36224 V. 1 POLICY NOTE MUNICIPAL DECENTRALIZACION IN NICARAGUA: FISCAL AND INSTITUTIONAL ISSUES1 November 2004 Main Findings and Recommendations · Nicaragua is pursuing decentralization in an aggressive manner. The Municipal Transfers Law approved in July 2003 will substantially increase the funds available to municipalities from about 1 percent of tax revenues to at least 10 percent in 2010. However, the new law does not specify expenditure responsibilities and previous legislation assigning expenditure responsibilities among different levels of government is ambiguous. · The experience of other Latin American countries that have pursued aggressive decentralization shows that haste, without an appropriate institutional framework and monitoring system, can lead to waste of resources, a worsening of the provision of public services, and increased macro-economic instability. · Fiscal neutrality should be pursued as of 2005. Delays in pursuing fiscal neutrality can threaten macroeconomic stability and make the task more difficult and less efficient in the future. Short-term measures under the control of the Executive should be implemented as soon as possible. In addition, an amendment to the Municipal Transfers Law and the Municipal Law could avoid ambiguities, limit transfers of expenditure and revenue to municipalities until proper conditions are in place and establish proper controls. · Fiscal neutrality should not be achieved as a purely accounting matter, but in an economically efficient and equitable way. Rather than simply cutting central government expenditures to compensate for the amount of transfers, expenditures executed at the central level should be transferred to municipalities. · Solutions to make decentralization fiscally neutral include: a) the central government should suspend expenditures in areas of municipal competence and let them be undertaken at the local level, b) part of the transfers could be used for investments that would be executed by central government agencies' with a proven track record and experience in procuring capital projects, c) new legislation could shift expenditure responsibilities from central government to municipalities, d) municipalities could be required to provide the counterpart financing of the Public Investment Program for projects being executed in their respective territories or contribute a share of project costs by a system of co-financing or matching-grants. The costs of these various options are considered in the document. Two other measures would promote fiscal discipline: using part of the transfers to reduce municipal obligations towards the National Social Security Institute and applying strictly the provisions of the Municipal Transfers Law, which make transfers contingent on the fulfillment of some 1This Policy Note was prepared by a team led by Amparo Ballivián (World Bank Country Manager, Nicaragua) and Antonio Spilimbergo (IMF, Fiscal Affairs Department) and including Jonas Frank (World Bank, LAC- PREM) and Armando Barrios (Consultant). - 1 - basic criteria. · There are other risks deriving from aggressive decentralization in Nicaragua. The monitoring and auditing mechanisms of local expenditures are weak, the central government's institutions in charge of municipalities have overlapping mandates and the central government does not have a clear and consistent plan on decentralization. · In the long-run, to ensure a proper decentralization process the government should: a) reform the central government's agencies dealing with municipalities by establishing clear policy leadership and aligning the mandates of agencies currently financing investments in municipalities with the objectives of fiscal neutrality, expenditure efficiency and national priorities, b) re-organize the legislation dealing with municipalities' responsibilities and c) support efforts to improve the ability to monitor and audit spending by municipalities. · Finally, the success of the decentralization process depends ultimately on the ability by the local population, which is the final beneficiary of local services, to monitor and to reward successful administrators. In the long run, the central government should intervene only to ensure that local population has the mean to carry on careful scrutiny. - 2 - ACRONYMS AND ABBREVIATIONS AMUNIC Asociación de Municipios de Nicaragua AMURACAN Asociación de Municipios de las Regiones Autónomas de la Costa Atlántica CERAP Comité Ejecutivo para la Reforma de la Administración Pública - Executive Committee for Public Administration Reform. CGR Contraloría General de la República ­ Comptroller General Office. CSD Comisión Nacional de Descentralización - National Decentralización Commission. FISE Fondo de Inversión Social de Emergencia ­ Emergency Social Investment Fund. GC Central Government . GCO General Comptroller's Office GDP Gross Domestic Product HIPC Heavily Indebted Poor Countries Initiative HLIC High Level Intergovernmental Commission IADB Inter American Development Bank IDR Instituto de Desarrollo Rural ­ Institute for Rural Development IMF International Monetary Fund INAA Instituto Nicaragüense de Agua y Acueducto ­ Nicaragua Aqueduct and Sewage Institute INE Instituto Nicaragüense de Electricidad ­ Nicaraguan Electrical Institute INIFOM Instituto Nicaragüense de Fomento Municipal ­ Institute for Municipal Promotion INSS Instituto Nicaragüense de Seguridad Social ­ Social Security Institute MG Municipal Government ­ Gobierno Municipal MHCP Ministerio de Hacienda y Crédito Público ­ Ministry of Finance NGO Non-Government Organization OAFE Oficina de Asuntos Fiscales y Económicos - Fiscal Affairs Office. PRGF Poverty Reduction and Growth Facility SECEP Secretaría de Coordinación y Estrategia de la Presidencia - Secretariat for Coordination and Strategy of the Presidency SNIP Sistema Nacional de Inversiones Públicas ­ National Public Investment System - 3 - Table of Contents I. Background................................................................................................. 5 II. Fiscal Issues................................................................................................. 7 II.1 Basic precepts ......................................................................................................... 7 II.2 Municipal Transfers Law........................................................................................ 7 II.3 The Size of the Fiscal Threat .................................................................................. 9 II.4 Putting the cart behind the oxen: Achieving fiscal neutrality............................... 11 II.5 What expenditures can be transferred to municipalities? ..................................... 13 II.6 What expenditures should be transferred to municipalities? ................................ 15 II.6.1 Expenditure transfers: fiscal considerations ......................................................... 15 II.6.2 Expenditure transfers: Efficiency and Equity Considerations .............................. 17 III. Institutional Issues.................................................................................... 20 III. 1 Current Institutional Framework Supporting Municipal Decentralization ........... 20 III.2 Evaluation of the Current Institutional Framework .............................................. 22 III.3 Reforming the Institutional Framework for Decentralization............................... 23 III.3.1 Guiding Principles for Institutional Reform for Fiscal Decentralization.............. 24 III.3.2 Proposed Mechanism for Financing, Mobilizing and Allocating Resources for Municipal Development........................................................................................ 25 III.3.3 Proposed Institutional Framework for Decentralization in Nicaragua.................. 27 IV. Issues for Further Analysis...................................................................... 32 IV.1 Borrowing Capacity.............................................................................................. 32 IV.2 Auditing ................................................................................................................ 33 IV.3 Other ..................................................................................................................... 33 V. Next Steps.................................................................................................. 34 VI. Conclusions and Recommendations ....................................................... 35 References.................................................................................................................. 38 ANNEXES ................................................................................................................. 39 Annex 1: Sources of Financing of Municipalities prior to 1954. ............................... 39 Annex 2: Municipal Transfers Law............................................................................ 39 Annex 3: Transfers by Municipality ­ Current and Capital breakdown..................... 39 Annex 4: Regular Contributions and Debt Service to INSS by Municipality............ 39 Annex 5: Projects that can be Transferred to Municipalities...................................... 39 Annex 6: Experience with Decentralization in Selected Latin American Countries.. 39 Annex 7: Summary Legal Framework for Decentralization....................................... 39 Annex 8: Main Institutions Financing Decentralized Investment in Nicaragua......... 39 - 4 - Policy Note Municipal Decentralization in Nicaragua: Fiscal and Institutional Issues I. Background 1. Nicaragua is aggressively pursuing fiscal decentralization. This process is an opportunity to enhance the quality of local services and is part of the democratization process of Nicaraguan society. However, decentralization also presents potential fiscal and efficiency risks if it is not properly designed. To ensure the success of fiscal decentralization, the process that has already been started needs to be complemented with measures to avoid fiscal problems and improve the effectiveness of public service delivery. This Policy Note aims to provide some guidance and advice aimed at easing the macroeconomic risks and improving the chances of success of the decentralization process in Nicaragua. 2. On July 2003 the Nicaraguan National Assembly approved Law No. 4662 mandating the transfer of 10% of tax revenue to the country's 152 municipalities by 2010, starting from at least 4% of tax revenue in 2004 (US$ 26 million). This measure followed several years of discussion and efforts to promote municipal decentralization in the country. Before the Law was approved at least three central-level institutions, the Institute for Municipal Promotion (INIFOM), the Emergency Social Investment Fund (FISE) and the Rural Development Institute (IDR), financed goods, works and services in the municipalities for a combined investment budget in 2003 of US$ 105 million. The municipalities also received direct transfers amounting to approximately US$ 24 million in 2003 -mostly financed by foreign donors- and raised their own revenues to the tune of US$ 28 million (C$ 432 million). 3. Law No. 466 was considered a big step forward in the decentralization process by mayors and some donors, who had been arguing for several years that the 1988 Municipal Law, its amendments introduced in 1998 and its regulation, had assigned responsibilities to municipalities without the concomitant resources to fulfill them. The new Transfers Law, however, mandates the Nicaraguan Treasury to transfer resources without raising new revenues or defining which central government programs need to be transferred in order to preserve fiscal balance. 4. Whereas previously the dilemma had been an excessively low and uncertain level of transfers, the problem now appears to be the opposite: a sudden inflow of resources that will challenge absorption capacity at the local level3 and, most importantly, threaten 2In Spanish: Ley de Transferencias Presupuestarias a los Municipios de Nicaragua, Ley No.466, adopted August 2003, hereafter called Municipal Transfers Law and not to be confused with the Municipal Law, passed on June 28, 1988 and amended on June 26, 1997 which is the basic legislation governing municipal activities. 3 A high level of intergovernmental transfers is not negative by itself and may constitute a favorable environment in some circumstances. It is desirable in principal-agent models where subnational government act as delegates of the center. Transfers can compensate inter-jurisdictional externalities associated with service delivery. And they typically compensate gaps between fiscal capacity and needs on local levels. - 5 - fiscal stability at the national level, unless the corresponding transfers of central government expenditure are made. The country, therefore, would need to take additional revenue measures or expenditure cuts at the central level, lest the hard gains in macroeconomic stability since 2002 and the PRGF program that accompanies them go astray. 5. Nicaragua is embarking on an ambitious fiscal decentralization program at a time when it is still recovering from a serious economic crisis. In 2001, the fiscal deficit of the central government was 12.4% of GDP. By 2003 it was brought down to 8.1% of GDP and it is projected to be 4.3% of GDP in 2004. Launching aggressive fiscal decentralization in such circumstances could undermine the government's efforts to stabilize fiscal accounts. Most national policy-makers therefore recognize that the only way to decentralize responsibly is to honor the principle of fiscal neutrality, that is, to avoid any impact from the Municipal Transfer Law on the national deficit. This implies identifying expenditures the central government currently executes but that are suitable to be carried out by municipalities. By transferring these expenditures in financial amounts equivalent to municipal transfers, the center may be able to neutralize the fiscal impact of decentralization.4 6. Section II considers possible expenditure transfers from the central level that could result in fiscal neutrality of the Law. Since some of the options explored have important implications for the institutional framework, section III of the paper explores some ideas in this regard. The document does not deal with revenue measures or with other issues surrounding decentralization, which are many ­technical assistance needs, financial controls, results monitoring, municipal indebtedness, criteria for allocating resources to municipalities, human resources policy, participatory planning and others which are mentioned only in passing. Some of these themes are briefly indicated in section IV, highlighting some issues to be further explored. Section V summarizes the conclusions and recommendations. 7. This Policy Note is based on four background papers5 and has benefited from extensive consultations with several stakeholders in Nicaragua, mostly in Government and the international community. Any policy decisions would need to be discussed and consulted further, including with municipal representatives, local community representatives and legislators. These efforts should contribute to rationalize public expenditure, enhance fiscal efficiency and keep the decentralization process moving forward in a sustainable way. 4 The authorities consider that revenue increase measures are not feasible within the present political environment. Since the present Government has already passed two tax increase laws, additional tax measures would likely yield marginally less revenue and the Executive branch controls only 10% of congressional votes, so it is unlikely to have the required political capital for further tax increases. Further, since 10% of any tax revenue increase would accrue to municipalities, achieving fiscal neutrality through revenue-side measures would require an increase of 11.11% in revenue.. 5 Spilimbergo, Antonio, "Fiscal Decentralization in Nicaragua," April 2004; Barrios, Armando, "Marco Institucional de la Descentralización en Nicaragua," June 2004; Frank, Jonas, "Decentralization in Nicaragua: The Challenge of Fiscal Neutrality," May 2004; and Budinich, Ema B., "Identificación de Alternativas de Descentralización de Gasto Público hacia las Municipalidades", 2004. - 6 - II. Fiscal Issues II.1 Basic precepts 8. The Nicaraguan Constitution establishes that Municipalities are the basic territorial units of the state and mandates that ordinary laws should specify their responsibilities. Article 177 of the Constitution specifies that a fixed percentage of central government revenues should be transferred every year to municipalities to fulfill their duties. The Constitution also states that municipalities with less own revenues should receive more transfers. The details on the implementation of this mandate and the municipalities' duties are left to ordinary laws, as is the precise percentage level of the transfers. Before the passage of Law No. 466, there was no law regulating such transfers. The sources of financing for municipalities prior to the passage of this Law are summarized in Annex 1. 9. Passage of Law No. 466 implies that the Nicaraguan National Assembly has not created one comprehensive municipal law regulating financing and expenditure responsibilities. Rather, it has introduced two sets of laws: one regulating only responsibilities -in a broad way and without specific reference to expenditure responsibilities-, the other one regulating only financing. This has created a hiatus between expenditures responsibilities and financing that has made inter-governmental relationships troublesome. II.2 Municipal Transfers Law 10. Law No. 466 mandates that 4% of tax revenue (0.8% of GDP) must be transferred to municipalities in 2004 and that this share must be increased by at least 0.5% every year thereafter -provided GDP grows by at least 1% in the previous year-, until it reaches at least 10% of tax revenue in 2010. The total amount thus calculated must be transferred to each of the 152 municipalities according to a sophisticated formula that takes into account fiscal efficiency, fiscal equity, population and budgetary execution, as explained in Annex 2. There are few constraints on the use of these transfers with the exception that the total amount transferred to each municipality must be spent on a varying proportion in current and capital expenditures (see table 1 for the rules on the subject and Annex 3 for the total transfer and the current/capital distribution by municipality). On average, the legal provisions imply that approximately 30% of the transfer must be used for current expenditures and the rest for (loosely defined) investment expenditures.6 Putting things in perspective the country has been moving from practically discretionary expenditure functions together with discretionary revenue sources, to a more stable fiscal system with predictable fiscal resources, which should allow for predictable and stable expenditure responsibilities. 6 According to the Municipal Code, municipalities are grouped into different categories based on their revenue raising capacity. Unfortunately, the Transfer Law allows fiscally weaker municipalities a higher margin for current expenditures, a rule that may undermine their possibility to perform their own fiscal adjustment at the local level. - 7 - 11. The Municipal Transfers Law approved in July 2003 substantially increases the funds available to municipalities. Out if 152 municipalities, more than 120 will double their budget in 2004. For one third of all municipalities, the transfers in 2004 will represent more than nine times their own revenues and for more than two thirds of them transfers will represent almost twice as much as own revenues. Table 1 Transfers per Category of Municipality Number of Minimum Transfers as share Amount per capita (2004 C$) Category /1 municipalities Percentage of the of municipal Municipal Revenues per category transfer for revenues Municipal revenues investment /2 and Transfers Managua 1 90% Category B 11 80% 0,24 144 178 Category C 9 80% 0,38 125 173 Category D 22 80% 0,77 103 183 Category E 43 70% 1,88 60 174 Category F 13 70% 2,84 49 189 Category G 19 60% 4,01 37 183 Category H 34 60% 9,17 22 226 TOTAL 152 0,94 94 183 Sources: Ema Budinich (2004). 1/ Administrative categories defined according to the amount of local revenues. 2/ Minimum share for investment according to the Municipal Transfers Law. 12. The new law has several shortcomings in its design. The main problems are: a) The law is not fiscally neutral. While it devolves new revenues to municipalities, it does not devolve any new expenditure responsibilities to them. If the central government continues paying for local projects and services, while municipalities use the transfers for the same purposes, the fiscal deficit inevitably increases. Municipalities are already using the transfers as additional funds and the central government had no plan to cut expenditures at the local level in 2004. b) By linking the level of transfers to the growth of GDP the central government may still transfer revenues it does not actually dispose of. This is because tax revenue is not necessarily elastic with regards to GDP growth. Although in the last eight years (1995-2003) the tax elasticity relative to GDP is 1,28 per average (in real terms), there may not be a positive relationship in every year. c) The law has no incentives to orient local investment decisions in the direction of national public investment priorities and the HIPC-related commitments to increase pro-poor public expenditure. d) The rules regulating current and investment expenditures will be difficult to monitor in practice. Whether municipalities will be able to spend according to the rules will largely depend on: (i) the structure of expenditures prior to decentralization and (ii) the type of expenditures to be undertaken by - 8 - municipalities (education, for instance, has higher share of current expenditures whereas the contrary is true in the road sector). e) Municipalities with weaker fiscal capacity (that is, less ability to generate own revenue) are given a higher margin for the use of current expenditures. This may undermine their ability to run their own fiscal adjustment programs under shortfall of revenue (transfer volatility inherent to revenue-sharing arrangements), particularly when faced with rigid expenditures (education, for instance). f) By earmarking an increasing share of fiscal revenues for municipalities the law makes the national budget less flexible. This is a general dilemma of general revenue-sharing arrangements, but is worsened in the case of Nicaragua by the fact that several expenditures at the central level are already earmarked.7 g) The law puts heavy emphasis on ex-ante controls rather than ex-post controls. Transfers should be automatic, that is, based on the effective revenue obtained on a daily basis. The use of quarterly transfers, or even of monthly transfers, induces stop-and-go activities in municipalities and undermines their ability to provide services which are stable (and hence need stable form of financing). h) The law does not contain mechanisms to improve local governance and to ensure proper auditing of local expenditure. i) Municipalities are allowed to borrow using future transfers as collateral. This poses a potentially large financial problem, especially because the legal framework regulating municipal borrowing is still weak. II.3 The Size of the Fiscal Threat 13. Table 2 shows the amount that is to be transferred to municipalities up to 2010. The following assumptions underlie this table: · Annual GDP growth increases over the following years. Following IMF projections, in 2004 growth would be 3.7% to increase up to 5% in 20098. This is above the threshold of an annual GDP growth of 1% that the Law uses as condition for an increase in transfers. 7The biggest of these is the Constitutional mandate to assign 6% of "the budget" for higher education. After years of discussion about whether the basis for calculation should be budget revenue or budget expenditure, the Supreme Court has recently ruled that this should be interpreted as revenue plus deficit. Another big case in point is the legal requirement to assign 4% of total revenue to the Supreme Court. 8In the absence of better projections for 2010, this exercised used the same annual growth rate assumed for 2009 (5.0%). - 9 - · Tax elasticity relative to GDP growth is 1. This is below the average of tax elasticity of 1.28 (in real terms) between years 1995 and 2003. In other words, tax revenue is likely to grow over-proportionately relative to GDP growth. Should this trend continue then the central government would be forced to transfer even a higher amount of resources than projected in the costing exercise. · The share of tax revenue transferred increases 1% per year. · Fiscal neutrality is achieved progressively, with full neutrality reached in 2007, in line with recent government decisions. Table 2 Central Government Transfers to Municipalities (in million C$) 2003 2004 2005 2006 2007 2008 2009 2010 TRANSFERS OF REVENUE Total Tax Revenue 9,851 11,416 11,838 12,253 12,743 13,316 13,955 14,653 % Transfer to Municipalities 1% 4% 5% 6% 7% 8% 9% 10% C$ Transfer to Municipalities 99 457 592 735 892 1,065 1,256 1,465 o/w for capital expend. (70%) 69 320 414 515 624 746 879 1026 o/w for current expend. (30%) 30 137 178 221 268 320 377 440 % absorption towards fiscal neutrality (PRGF) 0% 0% 33% 66% 100% 100% 100% 100% TRANSFERS OF EXPENDITURE (C$) 0 0 195 485 892 1,065 1,256 1,465 Pour memoire: Annual GDP growth (constant prices) 3.7% 3.5% 4.0% 4.5% 4.8% 5.0% 5.0% Transfers of revenue in US$ 7 29 35 42 48 55 62 69 Transfers of expenditures in US$ 0 0 12 28 48 55 62 69 Transfers of expenditures - % of total exp. NFPS GDP (US$) 4,134 4,363 4,524 4,683 4,870 5,089 5,334 5,600 Revenue transfers/GDP 0.16% 0.66% 0.78% 0.90% 1.00% 1.08% 1.16% 1.23% Exchange rate 15.1 15.90 16.70 17.53 18.41 19.33 20.29 21.31 Source: Authors' calculations. 14. Table 2 shows that the central government would have to transfer expenditures worth around C$195 million (approximately US$ 12 million) in 2005 and more than C$ 1,465 (US$ 69 million)9 by 2010. However, as mentioned above, the central government has some flexibility in selecting its own schedule for the increase of transfers so as to meet the 10% target by 2010, either by proceeding more aggressively or by delaying the transfer of funds. This, then, would directly affect the pace, scope, and depth of the expenditure decentralization program the central government would need to design in order to make transfers fiscally neutral. Figure 1 depicts this graphically. 9Table 1 assumes that the nominal exchange rate also rises by 6% every year, from an average of 15.90 C$/US$ in 2004. - 10 - Figure 1. Schedule of Municipal Transfers, 2004-2010 1,400 Slower fiscal 1,200 decentralization Target: 10% Transfers to Faster fiscal 1,000 municipalities decentralization 800 For capital Cordobas expenditures 600 Starting (70%) Point: 4% Million 400 For current expenditures 200 (30%) 0 2004 2005 2006 2007 2008 2009 2010 Years II.4 Putting the cart behind the oxen: Achieving fiscal neutrality 15. Under normal circumstances, decentralization involves a process of deciding what activities are best undertaken at the local level, estimating their cost and transferring to municipalities the resources and the mandate to undertake those responsibilities. Most of the time these decisions are heavily influenced by politics, but the basic sequencing of events is respected, that is, the allocation of resources follows the assignment of responsibilities. This is not the case of Nicaragua, where Law No. 466 transfers resources only, on the erroneous premise that a previous Law had given municipalities an unfunded mandate. Therefore, the challenge of the Nicaragua authorities is the inverse of the usual, i.e. to transfer expenditures that would match a pre-determined level of resource transfers.10 The Law has put the oxen behind the cart and now the challenge is to put the cart back behind the oxen. 16. Notice that instead of referring to expenditure cuts, table 2 refers to expenditure transfers. This reflects the principal recommendation of this Concept Note, i.e. rather than cutting expenditures elsewhere, which would achieve neutrality in a purely accounting sense without regard for economic efficiency, the central government should transfer the responsibility for expenditures it currently undertakes to the municipalities. This recommendation is based on: 1) the hypothesis that current expenditure responsibilities of municipalities hardly warrant transfers of up to 10% of government revenue (or 1.15% of GDP)11, and 2) the recent contraction of central government expenditures (from 26.6% of GDP in 2001 to 21.3% of GDP 10Or be larger. Some of the options for transferring expenditure responsibilities to municipalities presented in sections II.5 and II.6 imply an amount that is larger than 10% of fiscal revenue. Transferring a larger amount may also provide a politically more feasible way of transferring expenditure. 11See paragraphs 18 and 19, and table 3. - 11 - projected in 2004), combined with other large and inescapable expenditure responsibilities of the central government (see footnote 7) presumably leaves little to cut at the central government level without disrupting services and leading to inefficient allocations of public expenditures. 17. Furthermore, such a large amount of revenue transfers without transferring expenditure responsibilities defeats one of the principal objectives of decentralization. Among other reasons, decentralization is desirable because some government services are better delivered at the local level and because social auditing mechanisms work better at this level. But for this to be the case, local governments must be given expenditure responsibilities that, before decentralization, rested at the central level. 18. Some advocates of the new Law claim that the Municipal Law gave municipalities an unfunded mandate. To evaluate this claim, consider the set of responsibilities assigned by the legislation to municipalities vis-à-vis those assigned to the central government, presented in table 3. Although we know of no study that has estimated the cost of delivering these responsibilities, it is hard to be persuaded that they would represent 1.85% of GDP (1.15% central level transfers and 0.7% of own resources). 19. The arguments on this issue are difficult because the legislation just refers to "competences" without specifying if those are expenditure responsibilities, policy- making responsibilities, regulatory responsibilities or responsibilities of any other sort. To complicate things further, other legislation also provides that some of these responsibilities are to be undertaken by central ministries or autonomous agencies (see paragraph 34). Hence, the laws and regulations governing decentralization in Nicaragua leave ample room for differing interpretation and both central government agencies -line ministries and autonomous entities- and municipal governments consider they have a legitimate claim to resources to fulfill the same responsibilities. 20. For these reasons, a second recommendation of this Note is to define clearly and unambiguously ­read exclusively- what are the expenditure responsibilities of municipalities. A clear definition of expenditure assignment would allow an estimation of the required transfers based on technical grounds and would improve accountability. - 12 - Table 3 Municipal and Central level Responsibilities Central Municipal Governments Government Exclusive: Defense, national security, Exclusive: Urban infrastructure development and Expenditure foreign relations, electoral system, justice maintenance, management of local public services, Responsibilities system, school education, higher (University) construction, maintenance and administration of libraries, education, central banking, regulation of promotion of handicrafts, construction and management of financial system, taxation and public public cemeteries, construction and maintenance of sports, borrowing, regulation of public services and parks and recreational facilities, civil register and local public infrastructure, civil security and social institutional capacity building. security. Assigned to both: Roads, public health, electricity, environment and natural resources conservation, agricultural and forestry development, culture, tourism promotion and development, public transportation development and regulation, solid waste disposal, execution and monitoring of local public infrastructure, water and sewage management. Resources Exclusive: Income and profit tax, value added Exclusive: Local sales tax, property tax, some minor local tax, excise taxes, custom duties, other small taxes and fees. taxes and fees. Shared: Minor taxes. taxes on sugar. Transfers Before 2004: discretionary transfers from central government to municipalities. The amount was discussed every year with the budget and usually amounted to 1 percent of fiscal revenues. Starting in 2004: by law the central government must transfer at least 4 percent of fiscal revenues. This amount is progressively increased by at least half percentage point till reaching 10 percent of fiscal revenues in 2010. Legal Limits to Common: The Law on Public Debt regulates all public borrowing. Approval by Presidential Decree, Borrowing following the determination of financial sustainability of aggregate public borrowing by the Ministry of Capacity Finance, is required for issuance of debt by any public institution. The Municipal Budget Law states that municipal public debt services should not exceed 20 percent of the annual municipal budget. Moreover, a municipal administration cannot leave debt for future administrations unless is authorized by a citizens' council. Auditing Common: De jure, the General Controller Office (GCO) has the general mandate to oversee all public expenditure. De facto, the GCO does not have the means to audit municipalities. Very few municipalities have internal auditing Source: Authors' elaborations based on Municipal Law, Municipal Transfers Law, Municipal Budgets Law and Law on Public Debt and budgetary allocations. II.5 What expenditures can be transferred to municipalities?12 21. There are several options to transfer expenditure responsibilities to municipalities. We have identified four basic options, which are not mutually exclusive: 12 For the purpose of this study, the option to amend the Municipal Transfer Law has been disregarded. Although a change in the Law should be given serious consideration on technical grounds, it is a matter of fact that transfers in the amount of 4% of central government revenue have already been executed. Authorities operate within this constraint, and this is simply the starting point to study the choices that Nicaragua has to solve the present dilemma situation. - 13 - a) Identify activities that are assigned to municipalities under existing legislation but are being undertaken by central government agencies and mandate, through a regulatory decree, all central government agencies to cease such activities. To avoid a sudden interruption of investments, the government could seek an agreement to account some or all foreign-financed projects currently implemented in municipalities in sectors for which municipalities are responsible, as part of the mandatory transfer. This agreement would not require any change in the current laws, but would require a particular interpretation of article 6 of Law 466.13 b) Part of the transfers could be used for investments that would be executed by central government agencies' with a proven track record and experience in procuring capital projects. We discuss this in further detail in Section III. c) Entrust municipalities with the responsibility to provide one or more of the public services or functions currently assigned to the central government. Depending on the sectors and the extent of transfer of responsibility, this option may be implemented just by decree or by an amendment to the Law of Municipalities. d) Get municipalities to finance part of the Public Investment Program (PIP) executed in their territories, by either: d.1) Establishing that municipalities must provide the counterpart financing of foreign-financed public investment projects being executed in their respective territories. d.2) Require municipalities to contribute an even larger share of project costs (i.e. larger than the counterpart requirement vis-à-vis the foreign financer), by a system of co-financing or matching-grants which could be variable depending on the national priority accorded to the sector or location. We discuss this option in more detail in Section III. 22. Along with any of the above options we recommend the adoption of a couple of complementary measures. First, the automatic reconciliation of at least part of the transfers to pay the obligations of the municipalities against the social security system (INSS). Municipalities are not regular in their obligations vis-à-vis the social security. The restructured accumulated debt is over C$ 100 million. At the moment, the central government does not have good legal instruments to foreclose the transfers, therefore, INSS is pursuing a negotiated solution with the objective of recovering about C$ 47 million in 2004. A more aggressive strategy, which does not require amendments to any laws, could increase this amount considerably. Regular contributions mandated by law should also be automatically discounted from the transfer. 13Article 6 states that expenditures at the municipal level financed with donors' resources should be additional to the amount of central government transfers. However, the way it is drafted would allow to argue that this applies only to [budgetary transfer financed with donor's money and not to foreign-financed projects.] - 14 - 23. While this measure would not contribute directly to fiscal neutrality, we believe this is an important complement to expenditure transfers because: i) it is one of the easiest ways to transfer current expenditure, ii) it sets a precedent for further automatic conciliation of other debt and iii) it marginally, but in a symbolically important way, eases the burden of INSS on the budget. Annex 4 shows the current debt and regular employer contributions of each municipality to the INSS. 24. Secondly, the government could opt for a very strict interpretation of the transfer law, which requires that municipalities must fulfill some conditions before receiving the funds. The law requires the municipalities to send appropriate documentation by mid-February to receive the funds. As the Law was implemented as of 2004, for the transfer corresponding to the first quarter only 2 out of 152 municipalities could present the appropriate documentation in time. Nevertheless, the Fiscal Transfers Committee decided to extend the deadline to present the appropriate documentation, by interpreting the Law in a very liberal way. In addition, the Ministry of Finance lacked the appropriate personnel to evaluate the completeness and appropriateness of the documentation, but political pressures determined that transfers had to be executed anyway. The message to municipalities was that they would receive all transfers even if they do not fulfill any criterion, as prescribed by the law. The central government could be very firm on this and interpret the law in a strict sense. While this would detain some of the short-term pressures on the National Treasury and would not solve the fiscal problem in a sustainable way, we recommend that this stance be adopted in conjunction with any of the above options in order to improve governance of the system of transfers. II.6 What expenditures should be transferred to municipalities? 25. We propose that the options presented in the previous section be evaluated on the basis of two main criteria: 1) their contribution to fiscal neutrality and 2) efficiency and equity in the delivery of government services, especially to the poor. Of course, the authorities might also consider as an important criterion the political feasibility of each option. It is also clear that a combination of options may provide the best solution for the particular case of Nicaragua. II.6.1 Expenditure transfers: fiscal considerations 26. We have attempted to quantify the savings to the central government budget of each option. This work is by no means exhaustive and plenty of additional estimates and data-finding are needed to arrive at more precise figures, but the intention is to provide orders of magnitude to guide decision makers. The results are presented in table 4, followed by explanations about each option. - 15 - Table 4 Summary Fiscal Impact of Options (million cordobas, constant) 2004 2005 2006 2007 2008 2009 2010 Fiscal Savings Option a : CG ceases activities of MG competence* 2,131 2,178 1,750 1,304 1,367 Option b: Vouchers (INIFOM, FISE and IDR only) 1,200 Option c: Transfer activities from CG to MG 0 368 882 1,273 1,717 1,991 2,091 Option d d.1 MGs pay counterpart 991 1,028 1,067 1,112 1,163 1,218 1,279 d.2 Matching-grants at 50% of PIP in Municipalities 2,008 2,082 2,161 2,252 2,356 2,469 2,592 Pour Memoire: Transfers to municipalities 408 529 659 801 957 1,129 1,317 NOTES: CG = Central Government, MG = Municipal Governments Option a): Central government is legally mandated to stop expenditures in areas of municipal competence. The figures in the table include investments only and are taken from Budinich (2003). Budinich's paper identifies investments that are currently financed partly or totally by the central government and are in the areas of municipal competence. Annex 5 gives the sum by year and central government institution. This analysis could be complemented with the identification and costing of current expenditures undertaken by central government agencies in areas that are by law assigned to municipalities, such as water and sanitation services, environmental protection, and rural roads. However, the transfer of current expenditure responsibilities must be approached with extreme caution in sensitive areas, such as health, where the poor may suffer from interruption of service if these are not well-planned and progressively implemented. Option b): The amount for option (b) corresponds to the investment budget of FISE, INIFOM and IDR in 2004, a part of which could be distributed to municipalities in the form of vouchers in the same proportions as mandated by the transfer formula and to count against them. Municipalities would use these vouchers to use to pay for projects according to municipal investment priorities, therefore also ensuring that the projects are demand-driven. This solution appears very appealing in the long run because it gives municipalities autonomy in choosing the investment plans while provides a good instrument of quality control. However, it could be difficult to implement in the short run because it may require some time to set up the institutional arrangements, particularly in the legal framework. - 16 - Option c): The data represents the transfer of certain responsibilities to the municipal level in seven sectors: health, education, social assistance, water, transport, rural electricity and environment, as reported in Frank (2004). Option d): Sub-option d1) is calculated as the total counterpart funds currently assigned by the national government for the execution of public investments at the municipal level as per the 2004 Public Investment Program. Sub-option d2) is calculated as half the total 2004 Public Investment Program executed in municipalities (SNIP data). It represents an application of the co-financing matrix explained in section III.4.2 of this Policy Note, with the average co- financing share set at 50%. This sub-option would only contribute to fiscal neutrality if the co-financing provided by national institutions can count against the transfers. II.6.2 Expenditure transfers: Efficiency and Equity Considerations 27. Theory and experience with decentralization in other countries14 tell us that one or several of the following criteria should be met when deciding to transfer responsibilities from the central to the local level trying to optimize service delivery: · Expenditures should be transferred only in areas where municipalities can provide services more efficiently than the central government. Following the subsidiarity principle, the lowest level of government capable to providing public services should be assigned the responsibility. In some sectors, municipalities can better reveal user's preferences than national level institutions. · Expenditure responsibilities to be transferred should be in line with the financing mechanism of a general revenue-sharing arrangement. Such a form of financing is primarily useful to carry out expenditures that are stable over time, that is, services that are delivered as routine exercises. This makes the use of resources less suitable for "one-shot" financing requirements. In addition, not all responsibilities carried out by local government may make sense in the overall, sector framework. · Expenditure responsibilities must be assigned so as to ensure accountability among users, policy-makers and service providers. The participation of municipalities in service delivery may not always work in favor of this. Decentralization implies that citizens, politicians and service providers have to find an arrangement in which individuals are clearly responsible for their actions. 28. Nicaragua needs to strike a balance between the criteria listed above. But not all principles can be honored and implemented at the same time and trade-offs between different solutions exist. The 2004 World Development Report "Making Services Work for the Poor" contains a useful tool to help with the decision-making process. The decision would depend on three dimensions: (a) how pro-poor or how clientelist is the 14See annex 6 for a summary of experience with decentralization in selected countries in Latin America. - 17 - political system, and therefore how much information clients need regarding their entitlements, (b) how homogenous or heterogeneous are the citizens and therefore how their needs differ in each jurisdiction, and (c) how difficult it is to monitor services for either governments or clients. For instance, when services are easy to monitor, i.e. those whose results are easy to gauge by a quantitative indicator such as provision of water or electricity or construction of a road, vouchers or private provision are best suited. On the other hand, when the quality and results of the services are difficult to monitor, such as education and health services, reliance on co-payments is advised, as they decrease the moral hazard problem inherent in the difficulty to monitor. These three dimensions and their consequences for what activities to decentralize are illustrated in Box 1. 29. Given the characteristics of the Nicaraguan population and its political system, it seems that the seventh and eight forms of provision are the best suited for the country. Following the analysis of the World Development Report 2004 (Box 1), in a system where pro-poor politics is still ephemeral but where clients are heterogeneous, then either decentralized, rule-based allocations, vouchers or private provision (option 7), or, conversely, the strengthening of client power (option 8) are recommended approaches for service delivery. 30. Municipalities already provide a set of primarily urban services, such as street lighting, recreational facilities, urban streets, among others. Were they to expand their role in service delivery they could venture into both the infrastructure, as well as social sectors. In the area of infrastructure, the adoption of responsibilities in secondary road maintenance would be suitable as monitoring is relatively easy and technical capacities are limited to deliver those services. In the social sectors, some social assistance programs currently managed by the Ministry of Family Affairs are a suitable entry point, given the knowledge of local circumstances and heterogeneous clients municipalities have compared to central level institutions. In both cases, particular service delivery models would need to be crafted that take into account management principles of option seven and eight as highlighted in Box 1. - 18 - Box 1 Accountability Relationships among Citizens, Politicians and Service Providers The 2004 World Development Report gives basic criteria for deciding what kind of services are best provided at different levels of government, characterizing the range of public services in eight categories, according to three dimensions: (i) whether the political system is clientelist or pro-poor, (ii) whether clients are homogeneous or heterogeneous and (iii) whether the provision of services are easy or hard to monitor. The appropriate service delivery mechanism for each category is explained in the following graph. Eight Sizes Fit All: Source: Development Outreach, March 2004. - 19 - III. Institutional Issues 31. Nicaragua has a history of rapidly changing institutional framework. As the new Municipal Transfers Law imposes new obligations and challenges, it will be necessary for Nicaragua to change yet again the way agencies operate and coordinate. There are new players with new interests (for instance the Transfers Commission) and a different allocation of responsibilities and fiscal resources implies that existing institutions will need to be reorganized. In the following section we explain the current institutional setup and describe its problems. We then propose some realignments, based on the best options to achieve fiscal neutrality as presented in the previous section, pro-poor spending considerations and international experience (see Annex 6 for a summary of such experience in Latin America). III. 1 Current Institutional Framework Supporting Municipal Decentralization 32. Annex 7 contains a summary of the legal framework for decentralization in Nicaragua. The leadership and coordination functions on reforming, modernizing and decentralizing the State in Nicaragua are shared between the Vice-Presidency, the Secretariat of the Presidency and the Secretariat for Coordination and Strategy of the Presidency (SECEP). The detailed development of these areas of reform is formally focalized in the Executive Committee for Public Administration Reform (CERAP), which includes a sector commission for decentralization (CSD). The CSD has the mission of designing and promoting decentralization in Nicaragua. The CSD is chaired by INIFOM's President and integrated by representatives of the Presidency's Secretariat, the ministries of health, education, environment and infrastructure, and the Presidents of FISE and IDR. 33. More recently, Law No. 466 created a Transfers Commission, with some functions on guidance and financing decentralization. It is integrated by the President of INIFOM, the Presidents of AMUNIC and AMURACAN (two associations of municipalities), the Director of the Public Investment System (SNIP), the President of the Municipal Affairs Commission of the National Assembly and the Comptroller General. Paradoxically, the Ministry of Finance participates only as observer in that Commission. Finally, three institutions carry out most of the financing of investments at local level in Nicaragua: (i) the Nicaraguan Institute for Municipal Promotion (INIFOM), (ii) the Emergency Social Investment Fund (FISE) and (iii) the Institute for Rural Development (IDR). Annex 8 contains a description of the main institutions financing decentralized investment in Nicaragua. A summary of these agencies' responsibilities vis-à-vis decentralization is provided in table 5. 34. Central government agencies, including the Ministries of Education, Agriculture, Environment, Health and the National Water Company have been opening regional offices throughout the country and developing a set of municipal delegation initiatives. These actions could contribute to intensify interaction between both government tiers and create local managerial capacities to delivery public services and programs. - 20 - 7 6 7 offices 8 plus 14 central central central No. (HQ regional) (HQ, regional, projects) None None One, One, One, catedidni ) ities HQ nent,a No. 223 at 205 200 perm mbersem nent,a perm mbersem active 453 1 1 Th employees (60% None Board None Board ers.htod an t ) . tsnem 15 228.9 10.6 0.0 0.6 talto Budge 218.3 513.8 513.8 469.4 468.8 ern ation million are ets gov ilities.bsi 2004 (C$ Total: Local: Ext.: Total: Local: Ext.: Total: Local: Ext.: None None None None dguB.n callo,s on resp s't Decentraliz 16 to ciation enm for rural local asso rnev service specially easra to ralizatiot and ices encies go 5 ork water, and areas al serv ag erscu al planning, and health, ent rur transfers financial ecendla - functions. in rural Table Framew in insk and odrp nistrativei of public cipinum 21- (housing, of public es,iv to nistrationi endations. cipinumot areas. erat ntse education, protection. unicipalities,m anagemm wor adm system fors alizationr Coordination adm ins to activities tructure tion coop of decent Activities municipalities above social tax cen the project fiscal, of recomm system andy relatedes eyth relatedrost Institutional to the investm infras operational, tho tod ce nt,e project and itigam tega enm gne in tion nte assista al nts.e keam and be,sn Str assi planning, regarding and andsm abilityts toylno transfers rnevog sistan refer those Summary as nagemam .ntem unicipalm nical al fiscal sanitation nte Nor zatioina investm tech ent. productive cologice kinga ecentralizationd governm applicaeth financial .ntem tional enting Na and eesyolp org rale cipinum to only Technical financial nageam investm Capacity-building Financing roads). Finance water Provide in developm Financing Financing Policy-m functional municipal Oversee municipalities Control, Budgeting. nageam plem em sev with Im Overseeing Overall of ugh en n erbmun roth fersres rtaked ities and unes ployee em ission activ tho of ities to erb ion/Bodyt ly Decentralizatio Comm activno ertakes on num und ission 17 refer Institu Data IDR 15 16 ereh Budget, 17 INIFOM FISE IDR National Comm Transfers CGR MHCP SECEP 35. Local governments still work in a very localized public sphere, detached from national priorities. That means that Nicaragua is not taking advantage of the interactions between both government tiers, in order to achieve improvements in allocative efficiency, responsiveness and democratization of public choices. Therefore, it is advisable that local governments be incorporated in the delivery of services traditionally provided by the central government (health, education, social programs, housing), through modalities such as delegation and management contracts, shared provision agreements, and so on, accompanied by credible mechanisms of monitoring and accountability. In section III.4 we present a possible framework to accomplish this. III.2 Evaluation of the Current Institutional Framework 36. The current legal framework does not contribute to the efficiency of local public expenditure. There are several reasons as to why the current inter-governmental relationship leads to inefficient outcomes: a) Several government agencies have various responsibilities for steering the decentralization process. There is neither an acknowledged leader nor a master plan subscribed by the mayor stakeholders. b) The respective responsibilities of municipalities, central government, and governmental agencies are unclear, leading to duplications and voids. c) Sector ministries undertake decentralization initiatives at the margin of the efforts to achieve fiscal neutrality. d) The coordination between local expenditure plans and national sector targets is weak. e) Many municipalities, but not necessarily the small and poor ones, have weak governance. The transfers are multiplying the funds available by several times for more than half of the municipalities, hence exacerbating resource-waste potential. f) The central government does not have an appropriate legal instrument to intervene whenever the level of local services falls below an acceptable level. The constitution states that the central authority is ultimately responsible for the welfare of citizens. However, the law does not specify the form this controlling function should take. g) Auditing of local expenditure is weak or non-existing. h) Incentives to collect local taxes are still too small. 37. All these problems need a reorganization of inter-governmental relationships. The authorities should re-thing the decentralization strategy with a view to clarifying the respective goals of each level of government. Five guiding criteria should be followed: fiscal neutrality--revenues transferred to municipalities will be matched by transfers of - 22 - expenditure responsibilities, efficiency--the quality of the municipal services should improve as control is shifted to local governments, equality--municipalities with less means to fulfill their mandates should be given relatively more resources, non- ambiguity--the assignment of expenditure responsibilities between municipalities and central government should be clear, and transparency--proper monitoring should ensure the transparency of resource allocation at all levels of government. 38. The essential missing element is a recognized policy leadership and institutional coordination agency, that: (i) sets overall direction and monitors progress in carrying out the implementation of a national strategy for territorial decentralization, (ii) can blend a clear vision on intergovernmental mechanisms, with the transfer of effective legal and financial autonomy to local governments, so that they deliver efficiently the services demanded by their communities and (iii) acts as a single valid interlocutor with the legislative branch and with foreign cooperation agencies. 39. Originally, INIFOM was recognized as the leading policy-making and coordination institution, but its deviation into financing activities has come at the cost of loosing leadership in this respect. At the same time, the Decentralization Commission has been loosing relevance in the past few years and the creation of the Transfers Commission has weakened even more its leadership. To fill this void, the authorities have committed under the PRGF program to create a unit within the Ministry of Finance (MHCP). In addition, since Law No. 466 gives several responsibilities to the Budget Directorate of the MHCP, we recommend that this Directorate be staffed with enough personnel to fulfill the mandate imposed by the law and that the Fiscal Affairs Office (OAFE) of the Ministry of Finance is assigned a few specialists to monitor fiscal and debt sustainability of municipalities. 40. New institutional rearrangements must follow a clear decentralization policy, including a system of performance monitoring and evaluation for the whole decentralization arrangement. Such policy needs recognized leadership and acceptance of each institution's unique role. There is also need to clarify the legal language regarding "competencies," attributing unambiguous responsibilities for who has to do what and with what financial resource. A new intergovernmental system that scales up virtuous initiatives, redirects what is wrong and enhances the performance of each level of government through rewards and sanctions is presented in the next section. III.3 Reforming the Institutional Framework for Decentralization 41. There are many ways that the Nicaraguan State could organize the institutional framework for municipal decentralization, so any proposed organization must respond to some guiding criteria. Any proposal with regard a new institutional framework has to make explicit its rules of engagement to provide consistency and sustainability to the process. Further, the institutional framework for decentralization must respond to the need for fiscal neutrality discussed in Section II. In Section III.3.1 we make explicit what should be the rules of engagement, and in Section III.3.2 we propose a new financing mechanism to achieve fiscal neutrality while promoting economic efficiency and equity. The institutional framework proposed in Section III.3.3 follows from these two. - 23 - III.3.1 Guiding Principles for Institutional Reform for Fiscal Decentralization 42. Table 6 presents a minimal set of principles that should be present in any institutional framework for fiscal decentralization, along with a set of proposed strategic guidelines corresponding to each rule. Table 6 Rules of Engagement and Strategic Guidelines for a Institutional Framework for Decentralization Rules of Engagement Strategic Guidelines 1. Integration between national and local Establish mechanisms through which national and priorities: The Nicaraguan State must achieve a local governments can be motivated to act reasonable degree of reconciliation between cooperatively in their collective actions. Activities the needs and preferences of each local requiring inter-territorial scope to achieve national community and the general interests of the goals can be prioritized following a strategy in nations. which joint investments are rewarded. 2. Interaction between territorial and sector Work in two complementary directions: First, dimensions: Reconciliation of local and sector sector ministries incorporate an explicit territorial goals. dimension in their programs and services (health, education, urban development, etc.). Second, provide municipal governments with incentives compatible with national sector policies. 3. Fiscal efficiency and discipline at all levels of Establish a stable budget ceiling to the national government: In a context of scarce fiscal resources allocated to municipal governments. The resources, its efficient deployment is just an aim is that this annual allocation be fiscally imperative. consistent with the macroeconomic balance strategy, but at the same time these resources are predictable enough to enhance the programming of local investments. This would discourage to some extent that local authorities focalize their investments only in the short run, given their political pressures. 4. Separation of functions within the public Actors and organizations in charge of establishing sector: specialization, and institutional checks norms and public policies should not delivery and balances. services. According to this rule, the public sector must separate as much as possible the service delivery functions from the strategic, financing and monitoring functions. 5. Accountability and citizen participation: First , opening regular channels to citizen Guaranteeing that: (i) weak actors are participation. Second, create accountability protected from abuse of a dominant position in instances from local governments to their the intergovernmental arrangement, (ii) all communities. Third, strengthen accountability public servants are responsible for the use of channels with regard to the performance of services tax-payers money and (iii) citizens have the providers. means to express their needs and preferences, and develop social monitoring. - 24 - III.3.2 Proposed Mechanism for Financing, Mobilizing and Allocating Resources for Municipal Development 43. The implementation of the above principles in practice, as well as the demands of fiscal neutrality, requires new mechanisms to allocate resources to municipalities. Efficiency and transparency of fiscal decentralization arises from being able to identify where and how fiscal resources for decentralization are allocated. Good decentralization implies some balance between autonomy and coordination in the public expenditure management process. Municipal governments can provide not only their informative advantage from being close to the people, but also mobilize their own resources in a coordinated manner with the national government to maximize the social welfare of collective actions. Matching-grants or similar co-financing mechanisms can help very much in such a cooperative interaction between both government tiers. In this section we sketch the main features of such a proposed financing mechanism. 44. The first aspect in the mechanism for financing municipal-level investments is the "sector prioritization component." This element would reward with different percentages of national co-financing those sectors and sub-sectors (specific programs and services within a sector) responding to national sector strategies or showing the greatest externalities. This component should express the recognition by national authorities that there are sector interventions that have a higher national priority. Within this same first component, in the exercise of the strategic and policy-guidance role of the national organizations in charge of each sector, their authorities could incorporate in this mechanism differing priorities to specific programs and services within a sector (i.e. preventive programs over curative interventions in the health system, primary over secondary programs in the education sector). 45. The second component of the co-financing mechanism to take into account is the "local prioritization component". This element allows pooling resources to finance projects, with different percentages of financial contributions by the interested actors. This means signaling with different co-financing weights the priorities established to different kinds of investments, according to some transparent and explicit criteria used to allocate resources. This component would express the recognition by the leaders of decentralization and the national sector authorities that there are jurisdictions with different levels of human development and unsatisfied basic needs in Nicaragua. 46. Two other characteristics of the co-financing mechanism should be, on the one hand, financial incentives oriented to give priority to some municipalities with higher social vulnerability, rewarding with a higher national co-financing share the municipal effort in the investment projects solicited by these communities and, on the other hand, financial incentives motivating investments undertaken by more than one local government in projects with inter-jurisdictional scope (i.e. a hospital or an educational center attending populations from several municipalities). 47. One way to achieve the above objectives would be to propose to municipal governments a matching grants or co-financing scheme where they compete in an open and transparent manner so as to reveal the inter-sector and intra-sector - 25 - preferences of each community (see table 7). This scheme would introduce new incentives to the public administration, whose actors would now be guided by the demand of resources for municipal investments. This would result in higher allocative efficiency and search for innovations to attract the attention of municipal authorities and their communities. In fact, these municipal demands for investment projects should express to a large extent the needs and preferences of the potential beneficiary communities for public services and programs. In this sense, the promotion of efforts in elaborating participatory municipal investment plans is essential. Table 7 Sample Co-financing Matrix for Municipal Investment Projects Sector 1 Sector 2 Sector ... Sector N Municipality 1 C$ 5 for C$ 3 for each C$ 1 for each C$ C$ 0 for each C$ C$ each C$ Municipality 2 C$ 4 for C$ 3 for each C$ x for each C$ C$ 0 for each C$ C$ each C$ Municipality ... ... ... ... .... Municipality N Weight N1 Weight Weight N... Weight NN N2 48. The above table means that the national government would offer, for instance, five córdobas to municipality 1 for each córdoba that said municipality contributes to an investment in sector 1. Likewise, the national government would offer 3 córdobas to any municipality for each córdoba that the municipality wants to assign to investments in sector 2, and so on. By changing the weights for different sectors, different investments within the sector and different municipalities, the government would provide incentives for allocation of municipal resources that are consistent with national, regional and sector priorities. These different weights would be established by the relevant sector Ministries in charge of policy-making in their respective sectors. For instance, the Health Ministry may decide that the national government should put three córdobas for each córdoba contributed by municipalities when it is a project for preventive health, but only one córdoba of additional national resources should be offered for a project of curative interventions, given the lower social cost-benefit ratios usually attributed to the last kind of project. 49. Furthermore, the national government could use the above co-financing matrix to promote the association of municipalities for some projects showing externalities or economies of scales beyond the borders of each municipal territory. This could be done by increasing the weights for each sector, sub-sector and municipality by a given percentage, when an association of two or more municipalities undertakes the project. 50. Finally, the national government could complement the above financing mechanism by giving municipalities vouchers as part of the transfers mandated by Law No. 466. These vouchers would be used by municipalities to put their share of the co-financing for investments as per the sample co-financing matrix. The national - 26 - financing and technical assistance agencies (FISE, IDR and INIFOM) would then cash the vouchers either directly with the international financing institutions that currently provide 99% of their investment budgets or through the National Treasury. This would decrease the direct fiscal cost of the Fiscal Transfers Law by whatever share of the transfer is decided to be given in the form of vouchers. We recommend that the Government seek a legal opinion to determine the feasibility of this scheme without modifying Law No. 466. III.3.3 Proposed Institutional Framework for Decentralization in Nicaragua 51. The following proposals for restructuring the institutional framework for decentralization and monitoring of the process are based on the proposed rules of engagement and strategic guidelines, as well as the implementation of the financing scheme proposed above. We propose some arrangements concerning three main functions: (i) governance and strategic guidance, (ii) institutions financing municipal development and (iii) monitoring and evaluation. (i) Governance and strategic guidance 52. According to the above rules and guidelines, it seems reasonable separating the spheres of stewardship, technical support and consultation within this function. a) Leadership and Strategic Guidance 53. The Latin American experience on this subject is very varied (Annex 6). There are no best practices and each country must find the solution best suited to its existing environment and political situation. For the case of Nicaragua, we suggest that the leadership of decentralization be entrusted to a High Level Intergovernmental Commission (HLIC), which would absorb the functions currently assigned to both the National Commission for Decentralization and the Transfers Commission, among others. It should have a few members in order to help the celerity of decision-making, but should include the Secretary of SECEP, the Minister of Finance and probably the Presidents of FISE, INIFOM and IDR. This governance and policy-making instance would focus its efforts on strategic activities, so it would be expected to meet a couple of times per year. One of the key responsibilities of the HLIC would be to approve the weights of the co- financing matrix proposed in the previous section. The Chairperson of the Commission should be appointed by the President and should be the main interlocutor for all the actors involved in the decentralization process including different ministries, municipalities, legislators and international agencies that finance decentralization in Nicaragua. b) Technical Secretariat 54. This instance should be mostly in charge of supporting the decision-making by the HLIC. To this end, it would be responsible to systematically monitor and evaluate the territorial decentralization and development process. Its actions should include four additional tasks: (i) provide technical assistance to municipal governments, (ii) deliver training courses for municipal government authorities and employees, (iii) prepare - 27 - proposals for HLIC approval of regulations for financial and administrative management of municipalities, in coordination with the relevant national agencies, and (iv) disseminate information to improve communication between the actors involved in the process, organizing the discussion among actors regard relevant issues for municipal development and decentralization, in particular local communities. 55. Following current Nicaraguan legislation, INIFOM should be in charge of these tasks. A group of senior level officials should be supporting the Technical Secretariat, including the Director of the National System of Public Investment and the Director of the National Budget Office. c) Consultation Committee 56. This instance should be integrated by an ample number of relevant institutions and actors related to municipal decentralization. These would include: the Municipal Decentralization Commission of the National Assembly, the municipalities associations, sector ministries, international financing institutions and any other actors involved in the decentralization process. Its purpose should be to serve as a forum for frequent exchange of information, coordination of activities and discussion of policy alternatives. Its deliberations would provide advice to the other instances in charge of decentralization. We recommend that this Consultation Committee be one and the same as the Decentralization Board ("Mesa de Coordinación-Descentralización"). (ii) New Role for Institutions Specialized in Financing Municipal Development 57. This section sketches the main institutional changes required by the institutions currently financing investments for social and productive purposes at the municipal level -INIFOM, FISE and IDR-, in the context of the implementation of the above principles and financing mechanism. a) Nicaraguan Institute for Municipal Promotion (INIFOM) 58. In the suggested new institutional framework, INIFOM would move towards a more technical role in the governance function related to decentralizations and local development. In fact, in its role as Technical Secretariat supporting the HLIC, it would require a large amount of efforts and resources because that Commission would discharge its decision-making responsibilities largely on the basis of diagnostics, recommendations and proposals formulated by the Technical Secretariat and interactions with the Consultation Committee (Mesa de Coordinación). Additionally, responsibilities regarding monitoring and evaluation should be attributed to INIFOM, as indicated in paragraph 61. 59. Complementing these responsibilities, INIFOM must continue to take advantage of its strengths and experience in institutional development of local governments. This is perfectly compatible with its new strategic role in the decentralization and local development process, to the extent that it contributes to developing capacities required to improve municipal management in Nicaragua. Nevertheless, it should be clear that this does not necessarily mean that INIFOM should be a direct provider of training and - 28 - technical assistance (budgeting and financial systems, management information systems, local tax administration, organizational development, among others) to municipalities. INIFOM's mission on this regard should focus on the identification of local institutional strengthening needs, the definition of programmatic contents for training and technical assistance in coordination with line ministries and other national agencies, setting standards for institutional development activities, management of databases on service providers, consultants, NGO's and their outputs, and technical and financial support for municipalities contracting those services, among others. 60. At the same time, INIFOM should be the sole provider of technical assistance, training and institutional development activities, absorbing those tasks presently realized by organizations such as FISE and others. Of course, it would be understandable if these organizations resist this, but considering that these are activities mostly contracted by these institutions in the market should make it easier to unify the contracting of these services under a single agency. Increased responsiveness and accountability in the provision of these services is more likely under a set of pro-efficiency incentives associated with competition among providers and strategic targeting following homogenous rules. 61. Perhaps an additional activity in which INIFOM could be involved in the promotion of projects and programs with inter-jurisdictional scope, and the associated facilitation of dialogues. b) Emergency Social Investment Fund 62. In this suggested institutional framework for decentralization and municipal development, FISE should focus its responsibilities in co-financing municipal projects of a social nature (health, education, social protection, social housing, rural water and sanitation and so on). This means that FISE would absorb the financing of municipal investments presently financed by institutions such as INIFOM and others. This rationale could even lead to attribute to FISE the management of the Supplementary Social Fund. 63. Line ministries in the social sectors would not relinquish their strategic, policy- making, impact evaluation and lobbying roles in their corresponding sectors. However, they could progressively delegate in a new "common pool of social resources" the task of financing projects at the local level, following the strategic guidelines and intervention priorities set by these ministries in their corresponding social sectors. 64. Under these new responsibilities, FISE would become a financial intermediary between national agencies in charge of social policies and municipal governments demanding resources for projects at the local level. It should be clear that, just like we argued above for the case of INIFOM, this does not imply direct execution by FISE. It would be just a provider of financial resources according to the terms of the co-financing matrix and following due evaluation of the social returns of projects. It is possible to expect that this would translate into higher consistency of the social investments by the Nicaraguan State, as well as costs savings, and a more effective effort to channel - 29 - resources towards populations with human development deficit. This could also provide a reasonable mechanism for sequencing of the decentralization process. c) Institute for Rural Development (IDR): Transformation into a Productive Development Fund 65. While the inclusion of IDR in a strategic analysis of decentralization in Nicaragua may seem out of place to some observers, to do so seems necessary in order to provide more integrality to strategic actions in developing productive investments at the local level. For this purpose, we posit that IDR could follow a role in the productive sectors similar to that described for FISE in the social sectors. That is, IDR could become a sort of Productive Development Fund, in charge of financing projects to enhance municipal competitiveness in sectors with local comparative advantage in Nicaragua. This option would be entire consistent with the goals of the National Development Plan. 66. In a first stage, IDR could be merged with other national agency programs financing agricultural and agro-industrial development, and environmental protection. Like in the proposal for FISE, line ministries and other national entities would retain their strategic, policy-making, impact evaluation and lobbying roles in their corresponding sectors. However, they could progressively delegate in a new "common pool of productive resources" the task of financing projects at the local level, following the strategic guidelines and intervention priorities set by line ministries in their corresponding productive sectors. 67. In a second stage, the authorities could evaluate the convenience of extending this model of financing to other economic sectors such as tourism, rural electricity and telecom, among others. Again, the aim is to provide higher consistency to sectoral strategies in Nicaragua, avoiding technical and managerial overlappings in the public sector interventions, and advancing toward the optimization in the deployment of scarce resources in the country. d) Further Considerations Regarding the Funds and Institutes 68. Whatever the scope of the decision to reform the institutional framework, the idea is to transform the organizations of the national government presently involved in different territories in Nicaragua into co-financing agencies for municipal projects. In practice this means a fundamental change in the mission and performance of these institutions, transforming them into financial intermediaries for municipal development projects. All these changes imply that they should make intensive use of figures such as contracting out investments and institutional development, while their roles as direct executors must be exclusively restricted to exceptional circumstances.18 69. Of course, by extension of the above arguments, in the medium and long terms, that could mean that FISE absorbs the investments at the municipal level of line 18For instance, in cases when public bindings show a lack of interest from potential suppliers. - 30 - ministries in the social sectors, while IDR would likewise absorb investments of line ministries and national organizations in the productive sectors. However, we do not recommend such a mayor change in a sudden way, as the quality and continuity of essential social services could be seriously jeopardized. Nevertheless, we recommend taking a steady, yet slow, course in the delegation of these financing obligations to the Funds. 70. Finally, we have examined the idea of merging the three institutions considered in this section and have concluded that this would not be advisable in the present political context of Nicaragua, given the extraordinary difficulties in merging institutions, even in the private sector where decisions are not so heavily regulated and monitored. Nevertheless, it would make perfect sense to merge the Boards of the three institutions. In fact, not doing so would defeat the basic arguments for the co-financing mechanism proposed above, because it would be extremely difficult to ensure the coherence of two co-financing matrices (one for the social sectors and one for the productive sectors) and would lead to the emergence of segmented priorities by municipalities. We therefore recommend that the authorities consider seriously the possibility of merging the Boards of the three institutions, giving the unified Board, among other, the task of deciding if and how much further to consolidate them. Furthermore, in order to avoid a multiplicity of collective bodies, it could be possible to have the High Level Commission and the merged Boards be one and the same. (iii) Monitoring and evaluation 71. Efficient decentralization requires a set of mechanisms and practices related to monitoring, control and performance evaluation of municipalities. The objectives would be to ensure the proper use of funds, to ensure adherence to rules and regulations, to obtain information on what is going well and what is not for discussions and decision- making purposes, to assess performance of local governments and to assess the impact of policies and practices on the intended beneficiaries. 72. To make this possible, it is necessary to reinforce Nicaragua's capacity at both levels of government, as well as national organizations supporting decentralization. In this sense, it is useful to distinguish four different but complementary areas to be reinforced as ex-post monitoring practices: a) Operational and financial monitoring in charge of the General Comptroller's Office (GCO). The possibility to create a specialized unit for municipal audits within the GCO should be explored, given the higher importance of this level of government in total public expenditures. b) Managerial controls would be under the responsibility of the specialized institutions in financing investments for municipal development (FISE and IDR). c) Impact evaluation in charge of sector ministries and the Technical Secretariat mentioned before (INIFOM). d) Social monitoring would be part of the co-shared responsibilities between governmental and non-governmental actors in society, which would exchange - 31 - information needed to evaluate performance, assess impact, ensure transparency and establish priority expenditure needs. 73. All this should be accompanied by sustainable efforts to introduce financial and managerial transparency in overall public administration, including the implementation of integral systems of financial and human resources management and the adoption of crystal-clear budgeting systems (minimizing both extra-budgetary accounts and practices of earmarking). IV. Issues for Further Analysis IV.1 Borrowing Capacity 74. The Municipal Budgets Law (Law number 376, article 20) and the Public Debt Law (Law No. 477) determine the borrowing capacity of municipalities. The Municipal Budgets Law states that municipalities can borrow only to the extent that debt service is not more than 20 percent of the current annual municipal budget. In addition, municipal governments cannot leave debt to their successors unless there is a popular consultation. Finally, the Public Debt Law requires that municipalities must obtain no-objection from the Ministry of Finance before borrowing. 75. Before 2004, local governments had limited capacity to borrow given their limited amount of resources. The only notable exception was the Municipality of Managua, which had access to the local financial market. The most common form of financing for other municipalities was accumulating large arrears with providers of goods and services. In addition, municipalities have accumulated large debts against the social security system by not paying social security contributions for their employees. 76. The legislation does not give enough guarantees against excessive local borrowing. Experience in several Latin American countries has shown that the weak regulation of local borrowing can easily undermine macro stability of the financial system. The problem is not immediate, as municipalities are receiving a limited amount of resources in the first years, but will become dramatically relevant when the transfers are at the full 10 percent of tax revenues. International experience (for instance in Brazil) has also shown that the financial problems deriving from weak local regulation can take many years to be fully evident. Several measures must be taken now to avoid future problems. Some of these are: a) Financial regulation should impose a high coefficient of risks for loans to municipalities. b) The Public Debt Law should be modified so that the Ministry of Finance must give an explicit authorization for municipalities to borrow. c) Arrears towards providers of goods and services should be explicitly considered a form of borrowing, subject to the same restrictions and regulations as other forms of debt. - 32 - IV.2 Auditing 77. The General Comptroller's Office has a broad mandate to oversee general government operations, including municipalities. In practice, the scarce resources do not allow sufficient central auditing of municipalities by the GCO. Less than 10 out of 152 municipalities have their activities audited externally and this is mostly done at external donors' request. Foreign donors generally audit the projects that they finance, but there is no systematic consolidation of such audit findings. 78. The Municipal Transfer Law mandates that municipalities should transmit the current and previous year's budgets to INIFOM and the Ministry of Finance. However, neither of these institutions has the resources nor the clear mandate to audit them. As indicated in paragraph 72, it would be desirable to create a specialized unit for municipal audits within the GCO. IV.3 Other 79. Several measures may improve the overall design of decentralization in Nicaragua, some of them depending on whether it is politically feasible to modify the laws governing decentralization: a) Given the particular situation in Nicaragua, where most municipalities are not sufficiently trained to assume new expenditure responsibilities in a sudden way, as well as the efficiency requirement to ensure that such transfers of responsibilities respond to a demand-driven process in which municipalities (both authorities and civic groups) are aware of the benefits and costs of gaining autonomy, it will be necessary to implement a flexible strategy by which Government identifies expenditures to be transferred and establishes conditions to be fulfilled by those municipalities interested in receiving both revenue and expenditure responsibilities. Local capacity-building processes should be linked to this strategy in a very flexible and demand-driven program, that would provide support to those municipalities interested in fulfilling conditions to become eligible for a fast-track decentralization scheme and that are prepared to assume both greater revenue and expenditure responsibilities from the central government. b) The present legal framework gives very little incentives to increase the efficiency of local taxation, in particular through the property tax. The low collection rate is due both to technical reasons and to scarce local political incentives to enforce the law. We recommend to increase substantially the incentives to collect local taxes. This could be done by increasing the weight of the "fiscal efficiency" criterion in the allocation formula for the transfers. Additionally, improvements in the cadastre systems are necessary to increase property tax collections. c) The legislation is silent on what happens if a municipality fails to deliver some basic services. We propose that the government explicitly sets minimum target - 33 - of efficiency that municipalities must accomplish. If the targets are not reached the central government should be given legal instrument to intervene to ensure that the population does not suffer further. d) We recommend simplifying the transfers formula in order to increase transparency. While the criteria are not being questioned, the reliance on highly subjective parameters like "potential" revenue collection is. This list of recommendations is meant to address long-term problems and require the reconsiderations and re-drafting of legislation concerning municipalities. 80. There are other issues that require close attention from authorities as they embark on their decentralization program, but that will not be given attention here: the need for training and technical assistance, tools for public sector management, including SIGFA, procurement, human resource administration and planning. In the definition of training requirements in each sector and for each transferred expenditure responsibility, emphasis should be given to municipalities being in the position to select and define their own training needs. It should not be a top-down exercise, but put municipalities "in the driver's seat". V. Next Steps 81. Once the authorities have made the basic decisions on the responsibilities to be transferred to municipalities, they may want to consider the following next steps: · Refine the costing exercise and draft the Presidential Decree according to the implications of this analysis, clarifying as precisely as possible the spending responsibilities and revenues of each level of government. · Define entry points and sequence of decentralization. This sequence should have a clear sector logic, that is, strengthen overall sector strategies. o Decentralization of secondary road maintenance programs should be given priority and should be the starting point for the process. This service can be managed with relatively low institutional capacity and is of genuine interest to municipalities. o Decentralization of social assistance could also be part of a first phase of decentralization. o All other sectors would need more detailed costing to define the technical and financial feasibility of decentralization. · Identify debt (suppliers debt, other) currently held by central government agencies in the current service delivery model. Define magnitude of debt and financing options in a decentralized setting. · Conduct a legal study on severance payments for civil servants in each sector, in particular health and education. This would provide government with a better idea - 34 - on the magnitude of the fiscal problem that underlies decentralization of current expenditure in those sectors. VI. Conclusions and Recommendations 82. Nicaragua is pursuing decentralization in an aggressive manner and could repeat the same mistakes of other Latin American countries without prompt corrective actions. Unfortunately, the current legal and institutional framework for decentralization does not bode well for the future. This paper has outlined the present situation, the present and foreseeable future problems and the possible corrective actions. The best option would be to modify the Municipal Transfers Law. Short of this possibility, several measures can be taken to ensure efficiency of local public expenditure, accountability and fiscal responsibility. 83. A gradual approach to devolve expenditure responsibilities is neither economically nor politically advisable. We argue against the policy option of start giving resources immediately to municipalities while postponing the devolution of expenditure responsibility to a later stage. First, economically it is not convenient to give additional resources to municipalities just to take them away after few years, because projects could be started to be left unfinished when the resources are used to comply with new functions. Second, if the concern is the limited capabilities of municipalities, it is not clear why the central government should send a blank check now. If anything, the opposite would make more sense, i.e. transferring additional resources only after municipalities have a proven record of efficiency. Third, well-established international evidence has shown that it is very difficult to cut public expenditure once interest groups are formed, so a `temporary' increase of public expenditure will probably become a permanent allocation of resources that the country cannot afford to waste. 84. The Nicaraguan Government should pursue fiscal neutrality while minimizing the sacrifice in efficiency of service delivery. Expenditures should be transferred from the central level to the municipal level, rather than simply cut. The decision to transfer revenue up front without expenditure responsibilities constrains the choices government has for the implementation of this recommendation. There are areas where municipal involvement cannot be recommended on efficiency and accountability grounds, but from the perspective of fiscal neutrality authorities have probably no other alternative than to shift expenditures to local levels. For instance, the payment of teacher salaries may not make sense in the school autonomy model that is being implemented, but the fiscal neutrality goal may require the involvement of municipalities in transfers to autonomous schools. 85. At the moment, nothing ensures that municipalities will spend the money on investment projects that are consistent with national priorities. This is unfortunate given that Nicaragua, by being part of the HIPC initiative, has taken the international commitment to allocate a significant share of public expenditures to pro-poor investments. In order to close the possible gap between municipal investment decisions and nationwide objectives, we propose that the central government implements a matching-grants program that reflects national priorities, as explained in section - 35 - III.3.2. This program would be financed with the resources currently allocated to municipalities through INIFOM, FISE and IDR. 86. This Policy Note has presented some options for transferring expenditure responsibilities to municipalities, along with their fiscal impact, and for revising the institutional framework in a way that accompanies these new responsibilities. They are strictly based on the assumption that no changes to the laws can be introduced in the short term. The principal recommendations can be summarized as follows: · The central government should immediately suspend expenditures in areas that are of municipal competence according the present legislation. A regulatory decree must unambiguously define the expenditure responsibilities that are assigned exclusively to municipalities, avoiding duplications with same expenditure responsibilities assigned elsewhere to central government agencies. · In the short and medium run, a combination of vouchers and a co-financing scheme for investment expenditures between the central and the municipal level seems best suited to ensure maximum benefits from the transfers in a way that promotes efficiency, given the characteristics of the population and the political system in Nicaragua. The authorities should seek a legal opinion regarding the feasibility of issuing vouchers that can be used to pay for services or projects to be delivered by national agencies, specially INIFOM, FISE and IDR, as part of the mandated transfer flows. · Areas best suited would be rural roads, expansion of water and sanitation services, environmental protection and social services, as these are already under the expenditure responsibilities assigned to municipalities by the Municipal Law. · The deduction of regular contributions and debt service from municipalities to the social security system from the transfer amount should accompany any expenditure-transfer option. · Law No. 466 should be interpreted strictly with regard to the documentation required to effect the transfers. · The minimum yearly increase foreseen in Law No. 466 should be applied. 87. This Policy Note has also presented some recommendations for reforming the institutional framework for decentralization, along with their rationale. These include: · The Government needs to create or define a unique policy-making body for the decentralization process. We recommend that this be a High Level Intergovernmental Commission (HLIC). This single policy-making body should define national decentralization policy as well as reconcile sector decentralization initiatives initiated by individual ministries. - 36 - · The Budget Directorate of the Ministry of Finance should be staffed with enough personnel to fulfill the duties assigned by Law No. 466 to this Ministry and OAFE should regularly monitor the fiscal and debt sustainability of municipalities. · INIFOM should circumscribe its activities to acting as Technical Secretariat of the HLIC and the organization and financing of technical assistance, training and monitoring of municipalities, as well as avoiding financing projects at the municipal level. · FISE should progressively become the single co-financing institution for social investments at the municipal level. · IDR should progressively become the single co-financing institution for productive investments at the municipal level. · The authorities should consider the possibility of merging the Boards of FISE, INIFOM and IDR. In addition, this merged Board could be one and the same as the HLIC. 88. As soon as it becomes feasible to amend the laws governing decentralization in Nicaragua, the principal changes we recommend are: · Define first what are the public services which are more efficiently delivered at the municipal level, calculate their cost and only then decide the share of tax revenue that should be transferred to municipalities. · Simplify the formula for allocation of transfers among municipalities, in order to promote transparency. · Establish clear distributional goals and criteria that compensate for the lack of fiscal capacity vis-à-vis expenditure needs. · Remove the ex-ante conditions for transfers and move to a system of automatic, daily disbursements. · De-link the growth of transfers from GDP growth. - 37 - References Budinich, Ema B., 2004, Identificación de Alternativas de Descentralización de Gasto Público hacia las Municipalidades. Report prepared for the Nicaraguan Ministry of Finance. Managua. Eaton, K., 2003, "The Dilema of Decentralization in Latin America: Risks and Opportunities", Latin America's Foreign Economic Relations, Ana Margheritis, ed., University of Miami Press. Wiesner, E., 2003, Fiscal Federalism in Latin America, Inter-American Development Bank, Washington D.C. World Bank, World Development Report (2004): Making Services Work for the Poor, Washington D.C. Ter-Minassian, T., 1997, Fiscal Federalism in Theory and Practice, International Monetary Fund, Washington D.C. Spilimbergo, Antonio, "Fiscal Decentralization in Nicaragua," April 2004. Barrios, Armando, "Marco Institucional de la Descentralización en Nicaragua," June 2004. Frank, Jonas, "Decentralization in Nicaragua: The Challenge of Fiscal Neutrality," May 2004. Development Outreach, March 2004. Web Pages Descentralization Comisión (Comision Sectorial para la Descentralización): www.csd.gob.ni. National System of Public Investment (Sistema Nacional de Inversión Pública): www.snip.gob.ni Ministry of Finance and Public Credit (Ministerio de Hacienda y Crédito Público): www.hacienda.gob.ni Rural Development Institute (Instituto de Desarrollo Rural): www.idr.gob.ni Emergency Social Investment Fund (Fondo de Inversión Social de Emergencia): www.fise.gob.ni - 38 - ANNEXES Annex 1: Sources of Financing of Municipalities prior to 2004 Annex 2: Municipal Transfers Law Annex 3: Transfers by Municipality ­ Current and Capital breakdown Annex 4: Regular Contributions and Debt Service to INSS by Municipality Annex 5: Projects that can be Transferred to Municipalities Annex 6: Experience with Decentralization in Selected Latin American Countries. Annex 7: Summary Legal Framework for Decentralization Annex 8: Main Institutions Financing Decentralized Investment in Nicaragua - 39 -