIES? 3.12 The enumeration of the competences to be transferred (listed in articles 15 to 22, Law 2004/018, and articles 18 to 24, Law 2004/019) is too broad and useless from an operational point of view. The devolution of competences and functions cannot be made simply by listing the competences or functions, but should be on the basis of the intrinsic components of each function or sub-function to be transferred (this relates to the economic classification—wages, operations and maintenance, capital spending and the like—see Table 3.1). 3.13 Without clarity in the allocation of functional responsibilities, decentralization becomes meaningless. Consequently, the devolution is reduced to local governments performing this or that program largely on the instructions (and often without the requisite financing) from the State (center). 3.14 Currently, local governments neither have responsibilities at the functional or sub- functional level, nor do they have adequate control over the economic categories of the budget—wages and salaries, hiring and firing decisions—related to the functions that have been transferred. In addition, the lack of a functional classification in the budget structure leaves the accounts of the municipalities opaque (see the discussion below on the budget)—and it is hard to piece together what has been spent, say on primary education or 33 other public functions. Together, these severely constrain local accountability—it would be hard to hold mayors accountable for the outcomes of primary education, say, if they are not fully responsible for the function, and do not control the key economic inputs— particularly, hiring and firing decisions and paying for the teachers. 3.15 It is useful to examine a typical budget structure—with functions or sub-functions as activities (see the rows in Table 3.1). One could then evaluate the economic components of each row—with the various components (e.g., wages, or operations and maintenance or capital expenses) along the columns. These functions could also be expressed together with the respective administrative arrangements—central ministries, regional departments, or municipalities. These administrative responsibilities could also be expressed along the columns. 3.16 In Cameroon, responsibilities of lower levels of government appear in designated cells, but not along the rows (functions) or the columns (economic or administrative categories). With this pattern of responsibility, it is hard for an electorate to hold a mayor responsible for the provision of basic health care, primary education or even clean water and sanitation—functions referred to in the constitution in Cameroon and very much in the eye of the political debate. 3.17 The PRSP envisages the use of performance-based contracts between local governments and the central administration, in order to ensure that the assigned public services are being delivered effectively. This is reflective of some of the most modern perspectives in intergovernmental finance, driven by the experiences in the EU (see Spahn, 2006). Table 3-1: Subnational Spending Assignments, Cameroon Functions Wages Other Capital ……. Ministries/Local Government Current agencies Education a. Primary cccc Ccccc cccc M Ccccc ccccc M b. Secondary cccc Ccccc cccc Ccccc Ccccc Heath care a. Clinics cccc Ccccc cccc Cccc Ccccc Water cccc Ccccc cccc M Cccc ccccc M Sanitation cccc Ccccc cccc M Cccc ccccc M Note: “câ€? represents a central assignment, and “Mâ€? is municipal. 3.18 A contract-based approach is an advanced form of decentralized operation. Ideally, it would reflect a contract or promise for improvements in “outcomesâ€?—such as reductions in morbidity or improvements in literacy. This typically would require a shift to “performance–based budgeting,â€? something that Cameroon is far from achieving. 3.19 But can performance-based contracts be used for transfers before performance- budgeting is instituted? To some extent this depends on how well the outcomes are 34 defined, and whether sanctions could be applied if the outcomes are not met. Some countries, like China, have begun to use these types of performance-based transfers, as has the EU, without requiring performance budgeting at the local level. 3.20 Contracts for intermediate “outputsâ€? are also possible, but these could distort the spending process, without improving outcomes. For instance, there could be contracts for building additional classrooms, but these may or may not improve literacy. Making such “contractsâ€? functional and effective will also require some preconditions, which may not be fulfilled in Cameroon at the present time, as described below. 3.21 First, there may need to be flexibility in the use of “economic inputs,â€? so that the local government or contract party is able to most efficiently produce whatever is contracted. Thus, for a given function (row in the example in Table 3.1) there would be flexibility in the use of the columns as well as responsibility for the function or sub- function. 3.22 Second, there would need to be full information on both inputs and outputs contracted. Again this is a big lacuna in Cameroon at present, given the poor state of information on local public finances at the present time, as mentioned in Chapter 1 and discussed further in Chapter 4. Without this degree of flexibility, it is hard to see a performance contract becoming actionable. D. SPENDING ASSIGNMENTS—NEXT STEPS 3.23 In the short to medium-term, it would be useful to consider which function or sub- function or sub-sub-function might be assigned to lower levels of government. While primary education is a typical sub-function that is normally assigned to lower levels, and central governments continue to set curricula and standards, there may be some game- play in the payment of teachers and effective responsibility for the education function. Thus, in the Cameroonian case, with a gradual devolution, one could think of beginning the process with simpler functions like water-supply and sanitation. 3.24 In some countries, such as Peru, a gradual decentralization process has involved a level of certification of competences and ability of the local government to carry out the functions.36 However, the argument that local governments lack the capacity to carry out certain functions cannot be made too strongly, as the transfer of human resources and financing can easily address any shortcomings that might be apparent under the present circumstances. A faster approach to decentralization, as in Indonesia, would ensure that commensurate human and financial resources are transferred at the same time as the functions.37The case of China is particularly interesting, as it remains a unitary state, but one that has decentralized very substantially (see Table 3.2). Effective coordination by the center is predicated on a standardized budget classification based on the IMF’s GFS2001 at all levels of government, together with a system of “nestedâ€? Treasury Single Accounts at each level of government to track the spending, and ensure effective management of public funds. The center also employs performance-based transfers to meet its own 36 See Ehtisham Ahmad and Mercedes García-Escribano (2011). 37 See Ehtisham Ahmad and Ali Mansoor (2002). 35 objectives, in addition to an advanced equalization framework that permits local governments to provide similar levels of service at similar levels of effort. 38 Table 3-2: China: Local Government Spending (in percent of total) Local 2009 Local 2000 Total 80% 69% Transportation 77% 66% Education 95% 89% Health Care 98% 90% Social 94% 99% Source: Ehtisham Ahmad (2011) 3.25 The peeling of administrative arrangements away from the center at the discretion of line ministries will not result in effective decentralization. It will neither transfer the competences that are required by local populations, nor will the local electorates hold the mayors or local officials responsible for the effective implementation. 3.26 Yet, expectations have been raised, and there is a danger that these will not be met. The Cameroonian government needs to specify the extent and strategy for decentralization, given the objectives in the Constitution, and establish a road map for its effective implementation. This would help to moderate expectations from the decentralization process and lead to more sustainable improvements in local public service delivery. 3.27 Clarity in the functions devolved, has to be accompanied by better and timelier information on what was spent and the outcomes of the spending. This information is not only needed for macroeconomic coordination by the central government, but is also a critical element in the political economy story of making the local governments accountable to their electorates. This is a critical element in the operations of “yardstick competition,â€? as the performance of local governments are benchmarked by the electorates in relation to what is seen in neighboring jurisdictions, or even neighboring countries, given the enhanced flow of information with the widespread use of information technology as well as international migration. 3.28 Even as more information is needed on “decentralizedâ€? operations, there should also be better accounting for “deconcentratedâ€? operations on behalf of the central government. The same standards as for “decentralizedâ€? operations should apply, with full information on what was spent and the results of the spending by the prefects and the sub- prefects. They too should be subject to “benchmarkingâ€? and if they are unable to perform with a minimum level of efficiency, there would be a strong case for “devolvingâ€? the functions to others who might have better incentives to operate efficiently. 38 See Ehtisham Ahmad (2011), “Should China revisit the 1994 fiscal reforms?â€? Working Paper, ZEF University of Bonn; and Asia Research Center, London School of Economics, also Chinese Journal of Reform, 2011 (in Chinese). 36 Chapter 4: BUDGETING, SOFT BUDGET CONSTRAINTS AND ACCOUNTABILITY A. INTRODUCTION 4.1 Poorly designed rules and regulations lead to incentives for soft budget constraints and lack of accountability at the subnational level in Cameroon. In this chapter we examine some of the budgeting, treasury and cash management functions and governance issues associated with operations at the subnational level. 4.2 We argued in Chapter 2 that legal underpinnings for local budget autonomy embody structural weaknesses, particularly the bailout clause, that leads to soft budget constraints. The deficiencies in the budget classification make it hard to specify precisely the spending autonomy for functions or sub-functions of spending, or hold local officials responsible for delivery or outcomes against those functions. The absence of standardized and timely information on budget allocations, as well as actual spending, leads to an open invitation to engage in rent-seeking—something that seems to have been exploited effectively by numerous local governments. While there are undoubtedly dedicated and honest officials, the perception of widespread corruption (together with the fact that mayors do not receive a salary) is a major problem for the decentralization process. 4.3 Similarly, weaknesses in the coordination of budgets at various levels of government make it difficult to effectively plan for rational spending. Ad hoc cuts in assured transfers, without consultation or notification, effectively force the stabilization function down to local levels, something that they are not equipped to handle. Thus, ill-defined revenue-sharing mechanisms, with poor coordination, shift a significant portion of the fluctuations in revenues (e.g., with respect to the RFA) on to local governments, usually without warning. As these governments are typically less able to borrow to smooth consumption, their assigned spending is subject to a risk of underfunding. There is thus an incentive to “play games with the center,â€? and withhold payments for centrally provided inputs (including utilities). 4.4 As we will see in Chapter 5, the RFA could be considered to be central or state revenue and it would effectively finance central or deconcentrated operations. In this case, the responsibility is effectively retained by the center, and the ad hoc adjustment should not affect devolved functions, or be a problem for the municipalities. However, it would have an impact on economic activity levels in the regions concerned, and affect them indirectly—so there is still a case for better information flows from the center to the local governments as well as vice versa. 4.5 We examine in detail the issue of the establishment of the Treasury Single Account (TSA), and what this implies for local operations. As with the tax reforms, it is rather more important for local governments to be able to access their budgeted resources efficiently and without delays than having to operate and manage their own treasury single accounts. This has become a divisive issue, as we see below, and the central government needs to put 37 in place mechanisms that ensure that local government operations are not hampered or delayed, or the procedures have not become too onerous. B. BUDGETING 4.6 There is a legal requirement that financial resources transferred to local governments must match the decentralized competences, and this corresponds to international best practice. However, there is no precision about the nature of the transferred resources, and it is not specified whether these should be taxes or transfers.39 There are also some legal qualifications that confuse the picture: x Law 2004/017, art. 24, states that the financial burden of the decentralized competencies must be evaluated before their transfer. Since evaluating this burden is quite subjective, the agency (line ministries) actually in charge of the evaluation would play an important role. x Law 2004/017, art. 23, introduces a general transfer, which corresponds to the partial financing of decentralization. However, art. 7 of decree 2009/248 states that it should be an “appropriateâ€? transfer. Neither of the formulations has to correspond precisely to the transferred functions, and could leave the local governments to make up for the difference out of their own (non-existent) resources. x Law 2004/017, art. 23, also stipulates that the general transfer should be decided annually with the central budget – which means that it will likely be adjusted to the financial circumstances of the central government, as has been seen recently (see below). Soft budget constraints 4.7 A “softâ€? budget constraint is introduced through a bailout provision in the law. Article 24, Law 2004/017, allows for a “special financial grant-in-aidâ€? whenever insufficient financial resources compromise local governments’ role in the delivery of local public services. This effectively removes any incentive that local governments may have to raise revenues or manage budgets efficiently. 4.8 The capital (investment) budget is always in balance since borrowing is considered a “normalâ€? revenue entry in the accounting system (see below). This is contrary to accepted international practice (e.g., the conventions of the IMF’s Government Financial Statistics Manual 2001), and removes the focus on the appropriate “bottom lineâ€? for the respective governments. 4.9 It is clear that there is a need to ensure that the individual components of decentralization are properly designed, including own-source revenues (as well as spending assignments) and that the measures are appropriately sequenced. However, interactions between policy instruments are critical and will determine whether the decentralization will 39 Law 2004/018, art. 144; Law 2004/019, art. 85; and Law 2004/017, art. 7 and 25. 38 be successful. Even if appropriate own-source revenues were to be assigned, the incentives to use the tools could be vitiated by the bailout clause. 4.10 To the extent that local governments provide basic functions, and do not have access to significant own-source revenues, or access to credit, it is appropriate that the stabilization function should remain with the central government—this is one of the tenets of normative federalism theory. Thus, local governments should be assigned stable revenue sources and transfers that do not fluctuate significantly over time. 4.11 In budgeting terms, the center should provide reasonable multi-year forecasts of the transfers that it proposes to include in its budget framework, perhaps with a floor that provides assurance and certainty for the local governments to carry out their budgets within a reasonable framework. This is critical to engendering accountability at the subnational level. Budget classification 4.12 The budget classification (see Tables 4.1 and 4.2) does not contain a functional classification. This makes it difficult to track the devolution of functions or how well the local governments are performing. As mentioned above, loans are included as part of revenues, and thus budgets would be balanced automatically. However, weaknesses in budget controls may allow a local government to run overdrafts, as well as arrears in terms of commitments made but with outstanding payments. 4.13 The operations of a Treasury Single Account (TSA) will limit overdraft operations— and municipalities will no longer be able to incur cash deficits or spend more than the balance of their funds in their accounts, unless there is a special provision by the central government. While the TSA allows for a tighter management of cash and deficits (on a cash basis), it can be overridden by the budgetary requirement to provide support to local governments that are unable to make key payments. This provision dilutes the potential controls that are now available to all levels of government, and that would be an important component in combating rent-seeking behavior and corruption. Further details on the TSA follow below. Budget cycles and coordination 4.14 The budgeting process does not encourage discipline at the local level. To begin with, central transfers (budgeted in mid-November) are not known at the time that the local budgets are formulated (early November). In addition, the actual allocations are not known until February or March of the year in which spending is incurred. This leads to something akin to wishful thinking for the local governments, and the budgeted numbers bear little resemblance to reality. 39 Table 4-1: Content of Local governments’ Budgets Expenditure Revenues Recurrent Art. 25 Operational costs Art. 11-13 Fiscal revenues 27-28 - obligatory 14 Proceeds from exploitation of 29 - optional municipal property and services 15-19 Endowments and grants 20 Other operating revenues Investment 26 Investment expenditure 21 Investment revenues 22 Loans Source : Law 2009/011 of 10 July 2009 on the financial regime of decentralized authorities Table 4-2: Budget Classification for Decentralized Governments Expenditure Revenues Recurrent 1.1 Reserves 6.1 Goods and services used 7.1 Fiscal revenues, CAC and municipal taxes 7.2 Proceeds from exploitation of municipal 6.2 Staff costs property services 7.3 Rebates and royalties granted by the state 6.3 Taxes and duties 7.4 Financial revenues 7.5 Subsidies received 6.4 Financial costs 7.6 Transfers received 6.5 Subsidies granted 7.7 Miscellaneous revenues and sundry profits 6.6 Transfers granted 7.8 Depreciation returns 6.7 Other expenses and sundry losses 6.8 Depreciations Investment 1.5 Repayment of long- and medium term 1.0 Endowments received debts 1.1 Reserve funds 1.6 Repayment of other long and medium 1.4 Equipment and investment grants received term debts 1.5 Long and medium term loans 1.7 Repayment of debts resulting from long- 1.6 Other long and medium term debts and medium term commitments made by 1.7 Repayment of debts resulting from long- and the municipality medium term commitments 2.0 Intangible assets and expenditure payable 2.5 Repayment of loans and other medium and in annual instalments long term credits 2.1 Acquisition of land/plots 2.6 Repayment of equity shares 2.2 Acquisition of other fixed assets 2.8 Depreciation of fixed assets 2.3 Ongoing acquisition of fixed assets 7.9 Production of assets by the decentralized 2.4 Advance payments and down-payments municipality for itself. on ongoing acquisition of fixed assets 8.4 Transfer of fixed assets 2.5 Loans and other long- and medium terms credits 2.6 Holdings and appropriations Sources: Law 2009/011 on the financial regime of decentralized municipalities; Decree 2010/1735/PM of 1 June 2010 establishing the budget nomenclature of decentralized municipalities; Decree 2010/1734/PM of 1 June 2010 establishing the sectoral accounting plan for decentralized municipalities. 40 Budget implementation 4.15 The mission obtained eleven “comptes administrativesâ€? or administrative accounts on budget implementation from FEICOM (the mission was unable to obtain the full data set available to FEICOM, or MINATD). However, it managed to obtain 56 administrative accounts in total, including some directly from municipalities during field visits to Douala and the East and Adamaoua regions. Although the accounts clearly had been audited and were certified, the mission has observed a number of inconsistencies and irregularities in the administrative accounts. Some are systemic, but others are suggestive of problems that need to be addressed through better public financial management processes and procedures, and technical assistance. The key findings from the administrative accounts follow. 4.16 The local budgets are often quite unrealistic. The revenues forecast tend to be optimistic and do not appear to be under the control of the local administrations. Indeed, for a “normalâ€? year for which full figures are available, the gap between budgeted figures and actual spending is enormous (see Chart 4.1). Part of the difficulty is that the budget cycles are not properly synchronized. This is relatively easy to adjust, e.g., with sharing of the likely central budget envelope in advance, perhaps along with the budget circular that is also sent to central spending agencies. Chart 4:1: Local Governments’ Budgets and Execution, 2007 (CFAF millions) 3000 2500 Appropriation Execution 2000 1500 1000 500 0 Sources: FEICOM and Staff calculations 4.17 The largest and most stable element in the budget process is the CAC (see Chart 4.2). 40 The average actual CAC for our sample of 56 municipalities is 99 percent of the budgeted CAC (median=89 percent). Significant variations exist though. Many municipalities have a realization between 60-80 percent while many receive more than they budgeted (i.e., >100 percent). As mentioned in the chapter on taxation, it is interesting that 40 The data on net CAC payments 2004-2010 received from FEICOM does not correspond with the CAC received according to the administrative accounts. 41 Yokadouma received 426 percent of the CAC it budgeted in 2006, but only 15 percent in 2008, perhaps in reaction to the unrealistic budgets in preceding cycles. Chart 4:2: CAC Element in Municipalities’ Budgets (in percent) 100 80 60 40 20 0 1 6 11 16 21 26 31 36 41 46 51 Budgeted share of total revenues Actual share of total revenues Sources: Administrative accounts and Staff calculations 4.18 As far as other revenues are concerned, on average, municipalities budgeted for local tax revenues at 22 percent of recurrent revenues (median=18 percent) and 18 percent of total revenues. The share of local tax revenues ranged from below 1 percent to 88 percent of total revenues. Overall, actual revenues are on average 57 percent of budgeted revenues (median=58 percent). Once again there are large variations with the lowest execution relative to the budgeted figure being 15 percent and the highest 99 percent. 4.19 The spending side is even less realistic in terms of budgeting, and the execution is considerably lower than for the projected revenues. Actual expenditure is on average 50 percent of budgeted expenditure (median=51 percent), and varies across municipalities— from a low of 10 percent to 86 percent (see Chart 4.3). Ebebda municipality (2009) had an actual expenditure execution rate of 17 percent despite a total surplus of CFAF 374mn. 4.20 The average investment budget execution is 38 percent (median=35 percent) with variations between 1 percent and 93 percent of budgeted figures. If one excludes investment spending on ‘equipment and movables’ the average budgeted share is 25 percent of total expenditures (median=20 percent). By excluding spending on ‘equipment and movables’ one gets an idea of the size of spending which is likely to have a more significant developmental impact (e.g., roads, sewage, electricity, irrigation). The actual spending on productive investment is only 14 percent of total expenditure (median 10 percent). This probably explains why the decentralization has so far had such little impact on growth. 4.21 The accounts for Guidguis in the Far North region suggest that 100 percent of its spending was on equipment and movables—but zero spending on pensions and utilities seems implausible, given that it had large surpluses in both years (CFAF 40-55mn). More 42 likely it had a significant buildup of arrears on current spending, including utilities and pensions, to say nothing of wages, and this is probably part of the bargaining process with the central government. 4.22 Given the unclear responsibilities at the local level, and abstracting from the possibility of graft and diversion of resources, it is clear that the focus of local spending is geared towards “consumption.â€? Thus, the highest budget execution rate (81 percent, with median of 90 percent) is for official ceremonies and entertainment (see Chart 4.3), together with the purchase of cars (accounting for much of the capital expenditures). Chart 4:3: Budget Execution (in percent) 160 140 120 100 80 60 40 20 0 1 6 11 16 21 26 31 36 41 46 51 Expenditure Revenues Entertainment and ceremonies Sources: Administrative accounts and Staff calculations Note: Revenues exclude loans and expenditures include debt payments 4.23 The buildup of arrears is suggested by spending patterns in relation to budgets on utilities, pensions and to a lesser extent, wages (see Chart 4.4-4.6). Almost one third (29 percent) of the municipalities had no actual spending on utilities, and 42 of 56 municipalities showed 50 percent or less of the budgeted spending. This is just not credible and is suggestive of the buildup of arrears. It is clear that an elaborate game is being played between “municipalitiesâ€? and higher levels of government. Payments are being withheld in critical areas, even when funds are available. 4.24 In most cases, wages were fully paid—but low or zero payment of wages appears to be strongly correlated with zero or low payments on utilities and pensions as well. Two municipalities (one of them Yaoundé VI in 2007) paid no salaries at all. A few municipalities paid less than 50 percent while others paid more than 100 percent of budgeted spending on salaries. The average execution rate for pension payments is 47 percent. 15 municipalities (27 percent) had no actual spending on pensions. This includes several with substantial surpluses, including Yaoundé VI with a surplus of CFAF 199mn, but zero spending on pensions. This is clearly an example of a buildup of arrears on account of pensions, in the expectation of future “bailoutsâ€? by the center. The bailout provision, thus, also generates an incentive to run arrears. 43 Chart 4:4: Municipalities’ Current Budget Execution, Utilities (in percent) 100 90 80 70 60 50 40 30 20 10 0 1 6 11 16 21 26 31 36 41 46 51 56 Sources: Administrative accounts and Staff calculations Chart 4:5 : Municipalities Current Budget Execution, Pensions (in percent) 150 125 100 75 50 25 0 1 6 11 16 21 26 31 36 41 46 51 56 Sources: Administrative accounts and Staff calculations 44 Chart 4:6 : Municipalities Current Budget Execution, Wages (in percent) 120 100 80 60 40 20 0 1 6 11 16 21 26 31 36 41 46 51 56 Sources: Administrative accounts and Staff calculations 4.25 Mayors do not receive a salary. This is a significant problem in the accountability of local governments. Unless the mayors are independently wealthy, there would be an incentive to divert resources into “benefitsâ€? for the mayor to compensate for a lack of cash payments. 4.26 Borrowing is treated as ‘revenues,’ but is relatively small. However, this can be a major problem in evaluating the effects of subnational operations in the future. As it is, many municipalities with outstanding debt are not making any payments. While 79 percent of the municipalities budgeted to service their debt, 27 percent actually spent nothing on debt repayments. The average execution rate of repayments for all the municipalities, which budgeted debt payments, was 41 percent (See Chart 4.7). There is clearly no hard budget constraint at the subnational level in Cameroon. 4.27 A third of municipalities budgeted in the expectation of a FEICOM grant. However, three quarters of these received nothing. This is very worrying, as FEICOM is responsible for passing on transfers as well as providing loans (see Chart 4.8). Overall, only 20 percent of the budgeted amount was actually received—and only three out of the total municipalities sampled received the amounts budgeted. 45 Chart 4:7: Debt Payments (CFAF millions) 70 60 Budgeted 50 Actual 40 30 20 10 0 1 6 11 16 21 26 31 36 41 Sources: Administrative accounts and Staff calculations Chart 4:8: Execution, FEICOM Grants Component (CFAF millions) 250 Budgeted 200 Actual 150 100 50 0 1 3 5 7 9 11 13 15 Sources: Administrative accounts and Staff calculations Stabilization 4.28 A more significant difficulty is that the central government currently passes on part of the stabilization function to the local governments. This comes about partly through the fluctuations in the share of revenues, as well as the recent ad hoc downward adjustment in the RFA. If the local governments actually carry out basic spending functions (not entirely clear at this stage), then it would not be sensible to force macroeconomic stabilization down to this level, thus jeopardizing basic service delivery at the local level in poor areas 46 4.29 Local governments have no autonomous sources of revenues (over which they control rates or base) and little scope for adjusting spending. The natural response is to run arrears, in particular where it affects the interests of the central government—such as for social insurance contributions as well as utility payments. 41 The Ministry of Finance had assumed local government arrears on these counts before the recent intensification of the decentralization process in 2010, but the evidence suggests that arrears have reappeared.42 It is unclear as to how much of the increase in the stock of unpaid payments orders is due to the operations of local governments. What is clear is that public financial management (PFM) needs to be strengthened across the board, despite some improvements in the treasury system. In particular, the coverage of the operations of the local governments should be part of any strengthening of the PFM arrangements in Cameroon. C. DEBT MANAGEMENT, ARREARS AND IMBALANCES IN THE INTERGOVERNMENTAL SYSTEM 4.30 There is clearly a “political economyâ€? game in play between the central and local governments. The significant buildup of arrears for utilities and social insurance have not been stopped by periodic central government attempts to clear them as often the root causes and incentives to generate them have not been addressed. Arrears to the utility providers generate very damaging circular debt, and social insurance benefits have to be paid whether or not the contributions have been collected. Since 2009, the macroeconomic crisis has led to a curtailment in revenues and transfers, adversely affecting the local governments. The latter retaliated by withholding payments for utilities and social insurance—the recent amnesty an excellent example of the “political powerâ€? at their disposal. Without an effective set of sanctions, or hard budget constraints, the periodic arrears “charadeâ€? is set to continue. 4.31 The repeated buildup of local government debt and arrears is an indication of imbalances in the system of intergovernmental fiscal relations, absence of a hard budget constraint, as well as adverse incentives that are generated by the debt amnesties and periodic central government assumption of local arrears and debt. Moreover, the legal framework, with its bailout clause, encourages irresponsible behavior. 4.32 Anecdotal evidence, based on interviews as well as some of the administrative accounts reviewed suggests that the issue of arrears has reappeared. While there are allocations for utilities and social insurance payments in the budgets of local governments, several local government accounts reported that no payments were made, clearly suggesting that arrears are developing (not that there was no electricity and water usage, nor that all personnel were laid off). As mentioned above, the aggregate stocks of arrears, as estimated by the IMF, have increased in absolute terms during 2010. 41 See for example AES-Sonel (La société nationale d’électricité du Cameroun) and the Ministry of Finance, September 16, 2010, contract pursuant to “Avenant No. 3 à la convention de compensation des dettes entre l’état et AES Sonel du 27/12/2005â€?), offsetting a debt of CFAF 6.4 billion to AES-Sonel by the state and local governments against arrears of CFAF 4.5 billion owed by AES-Sonel, with a payment of the balance to AES- Sonel. The debt vis-à-vis pensions is altogether more significant, CFAF 6.2 billion on account of arrears from the local governments, and CFAF 211.8 billion in total—this was regularized on 17 February 2010 between the Ministry of Finance and CNPS pursuant to “Avenant No. 3 à la convention des dettes réciproques Etat/Caisse Nationale de Prevoyance Sociale du 16/09/2005â€?. 42 See IMF Staff Report for the 2011 Article IV Consultation, Text Table 3. 47 4.33 The current cash-based system does not track stocks of debt, and even the accounts payable are generally not known by those who should be aware of the magnitudes, including treasurers. Even FEICOM is generally not aware of total liabilities. Arrears are clearly building up but are not being systematically tracked. 4.34 The incentive structure is driven by soft budget constraints as shown above—and the amnesty in 2009 leads to expectations that this will be repeated. The absence of own- source revenues is severely problematic. So sanctions are not credible and if they existed they could not be effectively enforced. The bailout clause in the law does not help. Indeed, FEICOM provides 100 percent loans to cover wage arrears (see Table 4.3). Table 4-3: Eligibility and Conditions for Loans from FEICOM Objective Loan in % of Amortization (in Financial costs total gross years) (in percent) expenditure Income generating projects a) Commercial facilities : 60 10 6 markets; bus stations; municipal hostels; slaughterhouses; community halls; agricultural processing units; creation of layouts; municipal forests; building of dwelling units. b) Utility equipment: 100 4 7 service vehicles, small rural facilities, trucks, public works and agricultural trucks. c) Social projects : 30 10 6 hydraulic infrastructure; energy infrastructure; road maintenance and works; bridges and other crossing works; urban development; schools; health centres. Non income generating projects Public facilities : 30 10 6 Buildings housing local services; cultural facilities. Operational loans 100 2 9 Various studies and seminars; municipality’s contribution to projects funded by FEICOM; municipality’s contribution to projects funded by other bodies, office supplies and furniture; payment of various bills. Cash advances 100 2 9 Salary arrears; cooperation or training trips for municipal officials. Source: FEICOM 4.35 FEICOM is assured of CAC resources, and easy rescheduling, so has no incentive to examine the full set of liabilities being incurred by local governments. Consequently, it has incomplete information on the stock of debt, and the net accretions. What information it has largely relates to its own lending and captive revenue sources (CAC allocations that it controls). In the new public financial management regime, when CAC allocations are paid directly by DGT, FEICOM will no longer have first access or control over the funds, and may be forced into taking a broader and more complete picture of local liabilities. 48 4.36 No single institution is able to monitor and assess all subnational liabilities. There is a rapid increase in FEICOM debt. Any lending agency should examine the full extent of local government debt and assess this in relation to ability to repay. D. TREASURY AND CASH MANAGEMENT 4.37 A Treasury Single Account (TSA) is a standard mechanism to control the spending process and prevent leakages and misappropriation. In a unitary state, it is not unusual to have a single TSA for all levels of government. This helps in providing full and timely information to all levels on the revenues received and on the spending process. But all this presupposes that the local governments will not face additional delays in accessing their funds—if they do, the entire TSA system can be discredited. As mentioned, in the Chinese case, separate TSAs have been established for all major levels of government—but this model may not be easy to implement in Cameroon. 4.38 Cameroon has made good progress towards the implementation of the TSA, and this now also includes local government funds. There is some concern, particularly among local officials and activists that this process runs against the spirit of decentralization. However, the membership in the TSA by junior levels of government does not affect their ability to utilize their funds. Indeed, there may be greater assurance that the funds have been collected and are available for use. The existence of a TSA providing greater oversight of the funds could lead to a simplification of processes and procedures, and make operations easier for the local governments. It would also make it harder for funds to be misappropriated at any level of government. 4.39 With multi-level governments, the TSA could include a “correspondentâ€? account for each government. This would permit operations through a separate circuit and approvals, but limited to the level of funds collected in the TSA on behalf of the relevant government. So the sanctity of the TSA would be safeguarded, without restricting access for the owners of the correspondent accounts (this could be on account of donors, like the World Bank or bilateral agencies, or even local governments). 4.40 There are generally two mechanisms to ensure quick payments through a TSA. The first is if there were to be an integrated Government Financial Management Information System (GFMIS) linking spending agencies to their oversight mechanisms, and with electronic payments to recipients of public funds and suppliers. It is important that there be a standard budget classification (e.g., based on the IMF’s GFSM2001 economic classification) as well as the UN COFOG functional classification, plus the administrative structures (outlined in Table 3.1) that form part of the chart of accounts. This permits funds to be tracked from budget allocation, commitments, through receipt of goods and payments, and can also be a mechanism to track arrears in payments. The mission was informed that the EU is initiating a project to install a GFMIS in Cameroon—however this is in its incipient stage and will take another three to four years to materialize. 4.41 In the absence of a properly functioning and integrated GFMIS and the existing bureaucratic processes and procedures, there is a danger that the TSA could increase the time it takes for local governments to access their funds. With cash-strapped local governments, this will just lead to arrears and further weaken the budget process. In the 49 context of Chart 4.9, and the payment circuit for a local government, the TSA with the current procedures adds another loop from the regional office of the treasurer to DGT and the TSA. 4.42 However, as a temporary measure, using the information on the accretion of funds for local governments, the DGT can establish a series of zero-balance accounts for local governments up to the level of the balances in the local government’s correspondent account in the TSA. Local governments can make payments against their ZBAs up to the reimbursable limits in their correspondent accounts. In this way, the principle of the TSA is maintained, and the local government can bypass the additional steps needed to access their funds. In Chart 4.9 the circuit relating to the zero-balance accounts (ZBA) option is shown with dotted lines. Chart 4:9: Spending Circuit with a TSA LG1 DGT Payment Order Regional DGT TSA office CA1 Payee ZBA Note: Narrow arrows refer to information flows; thick arrows to flows of funds. Full arrows refer to current circuit; dotted arrows to ZBA option. 4.43 The local government receives an authorization to issue checks up to a certain level—the payment order (once cleared by the treasurer in the normal manner) can then be submitted to the local bank or branch holding the zero-balance account. The local government (LG1) issues a payment order that is first examined by the regional treasurer. At present the latter passes the approved payment order to DGT (full lines). With a ZBA system in operation (dotted lines), this step can be simplified, and the payment order can go directly to the branch with the ZBA subject to the balances in their correspondent accounts (CA1) (mimicking the situation in which the local government kept funds in the account). The main difference is that the branch makes the payment and gets reimbursed by the TSA 50 overnight. This simplification of the payment circuit would go a long way to alleviating the concerns of the local governments vis–à-vis the TSA reforms. Of course there would be a flow of timely information back from the ZBA and the TSA correspondent account to the local government—this is not shown in Chart 4-9, to avoid complicating the representation of the main flows. Additional details of this model for operating the TSA can be provided at a later stage. 4.44 The government should initiate a reassessment of spending and own-revenue assignments. The Ministry of Finance should monitor all sources of subnational liabilities and risks, including the buildup of arrears, including for social insurance and utilities. The legal framework needs to be modified to remove bailout provisions. E. INSTITUTIONAL ARRANGEMENTS 4.45 There are serious questions about the adequacy of the information and coordination mechanisms that relate to the fiscal aspects of decentralization in Cameroon. These are separate from the institutional and management issues that are dealt with by MINATD. FEICOM 4.46 FEICOM provides several functions that have become redundant with the new streamlining of the Treasury functions and the establishment of a TSA. The redistribution of the CAC can be set up to credit directly the local governments’ correspondent accounts in the TSA, establishing limits for the zero-balance bank accounts in commercial banks. It would be a contradiction of the TSA principle to actually move funds to FEICOM’s account. However, overall coordination and cash planning becomes critical. A new Fiscal Coordination Unit (FCU) that should be established in MINFI would facilitate this coordination. It would be in a better position to utilize the generation of information on a real time basis in the TSA, for both the revenue and spending circuits. 4.47 It may also be appropriate to separate the lending function from the determination of the grants system. Running these two functions through FEICOM, along with the allocation of the CAC, leads to the rather lax attitude to managing debt that has been observed in Cameroon. 4.48 The options for FEICOM are two-fold. One would be to become a development bank operating on market principles. If subsidies have to be provided for specific purposes, these should be run through the budget (with coordination by the FCU) and accounted for by MINFI. Alternatively, the FEICOM could be converted into a Grants Commission. There is a conflict of interest in combining the two different functions into a single agency. Thus, either a Grants Commission (see below) or a Development Bank would need to be constituted afresh. Fiscal Coordination Unit (FCU) 4.49 Timely information is needed for setting the stage for improved governance by the local governments. The absence of complete, timely and accurate information on the fiscal operations of local government, as seen above, is a major problem for overall macroeconomic management. This is because the information needed for such accounts is 51 spread over various agencies that do not appear to cooperate very effectively, including MINFI and FEICOM, as well as the local governments themselves. With the move towards a simplification of revenue declaration, and the consolidated collection of revenues by DGI, the information on the revenue side is better covered by MINFI. Information on forestry revenues generated is also partly captured by DGI and MINFOF, and it is sensible to consolidate the fiscal information in MINFI. 4.50 The spending circuit is now also likely to be better covered with the extension of the TSA. Again, it may be better to centralize the spending information through MINFI, supplemented by reports from local governments on spending out of resources that they generate directly. 4.51 Full information on spending and revenue circuits will be needed for a more effective management of debt and liabilities. This information is also useful for a more comprehensive system of equalization transfers. The relativities would be established by the Grants Commission, but the payments would be made directly through the TSA. 4.52 In many countries there is a FCU in the Ministry of Finance for coordination with local governments charged with the collection and standardization of data, including consolidation of information from different sources. This is a critical component of macroeconomic management. Such a unit may also be involved in debt “workoutsâ€? or restructuring that may be needed from time to time—a sort of a “local IMFâ€? for subnational governments. 4.53 The FCU would coordinate with bodies making the loans—including FEICOM and commercial or multilateral development banks, if these make direct loans to the subnational entities. F. A SHORT TO MEDIUM-TERM STRATEGY FOR PFM REFORMS IN CAMEROON 4.54 The problems with the PFM system identified at the local levels in Cameroon are linked to deeper and more pervasive problems at all levels, including at the center. It is clear that the issue of tracking and addressing arrears is not just a problem at the local level, but also at the center (IMF, 2011). However, a solution at one level of government cannot be found in isolation of adverse incentives and lacunae at the other. 4.55 Many of the recommendations proposed for the central government’s PFM reforms by the IMF Article IV report are sensible, but need to be extended to include the decentralization process and local governments, including the monitoring of commitments and budget reporting systems. While centralization of commitment management and reducing the number of authorizing officers is only feasible at the level of the central government—what is clearly needed are: ƒ Creating a common platform of fully developed Government Financial Management Information Systems (GFMIS) at each level of government, ƒ Establishing linkages to other levels and using a common platform, and particularly a common chart of accounts and fully specified budget classification system (based on GFS2001 and COFOG); and 52 ƒ Initiating standardized monitoring of arrears and buildup of liabilities at the subnational level—this should also be a function of the central debt management department. 4.56 Institutional clarity is also needed—including on: ƒ Macro-fiscal coordination, leading to closer and better integration of the budget processes at each level of government; and ƒ Separation of the debt/lending and grant allocation functions—with FEICOM having to choose between one or the other and a new institution being created for the other function. 4.57 While the measures above are essential elements of a solution, they address symptoms of the problem and instruments to treat the symptoms—they do not address the underlying imbalances and adverse incentives that cause the problems in the first place. The solutions in this case involve a reconsideration of the strategy, and making the decentralized entities more responsible for their actions. As discussed in the other chapters, many of the issues are of profound policy importance, and not solely linked to PFM per se. Key elements in this strategy are to: (1) address the legal lacunae identified in this chapter, and (2) remove the automatic bailout clause from the legislation. 4.58 Complete, standardized, and timely information on the budget process, using both the economic and functional classifications, is essential for the accountability process to begin to work better. This should be mandatory for the decentralized organs as well as the deconcentrated parts of government so that a full picture of who spends what and where is available to the governments at each level, as well as local electorates. 4.59 While performance budgeting is a distant goal for countries like Cameroon, as functions or sub-functions are identified for devolution, simple criteria for measurement of outcomes should also be identified—such as literacy or morbidity rates. These could be tracked separately and used to inform electorates as to how effective governments are in the provision of public services and meeting assigned MDGs. 4.60 Similar criteria are also needed for the evaluation and measurement of deconcentrated operations of government at the local level. Thus, if local administrators are unable to meet established criteria, actions should be taken to change the administrators or devolve the functions. 4.61 MINATD would play a major role in this area, together with MINFI. 53 Chapter 5: SUBNATIONAL REVENUES AND ADMINISTRATION A. INTRODUCTION 5.1 The overlapping assignments in the 2004 legislation were supplemented by legislation on local tax instruments in 2009. This may have been an attempt to redress the dismally low non-oil tax/GDP ratio for general government. But, as we will see below, the structure remains opaque and cumbersome. The main sources of local revenues are still based on central revenue-sharing, supplemented by a myriad of taxes and fees that are likely to be relatively small in relation to the effort needed to collect them. Overall, the tax structure that has evolved does not add much to the incentives for local governments to utilize them, and is unlikely to result in improvements in revenue performance. Nor will it significantly improve the accountability of the Cameroonian local governments. Moreover, it may widen local disparities. 5.2 In this chapter, we examine an ideal-type structure of subnational revenues; and the current practice of shared and assigned local revenues. We also examine the issue of revenue administration. We will argue that the central tax administration can be used to provide services to local governments on an agency basis without affecting local “controlâ€? over their tax bases. 5.3 Under the Cameroonian classification, local borrowing is also treated as revenues. This is an error and removes focus on the bottom line for local governments, is contrary to established international standards, and was discussed in the chapter on budgeting and deficit financing. B. WHAT LOCAL TAX DESIGN SHOULD LOOK LIKE 5.4 A typical arrangement to finance local spending is based on the “normativeâ€? benefit principle—i.e., those who gain from public services should pay for them, and a tax/cum user-charge system would be designed to achieve this criterion. A purely user-charge based system may impose unacceptably high burdens on the poor, and a properly designed tax system could provide a modicum of local redistribution. However, the tax system should be such that it does not shift burdens on to other jurisdictions that do not benefit from the benefits provided. This gave rise to the “normativeâ€? principle that the best local taxes are those that are imposed on immobile tax bases, particularly property, and locally generated incomes, that are also borne by more prosperous members of the community. 5.5 The more recent political economy, or “positiveâ€?, approaches to local taxation focus on the accountability of local officials. Accountability requires a hard budget constraint involving own-source revenues (see below and the chapter on budgeting). Accountability may be enhanced by the use of mobile tax bases as sources of local revenues—contrary to the prescriptions of the normative theory—because this imposes “market disciplineâ€? on local officials (see Ambrosiano and Bordignon, 2006). Provided administrative arrangements can be met (usually through centralized collections with local setting of piggy-backed rates) the positive approaches can be used to impose discipline on the behavior of local officials. Under the positive approaches, interactions between policy 54 instruments become critical—so that if there are “gap-fillingâ€? or deficit financing transfers, they can vitiate any improvements in incentives through tax policy design. 5.6 A system of financing local expenditures is likely to include a range of instruments and administrative arrangements. This can range from fees and charges, based on the benefit principle, to own taxes in which the jurisdiction is able to control either the rate or the base of the tax. In addition, there may be revenue-sharing arrangements, particularly in connection with tax instruments administered by other levels of government. User charges usually meet the benefit criteria, but may be quite regressive, if not supplemented by more redistributive taxes. 5.7 Greater complexity is introduced when the central government wishes to maintain minimum standards for social services, such as for literacy or health status, or other MDGs, and the tax bases across local jurisdictions vary considerably. This gives rise to the need for equalization, so that local or subnational jurisdictions are able to provide similar levels of service at similar levels of tax effort. This inter-jurisdictional redistribution can come about through the revenue-sharing system or transfer systems. Thus, if a specified proportion of a central income tax is shared equally on a per capita basis (as is the case in Cameroon), this introduces an element of redistribution—as it tends to be borne by richer individuals in the more prosperous jurisdictions and the benefits would be shared with the poorer jurisdictions and lower strata of society. 5.8 The revenue-sharing could be on an origin basis, or alternative formulations related to population, or explicit equalization criteria (in the Australian case, the VAT revenue- sharing fully finances the equalization transfer system). In other cases, a piggy-backed tax or joint occupancy of a revenue base can convert a shared tax into own-source revenue. 5.9 The key principle underpinning the Cameroonian legislation, in keeping with the “normativeâ€? approach, appears to be that there should be sufficient funding for the devolved responsibilities. The principle of ensuring that funding follows functions is a good one. However, the lags in the promulgation of the tax assignments have acted as a “brakeâ€? on the devolution of functions. 43 The myriad of tax and user charge instruments potentially yield little revenues. The automatic coverage of deficits leads to little incentive to use the instruments that have been established in the legislation. Moreover, complex overlapping and inappropriate sharing of typically local taxes further weakens incentives to use assigned tax handles. C. CURRENT PRACTICES 5.10 The 2009 legal framework, specifying taxes and user charges assigned to metropolitan areas, as well as rural and urban municipalities, is summarized in Appendix tables A6-8.As with other legislation, there is a lack of consistency among the different articles. More importantly, the legislation does not appear to be linked to a “desirableâ€? 43 “While waiting for local governments to dispose of sufficient own financial resources, the government services territorially deconcentrated will be only progressively transferred to the decentralized local governmentsâ€? (Law 2004/017, art. 80). 55 model of decentralization for Cameroon, and properly sequenced to achieve the stated goals. 5.11 Table 5.1 shows the actual collections of own and shared taxes reported by the sample of 54 municipalities for 2008. Table 5.2 shows municipalities in the metropolitan areas of Douala and Yaoundé (the latter is for 2007). We see that the main subnational sources of tax revenue are from shared or concurrent sources: such as the CAC and the patent, as well as the forestry tax. The patent is incorrectly classified as local tax revenue, since it is a shared source of revenue. 5.12 Twenty percent of the collections of six local taxes and one local fee go to FEICOM, for redistribution to rural or urban municipalities or special (inter-communal) purpose districts: x Taxes: patent; licenses; tax on immovable properties; and CAC which is a shared tax on personal and corporate income and on VAT; x Parking fees. 5.13 The tax collections are centralized and redistributed to all rural and urban municipalities as follows: x 70 percent of the CAC; x 50 percent of the forestry license fee annually assigned to LGs; x 100 percent of the car tax. 5.14 The redistribution criteria and formula are fixed through central regulation. However, the terms of the law are not very explicit. CAC 5.15 The CAC is based on a 10 percent share of the principal taxes (PIT, CIT, and VAT), 44 after deducting 10 percent of the collection as administrative costs. 45 The remaining 9 percent is distributed to local governments, although the legal basis for the redistribution remains opaque. 46According to Article 116 of the 2009 law, (i) 20 percent of the tax yield is pooled for local government capital expenditure in the FEICOM; (ii) another 70 percent is earmarked for redistribution. Literally, 90 percent of 9 percent (taking into account 10 percent for administrative costs) is redistributed according to criteria other than origin. But decree 2007-1139-PM specifies that the CAC is to be distributed 10 percent to the central government for administrative costs, 20 percent for the capital expenditure pool and 70 percent for allocation of local governments with a formula still to be legally defined. In practice, the sharing is based on an “equalâ€? per capita distribution based on the population figures from the last census, with the exception of a larger than population share for Douala and Yaoundé, perhaps as the result of a strategy to generate “growth centersâ€?. 44 Law 2009/019, art. 54. 45 Law 2009/019, art. 55. 46 Law 2009/019, art. 56. 56 Table 5-1: Administrative Accounts – Revenues (54 sample accounts, excl. Yaoundé and Douala) Averag Median e (% of REVENUES Average (CFAF) (CFAF) Min (CFAF) Max (CFAF) total) Global tax (Impôt liberatoire) 6 574 976 4 919 045 162 000 31 967 720 5 Proceeds from businesses (Patente) 2 009 637 800 754 0 15 113 493 2 Liquor licence (Contributions des licences) 597 565 151 175 0 7 466 612 0 Cattle tax 212 722 0 0 1 882 720 0 Firearms tax 1 889 0 0 47 500 0 Transfer fees 18 404 0 0 957 000 0 Forest taxes 32 914 761 0 0 865 498 822 27 Sanitary inspection tax 137 748 0 0 6 391 200 0 Certification of motor vehicles and machines registration 23 921 0 0 725 000 0 Land tax 55 063 0 0 1 581 357 0 Other fiscal revenues 490 968 0 0 13 267 959 0 Total Fiscal Revenues 42 383 231 7 089 135 183 218 882 034 792 35 CAC 46 843 611 31 277 190 9 172 191 277 786 081 38 Total Direct Council Taxes 1 271 447 0 0 47 344 047 1 Stamp duties 631 119 290 000 0 3 213 000 1 Highway dilapidation tax 3 052 100 0 0 161 018 015 2 Tax on seasonal livestock movement 18 731 0 0 660 000 0 Transit tax on livestock 9 679 0 0 405 500 0 Municipal custom tax 12 757 0 0 407 188 0 Taxes on the exploitation of mineral, fishery resources etc 37 866 0 0 1 015 000 0 Other Indirect Taxes 457 257 0 0 12 696 275 0 Total Indirect Council Taxes 10 451 536 3 853 870 0 167 558 015 9 Rent of municipal buildings 1 921 092 197 000 0 22 222 550 2 Total Proceeds from Exploitation of Council Property and Services 17 309 701 2 152 875 0 217 669 641 14 Total Rebates and Royalties Granted by the State 90 358 0 0 2 439 900 0 Total Financial Revenues 1 852 0 0 100 000 0 Total Revenues from Operating Subventions 900 000 0 0 16 200 000 1 Transfers received from the state 0 0 0 0 0 Transfers received from other national bodies 0 0 0 0 0 Transfers received as international council cooperation 0 0 0 0 0 Miscellaneous transfers 18 519 0 0 1 000 000 0 Total Transfers Received 18 519 0 0 1 000 000 0 Fines for simple offences 221 987 0 0 4 086 125 0 Total Miscellaneous Revenue and Sundry Profits 562 749 0 0 17 856 700 0 Total Reserves Earmarked for Operations 2 739 282 158 837 0 36 147 657 2 TOTAL RECURRENT REVENUES 122 572 285 68 065 748 11 382 064 991 975 354 100 57 Table 5-2: Administrative Accounts from Douala VI and Yaoundé VI, Revenues Yaoundé VI Yaoundé VI MUNICIPALITY Douala VI 2008 2007 Douala VI 2008 2007 Actual (% of Actual (% of REVENUES Actual (CFAF) Actual (CFAF) total) total) Global tax (Impôt libératoire) 225 066 400 222 147 670 35 40 Proceeds from businesses (Patente) 0 0 0 0 Liquor licence (Contributions des licences) 0 0 0 0 Cattle tax 0 0 0 0 Firearms tax 0 0 0 0 Transfer fees 15 950 0 0 0 Forest taxes 0 0 0 0 Sanitary inspection tax 0 0 0 0 Certification of motor vehicles and machines registration 0 0 0 0 Land tax 0 0 0 0 Other fiscal revenues 0 0 0 0 Total Fiscal Revenues 225 082 350 222 147 670 35 40 CAC 152 450 195 0 23 0 Total Direct Council Taxes 9 723 790 0 1 0 Stamp duties 811 000 16 818 400 0 0 Highway dilapidation tax 115 000 280 000 0 0 Tax on seasonal livestock movement 0 0 0 0 Transit tax on livestock 0 0 0 0 Municipal custom tax 0 0 0 0 Taxes on the exploitation of mineral, fishery resources etc 0 0 0 0 Other Indirect Taxes 1 856 125 0 0 0 Total Indirect Council Taxes 127 364 435 105 796 025 20 19 Rent of municipal buildings 0 41 379 600 0 8 Total Proceeds from Exploitation of Council Property and Services 0 57 628 126 0 10 Total Rebates and Royalties Granted by the State 0 0 0 0 Total Financial Revenues 0 0 0 0 Total Revenues from Operating Subventions 135 000 000 165 000 000 21 30 Transfers received from the state 0 0 0 0 Transfers received from other national bodies 0 0 0 0 Transfers received as international council cooperation 0 0 0 0 Miscellaneous transfers 0 0 0 0 Total Transfers Received 0 0 0 0 Fines for simple offences 110 800 76 000 0 0 Total Miscellaneous Revenue and Sundry Profits 110 800 76 000 0 0 Total Reserves Earmarked for Operations 0 0 0 0 TOTAL RECURRENT REVENUES 649 731 570 550 647 821 100 100 58 RFA 5.16 The redevance forestière annuelle (RFA) is mentioned in the list of local taxation, but is effectively a shared central tax. The sharing proportions have been changed unilaterally by the center since the crisis in 2009 (see the chapter on forestry below). 5.17 The wording and legal framework governing the distribution of the RFA is ambiguous and unclear. According to the 2009/019 Law (art. 52), a share of 40 percent is allocated to municipalities, 47 half (20 percent) according to the origin principle, and 20 percent to FEICOM 48 for redistribution according to future regulation. 49The rest (that is, 60 percent) is managed by and remains with the state as indicated in the General Taxation Code. This however is not clear from the 2009/019 Law, art. 52(2-3) which states that the balance to be centralized is to be distributed between all local governments according to rules and formulae which should be fixed by central regulation. The wording is confusing and the distinction between the 20 percent redistributed by FEICOM and the 60 percent going to the Treasury is not clear. 5.18 Moreover, according to article 116(2) of the 2009/019 Law, 50 percent of the forest license attributed to the local governments should be redistributed to all rural and urban municipalities. Given the ambiguous wording in article 52, it is not clear which 50 percent is sanctioned by law: (1) fifty percent of the 40 percent attributed to municipalities – that is, half of 40 percent = 20percent, as above in art. 52(2); or (2) 50 percent of the rest, that is, of the 60 percent. But article 52(3) suggests that these 60 percent are to be distributed on the basis of a future regulation mentioned in article 52(2). Further, Joint Arrêté 0520 of July 28 2010 (Art. 2) states that 20 percent goes to producing municipalities, 20 percent to FEICOM for redistribution among remaining municipalities, and 10 percent to neighboring village communities. The remaining 50 percent goes to the treasury but is regulated by the taxation code and is not mentioned in the Arrêté. Thus, Decree 2011/1731/PM of July 18 2011 states for the first time the rules for the full distribution of the RFA. The car tax 5.19 The redistribution of the car tax is clear in the legislation, unlike the other items. 50 Yet it is not clear why this tax should be shared when the costs of car use are largely borne by the locality where it is registered and used. Municipal taxes 5.20 Municipalities can decide the tax rates, within the legally specified brackets, for three taxes (Law 2009/019): x patents (also referred to as “proceeds from businessesâ€? in the administrative accounts); x licenses (including for liquor); and 47 Art. 52(1). 48 Art 52(2, 3). 49 This regulation has only just been issued-- Decree 2011/1731/PM of July 18 2011. 50 Law 2009/019, art. 51 and 116. 59 x the global tax at source on activities (produits de l’impôt libératoire), excludes VAT, and activities under patents and licenses which are managed by the central tax administration. The principle of setting the tax rate within a band is a good precedent for the establishment of own-source revenues. However the overall incentive structure should not prevent the use of such handles. 5.21 Four other taxes (immovable property, sale of immovable property, games and betting and lotteries, and taxes on cars) are only local in the sense that the yields are attributed to local governments, but totally managed by the center that also fixes the tax rates. Financing decentralized regions 5.22 Decentralized regions are assigned 14 tax sources, but the tax rates are determined at the national level, so these are not own-taxes in our parlance. Article 118, Law 2009/019, lists these taxes but does not give further details as to how the tax bases are defined and calculated, and does not specify the rate schedules. Thus, the “decentralizedâ€? regions are effectively even less decentralized than the municipalities. The 14 taxes for which revenues are assigned include: x Stamp duties (droit de timbre sur les cartes grises); x Airport passenger document fees (droit sur les timbres d’aéroport); x Taxes on vehicle axles (taxe à l’essieu); x Forestry, wildlife and fishing, (les ressources de la forêt, de la faune et de la pêche); x Charges on : o Water resources (les ressources en eau); o Petroleum resources (les ressources pétrolières); o Mining (les ressources minières); o Fishing, farming and cattle breeding (les ressources halieutiques et d’élevage); o Tourism activities (les ressources touristiques); o Aircrafts (redevances aérospatiales); o Natural gas (les ressources du secteur gazier); x Road traffic charges/tolls (redevances d’usage de la route); and x Fees relating to dangerous, disturbing or unsanitary economic activities (l’exploitation des établissements classés dangereux, insalubres ou incommodes). D. ASSESSMENT 5.23 In this section, we focus on an evaluation of the main revenue sources that are available to the local governments, in particular, the CAC and the RFA. We also refer to other tax handles that might usefully be developed, especially property taxation. 60 CAC 5.24 In practice, the implementation of the CAC on the basis of the gross tax paid by the firm/individual (VAT, income taxes), and as discussed above, appears to be shared as follows: a. 91 percent to the central government (state); b. 9 percent to be shared between state and local governments; of which:  10 percent to state (this additional collection charge is high for a tax that is already collected at the center whether or not there is a local share);  20 percent to FEICOM for loans;  40 percent of the balance goes to Douala and Yaoundé on basis of gross collections;  60 percent of what remains is shared among all local governments (including Douala and Yaoundé) on a population basis. 5.25 The collections of the CAC suggest a “headquarters biasâ€? (as firms declare the VAT and corporate tax liabilities at headquarters, likely to be based in Yaoundé or Douala). The regional CAC allocations are shown in Chart 5.1. Chart 5:1: CAC Allocations, 2010 (CFAF per capita) 12 14000 Wellbeing indicator 10 12000 CAC allocation 10000 8 8000 6 6000 4 4000 2 2000 0 0 Sources: ECAM 3, National MDG Report 2008, MINATD and Staff calculations 5.26 As expected the allocations are heavily tilted towards Douala and Yaoundé. The allocations also appear to be negatively correlated with need, using the ordinal rankings of the wellbeing index (which may understate the intensity of the needs). 5.27 The CAC remains the most important and reliable source of revenue for subnational jurisdictions. But there are a couple of anomalies. In the forest zone, for the municipality of 61 Yokadouma, which benefits from RFA proceeds, the CAC represents only 2 percent of revenues while in several other rural municipalities it is over 85 percent. However, Yokadouma received 426 percent of the CAC it budgeted for in 2006 but only 15 percent in 2008.51 This may have been an implicit adjustment for the poor allocation of earlier years, but does not say much for the budgetary allocation process. 5.28 The second anomaly relates to the municipalities in the metropolitan regions (in our sample, Yaoundé VI and Douala VI). The CAC received by metropolitan municipalities is a lower proportion of revenues than for other municipalities, even though the aggregate CAC going to Yaoundé and Douala is greater than for other areas. This may reflect the fact that most of the CAC goes first to the metropolitan area, which then hangs on to the resources, delivering it partially on the basis of subventions to municipalities within the metropolitan areas. This leads to an even more circumscribed decentralization in metropolitan areas than in other municipalities, but may change as a result of the amendments proposed in July 2011. Redevance Forestière Annuelle (RFA) 5.29 The RFA is considered “State moneyâ€?, although shared with local governments, and its use is subject to central laws and regulations concerning management and use. As we have seen above, there is considerable lack of clarity in the legal framework. It is not surprising therefore that most of these strictures are not respected by the local municipalities/communities, and open up possibilities for rent-seeking and diversion of funds. Some of the main difficulties are as follows: x The legal texts are not known, including to the municipal treasurers, or disseminated before application, as for example Arrêté No 122 concerning the usage of the RFA (Ndjanyou and Majerowicz, 2004). x Some of the procedures to implement the laws are complex, confused or open to interpretation. Specific difficulties relate to commitments and payments. State functionaries are insufficiently trained, or lack the means of applying the rules. This facilitates the operations of the vested interests and adds to the discretion of mayors (Nzoyem et al., 2003). x The committees which oversee the local components of the RFA are very heterogeneous (Nzoyem et al. 2003, Lescuyer et al. 2008, Morrison et al. 2009). The interpretation and municipal execution of the law then falls to the prefects and sub- prefects, leading to implicit central controls over its usage. x Arrêté 0520 was adopted to clarify the legal usage of the RFA and enhance its effects on the wellbeing of the populations. But mayors of the forestry regions have not accepted this, as it requires a major change in procedures at the local levels. According to art. 12, line 3, the Mayor is required to produce annually an administrative account for all operations of the RFA, along with semester reports on the state of implementation, in order to better track the use of the RFA (seen distinct 51 The former mayor and treasurer of Yokadouma were recently sentenced to 30 years in prison for embezzlement of public funds (over CFAF 1bn) between 2002 and 2007: see http://www.cameroon- tribune.cm/national/7-regions/65913-yokadouma-30-ans-de-prison-pour-l’ancien-maire-et-son-receveur. 62 from other local resources). This is hard to do with the current level of skills and knowledge of accounting standards and budget practices. The mere publication of a legal text is not sufficient to put it into operation, especially when this requires considerable efforts on the part of the local officials that have neither incentives nor ability to do so. x Arrêté 0520 requires putting in place new institutional arrangements to regulate the usage of the RFA, e.g. 80 percent for investment. Mayors viewed this regulation as illegal, imposing constraints that are significant and costly, or unrealistic. This risks being totally ignored, or implemented partially at best. 5.30 It is appropriate to share the RFA, provided that the resources accrue to the originating localities on an assured basis without conditions. The sharing represents recognition that natural resource extraction poses costs in the relevant areas, as well as the political economy of providing resources to the generating regions. However, the sharing injects a large magnitude of revenues into the budgets of local governments, which may not be able to effectively absorb the funds. It also creates fluctuations and uncertainty in financing basic spending. 5.31 Given that the poorer municipalities in the forestry regions of the Eastern region should generate the RFA, one might expect that it would offset to some extent the biases in the CAC and in the FEICOM grants arrangements (see chapter 6). However, as we see in Chart 5.2, it really does not make an appreciable difference. The RFA is significantly less than the CAC in most regions, except the East, where it is largely used to finance “deconcentrated spendingâ€? (see chapter on Forestry below). Hence, the central government is able to utilize a purely local resource for deconcentrated spending under its control. Consequently, during 2009 and 2010, given the problems faced by timber companies, the central government had no compunction in reducing the RFA by half, to the benefit of the companies, without consulting the local governments. 5.32 The ad hoc reduction in the RFA introduced uncertainties for local governments that they are ill equipped to handle (see Cerutti and Lescuyer, 2011). However, if the funds are being used for largely deconcentrated functions, then the criticisms are less valid. But this underlines the reality that the RFA is not even a shared revenue, and is largely a mechanism to finance deconcentrated spending and subject to the same constraints as other central funds. 5.33 The reduction of the RFA in 2009 was accompanied by an increase in export levies (see Chapter 7) due to a relaxation in export restrictions. These export levies accrue to the central government, more than compensating for the reductions in the RFA. Thus, the costs of the stabilization appear to have been borne exclusively by the subnational governments, and the interests of the center were protected during the downturn. Consequently, the RFA design at present may increase the incentives to operate informally at the local level (returns which may or may not accrue to the local communities—depending on whether these are siphoned off by intermediaries and local officials). In both cases, the state/central government loses. 63 5.34 The net effects for informal logging and charcoal activities remain to be determined on an empirical basis. The presumption is that if rents revert to communities, they will be more careful about wasteful logging and charcoal activities that adversely affect the habitat and hence their “takeâ€? over a period of time, although it requires a sufficiently long time preference. However, the mechanisms for such “communityâ€? controls are not in evidence, and this may be a useful area for further research. Chart 5:2: RFA and CAC, 2010 (CFAF per capita) 12 12000 Wellbeing indicator 10 CAC 10000 RFA 8 8000 6 6000 4 4000 2 2000 0 0 Sources: FEICOM, CIFOR, ECAM 3, National MDG Report 2008, MINATD and Staff calculations. Note: RFA is with 50% reduction and with the distribution ratio changing mid-year. Other taxes, fees and charges 5.35 It is interesting that the myriad of other taxes, fees and charges produce little or no revenues—with the partial exception of the taxe globale. The main items are discussed below. (1) Taxe Globale 5.36 The taxe globale, or local business tax, goes directly to local governments, and although it is shared, is their main own-source revenue. Should the municipalities themselves administer this? Some may indeed have the administrative infrastructure to do this, but it is relatively straightforward to use the information available to the central VAT administration to improve the effectiveness of the business tax. The issue is whether the DGI has incentives to collect this tax, as the amounts collected are. Local governments may continue to use the good offices of the DGI in order to collect the relevant amounts from identified businesses. This collaborative mechanism has actually been used by some of the forward-looking rural municipalities, e.g. in the North West region. 5.37 Local information, together with the arms-length enforcement mechanisms that are at the disposal of the DGI, may be effective in jointly administering the taxe globale, although a fee for service mechanism may be needed in order to offset the costs to DGI. 64 5.38 There is no justification for sharing the taxe globale. It should accrue directly to the local jurisdiction in which it is levied. (2) Automobile registration 5.39 Another example of an inadequate assignment is with respect to automobile registration—the revenue accrues to FEICOM rather than the locality in which the automobile is registered. This is one of the simplest tax handles at the disposal of local governments, and there is little justification to redistribute the revenues collected to other jurisdictions. Hence, there is no need to involve FEICOM in what should be a convenient local own-revenue base. 5.40 The above examples illustrate that the local governments in Cameroon really are bereft of effective own-sources of revenues. A major effort needs to be made to clarify and determine which heads of own-source revenues would be most appropriate in the Cameroonian context. Capacity constraints can be overcome by using modern methods of self-assessment and audit. Some of the functions can be outsourced, including to the DGI. (3) Property taxes 5.41 Property taxes generate relatively little revenues in Cameroon, even in the better off metropolitan areas, including Yaoundé and Douala. A large share (60 percent) is assigned to the metropolitan area rather than the decentralized municipalities—but the data do not reflect that much if anything at all, gets through to the municipalities. 5.42 The disappointing performance with the property tax may reflect difficulties with establishing the cadastre, but also the system of valuation, and tracking and recording changes in prices and property sales. The incentives for local governments would clearly increase if these revenues are not shared. 5.43 Considerable effort will be needed to redesign the property tax to make it a significant source of own-revenues at the local levels—hence meeting the goals of the PRSP in significantly increasing the non-oil to GDP tax ratio, and also making it a viable source of local own-source revenues. Technical assistance by the Bank in this area may be justified. E. TAX ADMINISTRATION 5.44 The extent to which local tax administrations can be strengthened, it would facilitate more effective management and collection of the purely local taxes. These include the, property tax and the land-based taxes. 5.45 However, the shared taxes should continue to be administered by DGI, and as mentioned in the chapter in forestry, all other shares should go into a pool for sharing. The tax-by-tax tracking for the local bases of the main taxes is neither desirable nor indeed feasible—with the exception of the patent. 5.46 In keeping with international best practice, and in order to ease the terms and cost of doing business in Cameroon, the DGI has recently moved to a single declaration form for all taxpayers making payments for the VAT, the income taxes, and patents above a certain 65 level. Amounts collected in the Large Taxpayer Unit (roughly 80 percent of all revenues) are credited to the Treasury Single Account (TSA) daily. Payments made at branch offices that are not connected electronically to Yaoundé are made every two weeks. 5.47 From individual tax returns, it is important to note that taxpayers must identify the amounts that must be made legally against each type of tax and recipient of the revenues—central government, local government, or FEICOM. Thus, at the end of each day, DGI deposits, and knows the amounts deposited, in the sub-accounts of each recipient agency in the TSA. 5.48 While the process of a single return and the TSA has been perceived by several agencies and civil society as an “unwarrantedâ€? recentralization of the revenue circuit, in fact, this need not be the case. In fact, the reform can considerably simplify existing processes and procedures, make life easier for taxpayers, and also provide greater certainty to local governments about the amounts of “theirâ€? funds accrued than they have at the present time. 5.49 A major advantage of the new circuit of the revenue payment going into the TSA is that it blocks the game-play, delays and possible misuse of funds that were previously deposited in several accounts. These previously only percolated down to local governments as a result of the generosity of third agencies (e.g., FEICOM) or the regional treasurers. 5.50 A precondition for the wider acceptance of the new procedures is that DGI should send daily reports of revenues collected (especially the patents), not just to DGT, but also to the governments on whose behalf the revenues have been collected, as well as to FEICOM. With the new arrangements, there is no need for transitional arrangements to be routed through FEICOM’s accounts, and monies can go directly to the local government account on the basis of the agreed formula. Of course, the allocation of funds collected by FEICOM itself, e.g., for lending and investment purposes, does not concern DGI (see the section on local government debt). 5.51 Chart 5.3 presents a possible simplified collections circuit for revenues collected through the single declaration and credited to the TSA. The broad lines indicate flows of funds, and the narrow lines indicate the information flows that should accompany the deposit of funds each day; to the DGT, local governments, as well as FEICOM. In this example, there are two local governments—each with a virtual or correspondent account in the TSA. There is one taxpayer, say in Douala. His payments are credited against the sums due to the central government, amounts due to Douala, the shares of all other local governments, as well as the payment due to FEICOM’s account. 5.52 The single payment is efficient, and provides the best assurance, including for local governments outside Douala, that the correct amounts are being credited to their accounts. The simplified circuit also obviates the need for payments to be credited in Douala, then sent to FEICOM to redistribute to other local governments. The sharing is now done automatically, as the monies are credited. 66 5.53 It is important that DGI and DGT provide full information to all the relevant stakeholders, particularly to the local governments. Also there needs to be better public awareness of the new system, and its potential advantages. However, information flows need to be improved, as, in the absence of such information, local governments are of the view that their funds are being “swallowed upâ€? by DGI. 5.54 In the longer run, the local governments should have the option to collect the patent themselves, if they have the capability, or to ask the DGI to do so on an agency basis. Chart 5:3: Revenue Circuit with a TSA TSA Correspondent account 1 Correspondent account 2 VAT DGI FEICOM CIT LG1 LG2 Note: the broad arrows reflect flows of payments, and the narrow arrows refer to the transmission of information that should accompany each payment. F. A RENEWED SUBNATIONAL TAX REFORM AGENDA 5.55 Despite the very complex and detailed local tax system described above, it neither provides a significant source of local revenues, nor does it function as an own-source revenue handle that could be used to hold local officials and politicians accountable. In relative terms, the main sources of subnational financing include shares in the main taxes (VAT and the CIT), and the grant elements of FEICOM transfers on account of lending to the local governments (see Chapter on Transfers). 5.56 There are a number of additional difficulties in the way the sharing is organized and administered. A particular difficulty arises with the attempts by the administration to simplify procedures for the large taxpayers that have to pay a number of taxes, by 67 facilitating a single declaration, together with “best practiceâ€? attempt to implement a Treasury Single Account (see the discussion in the chapter on budget processes). A real concern of local administrations is that these well-meaning reforms could further reduce information on what has been collected on their behalf, as well as greater uncertainty, complexity and delays in accessing “theirâ€? resources. These concerns are genuine and could further weaken the decentralization process unless addressed—and timely information on the collections needs to be provided to the municipalities and other stakeholders. 5.57 We focus on the design of shared revenues, as well as options for enhancing the own-source nature of the myriad of taxes, fees and charges that are ostensibly assigned to local governments. A key element in this process would be to rationalize the fees and charges, and remove the “nuisance fees and chargesâ€?, especially those that act to restrict inter-regional trade and commerce and others that raise little or no revenues. Clarifying shared revenues 5.58 Some of the sharing arrangements are complex and not particularly well thought through. The sharing formulae are on a gross basis, as declared by the taxpayers. This generates difficulties in particular with respect to the VAT and the CIT. VAT 5.59 For the VAT, a gross return basis overstates the amount of revenue that is assigned to the “producing regionâ€?, or where the taxpayer is based. This is because it does not net out the refunds and credits due to exports out of the country as well as intermediate purchases in other regions. The gross basis can lead to more of the VAT being distributed than actually collected on behalf of the local governments. The best base for the VAT sharing is “consumptionâ€?. However, this is complex to implement and estimates of regional or local consumption are not likely to be easy to construct for Cameroon. 5.60 An alternative is to put the VAT revenues into a pool (including the CIT and perhaps also other central taxes) and a certain portion of this be allocated to close the vertical imbalance between assigned spending and the own-source revenues available in aggregate. The shares across regions or municipalities could be determined on the basis of relative spending needs and revenue capacities, perhaps on an equalization basis, as in Australia and many Scandinavian countries (where the equalization is to the municipalities, rather than the intermediate governments, as in Australia).52 This is explored in more detail in the Chapter on transfer design. Corporation Income Tax (CIT) 5.61 As taxpayers are required to file only one return, firms that may have a large number of branches tend to file this at their headquarters, typically in Douala and Yaoundé. This overstates the revenues being generated from these two metropolitan areas. This “headquarters biasâ€? in the tax payments, especially in the large taxpayers unit, is quite 52 See Bob Searle, 2010. 68 common, and there are techniques to “apportionâ€? the generated revenues on the basis of capital invested, labor, or output. 5.62 The “headquarters biasâ€? is to some extent taken into account by FEICOM, and there is a specific compensatory allocation to offset this bias. However, the information available to FEICOM to undertake this form of adjustment must be extremely limited, as it lacks access to the DGI master files needed in order to do this effectively. As in other countries, the tax administration, or DGI, would be in a better position to make any apportionment, given the data at their disposal. However, if an apportionment is made by another agency, including FEICOM, the basis and information to make this allocation should be available to all local governments, as it would affect their resources. Enhancing local own-source revenues 5.63 Given the structure of decentralized local governments, a tax-by-tax sharing of the main revenues, such as the VAT or the CIT is not appropriate. It is also meaningless for local governments to try and track the revenues of the VAT or CIT generated in their municipalities. Local governments cannot estimate net revenues for the VAT, and the headquarters problem for the CIT will also bias estimates. The current arrangements for these “mainâ€? revenues to go into a pool to be shared with all subnational governments (to close both vertical and horizontal imbalances) will be discussed later in the chapter on transfers. 5.64 The personal income tax (PIT), however, could be used to “piggy-backâ€? a municipal share—this could be on the central return, and the local return would be within a specified range, and applied to the central base. The local return could be made directly to local governments, although the audit and verification function would remain the responsibility of DGI. It is likely that much of the additional “piggy-backedâ€? PIT would benefit Yaoundé and Douala—putting greater emphasis on the better design and implementation of an equalization framework for the transfer system. 5.65 An assessment of the myriad of local taxes and fees should be conducted. It has been difficult for the mission to obtain verified accounts of all municipal operations for any of the recent years, but it appears that many of the numerous local tax heads (apart from the taxe globale) generate little or no revenues. Some of the nuisance fees and charges could be eliminated, in order to better dispose the local governments to participate in growth generation activities. 5.66 There should be no local taxes on the movement of goods. Some localities have been advised to implement taxes on the flows of goods and service, and this is reflected in Law 2009/019, art. 107-110. However, this could lead to effective export taxes imposed at the local level, and would complicate the working of the VAT. Such practices should be stopped, and the provision removed from the law. If an additional commodity tax is needed, it should be an increased share from the VAT, or a piggy-backed arrangement at the regional level, so as not to discriminate against productive activities. 5.67 Additional “own-sourceâ€? revenues should be just that. The sharing of the taxe globale, property and vehicle taxes should be abandoned, and the entire revenue 69 collection should go to the municipalities. An equalization fund could be financed by a share of the main central taxes, including the VAT and CIT. 5.68 A rationalization of the numerous fees and changes is needed as the amounts collected are relatively small. As discussed above, some of these can act as impediments to the flow of goods and services across local jurisdictions. 5.69 The property tax could be developed as the basis for local own-revenues. Amounts collected are meager, and are shared with FEICOM. Consideration should be given to making the property tax entirely “localâ€? with setting of the relevant rate (say within a band that could be legislated nationally, if needed, as is common in several unitary states where the legislative authority remains with Parliament). Work needs to be initiated to establish a cadastre (this could be maintained at the central or regional level), as well as valuation mechanisms and a system to track property transactions. 5.70 For the more advanced municipalities, such as Douala and Yaoundé, consideration could be given to applying the techniques of modern tax administration—using self- assessment and applying sanctions for misdeclaration (i.e., purchasing the property at above declared values). 5.71 Once the tax policy framework has been clarified, it would be appropriate to review the legal framework. Key elements would be to: 1. Redefine the policy scope of own-source revenues for each level of government, and the degree of control over rates and bases, and whether revenue-sharing is needed, why and how; 2. Remove contradictions and ambiguities in the law; and 3. Formulate regulations to make the tax and revenue-sharing system clear for all levels of government and the broader public, so that they are aware of their liabilities and consequences of the revenue system. 70 Chapter 6: TRANSFER DESIGN A. INTRODUCTION 6.1 A properly designed transfer system has an important role to play in ensuring that different local governments have the ability to provide similar levels of public service at similar levels of tax effort. It also ensures that central governments can effectively finance required spending by decentralized governments towards meeting central objectives. Badly designed transfer systems can subvert incentives for accountability at local levels, and end up exacerbating inequalities. In evaluating the transfer system, it is important to examine both current and capital transfers, their design and logic, and joint overall effects. The Cameroonian system is deficient in each of these dimensions, based on the observations during the mission—although the absence of information limits empirical illustration at this stage. B. CURRENT TRANSFERS 6.2 The main transfer mechanisms in Cameroon are the revenue-sharing arrangement, the CAC, as well as capital grants. The CAC has been discussed extensively in the revenue chapter above, but is a key element of the transfer system as well. The grant element of FEICOM loans is the second mechanism to provide transfers, although deficit-filling transfers from FEICOM are perhaps the most pernicious. 6.3 We have relatively little to add to the discussion in the previous chapters of the automatic gap-filling deficit transfers, other than that these represent a dreadful practice. Such transfers can completely negate all the other incentive improving measures discussed in the rest of the report. The legal basis for providing transfers to meet deficits should be removed on an urgent basis. 6.4 In Cameroon, there is a considerable emphasis on equality in the design of the allocation of the CAC. The basis appears to be per capita equality of transfers—and this is largely achieved through the revenue-sharing arrangements, CAC, discussed in previous chapters, with the exceptions for Douala and Yaoundé. Equal per capita transfers are often justified on the grounds of being “easy to understandâ€? and politically attractive as a proxy for cost of service provision. The exception is to provide additional transfers, including revenue shares, to the well-to-do areas, Douala and Yaoundé, as part of a broader strategy to address growth and development needs. 6.5 However, the use of the population basis for transfers in the Cameroonian context poses a number of significant difficulties. In discussions with mayors it was often not clear which estimate of population should be used—whether the information from the 1987 census, or more recent estimates—given the large migratory movements and informal sector activity. The total population in the CAC arrêtés (pre-2011) was just under 14 million, although it is now almost 20 million. From the administrative accounts, we see that population figures cited by individual municipalities are generally larger than those mentioned in the official data. Thus, a major source of dissatisfaction with the transfer 71 system was that it is based on out-of-date information on population. Attaching the revenue function to the population figure also runs the risk of politicizing the census. 6.6 The costs of providing public services in remote areas, such as in the northern regions and in the forested East region, may be higher than the average for all regions. Costs may also be higher in densely populated urban regions. Thus, an equal per capita transfer may not generate equal access to public services. Further, the basis for the transfer is likely to generate expectation of entitlements rather than a focus on the delivery of services. 6.7 A more serious difficulty is that the equal per capita transfer would seriously bias “effective service deliveryâ€?, if there are uneven levels of basic infrastructure. Without adequate school buildings, spending current dollars may not be as effective as similar amounts in other regions with schools. C. CAPITAL TRANSFERS 6.8 The criteria for the allocation of capital transfers are not clear. There does not appear to be an explicit attempt to link the transfers to the local priorities identified by PNDP, for example. Even if there were to be a serious effort to build on local priorities, criteria would need to be established for the allocation of investment funds across jurisdictions 6.9 One possible linkage could be in relation to the local dimensions of a growth and employment strategy, as hinted in the PRSP. One could also examine “gapsâ€? in basic capital infrastructure, and meet these through specially targeted programs, or place them in a modified “equalizationâ€? framework (this is explained below, and adapts the methodology normally utilized for current equalization). D. OVERALL EFFECTS 6.10 The transfer mechanism is opaque. The gap-filling provisions override efficiency or equalizing trends of other instruments. 6.11 Ostensibly, the per capita basis for the allocation of the CAC is to ensure “equalityâ€?, but this is not achieved, given differences in infrastructure and unequal capital endowments. Moreover, there are significant spatial price differences, and this has an impact on the net access to services that could be purchased by the “transfersâ€?. 6.12 The overall effect of the CAC allocation and capital grants is shown in Chart 6.1. There is no apparent pattern in the overall transfer/revenue-sharing system, and if anything, the allocations are inversely correlated with need. While there has been considerable work undertaken to define local preferences, including work undertaken by the PNDP, it is not clear how much of a role these assessments of need play in the allocation of capital transfers. Indeed, it is hard to discern a model underlying the system of capital grants. It is not based on any “equalizationâ€? framework, nor do the transfers appear to be linked with an explicit growth and development strategy. 72 6.13 The addition of the RFA to the total pool of local resources is not strictly valid, given that there is considerable uncertainty over whether these resources are actually under local control. However, even if the RFA were to be included, the overall pattern of resource allocation across local governments does not change appreciably. This is seen in Chart 6.2. Chart 6:1: CAC and FEICOM Grants, 2010 (CFAF per capita) 12 14000 Wellbeing indicator 12000 10 CAC and FEICOM grants 10000 8 8000 6 6000 4 4000 2 2000 0 0 Sources: ECAM 3, National MDG Report 2008, MINATD, FEICOM and Staff calculations Chart 6:2: Overall Local Resource Availability, 2010 (CFAF per capita) 12 12000 Wellbeing indicator CAC 10 10000 FEICOM grants RFA 8 8000 6 6000 4 4000 2 2000 0 0 Sources: FEICOM, CIFOR, ECAM 3, National MDG Report 2008, MINATD and Staff calculations. Note: RFA is with 50% reduction and with the distribution ratio changing mid-year. 6.14 The bias towards Douala and Yaoundé is clear in both the CAC and the FEICOM grants. Even if the resources do not effectively make it to municipalities in either metropolitan area, the resources are transferred to the region, and could act as a magnet 73 for migration from other areas, given the limited employment and growth opportunities elsewhere. The net effect may well be an increase in informality in these metropolitan areas, as well as deterioration in the quality of public services, both of which appear to have occurred. This is also a phenomenon common to other West African metropolitan areas, including Lagos. The breaking infrastructure and stretched public services may indeed be an impediment to growth—the policy having the opposite effect to that intended. 6.15 Another area for concern is the provision for marginalized and ultra-poor minorities, for instance in the resource rich but poorly served forest region. There may not be enough political support at the local levels for targeted provision for marginal groups (such as the Baka). Thus one could argue that the provision for such groups should be handled directly by central spending. The difficulty is in ensuring that the central transfers are adequately targeted in order to do better than local provision. 6.16 In order to examine this proposition, we examined the pattern of direct central spending in the regions (seen in Chart 6.3). In this respect, the higher spending in the East and Center over Yaoundé and Douala is an improvement over the system of transfers and CAC. However, the relatively low spending in the North suggests that the criteria for allocation of central spending may not include deprivation and need. Chart 6:3: Central Social Spending at Subnational Level, 2011 (CFAF per capita) 70 4500 Poverty (in percent) 60 4000 Appropriation 3500 50 3000 40 2500 30 2000 1500 20 1000 10 500 0 0 Sources: ECAM 3, 2011 Budget, MINATD and Staff calculations 74 E. SPECIAL PURPOSE TRANSFERS OR DIRECT PROVISION 6.17 Central governments often try to meet infrastructure gaps or other objectives by central transfers (to either deconcentrated or decentralized entities). A big constraint faced by countries with weak PFM systems and poor information on who spends what, as in Cameroon, is that it is hard to ensure that the funds are not diverted to other heads that may be more important for local officials—or just stolen. 53 6.18 As described in Ahmad (2009), it may be possible to offset some of the PFM disadvantages by inducing competition among recipient jurisdictions, using simple performance criteria. The basic idea is that a medium-term budget framework is put in place, and the transfers in period t+2 are made conditional on achieving targets set for period t+1. Thus, if growth and employment generation is an objective, and is not achieved by the additional transfers given to the metropolitan areas, it may be useful to reconsider the strategy in the coming period. Also the relationships between the metropolitan administration and the decentralized subordinate municipalities would clearly need to be clarified. Eventually, when the PFM systems are strong enough, and the court system begins to function efficiently, one could consider “contractâ€? based transfers (Spahn, 2006). 6.19 This is an area of increasing policy interest, and further work in the Cameroonian case would be justified. F. SOME ALTERNATIVE EQUALIZATION TRANSFER DESIGNS 6.20 The CAC does not equalize. The authorities may wish to examine a more modern system of “equalizationâ€?, and the objective could be to “provide resources so that local governments could provide equal standards of service at equal levels of tax effortâ€?. This is the modern mechanism that has been used across States in Australia, provinces in China, and municipalities and districts in Denmark, Hungary and Indonesia. Simplified versions could be used at municipal level; with a few criteria. Clearly, population is still important as a factor, but using it in a standardized manner to evaluate relative costs or needs diffuses the perceived concerns with the population variable. 6.21 Very simply, the equalization framework would be based on “standardizedâ€? factors. This ensures that local governments would not be able to influence the magnitude of the transfer by their actions or lack of actions (see Ahmad and Searle, 2006, for a description of alternative models). 6.22 The standardized transfers thus become more or less “lump sumâ€? and do not distort incentives at the local level. The standardized spending responsibilities would address differential costs of provision for services assigned to them, with higher costs in remotely populated areas, as well as densely populated urban districts. Similarly, the own-revenue potential would be based on standardized revenue (spatial distribution of bases, assuming average rates), and the fact that a local government chooses not to exploit a revenue base would not lead to a higher grant. Thus, there would be an incentive to better utilize assigned revenue bases. 53 We have referred earlier to questions concerning the use of funds in Douala. 75 6.23 Overall, a modern equalization framework should shift the focus from “entitlementsâ€? to a political focus on service delivery by local governments. This would help with local oversight and could generate “yardstick competitionâ€? for greater accountability of local officials. G. FORMULA-BASED GENERAL PURPOSE TRANSFERS 54 6.24 There are five types of formulae currently used around the world for general transfers, and the choice for Cameroon would be to move from (1) to the more general and appropriate allocations that imply a greater degree of redistribution than the current framework. (1) Transfers based on equal per capita allocations 6.25 This is the simplest system for allocation of grants requiring only information on population. It is used in a number of countries, such as Germany for the allocation of a share of VAT and Canada for the allocation of the block grant for health and social services. It assumes that population is a suitable indicator of local expenditure needs. It also has minimal equity content since it gives the same per capita amount to poor and rich areas, although it does not consider revenues. The formula would be as follows: TRi = (Pi/P) x TR where TR is the transfer; P is population; i stands for local unit i. Variables without i refer to the country total. This is the closest to the current Cameroonian arrangement. (2) Formulae based on general indicators of expenditure needs 6.26 These formulae are very popular and derive from the previous one by adding other indicators of needs such as poverty incidence, area, population density, infant mortality, and (inverse) of GDP. The indicators are not related to distinct expenditure responsibilities assigned to local governments, but to their total expenditure. 6.27 An illustration of a simple formula follows, with three equally weighted Å?ŶĚÅ?Ä?ĂƚŽƌÆ?͸ĨŽƌ ĞdžĂŵƉůĞ͕ Å?ĞŽÅ?ƌĂƉŚÅ?Ä? ĂƌĞĂ ĂŶĚ ƚŚĞ ŶƵŵÄ?Ğƌ ŽĨ ƉŽŽƌ ƉĞƌÆ?ŽŶÆ?Í• Å?Ŷ ĂĚĚÅ?ÆšÅ?ŽŶ ƚŽ population: TRi = Pi/P x 1/3 TR + Povi/Pov x 1/3 TR + Ai/A x 1/3 TR where in addition to previous symbols Pov is the number of poor persons and A is area (km2). 54 This section is based on Ahmad and Brosio (forthcoming), “On the design of transfersâ€?, (LSE). 76 (3) Formulae based on specific indicators of expenditure needs 6.28 These formulae are more complex since they use distinct indicators of need for each local expenditure responsibility. They represent a considerable improvement over the previous system but require more information and may be subject to obfuscation if not managed in a transparent way. An example is provided by South Africa, where the general purpose transfer to the provinces is allocated according to a system that has six components: (i) an education component, representing 51 percent of the total transfer allocated according to population in school age and to school enrolment; (ii) a health component, representing 26 percent of the total transfer allocated according to population with and without medical aid; (iii) a basic component, representing 14 percent of the total transfer allocated according to population; (iv) a poverty component, representing 3 percent of the total transfer allocated according to the number of poor persons (quintiles 1 and 2); (v) an economic activity component, representing 1 percent of the total transfer allocated according to GDP; and (vi) an institutional component, representing the remaining 5 percent equally distributed as a lump sum among provinces. 6.29 The formula for the education component would be the following: ETRi = SAPi/SAP x 0,5 ETR + Eni/En x 0,5 ETR where, in addition to previous symbols, ETR is the education component of the transfer; SAP is school-age population; and En is the number of pupils enrolled in schools. Similar formulae would apply to other expenditure functions. (4) Formulae based only on fiscal capacity 6.30 In this case, the transfer does not take account of expenditure needs, but only differences in fiscal capacity. An example is provided by the Canadian system of general- purpose transfers to provinces that consider only differences in tax capacity and assume, correspondingly, that each province has the same per capita expenditure needs. It has to be noted that the Canadian provinces are very large in terms of area, which reduces the variance in expenditure needs. Furthermore, the general-purpose transfers to provinces are supplemented with specific transfers that consider needs. The formula would be as follows: TRi= t x (B/P – Bi/Pi) x Pi where, in addition to the previous symbols, B is the effective tax base (not the assessed tax base, but the base that potentially can be assessed; and t is the average effective tax rate on the concerned tax base). 6.31 It has to be noted that since B/P – Bi/Pi measures the difference between the per capita national average tax base and that of region i, the formula brings the fiscal capacity of those subnational governments that are below the national average up to the national average, i.e., it provides 100 percent equalization with reference to the national average. Equalization can obviously be less intense. 77 (5) Formulae that consider both expenditure needs and fiscal capacity 6.32 These formulae are a combination of the third and fourth formulae. Expenditure needs are estimated for each local expenditure responsibility and aggregated; the transfer will result from the difference between total expenditure needs and fiscal capacity, as in the following formula: TRi=Nij – FCi where, in addition to previous symbols, N is expenditure need; FC is fiscal capacity; and j stands for expenditure responsibility. 6.33 Such formulae are used in an increasing number of countries such as Australia, Denmark, Japan, Korea, and the United Kingdom. They are very complex and require a considerable amount of information, thus requiring considerable length of time for their complete implementation. If properly built and implemented, they can be both efficient and equitable. Management of Transfers: a new Grants Commission 6.34 A Grants Commission could be established to determine the relativities for making equalization transfers to local governments, in coordination with the local governments. It does not make payments directly, which are routed through the Treasury, but establishes the basis and monitors and collects the information needed to make the system work. 6.35 In countries such as Australia, the Grants Commission is an independent agency with representation by the local governments. In countries such as Mexico or China, the Grants Commission function is managed by the Ministry of Finance and could be a separate section within the FCU. 6.36 As we have argued in previous chapters, a decision is needed whether FEICOM should become the Grants Commission for Cameroon, or the development bank providing loans to local governments. The latter should be managed on a commercial basis, with clear criteria, and this function should not be mixed with the management of the transfer system. H. AN AGENDA FOR TRANSFER REFORM A number of reforms are needed, and these link with both clarity in spending responsibilities, availability of own-source revenues, and elimination of deficit financing provisions. Key elements of a redesign of transfers include the following: ƒ Deficit filling transfers need to be stopped, and the legal framework should be reformed accordingly on a priority basis. ƒ The overall scheme of transfers in relation to the strategy for decentralization, particularly the role of growth centers, needs to be carefully defined. 78 ƒ If growth-generating centers are to be encouraged, this should be facilitated through capital transfers. The effectiveness of these transfers needs to be carefully evaluated, and if possible, inter-temporal performance-based transfer mechanisms should be used. ƒ The allocation of the VAT could usefully be changed to consumption from the present population basis. Piggy-backing could be considered for metropolitan shares in the income taxes. As own-source revenues are developed for local governments, the need for a relatively simple equalization transfer system will become more urgent. ƒ The role of managing the equalization transfer system needs to be separated from the lending function. Accordingly, FEICOM’s functions would need to be formalized in one direction or the other, as argued in previous chapters. 79 Chapter 7: FORESTRY, DECENTRALIZATION, DEFORESTATION AND RENT-SEEKING A. INTRODUCTION 7.1 Forests in Cameroon reflect the important interface between efficient ecological management and decentralization that should provide incentives for local governments to manage the process effectively and also achieve poverty reduction, local development and conflict abatement. However, incomplete decentralization (see Devarajan et al., 2009) encompassing inappropriate institutional arrangements, assignments, and poor monitoring and information flows might accentuate rent-seeking and negate the desired positive outcomes. 55 Indeed, the opportunities for officials and vested power-groups to “captureâ€? the rents might be enhanced. This leads to incentives for faster forest exploitation, including for charcoal, with an adverse impact on the environment, and the ultra-poor and minority groups. In addition, by ratcheting up the expectations of improved public service delivery, the shortfalls in this regard might generate a backlash from “sensitive regionsâ€? that the revenue-sharing is designed in principle to prevent. This is a critical element in keeping the country together. Thus, if the benefits from revenue-sharing do not reach the local communities in the natural resource rich regions, the decentralization process could actually accentuate the divisions and resentments and act to increase tensions. 7.2 The forestry story encapsulates the difficulties with the “incompleteâ€? decentralization that have been identified in the rest of the report. While the mission’s work in this area builds on the expertise and considerable experience of the Center for International Forestry Research (CIFOR), additional work is needed especially in relation to some of the factors that accentuate local “captureâ€? and potential over-exploitation of the forests. These include activities related to “charcoalâ€? as well as the most commonly recognized factor: commercial farming. A key issue arising is how the interests of the indigenous, albeit minority communities, such as the Baka, might be protected. 7.3 The discussion also provides some guidance vis-à-vis the sharing of other natural resources. However, the non-renewal aspects of mining activities provide an additional factor that would need to be incorporated in a general approach. B. BACKGROUND 7.4 Despite the assessment in this report and recommendations for improvement in various respects, Cameroon has one of the most advanced forest management frameworks in Africa. It has 20 million hectares of forest (Morrison et al., 2009), of which productive closed forests on drained land cover about 17.5 million hectares (ha). About 7.7 million hectares are designated for production, of which 6.8 million are set aside for timber harvesting. There is a permanent forest domain (PFD), incorporating 5.6 million hectares as 55 Locally elected leaders have more to gain from liquidating natural resources during their tenure, and buy votes or put funds in their pockets. This leads to incentives to condone or engage in illegal and excessive logging. This tendency exists even when there are external contracts, as then informal logging becomes their main source of additional income. Only very strong external checks and balances can counter this tendency. 80 Forest Management Units (FMUs), and a non-permanent forest domain (NPFD) that includes 0.4 million ha as community forests. 7.5 The PFD includes production and protection forests. About 100 logging concessions and 15 council reserves cover the bulk of production forests, while protected forests include national parks and wildlife reserves. The law mandates officially approved plans and management for both production and protected forests. As of 2009, about 65 concessions had approved plans, covering an area of about 4.5 million ha representing about 65 percent of the total concessions' area. 7.6 Timber harvesting also takes place in the NPFD, through logging titles, such as sales of standing volume (SSV) and timber recovery permits (TRP). The latter are short-term logging titles lasting from a few months up to a maximum of three years. 7.7 One of the primary factors leading to the loss of forest cover is agriculture, which accounts for more than 80 percent of the loss in forest cover in Cameroon (CBFP, 2006). This includes not only slash-and-burn shifting cultivation, long recognized as a key driver of deforestation in areas of dense forest, but also cash crops, most notably cocoa. The cocoa crop has been promoted by the state from independence up to the 1980s with considerable impact on forest integrity in Cameroon (The World Bank, 2009a). In addition, there are likely to be formal and informal incentives to engage in “charcoalâ€? production. The available information on this is patchy. 7.8 The bulk of the officially recorded annual timber production – an average of about 2- 2.2 million cubic meters over the last decade – is harvested in logging concessions inside the PFD. However, a recent estimate of the unrecorded production, largely occurring in the NPFD through small-scale, chainsaw logging, almost double official estimates. This places the total annual timber production at about 4.2 million cubic meters (Cerutti and Lescuyer, 2011). 7.9 Fuel wood harvesting, which is often linked to agriculture and associated population increase, also poses a threat to forest integrity. Cameroon produced an estimated 9.7 million cubic meters of fuel wood in 2009 (FAOSTAT, 2010), representing about 80 percent of its total official round wood production. The percentage decreases when unrecorded timber production is considered. 7.10 A full assessment of informal circuits should incorporate the incentives for both centrally appointed officials as well as “elected mayorsâ€? to capture the forestry resource rents. This can occur at various stages, including outside the formal channels, given opportunities and incentives, combined with a low probability of detection and punishment. 81 C. A DESCRIPTION OF FORESTRY TAXATION 7.11 Before 1998, forestry taxes were collected by the Ministry of Forests (MINFOF). 56 The 1997/1998 fiscal law, and related implementing decree, 57 transferred to the MINFI all the fiscal competences, including forest-related taxes of the MINFOF. This was designed to streamline revenue collection at the national level. But it required reconciliation of data on timber production and the collection of corresponding revenues. With that aim, in March 1999 a MINFOF-MINFI inter-ministerial program, the Forestry Revenue Enhanced Program (Programme de Sécurisation des Recettes Forestières,PSRF), was created within the MINFI to ensure a rigorous monitoring of fiscal revenues in the timber sector and to improve and increase its contributions to the state budget. The PSRF was restructured in 2001 to streamline and centralize the payments by forest companies of all forest-related taxes. However, collaboration between MINFI and MINFOF remains sporadic and weak. This leaves an information gap between production data (collected by the MINFOF) and fiscal data. 7.12 Private logging companies are divided for tax purposes according to their annual turnover (CFAF >1 billion reporting to the Direction des Grandes Entreprises or DGE; CFAF <1 billion reporting to the PSRF). In 2006, a further subdivision was introduced, assigning companies with an annual turnover between CFAF 100 million and CFAF 1 billion to the Centre des Impôts des Moyennes Entreprises (CIME), leaving only the very small forest companies (CFAF <100 million) declaring their taxes to the PSRF. 7.13 These administrative changes took their toll on the quality of data collection, as relevant agencies were not ready to collect data at the level of detail needed for meaningful statistics to be produced. In addition, responsibility for forestry revenues and management was dissipated, and coordination became difficult, especially between MINFI, MINFOF and the PSRF, as the DGE and the CIME (both reporting to MINFI) had no responsibilities vis-à-vis the MINFOF, while the PSRF remained an inter-ministerial programme. 7.14 In 2007, to improve the quality of data and coherence of analysis, development partners asked the MINFI to officially assign to the PSRF the task of collection and analysis of all data pertaining to the forestry sector.58 Thus, the DGE and the CIME had an obligation to report their data to the PSRF. The latter was made responsible for overall analyses on the sector. 7.15 In 2009, however, the Prime Minister’s Office instructed MINFI to reconsider the mission of its various programmes in the light of the new fiscal regime of the State (i.e., after reaching the completion point of the relief under the Highly Indebted Poor Country Program), with the objective to relocate each programme within its competent technical ministry, i.e. the MINFOF in the case of the PSRF. This instruction has not yet been 56 The ministry dealing with forest and forestry was created in 1992 with the name of Ministry of Forests and Environment, acronym MINEF. In 2004, the MINEF was divided into the Ministry of Forests and Fauna (MINFOF) and the Ministry of Environment (MINEP). 57 Decree 08/009/PM of 23rd January 1998. 58 Lettre n° 07/1313/CT/MINEFI/SG/C/MRP du 28 février 2007. 82 implemented, but adds to the confusion on the future mission of the PSRF, as well as affecting its present performance. Given the already weak collaboration, and the chronic distrust among different ministries, one might argue that the relocation of the PSRF within the MINFOF risks increasing the disconnect between the MINFI and the PSRF/MINFOF, with a likely negative impact on the quality and timeliness of data collection, exchange, and analysis. 7.16 To summarise, at the time of writing the DGE, the CIME and the PSRF are in charge of collecting forestry related taxes from different logging companies according to their annual turnover. The PSRF should then regroup data from the DGE and CIME, as well as from the MINFOF, analyse all data on due and paid taxes, and produce annual reports on the performance of the forestry sector. D. COLLECTED TAXES 7.17 In quantitative terms, forest-related taxes changed over the last two decades according to reforms in the fiscal system and in the forestry sector. The main changes, vis- à-vis taxes, inter alia, are i) the public auction system for the attribution of logging concessions and sales of standing volumes introduced by the 1994 law; ii) the (partial) 59 log-export ban introduced in 1999; and iii) industrialization induced by (ii). 7.18 The main source of revenue before 1998/1999 was the log export tax. Subsequently, the Annual Forestry Fee, or Redevance Forestière Annuelle (RFA) had become the main revenue source. It is calculated on the basis of bids made by logging companies and the surface of their titles (see below). In recent years, the export tax has reappeared while the RFA has declined. 7.19 For administrative purposes, forestry taxes are divided among specific and "generic" categories. Specific taxes refer to volumes harvested or exported and to harvested surfaces. These are the RFA, the stumpage fee (taxe d'abattage, TA), the sawmill entry tax (taxe d'entrée usine or TEU), the log-export tax (droits de sortie, DS), and the export surtax (surtaxe progressive until 2001 and surtaxe à l'exportation, SE, after 2001). The list of forestry taxes is shown in Table 7.1. 59 The 1994 legislation provided for a total log-export ban to be implemented five years after its adoption, but in 1999, the MINFOF decided for a partial ban. 83 Table 7-1: Forestry Taxes in Cameroon Name of tax Type of tax Subject to Calculation Scale Total taxation basis Collections 2010 Annual forestry Area-based Companies CFAF per Minimum price CFAF 5.21bn fee (redevance annual fee holding hectare plus logging forestière logging titles according to company’s annuelle)* type of bidding price: logging title a) Sales of Standing Volume titles: CFAF 1,000/ha plus bidding price b) Forest Management Units titles: CFAF 2,500/ha plus bidding price Stumpage fee Volume- All logs CFAF per m³ Zone 1 (closest to CFAF 5.43bn (taxe based fee declared to according to Douala): 3 d’abattage) authorities logging zone percent of each after and each logged species’ harvesting has species’ FOB FOB value. taken place value. Zone 2: 2.5 percent Zone 3 (furthest from Douala): 2 percent Sawmill entry Volume- All logs CFAF per m³ 2.5 percent of CFAF 0.59bn tax (taxe based tax entering the each species’ FOB d’entrée usine) sawmill value Log-export tax Volume- All timber CFAF per m³ 17.5 percent of CFAF 7.99bn (droits de based tax exported in and species’ the species’ FOB (total export- sortie) log format value value related Export surtax Volume- All timber CFAF per m³ a) Ayous: CFAF taxes) (surtaxe à based tax exported in and species’ 4,000/m³ l’exportation) log format value b) Species of primary category: CFAF 3,500/m³ c) Species of secondary category: CFAF 500/m³ * Part of proceeds goes to decentralized local governments 7.20 Generic taxes, on the contrary, are not related to volumes or surfaces, but cover the taxes that all companies have to pay to the Treasury irrespective of their domain of operations. These include, but are not limited to the VAT, (Taxe sur la valeur ajouté); corporate income tax (impôt sur les societies, IS); wage tax (retenue sur taxe salariale, TS); the property tax (revenue sur capitaux mobiliers, RCM); and the PA (précompte sur achats). 84 The RFA 7.21 As discussed in previous chapters, the RFA is an area-based fee. According to the rules set by the 1994 Forest Law, long-term “Forest Concessionsâ€? awarded on defined Forest Management Units (Unités Forestières d'Aménagement, UFA) and short-term Sales of Standing Volume (Ventes de Coupe, VC), two of the most important logging titles with regard to volume logged, must be allocated through an auction. The bidding price is made up of two parts: a minimum price set by the administration, plus the company’s offer. The minimum bidding price is set at CFAF 2,500/ha for VC, and CFAF 1,000/ha for UFA. The total RFA due annually is thus the concessionaire’s bidding price times the entire surface (in ha) of the allocated logging title. The taxe d'abattage (TA) 7.22 The taxe d'abattage or TA is a volume-based tax that must be paid by all logging companies on the volume (cubic meters) per species declared to the administration, after harvesting has taken place. To apply the rate of TA to each logging title, and to account for different transportation costs, the country has been divided into three “logging zonesâ€? to which different rates apply. In Zone 2, covering central Cameroon, the felling tax is calculated as 2.5 percent of each logged species’ FOB or default value. In Zone 1 (region nearest to the port of Douala), the TA rate is 3 percent, while in Zone 3 (furthest from Douala), it is 2 percent. The taxe d'entrée usine (TEU) 7.23 The taxe d'entrée usine (TEU) was introduced in 2000 to replace the export tax on processed products (prélèvement à l’exportation applicable aux bois ouvrés et semi-ouvrés des unités de transformation soumises au régime du droit commun). TEU is also a volume- based tax, on a per species basis, on logs entering the sawmill, on the basis of 2.25 percent of each species’ FOB value. The droits de sortie (DS) and the surtaxe à l'exportation (SE) 7.24 The log-export tax DS (droits de sortie) is calculated as 17.5 percent of the FOB value. Until 1999, the DS accounted for much of the forestry sector’s revenues. In 1999, the partial log-export ban targeted some of the most valuable exported species and the DS rapidly decreased. In recent years it increased again, especially as a consequence of the crisis, as the market for logs decreased less than that for processed products. In 2000/2001, the Finance Law also introduced the surtaxe à l’exportation (SE). The levy applies to all timber exported in log format and it is split into three different types: the first is made up of ayous (Triplochiton scleroxylon, the most exported species for many decades), with a surtax of CFAF 4,000 per cubic meter exported, and the other two sets regarding species of the primary and secondary promotional categories, with surtaxes of CFAF 3,000 and CFAF 500 per cubic meter exported, respectively. 7.25 It is worth noting that with the reduction in the RFA following the international downturn in 2008, to protect industries, the export tax was reintroduced. This led to a reduction in the RFA (shared with local governments) relative to the export tax. 85 Chart 7:1: Specific and Export-Related Taxes, 1998-2010 (CFAF billions) 30 TA TEU 25 RFA Total exports related 20 15 10 5 0 98/99 99/00 00/01 01/02 02/03 2004 2005 2006 2007 2008 2009 2010 Sources: CIFOR E. DESIGN AND MANAGEMENT OF THE RFA 7.26 The forestry taxation system in Cameroon is fairly centralized—with collection and use determined by the central government. This led to conflicts and tensions between the decentralized entities and local communities on the one hand, and the logging companies and the central government on the other. Prior to 1994, very little of the forestry revenue benefitted the local communities, which also bore the costs of the logging and deforestation. The RFA was a partial response to the tensions, and stipulated a share to be redistributed to the decentralized administration (municipalities) and rural villages. 7.27 Thus the RFA, which was one of the key provisions of the 1994 Forestry Law, 60 was introduced with three main objectives, to: (i) Make a consistent contribution to the State budget; (ii) Contribute to poverty alleviation; and (iii) Enhance equity in the redistribution of forest-related resources. 60 The legal framework of reference for the RFA consists of, but is not limited to: i) Article 68 of Law No 94/01 (January 20, 1994) establishing forestry, wildlife, and fisheries regulations; (ii) Article 122 of Decree No 95/531 (August 23, 1995) establishing the procedure for implementing Cameroon’s forestry regime; (iii) Article 10 of Decree No 98/009/ PM (January 23, 1998) establishing the tax base and procedure for recovery of duty, fees, and taxes related to logging activities; (iv) Joint Arrêté No. 000122/MINEFI/MINAT (April 28, 1998) establishing the procedure for the use of logging revenue intended for neighboring village communities, replaced by Joint Arrêté No. 0520/MINATD/MINFI/MINFOF (July 28, 2010), establishing the procedure for the use of logging and hunting revenues intended for neighboring municipalities and communities; (v) Law 2009/019 of 15 Dec 2009, on fiscalité locale. Worth mentioning also is Memorandum No. 2978/MINEF/ DF AP/AC (October 14, 1999) concerning management of revenue from wildlife, as well as several successive finance laws from 1995 to 2000. 86 7.28 Although RFA funds are distributed to subnational entities, the amounts are not clearly defined and have been changed without warning by the center. Moreover the funds are not entirely under the discretion of the local and regional governments, as described in the chapter on the legal framework, and there is an expectation that these will be used to finance deconcentrated operations in health and education that would normally, in other parts of the country, be financed directly by the center. 7.29 Indeed, as seen in Chart 7.1 above, since the crisis in 2009, there has been a shift from the RFA to export taxes, and this more or less offsets the revenue losses incurred through the 50 percent reduction in the RFA. However, the export taxes accrue entirely to the central government. Hence, a consequence of the measures is that there has been a centralization of forestry resources, leaving the local governments to directly bear the effects of the stabilization. Distribution of the RFA 7.30 Until June 2010, the redistribution of the RFA took place on a 50/40/10 ratio, i.e., 50 percent of the RFA was allocated to the Treasury, 40 percent to municipalities where the logging titles were located, and 10 percent to rural villages adjacent to the concessions. While the 40 percent is incorporated into municipal budgets, the use of the funds is determined through a committee (described in the legal chapter above) that overrides local controls, the 10 percent to villages can only be spent on projects approved by advisory boards, which include the representatives of the central government (MINFOF) and the prefects. However, given the complex nature of these boards, mayors are able to circumvent the requirements, leading to additional possibilities of “capture.â€? 7.31 Over the 2000–2009 period, about €180 million was paid by logging companies as RFA, which means that about €90 million was redistributed annually (40 percent + 10 percent), to an average of about 50 rural municipalities with about 1.1 million citizens.61 While these revenue flows can be considered significant, local development impacts were slow to materialize given local governance deficiencies. Studies based on fieldwork in rural municipalities and their consistent villages for the period 2000-2004 (Morrison et al, 2010) suggest that of the US$7 million allocated to the municipalities, US$2 million was not accounted for; and of the US$1.7 million allocation to the villages, US$1 million was unaccounted for. Moreover, the projects funded were not those requested by the village representatives and the costs were higher than the accepted norms for such projects funded by local sources. 7.32 The municipality’s nominal share was further reduced when the RFA sharing rule was updated to a 50/20/20/10 ratio following the law on fiscal decentralization adopted in 2009.62 In June 2010 a Joint Arrêté 63 clarified the way the RFA (and other taxes) were to be managed. The new ratio ensures that 50 percent of the RFA remains allocated to the Treasury, and 20 percent goes to FEICOM, or "any other body in charge of the 61 Data for population are only indicative, as they are not updated after the 2005 census. 62 Law 2009/019 of 15 December 2009 on local tax system. 63 Joint Arrêté No. 0520/MINATD/MINFI/MINFOF (June 3, 2010), establishing the procedure for the use of logging and hunting revenues intended for neighboring municipalities and communities. 87 centralization and equalization of taxes and fees due to municipalities," 20 percent to municipalities, and 10 percent to rural villages adjacent to the contract zones. How the RFA is paid 7.33 The RFA is paid quarterly by logging companies. The amounts due (normally per municipality) are calculated by the MINFOF on the basis of the companies' bids and UFA's surfaces, and then communicated to the PSRF. There still remain some inconsistencies in the data produced by MINFOF, especially with regard to the last set of municipalities created in 2007, which had no clear boundaries established by the Ministry of Territorial Administration and Decentralization (MINATD), and thus no official surfaces on which to redistribute the RFA. Overall, however, in recent years the quality of the basic data produced by MINFOF on the expected amounts of RFA has been improving. 7.34 The rules for the payment of the RFA changed a few times over the last decade. Until April 2007, the RFA was paid by logging companies by issuing three different cheques in each quarter: with 50 percent of the total amount sent directly to the state treasury, while the other two, delivered to the MINFI were 40 percent for concerned rural municipalities and 10 percent for villages neighbouring concessions. 7.35 In April 2007, the MINFI implemented a 2004 decision requiring rural municipalities to open two ad hoc bank accounts (one for funds destined for municipalities, and another for funds destined for villages) into which logging companies must directly transfer the RFA amounts (40 and 10 percent at the time, and 20 and 10 percent to municipalities and related villages respectively since 2010). 7.36 At the time of writing, almost all concerned municipalities opened special bank accounts, although it remains to be clarified how many municipalities opened two bank accounts, and how many opened only one account. As argued in this report, the multiplicity of bank accounts may complicate rather than facilitate tracking and controls, and is not in accordance with international best practice. Role of FEICOM 7.37 Since the RFA’s inception, its redistribution among all municipalities over the national territory has been on the national political agenda as an issue of inter-regional equity. Such a concern is a long-standing one between the center and decentralized municipalities and villages. In some cases, discussions focus on the inequitable distribution between municipalities that do not border logging titles paying the RFA (i.e. UFA and VC) but are nonetheless located in forested areas, such as those bordering protected areas (on which the RFA is not paid). In other cases, the focus is on municipalities that obtain only very limited amounts of RFA (and/or have hundreds of villages) versus municipalities that because of their territorial size and their location in remote regions (notably the East) receive large amounts of RFA to be redistributed among a small number of villages. 7.38 In 2000, when the MINFOF started to auction an increasing number of UFA and VC, and larger amounts of RFA were redistributed, the equity issue received so much political attention that the 2000/2001 fiscal law included provisions for an equalization fund (fond 88 de péréquation) to redistribute the RFA across a larger number of municipalities, including non-forested ones. However, at the time there was no political consensus built around the fund, as many mayors, notably those receiving large amounts of RFA in the East and South, vigorously opposed the idea of the equalization fund. Their argument was that if the money were to be redistributed on a population basis, as in the case of the CAC, their municipalities and villages would have never developed, as they were the least populated but also the least developed of the country. Consequently, the fund never went beyond its introduction in the fiscal law. In other words, the decree that should have identified the manager of the fund and laid down its practicalities ("les modalités de fonctionnement du fonds de péréquation sont fixées par voie réglementaire") was neither discussed nor adopted. 7.39 Over the years, the regional distribution and equity issue resurfaced time and again in debates and discussions around the forestry sector, and recently it acquired a new strength in the light of the law on fiscal decentralization adopted in 2009, but also after the drastic measures (among which the reduction by 50 percent of the RFA) proposed by the MINFOF and accepted by the Prime Minister’s Office. This was to help the forestry sector face the 2008 financial and economic crisis. In December 2009, the law on fiscal decentralization established that half of the 40 percent of the RFA received by each municipality (i.e., the 40 percent in the 50/40/10 ratio) must be recentralized for redistribution to all municipalities in the country ("la quote-part communale de la RFA est répartie ainsi qu'il suit: (i) 50% au titre de la retenue à la base au profit de la commune de localisation; (ii) 50% au titre du reliquat centralisés par le FEICOM ou tout autre organisme chargé de la centralisation et de la péréquation des produits des impôts, taxes et redevances dues aux communes"). Thus the ratio became 50/20/20/10. It is not evident why this fund should be financed from the RFA and not general revenues per se, as the current arrangement could increase the incentive to engage in illegal or informal logging that would not be shared with non-forest zone areas. 7.40 As in 2000/2001, the redistribution of the RFA was included in the 2010 fiscal law, but the practical modalities of redistribution were not clarified, including which municipalities would be eligible. An implementing decree is needed to make it fully functional. However, the major difference with the aborted 2000/2001 fund is that, although the implementing decree is still missing, the new 50/20/20/10 ratio does have a legally designated manager (i.e., FEICOM) and has been implemented by MINFI since 2010. In other words, the 20 percent for the FEICOM is already being deducted from the RFA due to municipalities and transferred into FEICOM's coffers. 64 Thus, until FEICOM explicitly sets the rules for the redistribution of the RFA (20 percent), the whole process started in 2009 can be described as just a recentralization of part of the RFA, with a lack of clarity as to how it is to be used. 64 Although some problems occurred during 2010, especially due to logging companies unaware of the new RFA ratios (50/20/20/10) and thus still transferring money (or checks) to the MINFI or to municipalities and communities, the MINFI reports that the situation has been improving and the amounts directly transferred to the FEICOM increased during the year. However, the situation of the RFA redistribution risks deteriorating in 2011. The fiscal law (and Joint Arrêté 0520) establishes a charge of 10 percent on all revenues entering the Treasury coffers and destined to decentralized municipalities. 89 7.41 As in the past, mayors of municipalities presently receiving the RFA are opposed (see also below) to the recent changes introduced. This is because their allocation of RFA is further reduced – both by the new 50/20/20/10 ratio and, at least in 2009 and 2010, by the 50 percent reduction of RFA because of the economic crisis. Mayors also complain that the whole process, in the name of decentralization, took place without effective consultation, and further restricts their ability to control their budgets. 7.42 Quantitatively, if the promise of MINFI and MINFOF to rescind in 2011 the special fiscal measures taken in 2009 and 2010 to cope with the financial crisis, the RFA amounts should increase back to pre-2009 levels, i.e., about CFAF 12 billion (although this may vary a bit as several concessions have recently been abandoned). This means that FEICOM should receive about CFAF 2.4 billion each year. The RFA allocated to the rural municipalities 7.43 Under the new rules, the sums transferred to rural municipalities (20 percent of the total RFA paid by the companies) are placed under the supervision of a Council Committee (Comité Communal), composed of seven members, one president (elected by the Committee members), one vice-president (elected by neighbouring communities) and a rapporteur (the mayor or a delegated person). Joint Arrêté N° 0520 fixes at 20 percent (maximum) the amount of the RFA received by any given municipality that could be used to top up the running costs of the municipality (appui au budget de fonctionnement), while the remaining 80 percent (minimum) must be invested (investissements). Investments must be made on the projects approved in the Council Development Plan, and the municipality also audits its implementation. 7.44 The mayor authorizes (ordonnateur) the expenses approved by the Council Committee, and s/he must prepare an annual administrative report of all operations to be approved by the Committee. The municipal tax collector (receveur municipal) is the financial agent (agent financier) of the Committee. As such, s/he is responsible for tax recovery (e.g., controls that bank transfers by logging companies are regularly made) and for the payment of services rendered to the municipality. 7.45 Contrary to the past, Joint Arrêté N° 0520 also explicitly mentions that forestry and hunting revenues must have an ad hoc budget line in the municipalities’ budgets (art. 14(4)). The RFA allocated to the communities 7.46 For the amounts (10 percent) allocated to the communities, Arrêté N° 0520 requires the creation of a Community Committee (Comité Riverain) in each municipality. The committee is in charge of preparing annual working plans with a detailed description of the projects to implement in their respective budgets. These plans are then transmitted to the Council Committee for approval. 7.47 The mayor and the municipal tax collector assume the same roles as for the Council Committees, i.e., ordonnateur and agent financier, and after a project is approved, the mayor issues a call for tenders for the work needed to realize it. After the project is 90 finished, the municipality approves the disbursement of funds to pay for the winning tenders. The collector of municipal taxes or the collector of the public treasury signs the check and the procedure is closed. 7.48 Joint Arrêté N° 0520 fixes at 10 percent (maximum) the amount of the RFA received by any given Community Committee that could be used to top up the running costs of the committee, while the remaining 90 percent (minimum) must be invested in social and economic activities (with a list of eligible activities provided in the Arrêté). The management of the RFA 7.49 Over the past decade, the RFA has become a major source of revenue for many rural municipalities. The dominance of the RFA in the municipalities’ budgets is obviously more pronounced for municipalities that have not developed alternative economic activities other than those related to timber exploitation. For instance, in a sample of eight municipalities located in the East Region of Cameroon, receiving about €4.8 billion of RFA in 2008 (i.e., about 25 percent of the total redistributed), the RFA ranged between 52 percent and 95 percent of total annual revenue (Table 7.2) Table 7-2: RFA and Municipalities’ Budgets Municipality Total of municipal RFA (in K€) % RFA in the revenue (in K€) municipal budget 1 1,250 1,183 95% 2 663 545 82% 3 839 629 75% 4 698 479 69% 5 482 333 69% 6 353 214 61% 7 2,202 1,221 55% 8 278 147 52% Source: CIFOR 7.50 The RFA constitutes a very significant potential for many municipalities to promote socio-economic development, although the effectiveness of its use has been questioned in much of the available literature. It has been difficult to estimate how the RFA amount was used at the municipal level prior to 2010 for two main reasons. First, the municipality’s elected members were usually very reluctant to provide the municipalities’ accounts, even if they are public documents. Second, even when accounts were made available, a significant part of the RFA was used for expenses, both tangible and intangible, which were often not traceable. In this sense, the obligation for Council Committees to record the RFA under a specific budget line, as introduced by the new Joint Arrêté N° 0520, should help improve transparency in the implementation by municipalities. 7.51 Overall, after more than a decade of RFA redistribution, several problems remain in the basic management of the municipalities’ budgets, with direct impacts on the effectiveness of expenditures. Often, there are compelling reasons for those problems. 91 7.52 For instance, one cannot underestimate the appropriate technical skills needed to manage a budget, with the additional accounting and reporting requirements of the RFA. Such skills are generally not available at the municipal level. 7.53 As mentioned in the legal chapter, the lack of clarity in the design exacerbates the difficulty in arriving at appropriate policy actions—for instance regarding investment and growth. Mayors often claim that it is not clear which activities could be funded from the RFA (Cerutti et. al.,2010). Arrêté N° 0520, links the use of the RFA to projects approved in to the municipal development plan and should improve the long-term effectiveness of the RFA. But this cannot be taken for granted. Indeed, how good these plans are and whether they amount to a vision for appropriate utilization of growth linkages is a much deeper issue than can be resolved at the level of individual municipalities. 7.54 In fact, rural municipalities often also lack the long-term vision needed for stipulated investments to be made, especially because longer gestation projects are not necessarily compatible with the five-year mandate of the mayors and the committees. And notwithstanding the good intentions that led to the adoption of the Joint Arrêté N° 0520, strategic development plans are being implemented in only a minority of municipalities. 7.55 Also, since the RFA started pouring into the municipalities’ coffers, several expenses that were previously covered by the budgets of concerned ministries, such as health, education, and associated infrastructure, had to be covered by the municipalities’ budgets because of faltering State resources. The substitution of the RFA for functions that should be covered by central funds further circumscribes the autonomy of the local governments, and makes the local budgets very tentative. However, the management of local operations becomes even more complicated if the RFA itself becomes volatile and uncertain. The volatility in the RFA 7.56 The RFA is paid by logging companies for the two larger logging titles, UFA and VC. The latter have higher minimum prices (CFAF 2,500/ha) because they are short-term titles and do not require the preparation of a management plan. The law provides for the ministry to periodically adjust the rates, especially in response to inflation or to particular economic conditions, but this has never occurred. 7.57 Overall, VC obtain higher bidding prices than UFA. The average price paid for 169 VC attributed between 2000 and 2008 is about CFAF 13,900/ha, while the average price of the 93 UFA attributed to date is about CFAF 2,500/ha. At least a couple of reasons could explain the difference. First, UFA are attributed for a period of 15 years, are renewable once, and can apply up to 200,000 ha, while VC have a maximum area of 2,500ha and must be harvested within three years. Second, UFAs require companies to make further investments to guarantee that their harvesting operations are sustainable (e.g., prepare a management plan with annual logging surfaces or engage in forest certification schemes). 92 In 2008, the RFA collected totalled about CFAF 12.7 billion for an area of about six million ha, including UFA and VC. 65 7.58 Two major issues could affect the RFA amounts collected annually. The first is more political in nature, while the second is largely linked to the forestry activities on the ground. 7.59 With the 50 percent reduction in the RFA that companies had to pay in 2009 and 2010, amounts collected decreased to about CFAF 5.7 billion. Rather than assuming the stabilization function, the central government maintained the sharing proportions inflicting further cuts on the concerned municipalities’ budgets. The cuts were not announced, or incorporated in the local budgets and municipalities only learnt about them as the inflow of funds was reduced. Both the MINFI and the MINFOF reportedly agree that in 2011 the RFA will go back to its normal level, 66 the uncertainty may induce local officials to run deficits or engage in informal activities. 7.60 The second reason affecting the volatility of the RFA relates to forestry operations and the logging titles granted. The RFA increase from 2000 to 2006 was due to the logging concessions in those years, and the increase in concessions is not likely to be maintained. 7.61 By the end of 2006, all available UFA had been attributed at least once and no major increase in the total RFA collected from UFA should be expected in the future, unless some new major UFA are created and auctioned.67 It is likely that the RFA amounts from UFA will likely decrease in the future, for several reasons. First, each year some companies default and others abandon their UFA. Several UFA were attributed to the highest bidders, but they were not necessarily those ready to engage in long-term sustainable forest management. In those cases, the winner exploits the UFA for a few years paying a relatively large annual amount of RFA, then abandons the concession, which is generally re- attributed for a lower price. For instance, the average price paid for six UFA abandoned in 2007 and later re-attributed, decreased from about CFAF 6,300/ha to about CFAF 1,800/ha, resulting in a decrease in the projected RFA of about CFAF 1.8 billion per year. In 2008 and 2009, about eight more UFA were abandoned, and their RFA will likely decrease when the ministry reallocates them. 7.62 Over the longer term, variations of revenues from the RFA may depend on the policy adopted by the MINFOF for VC, as well as decisions on other logging titles. For more than a decade after the adoption of the 1994 forest law, the stated policy was to grant VC until all UFA had been auctioned. This strategy made sense at the time, and was reportedly adopted in order to maintain a fairly regular timber production and tax income over the years (through the RFA), while progressively moving the production away from titles that did not have sustainable forest management requirements, such as the VC, while increasing the number of UFA auctioned. 65 Although six million ha is the area on which the RFA is paid, it does not represent the annual area with harvesting operations. The latter is much smaller as logging companies must respect annual logging surfaces of th about 1/30 of the total area of their UFA. 66 This appears to have happened during 2011. 67 There remains an area of about 800,000ha in the East Region of Cameroon that has been earmarked for auctioning. 93 7.63 Official planning documents from the MINFOF in 1999 and 2004 envisaged that timber production from VC would cease by 2007. This has not happened. Ventes de coupe decreased from about 130 in 1997/1998 to about 15-20 in 2007/2008, but by 2009 the trend had actually been reversed, with the ministry granting about 40 VC and planning more auctions for 2010 and 2011. 68 7.64 Also, it is worth mentioning that the number of VC in recent years has been increasing in parallel with the illegal granting of other logging titles, notably Timber Recovery Permits (Authorisations de Recuperation de Bois, ARB), 69 which are granted in a discretionary manner and do not pay the RFA. Although they require the same industrial capacity to be harvested, they often cover similar surfaces to the VC, and are more often than not granted to the same companies. 7.65 Apart from governance considerations, ARBs represent a clear loss of revenue for the Treasury and, in theory, also a potential loss to rural municipalities. 70 If ARB are indeed being used as a replacement of, or together with, VC, it would be wise for the MINFI to exert pressure on the MINFOF to auction them and have them pay the RFA as for VC.71 7.66 For municipalities receiving the RFA and heavily dependent on it, especially at a time when decentralization is increasing, it is of the utmost importance for the MINFOF and the MINFI to include concerned mayors and Council Committees in all discussions related to future decisions with impacts on the RFA. After more than a decade of redistribution, municipalities have developed a dependency on external sources of revenue, such as the RFA, and this may weaken the municipalities’ incentives to diversify over the longer term. When the RFA was halved in 2009 and 2010 without any prior notice or discussion with the mayors, for instance, many municipalities stopped covering the “central mandatesâ€? in education and health, which were almost entirely funded through the RFA. 7.67 Most of the spending from the budgets of municipal governments is directed towards the operations of the mayor and the administration (almost 60 percent) and the rest towards some sort of projects or investments. The operational spending is composed of various items, determined by the mayor and difficult to specify ex ante, and salaries (Cerutti et al., 2010 and Ndjanyou and Majerowicz, 2004). Almost two thirds of the spending for investments is for the benefit of the mayor (construction of offices, purchase and upkeep of vehicles…) and the rest towards small infrastructure in roads and health care at the instance of the mayor (Nzoyem et al. 2003, Ndjanyou and Majerowicz 2004). 68 72 VC were granted in 2011. 69 These are also commonly referred to as petits titres. 70 This is still a hypothesis. Preliminary results of a recent research we are conducting seems to suggest that the local population does prefer ARB over titles that pay the RFA, and this is because the companies with ARB commonly pay a volume based tax in situ (i.e. cash delivered directly to the population). 71 Take the Lom Pangar dam as an example, with an expected volume of about 1 million cubic meters. A procedure to grant about 40 ARB (about 80,000ha) was started in 2009 but then not finalized by the MINFOF. If those surfaces were granted as VC (or as ARB via a transparent auction), the Treasury may collect larger revenues than through a discretionary attribution. 94 7.68 The new Arrêté 0520 imposes a maximum of 20 percent towards the operational budget, and 80 percent at the minimum for investments. But it does not specify what constitutes investments, and together with poor accounting practices and little or no incentives for accountability, this opens a big loophole for the funds to be “capturedâ€? by vested interests. F. A NEW DESIGN FOR ASSIGNMENT AND SHARING OF FORESTRY REVENUES 7.69 It is clear that an efficient design of the assignments and sharing of forestry revenues will have a major impact on the poorest groups in Cameroon, generation of development in some of the most backward areas of the country, and incentives to limit or enhance deforestation and climate change. In this regard, what happens in Cameroon will also be relevant for other countries facing difficulties with the management of tropical forests. And with some modifications to address non-renewable resources, the underlying logic and models could become relevant for the treatment of other natural resources in Cameroon, and elsewhere. Enhancing accountability 7.70 From the recent experience, it appears that the central government continues to think of the RFA as effectively a “centralâ€? resource to be used largely for “deconcentrated functions,â€? albeit run through the municipalities and local governments. Pretending that the additional resources accrue to the local and municipal governments and then operating as if the resources were still in the central domain is self-defeating. The reductions in the RFA have an impact on local service delivery, even though they are mostly provided by deconcentrated organs. It is not just the RFA, but also funds from FEICOM that are not allocated as predicted on a reasonable assumption of likely volume of logging. Indeed, this might increase the incentives for mayors and local officials to divert or “captureâ€? the resources. Communities see few benefits, and may be tempted to engage in illegal or informal logging and charcoal activities, as this would be one of the ways to make ends meet in a desperately poor part of the country. 7.71 It is clear from the administrative accounts, that the bulk of service provision in the forest region appeared to be through the deconcentrated functions, and the RFA share for the forest municipalities was more than they needed for their meager responsibilities. Thus the temptation to divert, or misuse the resources was quite strong (we have noted the recent jailing of the mayor and treasurer of Yokadouma for embezzlement that predates the ad hoc reductions in the RFA). Reducing the RFA—even if much of the service delivery is by deconcentrated agencies—leads to resentment, as has been seen since 2009. The local population feels that their natural resources are being “exploitedâ€? by other regions. And, as mentioned in the chapter on the budget framework, resources flowing to localities (deconcentrated or decentralized) should not fluctuate, since they finance essential public services. Thus, it is likely that the existing asymmetric allocations on the revenue side will remain for political economy reasons. 7.72 To the extent to which the natural resource sharing arrangements benefit from higher revenue shares, for instance because of shared royalties, taxes or rents, these jurisdictions should also be given greater spending responsibilities to match their higher 95 resources. Otherwise, these resources are likely to be wasted. Thus, for local governments within the forest zone that benefit from the RFA, there should be an explicit and full functional transfer of responsibilities that might be greater than in other regions—including say for all primary and secondary education and basic health care. Central government responsibilities in terms of coordination and management of externalities would remain, consistent with subsidiarity principles. Worries about capacity constraints are probably justified, although the additional resources available should ease the hiring constraints so that qualified personnel can be hired. Without this asymmetry in spending, the asymmetry in revenue sharing could lead to a misallocation of resources. The minimum standards that must be met for these asymmetric functions in the forest region could also be specified, along with sanctions for non-fulfillment. 7.73 For mayors, clarity in the responsibilities that they are accountable for is crucial. And for electorates to hold them responsible, it is necessary to have clarity regarding the funds to be received through the “sharingâ€?, plus an element of “own-source revenuesâ€? that makes them aware that their resources or time are being used to some end and that they should demand to see the benefits. This puts greater emphasis on the budgeting and reporting mechanisms that also need to be standardized so that inhabitants of one locality can judge how their mayors and elected governments are performing relative to others. 7.74 Aerial/satellite monitoring of logging is now technologically feasible (as in Indonesia). This can lead to greater accountability of the local officials, both deconcentrated and decentralized. While this is a subject that requires additional research, some possible components of the analysis are described below. Political economy 7.75 The incentives facing mayors and their electorates will influence whether or not there will be a tendency to “evadeâ€? the formal logging channels. On the one hand, a larger share in the rents accruing to the local governments would induce them to preserve the sources of the rents, provided there was better accountability. But in a game-theoretic framework, as long as there are rents to be had, increasing the “revenue-shareâ€? to local governments may just act to increase the “reservation paymentsâ€? demanded from informal activities—hence ironically may increase the incentive to engage in the less controlled charcoal activities or illegal logging, which could lead to more rapid degradation of the forest with a deleterious effect on the integrity and sustainability of the forest resources. On the other hand, excluding local governments and residents from a share of resources generated leads to resentments, and probably also to illegal logging. Thus, small moves in the “right directionâ€? may still enhance the deleterious effects, and the overall set of incentives needs to be carefully modeled. 7.76 Revenue-sharing on an origin basis is appropriate for natural resources, so that the local electorates are assured that a share of what is perceived as local endowments directly benefit the regions and localities. This is important for political economy purposes, and for minimizing centrifugal forces, provided that the funds to be shared actually materialize and lead to an improvement in local service delivery. 96 7.77 The use of own-source revenue at the margin 72 is essential for accountability and hard budget constraints, as the mayors and local governments must be held accountable to electorates and markets in a meaningful way, especially if there is recourse to borrowing for investment. Some of the options for own-source revenues are described in the sections above. Reaching the most vulnerable 7.78 But can the government ensure social protection for the poorest/most vulnerable groups? Quite often these are minorities even in the localities in which they live, such as Baka communities, and they may not have many advocates or sympathy at the local level— in the municipal councils, or in the office of the mayor. This is an important subject for both analytical and empirical research. 7.79 In this context, direct programs of support may be needed from the central government or international community. But the problem lies in ensuring that the earmarked funds actually get to the target population—and this may be difficult with the weak information flows and budgeting and accounting frameworks that have been observed in the Cameroonian context. Options that have been suggested is to tie future resource flows to observable outcomes—not necessarily related to own-source revenue generation (that will by definition be somewhat limited). Further, the potential sanction of cutting off future flows of funds must be credible—for if it is not, the recipient municipalities will engage in a “gameâ€? with the center or the donors and there will be little palpable improvement in the conditions of the vulnerable groups (see Ahmad and Martinez, 2010). Non-renewable resources 7.80 The analysis of this section could easily be extrapolated to the treatment of non- renewable resources, and perhaps apply with even greater force. This is because revenue- sharing remains important for the regions that may lose “theirâ€? resources in short order. Again, a combination of revenue-sharing, without forcing price fluctuations to feed down to the local levels, together with own-source revenues will be critical to ensure accountability with the use of natural resource revenue funds. Improved governance and information flows 7.81 In virtually all cases there is a premium on improving the budget and reporting frameworks and ensuring that the possibilities of leakage are minimized. The full flow of information is essential on the part of all parties, particularly the central government and its agencies, including especially FEICOM. Similarly, the local governments have to be held to a tighter standard on reporting on funds generated from the center as well as the local communities, in order for the political sanctions and operations to be fully effective. 7.82 Cameroon, as discussed above, has a long way to go in this respect. However, some important steps in the right direction have been taken recently. 72 As explained above, this involves the control over the rate or base of a revenue source by the local government, so that “at the marginâ€? it is able to increase revenues as needed. 97 References Ambrosiano and Bordignon, 2006, “Normative Versus Positive Theories of Revenue Assignments in Federationsâ€?, in Ahmad and Brosio (2006), op cit. Ahmad, Ehtisham and Ali Mansoor, 2002, “Indonesia: Managing Decentralization,â€? in Ehtisham Ahmad and Vito Tanzi, 2002, Managing Fiscal Decentralization, Routledge. Ahmad, Ehtisham and Giorgio Brosio, eds., 2006, Handbook of Fiscal Federalism, Edward Elgar. Ahmad, Ehtisham and Bob Searle, 2006, “On the Implementation of Transfers to Sub- national Governments,â€? in Ahmad and Brosio (2006), op cit. Ahmad, Ehtisham and Giorgio Brosio, eds., 2009 Does Decentralization Enhance Service Delivery and Poverty Reduction?, Edward Elgar. Ahmad, Ehtisham and Leo Martinez, 2010, “Intertemporal competition and aid,â€? Economics and Politics. Ahmad, Ehtisham and Mercedes García-Escribano, 2011, “Constraints to Effective Fiscal Decentralization in Peru,â€? in Jorge Martinez-Vasquez and Francois Vaillancourt, eds., Global Perspectives on the Obstacles to Fiscal Decentralization, Edward Elgar. Ahmad, Ehtisham and Giorgio Brosio (forthcoming); “On the design of transfersâ€?, Working Paper, LSE. Ahmad, Ehtisham, 2011, “Should China revisit the 1994 fiscal reforms?â€? Working Paper, ZEF University of Bonn; and Asia Research Center, London School of Economics, also Chinese Journal of Reform, 2011 Bardhan, Pranab, 2006, “Decentralization and Developmentâ€?, in Ahmad and Brosio (2006), op cit. Cerutti, P.O., Lescuyer, G., Assembe Mvondo, S. and Tacconi, L., 2010, “The challenges of redistributing forest-related monetary benefits: a decade of logging area fees in Cameroonâ€?, International Forestry Review, 12(2): 130-138 Congo Basin Forest Partnership (CBFP), 2006, Les Forêts du Bassin du Congo – Etat des Forêts 2006, Kinshasa, Democratic Republic of Congo Cerutti, P.O. and Lescuyer, G., 2011, “Le marché domestique du sciage artisanal au Cameroun: état des lieux, opportunités et défisâ€?, CIFOR Occasional Paper, Bogor, Indonesia, Center for International Forestry Research (CIFOR) 98 Dafflon, Bernard, “The Assignment of Functions to Decentralized Governments: from Theory to Practiceâ€?, in Ahmad and Brosio (2006), op cit. Devarajan, Shanta, Stuti Khemani and Shehkar Shah, 2009, “The politics of partial decentralization,â€? in Ahmad and Brosio (2009), op cit. Government of Cameroon, Growth and Employment Strategy Paper, 2010-2020, August 2009 International Monetary Fund, September 2011Country Report No. 11/266, Cameroon:Article IV Consultation – Staff report, Washington DC Lescuyer, G., Ngoumou Mbarga, H., and Bigombe Logo, P., 2008, “Use and misuse of forest income by rural communities in Cameroonâ€?, Forest, Trees and Livelihoods, 18:291- 304 Morrison, K., Cerutti, P.O., Oyono, P.R. and Steil, M., 2009, “Broken Promises: Forest Revenue-Sharing in Cameroonâ€?, WRI Forest Note, Washington D.C., World Resources Institute (WRI) National Institute of Statistics, Conditions de vie des populations et profile de pauvreté au Cameroun en 2007, Rapport Principal de l’ECAM3, Décembre 2008 National Institute of Statistics, Conditions de vie des populations et profile de pauvreté au Cameroun en 2001, Rapport Principal de l’ECAM 2, Août 2002 Ndjanyou, L. and Majerowicz, 2004, “Actualisation de l’étude sur la fiscalité décentralisée du secteur forestier camerounais“, Boulogne, France, Institutions & Développement Nzoyem, N., Sambo, M. and Majerowicz, C.H., “Audit de la fiscalité décentralisée du secteur forestier camerounais“, Boulogne, France, Institutions & Développement Rapport National sur les objectifs du millénaire pour le développement (OMD) 2008 Searle, Bob, 2010, “VAT, revenue sharing, and intergovernmental transfer design: the Australian experience,â€? in Ehtisham Ahmad and A. Al-Faris, 2010, Fiscal Reforms in the Middle East, Edward Elgar. The World Bank, 2003, World Development Report 2004: Making Services Work for Poor People, Oxford University Press for the World Bank. The World Bank, 2009a, The Rainforests of Cameroon – Experience and Evidence from a decade of Reform, Direction in Development – Environment and Sustainable Development, Washington D.C., The World Bank The World Bank, 2009, World Development Indicators 2009 99 Appendix Chart A 1: International Comparisons Poverty (% of population ) 60 50 40 30 20 10 0 Lower middle World Cameroon Sub-Saharan income Africa Source: World Bank (World Development Indicators 2009) Primary School Completion Rate (% of relevant age group) 100 80 60 40 20 0 Lower middle World Sub-Saharan Cameroon income Africa Source: World Bank (World Development Indicators, 2009) Literacy (ages 15+; % of population) 100 80 60 40 20 0 World Lower middle Cameroon Sub-Saharan income Source: World Bank (World Development Indicators, 2009) Africa 100 Access to Improved Water Source (% of population) 100 80 60 40 20 0 Lower middle World Cameroon Sub-Saharan income Africa Source: World Bank (World Development Indicators, 2009) Access to Improved Sanitation Facilities (% of population) 70 60 50 40 30 20 10 0 World Lower middle Cameroon Sub-Saharan income Africa Source: World Bank (World Development Indicators, 2009) Child Mortality (under 5 yrs; deaths per 1000 live births) 160 140 120 100 80 60 40 20 0 Lower middle World Sub-Saharan Cameroon income Africa Source: World Bank (World Development Indicators, 2009) 101 Births Assisted by Qualified Staff (% of births) 80 60 40 20 0 Lower middle World Cameroon Sub-Saharan income Africa Source: World Bank (World Development Indicators, 2009) Life Expectancy at Birth (years) 75 70 65 60 55 50 45 40 Lower middle World Sub-Saharan Cameroon income Africa Source: World Bank (World Development Indicators, 2009) Prevalence of Undernourishment (% of population) 35 30 25 20 15 10 5 0 Lower middle World Cameroon Sub-Saharan income Africa Source: World Bank (World Development Indicators, 2009) 102 Table A 1: Cameroon: Spatial dimensions of Poverty Incidence Average gap Severity % of % of poor Average Q5/Q1 of between of population population annual poverty annual poverty expenditure (in %) consumption (in %) per head of of poor household households (CFAF) and poverty ceiling (in %) Urban 12,2 2,8 1,0 35,3 10,8 662 289 6,5 Rural 55,0 17,5 7,2 64,7 89,2 313 338 5,2 Regions Douala 5,5 0,9 0,2 10,0 1,4 745 132 5,7 Yaoundé 5,9 1,0 0,2 9,6 1,4 761 813 5,3 Adamaoua 52,9 14,5 5,4 5,2 6,9 349 144 6,6 Center 41,2 9,5 3,1 7,6 7,9 354 190 4,3 East 50,4 15,7 6,2 4,7 5,9 339 014 5,9 Far North 65,9 24,6 11,2 18,1 29,9 285 932 6,7 Littoral 30,8 7,7 2,7 3,5 2,7 413 866 5,5 North 63,7 21,0 8,6 9,8 15,7 305 186 6,7 North 51,0 16,6 6,8 10,2 13,0 367 278 7,6 West West 28,9 6,6 2,3 10,6 7,7 415 661 4,7 South 29,3 7,4 2,6 3,3 2,4 458 844 6,0 South 27,5 6,9 2,5 7,5 5,2 472 595 6,0 West Total 39,9 12,3 5,0 100,0 100,0 439 787 7,5 Source: ECAM3, INS 103 Table A 2: Summary Table of Indicators Région d’enquête DOUALA YAOUNDE ADAMAOUA CENTRE EST EXTREME LITTORAL NORD NORD OUEST SUD SUD CAMEROUN NORD OUEST OUEST URBAIN RURAL ENSEMBLE Indicateur DEPENSES EN FCFA Dépense moyenne de consommation par 576 331 584 411 252 995 267 928 248 207 238 320 093 221 273 303 343 363 510 562 231 327 513 tête 775 064 495 013 741 426 871 Dépense moyenne par unité de 745 132 761 813 349 144 354 190 339 285 932 413 866 305 367 415 458 472 662 289 313 439 787 consommation 014 186 278 661 844 595 338 Dépense de consommation médiane par 1 754 1 816 438 1 029 616 939 056 955 790 219 974 359 842 804 1 060 1 224 987 1 644 847 1 056 520 ménage 632 536 409 426 953 025 519 824 651 PAUVRETE Incidence de la pauvreté 5,5 5,9 52,9 41,2 50,4 65,9 30,8 63,7 51,0 28,9 29,3 27,5 12,2 55,0 39,9 Profondeur de la pauvreté 0,9 1,0 14,5 9,5 15,7 24,6 7,7 21,0 16,6 6,6 7,4 6,9 2,8 17,5 12,3 Sévérité de la pauvreté 0,2 0,2 5,4 3,1 6,2 11,2 2,7 8,6 6,8 2,3 2,6 2,5 1,0 7,2 5,0 Q5/Q1 (rapport entre le 5e et le 1er 5,7 5,3 6,6 4,3 5,9 6,7 5,5 6,7 7,6 4,7 6,0 6,0 6,5 5,2 7,5 quintile) Part du cinquième le plus pauvre de la 11,8 11,5 9,5 10,6 13,3 7,5 14,4 13,5 11,1 12,1 9,1 7,2 4,9 10,0 6,4 population dans la consommation totale EDUCATION Taux net global de scolarisation (6 – 14 ans) 98,0 97,2 62,8 95,5 75,3 51,5 94,6 60,5 91,3 95,5 95,8 94,4 93,3 73,9 79,8 Taux net global de scolarisation (6 – 11 ans) 98,1 97,6 57,0 95,1 72,5 46,7 94,9 54,1 89,7 95,2 95,5 93,7 92,9 70,4 77,2 Taux de redoublement primaire 6,7 6,8 14,6 9,3 10,4 11,3 10,3 13,4 8,3 13,8 9,6 6,9 8,3 10,9 10,1 Taux d’alphabétisation des adultes (15 ans 94,9 96,6 42,4 89,2 60,1 28,3 88,8 40,7 72,6 83,8 91,6 86,0 89,1 58,9 70,6 et plus) Taux d’alphabétisation des adultes (15-24 97,9 98,4 59,5 97,6 72,6 47,5 97,3 58,4 93,4 98,0 96,8 96,7 94,4 75,1 83,1 ans) Taux d’alphabétisation des femmes de 15 à 1,01 0,98 0,61 1,01 0,80 0,53 1,00 0,55 0,97 0,99 1,02 0,97 0,98 0,78 0,87 24 ans par rapport aux hommes Part de la dépense d’éducation dans les 6,4 6,7 2,3 4,3 2,5 1,4 4,3 1,9 5,2 5,5 4,1 5,0 6,0 3,1 4,7 dépenses totales des ménages Dépense moyenne d’éducation par enfant 105 541 99 802 23 365 32 360 23 576 14 880 36 830 17 37 233 39 433 32 162 48 582 80 207 24 455 47 490 et par an (en FCFA) 841 Distance moyenne à l’école primaire 1,5 1,2 1,7 1,7 1,1 1,4 1,3 1,6 1,3 1,2 1,3 1,5 1,2 1,5 1,4 publique la plus proche (km) SANTE Taux de morbidité 25,4 21,3 37,0 26,8 17,9 20,0 25,4 21,2 25,2 30,9 25,6 24,3 24,7 24,3 24,5 Part de la dépense de santé dans la 4,6 3,7 2,4 3,4 3,0 3,1 5,0 3,4 3,9 4,9 3,1 4,5 4,1 3,7 3,9 dépense totale Dépense annuelle moyenne de santé par 26 342 21 721 6 082 9 235 7 480 6 400 16 073 7 559 10 724 14 866 10 800 16 171 20 583 8 512 12 775 tête (en FCFA) Distance moyenne au centre de santé le 2,3 2,1 6,9 7,9 6,7 8,0 4,2 6,9 6,3 5,2 4,9 6,8 2,3 7,8 5,6 plus proche(km) EMPLOI Taux d’activité BIT (10 ans et +) 63,3 62,3 79,5 76,4 85,3 82,7 77,4 84,2 76,5 84,2 67,0 76,3 63,6 84,1 76,4 104 Taux d’activité élargi 75,8 75,9 84,4 85,8 90,6 89,6 85,6 89,6 84,9 90,6 79,8 85,9 75,7 90,8 84,7 Taux de chômage BIT 10,4 8,7 1,2 2,6 0,9 0,9 2,0 0,7 1,0 1,6 1,8 1,3 7,2 0,7 3,1 Taux de chômage élargi 16,2 16,2 2,0 4,4 2,1 1,7 4,2 2,0 1,9 2,8 4,5 2,5 12,6 1,4 5,4 Taux de chômage des 15 à 24 ans 26,1 27,1 2,5 10,0 2,9 3,5 5,5 3,7 3,0 4,1 10,8 4,1 19,4 2,9 8,2 Pourcentage des enfants en activité 10 9,1 60,9 36,8 61,9 48,5 44 53,4 36,8 57 26,1 30,1 17,9 51,1 41 économique Rapport de dépendance (10 ans et +) 2,3 2,4 1,7 1,7 1,6 1,8 1,7 1,7 1,8 1,6 2,1 1,7 2,3 1,6 1,8 Taux de sous-emploi BIT(10 ans et +) 44,4 49,8 64,0 61,8 70,5 86,9 61,9 84,3 87,0 72,6 60,9 69,5 53,8 78,8 71,0 Taux de salarisation (10 ans et +) 41,2 45,1 9,4 13,3 11,5 5,7 20,0 5,8 10,5 12,8 23,1 17,4 35,4 8,0 17,0 Taux d'informalité (10 ans et +) 75,6 73,4 94,8 92,8 93,4 97,4 84,5 96,3 93,2 92,2 81,9 89,8 78,6 95,3 89,8 Taux de pluri activité (10 ans et +) 9,6 13,1 43,7 43,1 57,7 48,6 33,3 43,7 48,5 50,6 50,9 25,4 19,8 47,9 38,6 LOGEMENT ET EQUIPEMENT DES MENAGES Pourcentage de ménages possédant 97,1 98,2 25,5 46,7 23,6 11,8 67,2 16,6 31,0 56,6 72,1 36,4 90,4 23,4 48,2 l'électricité Pourcentage de ménages ayant accès à 81,6 80,4 41,8 35,5 16,8 28,3 55,0 20,6 52,2 28,7 35,4 43,2 75,1 27,7 45,3 l'eau potable Pourcentage de ménages utilisant le gaz de 45,9 54,8 3,3 7,3 3,7 0,5 11,9 1,1 4,6 8,2 8,4 12,0 37,0 2,5 15,3 cuisine Pourcentage de ménages disposant d’un 77,1 74,1 23,1 37,0 13,4 4,8 52,6 5,8 22,5 26,8 32,6 28,9 66,4 14,2 33,6 WC moderne Pourcentage de ménages ayant accès aux 77,5 72,2 1,8 1,1 1,6 1,6 14,8 3,9 5,0 9,7 12,3 9,6 52,1 2,0 20,6 bacs à ordure Pourcentage de ménages habitant des 70,5 72,4 41,7 30,6 23,6 10,1 38,4 20,0 43,1 48,9 31,5 28,3 71,0 20,5 39,2 logements en matériaux définitifs Nombre moyen de personnes par pièce 2,0 2,1 2,2 1,6 2,1 2,2 1,5 2,3 1,8 1,9 2,0 1,8 2,0 2,0 2,0 Pourcentage de ménages victimes de 15,5 15,1 6,9 23,2 18,2 28,8 12,6 29,9 25,4 9,5 26,8 19,2 15,9 23,1 16,8 coupure d'eau pour facture impayée (au cours des 12 derniers mois) Pourcentage de ménages ayant au moins 30,9 24,0 10,7 17,9 29,4 19,5 29,0 20,6 28,1 26,1 23,5 40,6 27,3 29,6 28,0 un mois d'arriéré de loyer (au cours des 12 derniers mois) Pourcentage de ménages équipés en radio 57,8 61,5 54,7 65,5 38,3 37,6 58,4 43,7 33,5 54,7 53,8 45,3 59,6 43,4 49,4 Pourcentage de ménages équipés en 71,5 76,3 19,3 22,3 15,8 6,3 32,4 8,8 16,8 29,1 39,9 22,4 64,6 10,8 30,7 téléviseur Pourcentage de ménages équipés en 2,7 3,2 0,9 0,3 0,2 0,3 0,3 0,5 0,4 0,5 1,4 1,2 2,5 0,2 1,1 téléphone fixe Pourcentage de ménages équipés en 86,9 91,7 27,4 40,3 23,6 14,6 55,6 17,6 32,4 53,1 43,4 37,0 81,4 23,4 44,9 téléphone mobile Pourcentage de ménages possédant un 6,2 8,1 1,1 0,9 0,7 0,2 0,9 0,9 1,6 1,5 1,1 0,9 5,4 0,5 2,3 ordinateur Pourcentage de ménages possédant une 4,9 7,9 1,8 0,9 0,6 0,7 2,4 0,8 2,3 2,4 2,3 1,7 5,7 0,7 2,6 automobile ENVIRONNEMENT ET CADRE DE VIE DES MENAGES Pourcentage des ménages victimes des 43,8 43,4 24,3 20,3 32,2 16,2 40,0 22,6 39,6 19,5 38,0 26,9 42,3 22,2 29,6 bruits 105 Pourcentage des ménages victimes des 23,4 10,7 5,6 2,6 19,9 38,3 12,1 13,9 11,1 4,4 2,6 11,4 15,6 15,0 15,2 inondations Pourcentage des ménages victimes de 80,3 81,3 72,4 79,0 68,0 47,0 68,3 69,8 71,0 59,0 82,7 79,7 77,4 65,5 69,9 cafards Pourcentage des ménages victimes des 89,4 83,0 90,6 88,9 93,0 96,7 81,9 94,7 78,1 75,8 94,3 89,4 84,4 89,5 87,6 moustiques Pourcentage des ménages victime des 77,6 74,6 76,1 80,8 81,6 81,2 74,2 77,4 92,4 82,4 81,2 90,4 75,2 85,3 81,5 souris/rats Pourcentage des ménages Victime de 15,1 13,2 5,6 6,6 7,2 7,7 6,2 9,6 6,3 17,2 22,1 3,8 12,1 8,6 9,9 violences physiques (vol, agression, etc.) Pourcentage des ménages logeant à 42,3 37,5 31,8 32,2 38,2 40,8 37,6 18,6 36,7 31,3 38,9 46,0 38,2 35,5 36,5 proximité d’un cours d’eau non aménagés Pourcentage des ménages logeant à 12,1 7,0 5,8 0,8 1,8 0,6 7,8 0,1 0,3 0,6 3,8 1,3 6,6 1,3 3,3 proximité d’une ligne de chemin de fer ou un aéroport Pourcentage des ménages logeant à 49,9 39,3 17,5 35,7 13,7 10,8 38,3 14,3 23,7 28,6 47,5 18,7 44,7 17,1 27,3 proximité d’une ligne de moyenne ou haute tension CORRUPTION ET GOUVERNANCE Pourcentage des ménages victime de la 29,5 23,7 19,2 20,3 12,0 7,8 15,4 9,7 10,1 16,7 28,1 18,7 24,3 12,7 17,0 corruption Pourcentage des ménages insatisfaits des 86,5 74,4 63,8 88,8 90,5 62,5 77,9 37,8 73,4 80,6 87,3 44,5 76,2 79,5 77,2 services d’AES-SONEL pour cause de délestages fréquents Pourcentage des ménages insatisfaits des 20,8 28,4 56,3 35,2 43,9 34,5 56,9 36,5 57,9 38,6 42,2 42,3 28,4 58,5 31,1 services de CDE pour cause de coupures fréquentes PATRIMOINE ET CAPITAL SOCIAL Pourcentage de ménages ayant accès à la 16,9 32,8 74,4 79,6 76,4 78,0 54,1 77,5 72,5 67,4 44,8 48,8 30,0 77,0 59,6 terre Pourcentage de ménages possédant un 9,3 12,9 11,2 11,6 7,0 5,2 12,6 13,8 9,4 14,8 10,3 8,1 12,8 8,8 10,2 immeuble bâti non occupé par le ménage Pourcentage de ménages ayant bénéficié 3,8 5,2 3,4 0,8 2,5 7,0 2,0 5,7 7,6 9,7 2,7 6,9 5,6 5,5 5,5 d'un crédit Pourcentage de ménages dont au moins un 52,2 56,4 22,9 41,7 17,4 19,2 69,3 16,2 52,2 63,4 57,8 47,9 52,4 37,2 42,9 membre appartient à une association Source : ECAM3, INSN B : Centre=Centre-Yaoundé ; Littoral=Littoral-Douala 106 Table A 3: Comparison of Competences between Deconcentrated Regions/Departments and Decentralized Regions Deconcentrated region Decentralized region Source : decree 2008/377 of 12 November 2008 Source : Law n° 2004/019 of 22 July 2004 establishing the regulations applicable to regions Level of « governor » Art. 18Security; Public order; Police; Gendarmerie. Art. 33Civil protection. Art. 32 Limits of traditional chiefdoms; Operation of traditional chiefdoms. Art. 33 All matters related to : Art. 18 Economic development - tourism. - promotion of small- and medium enterprises; - organization of fairs and exhibitions; - promotion of arts and craft; - promotion of agricultural, pastoral and fisheries activities; - encouraging the creation of regional economic associations; - support to micro-projects generating revenues and employment; - promotion of tourism. Art. 19 Environment and Natural Resources - management, protection and maintenance of protected areas and nature areas relevant to the capabilities of the region; - preservation and other local measures to protect the nature; - management of waters of regional interest; - creation of woods, forests and protected zones of regional interest following a plan duly approved by the representative of the State; - creating firebreaks and controlled fires in order to fight against forest- and bush fires; - management of regional natural parks, following a plan submitted for approval to the representative of the State; - development, implementation and monitoring of regional action plans or schemes for protection of the environment; - development of regional plans for emergency interventions and risk prevention. Art. 34 Regional development Art. 20 Regional development, public works, city planning, - Regional planning; housing - Monitoring the development and - development and implementation of regional development implementation of regional and municipal plans; development plans as well as regional town - awarding, in relation with the State, planning contracts for the and country planning. realization of development objectives; - participation in the organization and management of public Art. 31City planning inter-city transports; - Land affairs and matters related to state- - coordination of development activities; owned property. - development, conforming to the national plan, of regional development schemes; - participation in the development of urban planning documents and of guideline policies for decentralized municipalities; - rehabilitation and maintenance of departmental and regional roads; - support to municipal activities related to city planning and 107 Deconcentrated region Decentralized region Source : decree 2008/377 of 12 November 2008 Source : Law n° 2004/019 of 22 July 2004 establishing the regulations applicable to regions housing. Art. 33All issues related to : Art. 21 Health and social action public health; - participation in the development of the regional section of the welfare; national public health policy; social security. - building, in accordance with national health policy, equipment, management and maintenance of health centres in the region; - support to health centres and social institutions; - implementation of preventive and sanitary measures; - participation in the maintenance and management of promotion and/or social reintegration centres; - organization and management of welfare benefits to those in need; - participation in the organization and management of drug supplies, in accordance with national health policy. Art. 33All issues related to: Art. 22 Education, elimination of illiteracy, vocational training education. a) education: - participation in the establishment and implementation of the regional section of the national education policy; - building, equipment, management, and maintenance of secondary schools and colleges in the region; - recruitment and management of support staff in secondary schools; - distribution and allocation of scholarships and educational assistance; - participation in the acquisition of manual and school supplies; - participation in the management and administration of secondary schools and colleges run by the State, through dialogue and consultation; - support to municipal activities concerning primary and maternal education. b) literacy: - development and implementation of regional plans for the elimination of illiteracy; - annual overview of the execution of literacy campaigns; - recruitment of staff in charge of fighting illiteracy; - training of trainers; - design and production of teaching materials; - realization of the plan towards full literacy; - establishment of educational facilities and equipment; - monitoring and evaluation of the plan to combat illiteracy; c) vocational training: skills and training profile; - participation in the development of the regional section of the school curricula relating to technical education and vocational training; - development of a plan for training; - maintenance of training centres and institutions in the region; - recruitment and management of support staff; - participation in the acquisition of teaching material, particularly school supplies and working material; - participation in the management and administration of the training centres run by the State through dialogue and 108 Deconcentrated region Decentralized region Source : decree 2008/377 of 12 November 2008 Source : Law n° 2004/019 of 22 July 2004 establishing the regulations applicable to regions consultation; - development of a regional plan for the integration of youth into working life; - supporting the establishment of partnerships between businesses and schools. Art. 33 All matters related to : Art. 23 Youth, sports, and recreation youth; - authorizing the opening of education centres, duly stamped by sports; the representative of the State; recreation. - support to regional sport associations; - realization, administration and management of sport- and regional socio-educational facilities; - organization, coordination and development of socio- educational activities; - promotion and management of physical and sport activities in the region. Art. 24 Culture, promotion of national languages a) culture: - promotion and development of cultural activities; - participation in the State’s supervision and monitoring of the conservation of areas historical sites and monuments as well as in the discovery of prehistoric and historic archaeological remains; - organization of cultural days, traditional cultural events and literary and artistic competitions; - creation and management of orchestras, traditional opera ensembles, ballet groups and theatre groups of regional interest ; - creation and management of socio-cultural centres and public libraries of regional interest; - collecting and translating elements of the oral tradition, including tales, myths and legends, in order to facilitate publication; - support to cultural associations. b) promotion of national languages: - functional control of national languages and development of a regional linguistic map; - participation in the promotion of national language publications; - promotion of spoken and written press in national languages; - setting up facilities and equipment. Note: Rather than following the order of the decree this table compares the competencies of deconcentrated regions with those of decentralized regions. 109 Table A 4: Comparison of Competencies between Administrative (Deconcentrated) Departments and Municipalities Department Municipalities Source : decree 2008/377 of 12 November 2008 Source : law 2004/018 establishing the regulations applicable to municipalities Level of « prefect » (under the direct authority of the governor : art. 35) Art. 47 Population, marital status, nationality; Traditional chiefdoms; Land affairs and matters related to state- owned property. Art. 48 Transports; Procurement; Fiscal census, preparation of assessments related to the global tax, taxes on arms and livestock. Art. 49 Civil protection; Employment and labour; Tourism; Arts and craft. Art. 50 States supervision over the municipalities; Advisory supports to municipalities and their institutions; Budget control and control of the legality of acts of municipalities and their institutions; Monitoring of decentralized cooperation. Art. 15 Economic development - promotion of agricultural, pastoral, crafts and fisheries activities of interest to the municipality; - development of municipal tourist sites; - construction, equipment, management and maintenance of markets, bus terminals, and slaughterhouses; - organization of local trade fairs; - support to micro-projects generating revenues and jobs. Art. 16 Environment and Natural resources - supply of drinking water; - cleaning of municipal roads, streets, and public spaces; - monitoring and controlling the management of industrial waste; - reforestation and creation of communal forests; - fight against insalubrity, pollution, and public nuisances; - protection of ground- and surface water resources; - development of municipal action plans for the environment; - creation, maintenance and management of green areas, parks, gardens of local interest; - management of household waste on the local level. Art. 49 City planning and housing. Art. 17 Planning, regional development, city planning, Art. 50 Monitoring and coordination of activities housing related to - development and implementation of municipal planning at the departmental level; investment plans; 110 Department Municipalities Source : decree 2008/377 of 12 November 2008 Source : law 2004/018 establishing the regulations applicable to municipalities Monitoring the relations with planning - awarding, in association with the State or the region, bodies; planning contracts for achieving development Monitoring development projects; objectives; Monitoring the promotion of and initiatives - developing land use plans, city planning documents, for local development. collaborative planning, and urban renewal and land development; - organization and management of urban public transports; - operations management; - issuing city planning certificates, authorizations for land allotments , settlement permits, building and demolition permits; - building and maintenance of municipal roads as well as the completion of related works; - development and servicing of living spaces; - street lighting; - addressing and naming streets, squares and public buildings; - building and maintenance of non-classified rural roads and ferries; - creation of industrial zones; - contributing to the electrification of zones in need; - authorization for temporary occupation and miscellaneous works. Art. 47 Population, marital status, nationality. Art. 18 Health and social action Art. 49Social and family protection; a. Health and population: All other social problems - marital status ; - building, equipment, management and maintenance of health centres of local interest, in accordance with national health policy; - support to health centres and social institutions; - sanitary controls in facilities dealing with manufacturing, packaging, storage, and food distribution as well as in facilities handling solid and liquid waste products from individuals or businesses. b. Social action: - participation, as appropriate, in the management and maintenance of promotion and social reintegration centres; - building, maintenance and management of public cemeteries; - organization and management of aid to the disadvantaged. Art. 19 Education, elimination of illiteracy, vocational training a. Education: - building, in accordance with national education policy, management, equipment, and maintenance of nursery and primary schools and pre-school establishments in the municipality; - recruitment and management of support staff in above mentioned schools; 111 Department Municipalities Source : decree 2008/377 of 12 November 2008 Source : law 2004/018 establishing the regulations applicable to municipalities - participation in the acquisition of teaching material and school supplies; - participation in the management and administration of secondary schools and State and regional colleges through dialogue and consultation. b. Elimination of illiteracy: - implementation of plans to eliminate illiteracy, in relation with the regional administration; - participation in the establishment and maintenance of educational facilities and equipment. c) Vocational and technical training: - development of a local plan for training and retraining ; - development of a plan for integration or reintegration into working life; - participation in the establishment, maintenance and administration of training centres. Art. 49 Organization of urban and rural activities; Art. 20 Youth, sports, recreation Youth and sports; - promotion and organization of youth and sports Games and recreation. activities; - support to sport associations; - creation and management of municipal stadiums, sport centres, swimming pools, playgrounds and arenas; - Inventory and participation in the equipment of sports associations; - participation in the organization of competitions. Art. 49 Organization of urban and rural activities Art. 21 Culture, promotion of national languages All other problems related to culture. a. culture: - organization, on local level, of cultural days, traditional cultural events and literary and artistic competitions; - creation and management, on the local level, of orchestras, traditional opera ensembles; ballet and theatre groups; - creation and management of socio-cultural centres and public libraries; - support to cultural associations. b. promotion of national languages: - participation in regional programs promoting national languages; - participation in establishment and maintenance of facilities and equipment. Art. 110 Urban metropolitan areas - creation, maintenance, and management of green spaces, parks and gardens in the metropolis; - management of rivers and lakes of interest to the metropolis; - monitoring and controlling the management of industrial waste; - cleaning the roads and squares of the metropolis; - collection, removal and disposal of household waste; - creation, planning, maintenance, exploitation and management of the metropolis’s equipment related to 112 Department Municipalities Source : decree 2008/377 of 12 November 2008 Source : law 2004/018 establishing the regulations applicable to municipalities sanitation, waste water, and rainwater; - development of community action plans for the environment, including the fight against pollution and public nuisances, and the protection of green areas ; - creation, maintenance and management of public cemeteries; - creation and management of all sportive installations of interest to the metropolis; - operations management in the interest of the metropolis; - creation of land reserves for the interest of the metropolis; - creation and management of cultural centres of interest to the metropolis; - construction, equipment, management, and maintenance of facilities of interest to the metropolis, including markets, bus terminals and slaughterhouses ; - participation in the organization management of urban public transports; - development and implementation of community investment plans; - awarding, together with the State of the region, contracts for the achievement of the metropolis’s development objectives; - urban planning and blueprints, land use plans or city planning document. To this effect, the urban metropolis gives its opinion on the draft regional planning document prior to its approval; - building, planning, maintenance, operation and management of primary and secondary public roads in the metropolis, their dependencies and their equipment, including street lighting, rainwater drainage, signage, safety equipment and art works; - creation and planning of urban public spaces of interest to the metropolis; - coordination of urban distribution networks of electricity, drinking water, telecommunications and all interventions related to the public road network; - traffic and travel plans for the entire urban road network. Note: Rather than following the order of the decree this table compares the competencies of deconcentrated regions with those of decentralized regions. 113 Table A 5: Comparison of Competencies between Decentralized Regions and Municipalities Decentralized regions Municipalities Source: Law n° 2004/019 of 22 July 2004 establishing the Source: law 2004/018 establishing the regulations applicable regulations applicable to regions. to municipalities. Art. 18 Economic development Art. 15 Economic development - promotion of small and medium enterprises ; - promotion of agricultural, pastoral, crafts and fisheries - organization of trade fairs and shows; activities of interest to the municipality; - promotion of arts and craft; - development of municipal tourist sites; - promotion of agricultural, pastoral and fisheries - construction, equipment, management and activities; maintenance of markets, bus stations and - encouraging the creation of regional economic slaughterhouses; associations; - organization of local trade fairs; - support to micro-projects generating revenues and - support to micro-projects generating revenues and jobs. jobs; - promotion of tourism. Art. 19 Environment and Natural resources Art. 16 Environment and Natural resources - management, protection and maintenance of - supply of drinking water; protected areas and nature sites relevant to the - cleaning municipal roads, streets, and public squares; capabilities of the region; - monitoring and controlling the management of - preservation and other local measures to protect industrial waste; the nature ; - reforestation and the creation of municipal forests; - management of waters of interest to the region ; - fight against insalubrity, pollution and public nuisances; - creation of woods, forests and protected zones of - protection of ground- and surface water resources; interest to the region following a plan duly - development of municipal action plans for the approved by the representative of the State; environment; - creating firebreaks and controlled fires in order to - creation, maintenance and management of green fight against forest- and bush fires; areas, parks and gardens of municipal interest; - management of regional nature parks, following a - management of household waste on the local level. plan submitted for approval to the representative of the State; - development, implementation and monitoring of regional action plans or schemes for protection of the environment; - development of regional plans for emergency interventions and risk prevention. Art. 20 Regional development, public works, city Art. 17 Planning, regional development, city planning, planning, housing housing - development and implementation of regional - development and implementation of municipal development plans; investment plans; - awarding, in relation with the State, planning - awarding, in association with the State or the region, contracts for the realization of development planning contracts for achieving development objectives; objectives; - participation in the organization and management - organization and management of urban public of public inter-city transports; transports; - coordination of development activities; - development of land use plans, city planning - development, conforming to the national plan, of documents, collaborative planning, urban renewal and regional development schemes; land development; - participation in the development of urban planning - building and maintenance of non-classified rural roads documents and of guideline policies for and ferries; decentralized municipalities; - building and maintenance of municipal roads as well as - rehabilitation and maintenance of departmental the completion of related works; and regional roads; - operations management; - support to municipal activities related to city - issuing city planning certificates, authorizations for land planning and housing. allotments , settlement permits, building and demolition permits; 114 Decentralized regions Municipalities Source: Law n° 2004/019 of 22 July 2004 establishing the Source: law 2004/018 establishing the regulations applicable regulations applicable to regions. to municipalities. - development and servicing of living spaces; - street lighting; - addressing and naming streets, squares and public buildings; - creation of industrial zones; - contributing to the electrification of zones in need; - authorization for temporary occupation and miscellaneous works. Art. 21 Health and social action Art. 18 Health and social action - participation in the development of the regional a. Health and population: section of the national public health policy; - marital status; - building, in accordance with national health policy, - building, equipment, management and maintenance of equipment, management and maintenance of health centres of local interest, in accordance with health centres in the region; national public health policy; - support to health centres and social institutions ; - support to health centres and social institutions - sanitary controls in facilities dealing with - implementation of preventive and sanitary manufacturing, packaging, storage, and food measures; distribution as well as in facilities handling solid and - participation in the maintenance and management liquid waste products from individuals or businesses. of promotion and/or social reintegration centres; - organization and management of welfare benefits b. Social action: to those in need; - participation, as appropriate, in the management and - participation in the organization and management maintenance of promotion and social reintegration of drug supplies, in accordance with national centres; health policy. - building, maintenance and management of public cemeteries; - organization and management of aid to the disadvantaged. Art. 22 Education, elimination of illiteracy, Art. 19 Education, elimination of illiteracy, vocational vocational training training a) Education: a) Education: - participation in the establishment and - building, in accordance with national education policy, implementation of the regional section of the management, equipment, and maintenance of nursery national education policy; and primary schools as well as pre-school - building, equipment, management, and establishments in the municipality; maintenance of secondary schools and colleges in - recruitment and management of support staff in above the region; mentioned schools; - recruitment and management of support staff in - participation in the acquisition of teaching material and secondary schools; school supplies; - distribution and allocation of scholarships and - participation in the management and administration of educational assistance; secondary schools and State and regional colleges - participation in the acquisition of manuals and through dialogue and consultation. school supplies; - participation in the management and administration of secondary schools and colleges run by the State, through dialogue and consultation - support to municipal activities concerning primary and maternal education. b) Literacy : b) Elimination of illiteracy - development and implementation of regional plans - implementation of plans to eliminate illiteracy, in for the elimination of illiteracy; relation with the regional administration; - annual overview of the execution of literacy - participation in the establishment and maintenance of 115 Decentralized regions Municipalities Source: Law n° 2004/019 of 22 July 2004 establishing the Source: law 2004/018 establishing the regulations applicable regulations applicable to regions. to municipalities. campaigns; educational facilities and equipment. - recruitment of staff in charge of fighting illiteracy; - training of trainers; - design and production of teaching materials; - realization of the plan towards full literacy; - establishment of educational facilities and equipment; - monitoring and evaluation of the plan to combat illiteracy; c) Vocational training : c) Vocational and technical training: - exhaustive survey of regional business and the - development of a local plan for training and retraining; development of a directory of existing vocational - development of a plan for integration or reintegration trainings indicating the required skills and training into working life; profile; - participation in the establishment, maintenance and - participation in the development of the regional administration of training centres. section of the school curricula relating to technical education and vocational training; - development of a plan for training; - maintenance of training centres and institutions in the region ; - recruitment and management of support staff; - participation in the acquisition of teaching material, particularly school supplies and working material; - participation in the management and administration of the training centres run by the State through dialogue and consultation; - development of a regional plan for the integration of youth into working life; - supporting the establishment of partnerships between businesses and schools. Art. 23 Youth, sports, and recreation Art. 20 Youth, sports, recreation - authorizing the opening of education centres, duly - promotion and organization of youth and sports stamped by the representative of the State; activities; - support to regional sport associations; - support to sport associations; - realization, administration and management of - building and management of municipal stadiums, sport regional sport- and socio-educational facilities; centres, swimming pools, playgrounds and arenas; - organization, coordination and development of - inventory of and participation in the equipment of socio-educational activities; sports associations; - promotion and management of physical and sport - participation in the organization of competitions. activities in the region. 116 Decentralized regions Municipalities Source: Law n° 2004/019 of 22 July 2004 establishing the Source: law 2004/018 establishing the regulations applicable regulations applicable to regions. to municipalities. Art. 24 Culture, promotion of national languages Art. 21 Culture, promotion of national languages a) culture: a) culture: - promotion and development of cultural activities; - organization, on the local level, of cultural days, - participation in the State’s supervision and traditional cultural events and literary and artistic monitoring of the conservation of historical sites competitions; and monuments as well as in the discovery of - establishment and management, on the local level, of prehistoric and historic archaeological remains; orchestras, traditional opera ensembles; ballet and - organization of cultural days, traditional cultural theatre groups; events and literary and artistic competitions; - establishment and management of socio-cultural - establishment and management of orchestras, centres and public libraries; traditional opera ensembles, ballet groups and - support to cultural associations. theatre groups of regional interest; - establishment and management of socio-cultural centres and public libraries of regional interest; - collecting and translating elements of the oral tradition, including tales, myths and legends, in order to facilitate publication; - support to cultural associations. b) promotion of national languages: b) promotion of national languages: - participation in the promotion of national language - participation in regional programs promoting national publications; languages; - promotion of spoken and written press in national - participation in the establishment and maintenance of languages; facilities and equipment. - establishment of facilities and equipment. 117 Table A 6 : Municipal Taxes (Law 2009/019) Type of tax levy Name of tax levy Subject to taxation Calculation basis Exemptions Scale Rate Municipal taxes levied by the Proceeds from Economic, commercial and Turnover (art. 10) 22 categories according Digressive scale Fixed by the recipient territorial state businesses industrial activities (art. 8 and according to 126 types of to a list (art. 11) (i) base rate up to CFAF 2 unit within the set ranges (art. 13) * 9) business licenses (list of exemption for 2 years: billion, 7 brackets of turnover (part of the proceeds marked professions) and corresponding new enterprises (art. 12) from 0,283% to 0,0875% with ranges (annex I) * goes to equalization) (ii) 5% reduction per additional CFAF 500,000,000 (iii) maximum reduction 30% over CFAF 2,000,000,000 (art. 13) Liquor license Manufacturing and wholesale Turnover (36) according to the According to list (art. 35) Non-alcoholic: 2x the amount As for the business tax (art. 36) * or retail of alcoholic and non- same rules as applicable to of the business tax; alcoholic beverages (list - art. proceeds from businesses and Alcoholic: 4x; 33 and 34)) global tax For retailers subject to global tax: non-alcoholic: 1x; alcoholic: 2x (art. 37) Global tax Taxpayer engaged in Four categories A, B, C and D Exclusive of payment of scale (art. 46): Fixed by the recipient territorial commercial, industrial, craft, with list of activities (art. 46) business license, A: CFAF 0 to 20.000 unit within the set ranges (art. 46) agro-pastoral activities not personal income tax and B: CFAF 21.000 to 40.000 relevant to the actual profit VAT (art. 45) C: CFAF 41.000 to 50.000 regime, the simplified tax D: CFAF 51.000 to 100.000 regime or the basic tax regime (art. 45) Tax on immovable Tax levies prescribed by the General Tax Code. All proceeds from these three taxes are allocated to the municipality where the property and, in case of gambling property and entertainment, the premises are located (art. 48, 49 and 50). * Tax on sales of immovable property Tax on gambling and entertainment Stamp duties on cars Prescribed by the General Tax Code: fully allocated to the Special Council Support Fund for Mutual Assistance (FEICOM) or for equalization for redistribution to * municipalities (art. 51) Revenue-sharing Forest taxes Prescribed by the General Tax Code; sharing of the taxes: 20% goes to municipalities * Distribution: 20% local municipalities ; 20% FEICOM or equalization (art. 52) Additional taxes = shared taxes Tax on personal income Taxes prescribed by the Tax Administration of the State (at least 10% of the revenues collected retained by the tax authorities 10% of the principal of the Corporation tax for the costs of managing these taxes (art. 55) concerned tax (art. 54) VAT * the proceeds from additional municipal taxes are shared between the State and FEICOM or all other bodies in charge of centralization and equalization and the municipalities and metropolitan areas following the procedures laid down by the regulatory framework (art. 56) 118 Table A 7: Municipal Taxes, Fees, and User Charges (Law 2009/019) Type of levy Name of levy Subject to taxation Calculation basis Exemptions Scale Rate Tax Local development tax In return for basic services Employees in public and None Employees in public and Employees in public and (art. 57-60) (public lighting, sewage, waste private sector: according to private sector: scale of 9 private sector : fixed from collection, operation of basic monthly salary (scale); classes CFAF 3,000 to 30,000/year ambulances, water supply, people subject to global tax or people subject to global people subject to global tax electrification) business tax: according to tax or business tax : or business tax : from CFAF principal tax (scale) scale of 9 classes 7,500 to 90,000/year Fee Slaughter tax (art. 63-65) Butchers slaughtering cattle in Per animal None Cattle, horses max. CFAF 1,000 municipal slaughterhouses Pigs max. CFAF 400 Sheep and goats max. CFAF 250 Tax Cattle tax (art. 66-72) Owner or keeper of bovine Per animal List CFAF 200 to 500 License /tax Firearms tax (art. 73 à 76) Keeper of firearms Per firearm according to list List CFAF 200 to 2,000 Fee/tax Health and sanitation tax Controls of industrial and (i) food sold in the open None 3 classes (i) CFAF 500 to (art. 7) commercial outfits handling (ii) food sold in commercial 1,000/quarter food products premises (ii) CFAF 1,000 to (iii) buildings 1,500/quarter (iii) CFAF 10,000 to 25,000/year User charge Pound fee (art. 78-79) Roaming animals, vehicles in List of 7 categories: 3 for None 7 categories CFAF 1,000 to 50,000 per violation of road regulations animals, 3 for vehicles, 1 for day other objects Concession Market tolls (art. 80-86) Occupation of a place in a (i) shops and stall constructed None (i) according to m2 (12 (i) from CFAF 5,000 to market in a permanent manner classes) 70,000/month (ii) occasional, without (ii) fixed daily fee (ii) CFAF 100 to 500 per permanent place vendor per day Fee Construction or building All construction (approved by Value of construction None 1% of the value of the permit (art. 87-90) the city plan) construction Fee Temporary occupation of Temporary occupation of public m2 per day List m2 per day max. CFAF 2,000 per public roads (art. 91-93) roads m2/day Fee Parking tax (art. 94-96) * Parking of vehicles used for Type of vehicle None motorcycle-taxis Max. CFAF 3,000/quarter urban transports in areas taxis CFAF 10,000/quarter managed by the municipality bus CFAF 15,000/quarter Fee Motor park fee Vehicles for public transports of Type of vehicle None (i) cars and vans (i) CFAF 1,000/day (art. 97 to 98) goods and people (ii) trucks and buses (ii) CFAF 2,000/day 119 Fee Parking fee Parking of private vehicles on Duration None parking (i) free or (i) CFAF 100/hour (art. 111 and 112) municipality’s parking spaces (ii) reserved (ii) CFAF 500/day or CFAF 15,000 /month User charge Quay fee Loading on bus stations and Load None (a) bus station (a) CFAF 250 per load (art. 99) jetties (b) jetty (scale for 3 (b) CFAF 200 to 1,000 per types of canoes) load Tax Entertainment tax Events organised for profit (list) Type of event None (a) occasional (a) CFAF 5,000 to (art. 100 to 101) (b) regular 50,000/day of performance (b) CFAF 10,000 to 100,000 /quarter/establishment Tax Stadium fee Sporting events which are not Per event None 5% of funds raised art. 102) free License/tax Advertising tax Advertising billboards etc. (list) Type of media (list) Illuminated type of media (list) (i) from CFAF (art. 103) signs (i) Vehicle 1,000/day/vehicle to CFAF indicating the (ii) Public address 30,000 /year/vehicle. location of a system (ii) CFAF 500/day commercial or industrial establishment Fee/tax Stamp duties Official documents issued by Document issued, invoice, or None Per document according Format smaller or equal to (art. 104) the municipality, invoices, requests to format (A4) A4: CFAF 200, larger than petitions and requests A4 : CFAF 400 addressed to the municipality (list) Fee/tax Highway dilapidation tax Users of vehicles and devices Work and equipment used None a) earthworks, sewage, (a) CFAF 15,000 to (art. 105 et 106) without tyres conducting works etc 200,000/m2 on public roads (b) damage cause by (b) CFAF 20,000 to caterpillars 100,000/m2 Tax/Duty Tax on seasonal livestock Passage of cattle from a Per animal being moved from The first 15 (a) cattle and horses Per animal/municipality movement neighbouring state the 16th day days (b) sheep and goats crossed (art 107) (a) CFAF 200 to 500 / (b) CFAF 100 to 300 Tax Tax on transports of quarry Transport of quarry products Type of None Type of vehicle CFAF 1,000 to 3,000 per products vehicle/truck/weight/trip truck per trip (art. 108 à 110) Tax Tax on salvaged forest Products originating from non- m3 None m3 CFAF 2,000/m3 products communal forests. (art. 113) 120 Table A 8: Tax Assignments between Urban Metropolitan Areas and City Councils Source Urban metropolitan area City council Taxes 4 taxes: proceeds from businesses, liquor 2 taxes: global tax, forest taxes. licenses, stamp duties on cars, tax on gambling and entertainment 3 shared taxes: The same 3 taxes: Additional taxes on personal income tax, Additional taxes on personal income tax, corporation tax and VAT. corporation tax and VAT. 2x revenue sharing: 60% land tax on real estate 20% land tax on real estate (20% FEICOM) 60% transfer duty on properties 20% transfer duty on properties (20% FEICOM) Fees and 9 items according to list art. 115 of which 4 13 items according to list art. 115of which 4 user for the use of the infrastructure of the for the use of the infrastructure of the city charges urban metropolis (sport stadiums, parking, council (stadiums, parking, market places, market places, pound fees). pound fees). 121 Annexe: Major problems from the 56 comptes administratifs 7.1 The problems observed in the sample of administrative accounts (comptes administratives) include the following: x Miscalculations/typos. Budgets are full of major as well as minor miscalculations and/or typing errors. They range from tens of CFAF to millions of CFAF. In rare cases they even make a difference between surplus and deficit. o One glaring example was for Dimako in 2007: Total revenues add up to over CFAF 400mn if one adds all the entries, yet total revenues of CFAF 290mn is written (which is the exact number of the total revenue from the previous year). Moreover, it is not specified where budgeted investment revenues of CFAF 96mn come from (i.e. only appears in total entry). o Ngaoundere II in 2007 did not receive any CAC according to its administrative account. This was probably a data entry error as the deficit would have been large otherwise, and the data available on the CAC suggest that it should have received CFAF 15mn in 2007. x FEICOM data formats. All municipalities use more or less the same budget format for their administrative accounts. Yet the ten budgets processed by FEICOM and received by the mission from them have lost a meaningful order. Entries for total amounts are missing and they are not ordered by type of income/expense, unlike in the formats followed in the accounts obtained directly from all other sources. It was not obvious why the data was reorganized by FEICOM, as it is impossible to keep track of local finances with such a system. x Budget execution is low. Overall budget execution is low relative to the budgeted amounts. Moreover, several occasions where there is large budgeted spending/revenue (ranging up to a couple of hundred millions; e.g., transfers for investments; various grants; investment spending) but zero or close to zero materializes. Similarly, several times budgeted ‘other’ entries are very large (CFAF 15-30mn), but actual spending/revenue is zero or close to zero. x Disproportionate/odd revenues. A number of municipalities were observed where a large share of total budgeted and/or actual revenues comes from a single (unexpected?) source. For instance, ƒ Diang, 2007, budgets CFAF 40mn proceeds from ‘garage transfers’ (representing around two thirds of total recurrent revenues) and in 2008- 2009 other proceeds from exploitation of council property represent equally large shares of budgeted revenues (low actual proceeds though); see previous point. 122 ƒ Dimako, 2006-2009: up to around 90 percent of budgeted and actual recurrent revenues come from ‘other revenue from exploitation of council property’. ƒ Ambam, 2007, received CFAF 161mn (or 67 percent of recurrent revenues) from highway dilapidation tax proceeds. Similar examples where ‘other indirect taxes’ represent large shares of revenues. x Changes and corrections by hand. Often changes are made by hand in the accounts, i.e., a line is drawn over an entry and a new figure put in its place. It is not clear which is the correct and audited number. x Population data. Population data cited in administrative accounts generally differs from official data (they are generally larger in the accounts). This is a major source of dissatisfaction with the transfer system based on out of date information on population. x Loans as revenue. Loans are seen as investment revenue such that budgets balance. This is contrary to the international practices and standards, e.g., the IMF’s Government Finance Statistics Manual 2001, and can lead eventually to unrealistic assessments of the fiscal position of the local governments. x Arrears. Arrears are very rarely reported explicitly. The numbers are generally left blank. However, the extent of arrears can be deduced from unbelievably low spending on pensions, utilities and debt repayments. This is reflective of “game playâ€? by the municipalities as they seek to push spending to the central government. x RFA: Yokadouma. Large forest (and other) revenues are reported, and not fully reflected in the spending—this may be because the revenues are for items that do not show up on municipal responsibilities. Staff costs high in absolute terms, as is spending on cars. But it is clear that the spending is low for “productiveâ€? investment. x Guidguis. In 2006 and 2007 the only expenditures were under ‘equipment and movables’ (Cars and vehicles, repairs, equipment (sport, parks, computer, for servicing, library, waste) etc.) in the investment budget, despite actual surpluses of over CFAF 40mn and CFAF 50mn respectively. Again this could reflect “game playâ€? with higher levels of government. x Reserves. Reserves earmarked for operational costs/overheads are sometimes counted into the total, sometimes not. x Operational Grants from the Metropolitan Area. Ngaoundere I, II and III operational grants from the Metropolitan areas appear erratic. Municipalities budget for such grants, but do not receive them each year. x Pages missing. In several administrative accounts there were missing pages. 123