82662 HONDURAS Public Expenditure Review Towards Restoring Fiscal Consolidation Report No: ACS6720 [82662-HN] Republic of Honduras Public Expenditures for Decentralized Governance in Honduras. Towards Restoring Fiscal Consolidation August 2013. LCSPS. LATIN AMERICA AND CARIBBEAN This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clear- ance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright. com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. HONDURAS Public Expenditure Review Towards Restoring Fiscal Consolidation August, 2013 Poverty Reduction and Economic Management Central America Department Latin America and the Caribbean Region Abbreviations and Acronyms ACCESO Programa de Acceso a Servicios de Salud con Equidad CONATEL Comisión Nacional de Telecomunicaciones y Administración Descentralizada (National Telecommunications Commission) (Program of Access to Health Services with Equity and DGVU Dirección General de Vivienda y Urbanismo Decentralized Management) (Housing and Urban Development Directorate) AIEPI Atención Integral de Enfermedades Prevalentes en la EDUCATODOS Programa de Educación Básica y Capacitación Técnica Infancia (Integrated Management of Childhood Illness) (Basic Education and Vocational Training Program) AIN-C Atención Integral a la Niñez en la Comunidad EFA Programa de Educación para Todos (Education for All) (Community-Based Integrated Child Care Program) EIB Educación Intercultural Bilingüe APREMAT Aprendamos Matemáticas (Let’s Learn Math) (Intercultural Bilingual Education) BANHPROVI Banco Hondureño para la Producción y la Vivienda ENCOVI Encuesta de Condiciones de Vida (Honduran Bank for Production and Housing) (Living Standards Measurement Survey) BCH Banco Central de Honduras ENEE Empresa Nacional de Energía Eléctrica (Central Bank of Honduras) (State-Owned Electric Power Company) BCIE Banco Centroamericano de Integración Económica ENESF Encuesta Nacional de Epidemiología y Salud Familiar (Central American Bank for Economic Integration) (National Epidemiology and Family Health Survey) CAINES Centros de Atención Integral a la Niñez EPHPM Encuesta Permanente de Hogares de Propósitos (Integrated Child Care Centers) Múltiples (Household Survey) CCEPREB Centro Comunitario de Educación Pre-Básica ERSAPS Ente Regulador de los Servicios de Agua (Community-Based Center for Pre-Primary Education) Potable y Saneamiento (Potable Water CCERP Consejo Consultivo de la Estrategia para la Reducción and Sanitation Regulatory Agency) de la Pobreza (Consultative Council for the Poverty FDI Foreign Direct Investment Reduction Strategy) FHIS Fondo Hondureño de Inversión Social CEB Centro de Educación Básica (Honduran Social Investment Fund) (Center for Basic Education) FIDAS Fondo Innovador para el Desarrollo y la CNPV Censo Nacional de Población y Vivienda Asistencia Social (Innovation Fund for (National Population and Housing Census) Development and Social Assistance) FONAPROVI Fondo Nacional para la Producción y la Vivienda PES Programa de Escuela Saludable (National Fund for Production and Housing) (Healthy School Program) HIPC Heavily Indebted Poor Countries Initiative PRAEMHO Programa de Alfabetización y Educación de Adultos HIV/AIDS Human Immunodeficiency Virus/Acquired Immune (Adult Literacy Program) Deficiency Syndrome PRAF Programa de Asignación Familiar HOGASA Hogares Gestores de Atención en Salud (Family Allowances Program) (Health Services Management Homes) PRIESS Programa de Reorganización Institutional y Extensión HONDUTEL Empresa Hondureña de Telecomunicaciones de los Servicios de Salud (HonduranTelecommunications Enterprise) (Institutional Reorganization and Extension of Health IAIP Instituto de Acceso a la Información Pública Services Program) (Institute for Access to Public Information) PRIMHUR Programa de Mejoramiento Habitacional Urbano IFPRI International Food Policy Research Institute (Urban Housing Improving Program) INE Instituto Nacional de Estadísticas PROEIMCA Programa de Educación Intercultural Multilingüe (National Statistics Institute) de Centro América (Central American Intercultural INFOP Instituto Nacional de Formación Multilingual Education Program) Profesional (National Vocational Training Institute) PROHECO Programa Hondureño de Educación Comunitaria INJUPEMP Instituto Nacional de Jubilaciones y Pensiones de (Honduras Community Education Program) Empleados y Funcionarios del Poder Ejecutivo PRS Poverty Reduction Strategy (Public Employees Pension Fund) PVMR Programa de Vivienda Minima Rural INPREMA Instituto Nacional de Previsión del Magisterio (Rural Minimum Housing Program) (Teacher Pension Fund) SANAA Servicio Nacional de Acueductos y Alcantarillados IPM Instituto de Prension Militar (Water and Sanitation Company) LPG Liquified Petroleum Gas SAP Suplidoras de Abastecimiento Popular MDG Millennium Development Goals (Popular Shops) MDRI Multilateral Debt Relief Initiative SIERP Sistema de Información para la Estrategia para la MOF Minstry of Finance Estrategia para la Reducción a la Pobreza MOH Ministry of Health (Povertyt Reduction Strategy Information System) NBI Necesidades Básicas Insatisfechas SOE State-owned Enterprises (Unsatisfied Basic Needs) SOPTRAVI Secretaría de Obras Públicas, Transporte y Vivienda PAPIN Programa de Apoyo a los Pueblos Indígenas y Negras (Ministry of Public Works and Housing) (Program to Support Indigenous and Black Peoples) SS Secretaría de Salud PEFA Public Expenditure and Financial Accountability UNAT Unidad de Apoyo Técnico de la Secretaría del PER Public Expenditure Review Despacho Presidencial (Technical Unit of the Ministry of the Presidency) Republic of Honduras Fiscal Year January 1 – December 31 Currency Equivalents (as of June 21, 2013) Currency Unit = Lempira (Lps) 1 US Dollar = 20.28 Lps 1 Lempira = US 0.049 Weights and Measurements Metric System Acknowledgments This report was prepared by a team led by Laura Zoratto (LCSPS) and Luc Razafimandimby (LCSPE) from February to September 2013, as co-task-managers, under the overall supervision and guidance of Oscar Calvo-González (Lead Economist and Sector Leader, LCSPR). The core team included, in alphabetical order, Calvin Zebaze Djiofack (LCSPE), Denis Medvedev (SASEP), Gustavo Ezequiel Miranda (LCCHN), Karina Ramírez (LCSPS), Natalyia Biletska (PRMPS), Nuria Tolsa Caballero (LCSPE), Ramón Arias (LCSPS), and Rong Qian (LCSPE). The main contributors of Chapter 2 are Lorena Viñuela (LCSPS) and David Carias (consultant). Authors of the background papers used as inputs for this report are as follows: for Water & Sanitation, David Mi- chaud (ESCUW), Antonio Rodríguez Serrano (TWILC), and Yoonhee Kim (LCSDU); for Health, Christine Lao Peña (LCSHH); and for Education, Juan Diego Alonso (LCSHE), Rocío Calidonio (LCSHE), and Martha Hernández (consul- tant). These sector-specific papers are published separately. The peer reviewers for this report are Jasmin Chakeri (OPSPQ), Enrique Blanco Armas (AFTP1), and Jonas Frank (PRMPS). This report was co-financed by The Swiss Agency for Development and Cooperation (COSUDE). Contents Executive Summary 11 I. Fiscal Context and Public Expenditure Trends 14 A. Composition of Expenditures 16 B. Managing the Deterioration of the Fiscal Stance 22 C. The Immediate Causes of Weak Public Financial Management 23 D. The Deeper Causes of Weak Public Financial Management 29 II. Fiscal Decentralization: Building Credible Reforms 32 A. Intergovernmental Framework and Local Conditions 33 i. Heterogeneity in Size and Capacity 34 ii. Municipal Responsibilities Focused on Local Services 35 iii. Features and Evolution of the Transfer System 36 iv. Recurrent Attempts at Deepening Decentralization 39 B. Municipal Finance in Numbers 39 i. Transfer Shortfalls: A Source of Allocative Distortions 40 ii. Municipal Revenue: Untapped Sources and Bright Spots 41 iii. Municipal Expenditure: Giving Precedence to Recurrent Expenditure 43 iv. Municipal Borrowing: Covering Transfer Delays 44 v. Local Public Investment: Increasing but Falling Short of Targets 46 C. Impact in Service Delivery: The Case of Water and Sanitation 49 D. Policy Recommendations 53 E. Appendix 54 References 58 Annex 59 Tables, Boxes, Figures, and Maps Table I 1: Operations of the Budgetary Central Government (% of GDP) 15 Table I 2: Debt Composition for Honduras (% of GDP) 15 Table I 3: Budgetary Central Government Expenditure Structure 20 Table I 4: Upward Adjustment of Aggregate Budget Ceilings 25 Table I 5: Current and Capital Expenditures—Approved and Modified Budget 26 Table I 6: Additional and Modified Budget by Functional Classification (% of Total) 26 Table I 7: Budget Execution by Functional Classification (%) 28 Table I 8: SIAFI Coverage 28 Table II 1: Average Population and Human Development Index per Category of Municipality, 2010 34 Table II 2: Functional and Expenditure Assignment across Levels of Government in Honduras 35 Table II 3: Expenditure Assignment across Levels of Government in Central America 36 Table II 4: Reforms to Allocation Formula (Law of Municipalities’ Article 91) 37 Table II 5: Tax Revenue and Transfer Shares of Municipal Revenue by Categories of Municipalities (%), 2002–2010 44 Table II 6: Borrowing per Capita and Debt Service Share of Municipal Expenditure, 2002–2010 47 Table II 7: Service Providers—Types and Population Served 50 Table II 8: Selected Characteristics of Distinct Service Providers 51 Table II 9: Comparison of Decentralized and SS Facilities based on Facility Characteristics and User Feedback 53 Table II 10: Correlation Coefficients 54 Table II 11: Multivariate Regression Analysis of Vertical Transfer Received (Fixed Effects) 55 Table II 12: Multivariate Regression Analysis of Municipal Expenditure (Fixed Effects) 55 Box 1: The National Security Tax—Implementation Challenges and Its Effects on Municipalities 40 Box 2: Deconcentrated Service Delivery Models and Community-driven Development Interventions 52 Figure I 1: Fiscal Deficits in Selected Central American Countries (% of GDP), 2012 14 Figure I 2: Central Government Deficit (% of GDP), 1990–2010 16 Figure I 3: Central Government Revenue and Expenditures (% of GDP), 2002–2012 16 Figure I 4: Budgetary Central Government Tax Revenue in Selected Central American Countries (% of GDP, 2008–2012 Average) 17 Figure I 5: Central Government Tax Revenue (% of GDP), 2002–2012 17 Figure I 6: Composition of Consolidated Central Government Expenditures (% of GDP) 18 Figure I 7: Composition of Consolidated Central Government Expenditures (% of Total Expenditures) 18 Figure I 8: Consolidated Central Government Expenditures in Honduras and Countries with High GDP per Capita Growth (% of GDP and % of Total Expenditures) 19 Figure I 9: Composition of Consolidated Central Government Expenditures on Infrastructure and Public Order and Safety (% of GDP) 19 Figure I 10: Gross Public Fixed Capital Formation (% of GDP), 2000–2011 21 Figure I 11: Composition of Budgetary Central Government Capital Transfers (% of GDP), 2009–2012 21 Figure I 12: Central Government Capital Expenditures (% of GDP) 21 Figure I 13: Central Government Capital Expenditures (% of Total) 21 Figure I 14: Gross Fixed Capital Formation (% of GDP) 21 Figure I 15: Gross Public Fixed Capital Formation (% of GDP) 21 Figure I 16: Budgetary Central Government Salary Expenditure by Sector 22 Figure I 17: Total Central Government Payment Arrears 22 Figure I 18: Composition of Central Government Payment Arrears, 2009 and 2011 23 Figure I 19: Aggregate and Composition Expenditure Outturn 24 Figure I 20: Total Budget at Different Stages of the Budget Cycle (Millions of Lempiras) 25 Figure I 21: Total Budget Execution (%) 35 Figure I 22: Budget Execution by Economic Classification (%) 27 Figure I 23: Arrears and Actual Payments (% of Modified Budget) 29 Figure I 24: Results of the Citizens’ Perception Survey 30 Figure II 1: Aliquot per Municipality per Capita, 2009. 38 Figure II 2: Total Transfers per Capita by Population Range, 2009 38 Figure II 3: Transfers per Capita and Population, 2010 39 Figure II 4: Transfers and Human Development Index, 2010 39 Figure II 5: Planned Transfer as Share of Central Revenue (%), 2002–2038 41 Figure II 6: Actual Transfer as Share of Central Revenue (%), 2002–2011 41 Figure II 7: Budgeted Transfers versus Actuals, 2009–2012 42 Figure II 8: Transfer Outturn (%), 2010 42 Figure II 9: Municipal Revenue as Share of Total Government Revenue and GDP (%), 2000–2010 42 Figure II 10: Local Revenue as Share of Total Government Revenue (%), 2003–2010 42 Figure II 11: Municipal Revenue Total and Detailed Composition, 2002–2010 42 Figure II 12: Municipal Revenue Total and General Composition, 2003–2009 42 Figure II 13: Tax Revenue per Capita and Population, 2010 43 Figure II 14: Tax Revenue per Capita and Transfer Share of Municipal Revenue, 2010 43 Figure II 15: Municipal Expenditure as a Share of Total Government Expenditure and GDP (%), 2002–2011 45 Figure II 16: Local Expenditure Total and as a Share of Total Government Expenditure (%), 2002–2010 45 Figure II 17: Municipal Expenditure per Capita and Population, 2010 45 Figure II 18: Municipal Expenditure per Capita and Transfer Amount per Capita, 2010 45 Figure II 19: Municipal Expenditure Total and Composition, 2003–2010 45 Figure II 20: Municipal Expenditure Total and Composition, 2003–2010 45 Figure II 21: Recurrent Expenditure as a Share of Expenditure and Transfers as a Share of Municipal Revenue, 2010 46 Figure II 22: Recurrent Expenditure per Capita and Transfer Amount per Capita, 2010 46 Figure II 23: Borrowing as Share of Municipal Current Revenue (%), 2010 47 Figure II 24: Debt Service as Share of Municipal Expenditure (%) , 2010 47 Figure II 25: Access to Improved Water and Sanitation Facilities (% of Total Population), 1990–2010 48 Figure II 26: Public Gross Fixed Capital Formation as a Share of GDP (%), 2000–2010 48 Figure II 27: Municipal Net Public Fixed Capital Formation and Transfer Assets, 2002–2010 48 Figure II 28: Municipal Net Public Fixed Capital Formation as a Share of Total (%), 2002–2010 48 Figure II 29: Expenditure in Capital Assets as a Share of Municipal Expenditure and Transfer as a Share of Municipal Revenue, 2010 49 Figure II 30: Expenditure in Capital Assets as a Share of Municipal Expenditure and Population, 2010 49 Figure II 31: Expenditure in Capital Assets per Capita and Infrastructure Gap, 2010 49 Figure II 32: Expenditure in Capital Assets per Capita and Population Density, 2010 49 Figure II 33: Employees per 1,000 Connections Based on Type of Service Provider 51 Figure II 34: Service Provider Performance by Management Model 51 Figure II 35: Number of Municipalities by Population Range, 2010 54 Figure II 36: Subnational Expenditure Share of Total Public Expenditure (%) in Developing Countries, 2010 56 Figure II 37: Subnational Revenue Share of Total Public Revenue (%) in Developing Countries, 2010 56 Map 1: Transfer Amount per Capita (Lempiras), 2010 57 Map 2: Recurrent Expenditure per Capita (Lempiras), 2010 57 Honduras Public Exphenditure Review 2008-2012 Executive Summary F iscal consolidation remains the central chal- (1.8 percent of GDP), pending payments for goods and lenge facing Honduras, mainly due to increas- services (1.7 percent of GDP), and transfers to munici- ing current expenditures. The fiscal deficit of palities (0.8 percent of GDP). the central government has been worsening since 2008, increasing from 2.4 percent of GDP The wage bill remains one of the highest in Latin in 2008 to 6 percent in 2012. The 2012 figure marked the America at 11.9 percent of GDP. The most significant worst performance in the region that year. The widening impact on the wage bill stems from growing salary expen- fiscal deficit has been driven by a significant increase in ditures in the education and health sectors. On average, current expenditures. For instance, in 2012, interest pay- these two sectors accounted for around 70 percent of ments increased by 37.5 percent, current transfers in- the central government wage bill during the 2008–2012 creased by 17.5 percent, and wages and salaries increased period, with education accounting for the majority (52 by 8.3 percent. In 2013, despite an announced govern- percent). In the education sector, roughly 93 percent of ment effort aimed at reducing the fiscal deficit to 4.5 per- the expenditures are allocated to the sector’s wage bill, cent of GDP, a mid-year assessment released in July 2013 including teaching and non-teaching staff. In contrast, points to a deficit of 6 percent once again. central government investment declined to a decade-low of 1.4 percent of GDP. Increases in interest payments are driven by new domestic debt. Between 2002 and 2008, interest pay- These increased current expenditures, which add rigid- ments declined from 1.6 to 0.6 percent of GDP, with ity to the budget, have occurred at the expense of in- interest payments on external debt falling from 1.4 to vestments. Current expenditures from the central govern- 0.3 percent of GDP. This reduction was the direct result ment increased from 17.5 percent of GDP in 2008 to 18.1 of debt alleviation through the Heavily Indebted Poor percent in 2012 due to the increase in interest payments, Countries and Multilateral Debt Relief Initiatives. By transfers, and expenditures in goods and services. But a 2012, however, interest payments rose sharply to 1.9 decrease in capital spending by the budgetary central gov- percent of GDP. The jump was driven by a drastic in- ernment was accompanied by a particularly sharp reduc- crease in interest payments for domestic debt, which tion in capital investment, from 2.8 percent of GDP in 2009 represented 1.4 percent of GDP in 2012 (or 78 percent to 1.4 percent in 2012 in order to provide for larger capital of total interest payments). The domestic debt of the transfers to other levels of government. The capital trans- central government grew from 2.8 percent of GDP in fers resulting from decreased capital investment were not 2007 to an estimated 15.2 percent in 2012. The new always used for capital formation, but rather to finance cur- debt is expensive due both to its short-term nature and rent expenditures. Compared with the second half of the the lack of depth in the domestic bond market, with its 1990s and the beginning of the 2000s, capital expenditures interest rates averaging 9 percent and at times exceed- of the central government have decreased from an average ing 15 percent. A substantial portion of the debt ac- of 30 percent of the total expenditures to just 20 percent in cumulation resulted from public sector liabilities added 2012. Moreover, gross capital formation of the public sec- in the 2006–2009 period (equivalent to 4.7 percent of tor declined to a decade low of 3.3–4.0 percent of GDP in GDP), which included the state’s pension contributions 2010–2011, compared to 5–6 percent of GDP in 2003–2004. The efficiency of spending remains a key constraint. As a result of budget modifications, cash rationing has be- Growth in general public services (executive and legis- come a constant and the level and composition of actual lative bodies, financial and fiscal affairs, external affairs, resource allocations often significantly deviate from the and other general public services) was the main determi- approved budget. nant of rising central government expenditures. They in- Honduras Public Exphenditure Review 2008-2012 creased from 8.0 percent of GDP to 12.6 percent between In order to set credible expenditure ceilings and 2009 and 2012. Spending on social sectors declined from maintain commitment control, the Ministry of Fi- 12.0 percent of GDP to 10.7 percent in the same period. nance requires a favorable authorizing environment Education still comprised the main share of total expen- underpinned by a strong political will. Although the ditures, although the total funds allocated to this sector adoption of laws and the implementation of informa- fell from 8.5 percent of GDP in 2009 to about 6.8 percent tion systems are all welcome developments, they alone in 2011. The reduction in education expenditures has not cannot address the central weaknesses, as they can eas- translated into allocative efficiency gains, as most of the ily be circumvented and adapted to traditional informal funds have been absorbed by increased general public practices. The Integrated Financial Management System services spending. Overall, limited improvements, espe- (SIAFI), for example, is bypassed at the commitment con- cially in health and education outcomes despite high allo- trol stage and used only at the payment stage. Until re- cations to these sectors, suggest the need to improve the cently, the system of programing and allocation of com- efficiency and quality of expenditures. Increased public mitment quotas established in the Budget Law in practice expenditures have not delivered in terms of growth and consisted simply of quarterly planning and allocation of improved public services, and whether the country is cash for payments broken down by month. To date, the ready for decentralization remains an open question. Ministry of Finance has been unable to exercise control over commitments, and unrecorded commitments still The lack of efficiency in service delivery is most nota- exist; as a result, in-year budget modifications across sec- ble in education. Central government spending on edu- tors and at the aggregate level are significant and budget cation, health, and infrastructure is consistent with the ceilings are not enforced. levels achieved in countries that experienced sustained periods of high GDP per capita growth, but the quality Growing fiscal deficits and weak public financial and efficiency of service delivery continues to lag behind. management practices have constrained the ability Although Honduras’ performance indicators in health and of the central government to implement and finance water and sanitation are comparable to similar Central the decentralization process. In 2010, the government American countries, performance in other infrastructure pledged to increase the share of central revenue trans- sub-sectors is variable and significantly behind its peers ferred to local governments from 5 to 20 percent by 2017 in education. Despite recent decreases, Honduras’ pub- and to 40 percent by 2038. However, the central govern- lic spending on education as a share of GDP remains the ment has complied with the mandated revenue transfer highest among Central American countries, and it is one percentage only three times in more than two decades of of the highest in the world. Analysis undertaken for the decentralization. education sector reveals the significance of the efficiency losses due to teacher absenteeism, “ghost teaching”, and Current decentralization targets would pose signifi- lack of instruction delivery within the allotted teaching cant fiscal challenges to the central government. The time. On average, correct answers in math do not surpass plans for deepening fiscal decentralization are not in- 40 percent, and improvement over the last thirteen years formed by an evaluation of the comparative advantage was virtually nonexistent. Spanish test scores were slight- of municipalities in delivering social and infrastructure ly higher, but still around the 40 percent mark. In light of services. For sectors in which there are clear scale econo- the new mandate for 12 years of compulsory education, mies and spillovers, devolution to the municipalities is spending in this sector is likely to increase even further. not advisable. Moreover, the gains expected from de- centralization in terms of greater accountability as well Across all sectors, public financial management prac- as increased production and allocative efficiency are not tices undermine efficient allocation of resources and likely to be realized in a context of weak institutional ca- aggregate fiscal discipline. The primary constraint pacity. The assignment of resources has to be consistent comes from nonbinding expenditure ceilings set by the with the distribution of responsibilities (“funding follows Ministry of Finance which are routinely breached by sec- function”). The current policy discussion has focused tor spending agencies. In addition, the absence of ex-an- solely on the distribution of financial resources without te commitment controls leads to a generation of expendi- attention to what functions could be realistically carried ture commitments without allocated budgetary funding. out by the municipalities. Improving transfer predictability is an important Municipalities are responsible for a sizable part of prerequisite for promoting effective local manage- public investment and additions to public fixed capi- ment of resources. Improving the predictability of tal, but many constraints for effective capital spend- transfers is more important than increasing their vol- ing remain. Public investment at the subnational level ume. Before embarking on new decentralization ef- is constrained by: (i) the low levels of management ca- Honduras Public Exphenditure Review 2008-2012 forts, special attention should be given to increasing pacity in most municipalities; (ii) limited coordination the capacity of municipalities to operate within the in investment projects across levels of government and existing framework and manage the current transfer among neighboring municipalities; (iii) unpredictability levels. Transfer shortfalls and delays distort allocation and lateness of transfers; and (iv) limited access to credit. decisions. Municipalities typically determine the shares To ensure a higher quality of decentralized capital spend- of recurrent and capital spending on the basis of bud- ing, sound public finance and investment management geted transfers and forecasted own-source revenue. systems, coupled with adequate monitoring by and coor- They tend to front-load recurrent expenditure and de- dination with the central government, are required. lay capital investments until they receive transfers from the central government. Consequently, shortfalls in The current transfer system only partially accommo- transfers are often translated into cuts to capital spend- dates for the heterogeneity in size, population, and ing. It is common for municipalities with little resources economic development of municipalities. Half of the of their own to resort to costly short-term commercial transfer pool is allocated in equal shares to all munici- loans to finance the operation of basic services when palities, and the rest according to population and poverty there are transfer delays. shares. The result is that per capita transfers are many-fold higher in municipalities with small populations than in ur- One unintended consequence of deepening decen- ban centers. Most small municipalities have few sources tralization efforts is an increase in expenditures at of revenue besides transfers, but some are rich and able to the local level based on expected transfers from the mobilize sizable resources. Presently, the distortions are central government. The reform could lead to an unin- not significantly pernicious because the overall volume of tended increase in local recurrent expenditure and higher transfers to the lowest tier is not large. Channeling more levels of debt because municipalities use budgeted trans- resources through that system, however, would magnify fers to determine how much they can spend on personnel distortions, as the fixed transfer per municipality would and to back short-term loans. be several times larger. High levels of indebtedness are a source of concern Devolving responsibilities to the municipalities is un- for a number of municipalities. In 2010, approximately likely to improve service delivery. The large majority of 28 municipalities took new loans in excess of the 20 per- municipalities lack the capacity and scale (with an aver- cent revenue ceiling, whereas 20 allocated more than age size of 27,000 inhabitants) to efficiently finance and one-fifth of their expenditures to debt servicing. While manage complex services. In the education sector, for in- information about cumulative levels of municipal debt stance, decentralization of responsibilities alone may not was not available, current trends suggest that the levels be sufficient and would not guarantee the active partici- are high for some municipalities. At least 25 municipali- pation of schools and civil society to ensure the efficiency ties have far exceeded their repayment capacity and ex- in the sector and improve learning. Initial evidence sug- hibit dangerous levels of debt (Vargas, 2011). gests that “teaming” local communities and municipali- ties can lead to the improvement of education delivery. An examination of expenditure composition sug- Honduras could take advantage of this and the PROHECO gests that most municipalities are not complying (Programa Hondureño de Educación Comunitaria) experi- with the legal earmarks established for transfers. ence to improve educational outcomes. They are required to invest 54 to 69 percent of trans- fers they receive from the central government and can The deterioration of the fiscal deficit is not a result only spend 15 to 30 percent in administrative costs, of fiscal decentralization. However, further decentral- depending on the size of the municipality. There is ization efforts could pose significant fiscal challenges to an inverse relationship between the ratio of recurrent the central government, as illustrated by the fact that the expenditure to total municipal expenditure and the central government has complied with the mandated level of transfer dependence. The data also suggest revenue transfer percentage only twice in the last decade that municipalities are choosing to use the majority of (and, as noted earlier, only three times in more than two their own resources to finance personnel and other re- decades of decentralization). Moreover, adding transfers current expenses. tends to be difficult and politically costly to revert. Honduras Public Exphenditure Review 2008-2012 Fiscal Context and Public Expenditure Trends Fiscal consolidation is the main challenge facing Hon- The increase in public debt, in particular domestic duras today. The fiscal deficit of the budgetary central debt, has raised pressure on public finances. Public government has been worsening since 2008, going from debt increased from 22.9 percent of GDP in 2008 to 34.8 2.4 percent of GDP in 2008 to 6 percent in 2012. The latter percent in 2012. This boost is mainly accounted for by an figure marked the worst performance in the region that year. The widening fiscal deficit has been driven by a sig- nificant increase in current expenditures. For instance, in Figure I.1: Fiscal Balance in Selected Central 2012, interest payments increased by 37.5 percent, cur- American Countries (% of GDP), 2012 rent transfers by 17.5 percent, and wages and salaries by 1 0.5 8.3 percent. In 2013, despite an announced government -6.0 -1.7 -4.5 -2.4 effort aimed at reducing the fiscal deficit to 4.5 percent of 0 GDP, a mid-year assessment released in July 2013 points -1 to a deficit of 6 percent of GDP once again. Operations -2 of the budgetary central government in 2008–2012 are shown in Table I-1. -3 -4 -5 -6 Honduras El Salvador Costa Rica Guatemala Nicaragua 1  SECMCA data covers the central government based on SNA (budgetary central government, excluding social security institu- tions). Source: Secretaría Ejecutiva del Consejo Monetario Centroamericano (SECMCA).1 Table I.1 : Operations of the Budgetary Central Government (% of GDP) 2008 2009 2010 2011 2012 Total Revenues and Grants 19.8 17.1 16.9 17.0 16.8 Honduras Public Exphenditure Review 2008-2012 Current Revenue 17.7 15.2 15.5 15.9 15.9 Tax Revenue 16.0 14.2 14.4 14.8 14.6 Capital Revenue 0.0 0.0 0.0 0.0 0.1 Grants 2.1 1.8 1.4 1.1 0.9 Total Expenditures 22.2 23.1 21.5 21.6 22.8 Current Expenditures 17.5 18.6 17.9 17.0 18.1 Consumption 12.0 13.8 13.1 12.0 12.4 Salaries 9.3 10.9 10.7 9.6 9.7 Goods and Services 2.6 3.0 2.5 2.4 2.7 Interest 0.6 0.7 1.0 1.4 1.8 Transfers 4.9 4.1 3.8 3.6 3.9 Capital Expenditures 4.9 5.1 3.7 4.6 4.5 Investment 2.4 2.8 1.7 1.7 1.4 Transfers 2.5 2.4 2.1 3.0 3.1 Net Lending −0.1 −0.7 −0.1 0.0 0.1 Overall Balance −2.4 −6.0 −4.7 −4.6 −6.0 Source: International Monetary Fund (IMF). Table I.2 : Debt Composition for Honduras (% of GDP) 2008 2009 2010 2011 2012 2013 Domestic 6.1 7.7 11.8 13.9 14.9 16.2 External 16.8 16.9 17.9 18.2 19.9 25.5 Total 22.9 24.6 29.7 32.1 34.8 41.7 Source: SEFIN/IMF. Data for 2013 is up to December. increase in domestic debt, from 6.1 percent of GDP in 2008 to the next in the past, they have not yielded a long-term to 14.9 percent in 2012. External debt also increased over reduction in fiscal deficits over the last decade. In 2010, the same period (2008–2012) although at a slower pace, the authorities embarked on an ambitious fiscal consoli- from 16.8 percent of GDP in 2008 to 19.9 percent in 2012. dation program which, by end-2011, succeeded in con- The debt structure has increased the government’s expo- taining the combined public sector deficit to 2.8 percent sure to refinancing risk given the declining average matu- of GDP. However, on the eve of the presidential election, rity (3.7 years) and high interest rate of domestic financing. fiscal slippages in 2012 returned the deficit to 4.5 percent of GDP (6 percent of GDP for the budgetary central gov- The performance of government finances has been ernment). This stop-and-go policy stance has produced volatile over the past decade, with episodes of fiscal rapidly increasing debt ratios. This section analyzes pat- slippages followed by fiscal consolidation. This trend terns in revenues and expenditures in the absence of solid has indeed prevailed in recent years, with a large fiscal def- fiscal consolidation, especially in the post-crisis period. icit coinciding with presidential and parliamentary elec- tion years. While corrective actions have successfully lim- Improved tax collection levels on par with the region- ited the accumulation of deficits from one administration al average have not translated into budgetary central Figure I.2: Central Government Deficit (% of GDP), Figure I.3: Central Government Revenue and 1990–2010 Expenditures (% of GDP), 2002–2012 / axis broken 12 26 24 Honduras Public Exphenditure Review 2008-2012 10 22 8 20 6 18 4 16 14 2 12 0 10 1990 1995 2000 2005 2010 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Revenues Expenditures Source: SEFIN. Source: SECMCA. Note: SEFIN data covers the budgetary central government. Shadowed years represent election years; the Y axis represents negative values. SECMA data covers the central government based on SNA (excluding social security institutions). government deficit reductions. With an average tax percent of GDP), fuel taxes (12 percent of total revenues; revenue of 14.8 percent of GDP in the 2008–2012 period, 2 percent of GDP), and import tariffs (6 percent of total Honduras ranked well above Guatemala (10.8), El Salva- revenues; 1 percent of GDP). dor (13.6), Costa Rica (13.7), Nicaragua (14.1), and Panama (11.5). The budgetary central government’s tax revenue Fiscal efforts have not been consistent enough to pro- tended to follow the economic cycle, increasing from 13.3 duce lasting results. Tax reforms have mainly served to percent of GDP in 2002 to a peak of 16.4 percent in 2007, address fiscal imbalances created by pre-electoral spend- then falling to 14.2 percent during the 2009 economic ing rather than to reduce long-term fiscal vulnerabilities. contraction and increasing again to 14.8 percent in 2011. Attempts to improve fiscal performance have not pro- However, this relatively strong revenue performance has duced the expected results due to their “ad-hoc” nature not yet resulted in a permanent decline in budgetary and a weak tax administration system. In addition, high central government deficits, owing largely to increasing turnover in the management structure of the Revenue Di- expenditures. rectorate—the Dirección Ejecutiva de Ingresos (DEI)—has led to a weakening of the institutions involved in tax re- The budgetary central government’s tax revenue ac- form. For instance, DEI changed directors three times in counts for the bulk of public sector revenue. Over the the last three years, and a commission appointed in 2012 last decade, public sector revenue in Honduras has aver- to manage the institution and strengthen revenue genera- aged around 30 percent of GDP. The central government tion was dissolved in early 2013. During its short mandate, has collected over half of this amount, while the remain- the commission developed a proposal to improve tax col- der comes from public enterprise2 revenue and public lection, reform the tax code, draft bills related to income sector workers’ pension fund3 contributions. Within the and sales tax laws, and institute other relevant measures. budgetary central government, more than 90 percent of If continued, the fiscal efforts introduced in recent years revenues come from tax collection, primarily sales taxes could generate a greater payoff. The challenge is to ensure (35 percent of total revenues; approximately 6 percent a continued consensus among all stakeholders that these of GDP), income taxes (26 percent of total revenues; 4.5 reforms are necessary and sustainable, and should thus re- main in effect during subsequent governments. 2  The electric utility, Empresa Nacional de Energía Eléctrica (ENEE), the port agency, Empresa Nacional Portuaria (ENP), and the water and sewerage utility, Servicio Autónomo Nacional de Acueductos y Alcantarillados (SANAA). A. Composition of Expenditures 3  The main funds include the teachers’ pension fund, Instituto Nacional de Previsión del Magisterio (INPREMA), the military person- nel pension fund, Instituto de Previsión Militar (IPM), and the pension Budgetary central government spending, which ac- fund for general government workers, Instituto Nacional de Jubila- ciones y Pensiones de Empleados y Funcionarios del Poder Ejecutivo counts for about 60 percent of general government (INJUPEMP). expenditures, is the main determinant of govern- Figure I.4: Budgetary Central Government Figure I.5: Central Government Tax Revenue Tax Revenue in Selected Central American Countries (% of GDP), 2002–2012 / axis broken (% of GDP), 2008–2012 Average 20 17 Honduras Public Exphenditure Review 2008-2012 16 15 15 10 14 5 13 0 12 Honduras El Salvador Costa Rica Guatemala Nicaragua Panama 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: SECMCA and IMF. Source: SECMCA. ment spending trends. In 2010, budgetary central of the funds have been absorbed by increased spending government spending and consolidated central govern- on public administration institutions at the national gov- ment spending reached 21.5 and 34.1 percent of GDP ernment level. respectively, while general government spending was estimated at 36 percent of GDP. Average spending by Consolidated central government spending on sec- the budgetary central government also accounted for tors supporting economic growth is on par or above the bulk of the increase in total spending, rising from an the levels and budget shares in countries that expe- average of 20.4 percent of GDP in the 2002–2006 period rienced sustained periods of high GDP per capita to 22.3 percent in 2007–2012. During the short-lived fis- growth.4 As noted in the background papers on sectoral cal consolidation periods, the budgetary central govern- public expenditures, there are limited improvements in ment expenditures were the key leading factor. Budget- the health and water sectors, but increased public expen- ary central government spending decreased from 23.1 ditures have yet to yield verifiable improvements, particu- percent of GDP in 2009 to 21.6 percent in 2011, which larly in education. This points to the lack of efficiency in was followed by an increase in 2012 to an estimated 22.8 sectors that are important for economic growth (World percent of GDP. Bank, 2008). In addition, another source of inefficiency is likely to be associated with the size of general public ser- The increase in consolidated central government ex- vices in Honduras, which is significantly larger—both in penditures has been mainly driven by growth in gen- percentage of GDP and total spending—than in countries eral public services. As Figure I6 and Figure I7 show, that have registered sustained high growth rates. the proportion of general public services (executive and legislative bodies, financial and fiscal affairs, external af- Consolidated central government expenditures on fairs, and other general public services), both in percent infrastructure and public order and safety have ex- of GDP and in percent of total spending, has risen from perienced drastic cuts in the recent years. Spending 8.0 to 12.6 percent and from 24.5 to 35.7 percent, respec- on safety, namely police and fire protection, has declined tively, between 2009 and 2012. More than one-third of more rapidly than public order spending, which is fo- the total consolidated central government expenditures cused primarily on judicial institutions. Infrastructure went to social sectors. Education absorbed the main expenditures, which comprised about one-quarter of the share, followed by health. However, the education share has declined over the same period due to a wage freeze 4  The data used is from Biletska and Rajaram (2007), which exam- ines public expenditure patterns in countries that experienced GDP discussed further in this chapter, and the share of social per capita growth rates of 4 percent or above for a sustained peri- security and welfare spending has slightly increased. This od of time and which was not interrupted by a prolonged growth plunge. In figures I-8, I-12, and I-13, the chosen years for these coun- suggests that the reduction in education expenditures tries correspond to periods of such levels of sustained economic has not translated into allocative efficiency gains, as most growth. Figure I.6: Composition of Consolidated Central Figure I.7: Composition of Consolidated Central Government Expenditures (% of GDP) Government Expenditures (% of Total Expenditures) 40 100 35 90 Honduras Public Exphenditure Review 2008-2012 80 30 70 25 60 20 50 15 40 30 10 20 5 10 0 0 2008 2009 2010 2011 2012 2008 2009 2010 2011 2012 Other Defense Public Order & Safety Agriculture Environment Infrastructure Social Security & Welfare Health Education General Public Services Source: SEFIN. Data classi ed according to the IMF GFS Manual (1986). Source: SEFIN. Data classi ed according to the IMF Government Finance Statistics (GFS) Manual (1986). overall budget and between 8.4–8.9 percent of GDP, were to regular health check-ups and to primary school. This drastically cut in 2012, to 21.2 percent of total govern- program is classified under capital transfers of the bud- ment expenditure and 7.5 percent of GDP. Furthermore, getary central government. patterns in the composition of infrastructure spending reveal that these cuts have been even more significant A large portion of transfers go to the municipalities, since subsidies for SOEs—which constitute the bulk of all of which have expanded their budgets follow- government expenditures in this sector—were subject ing the passage of Legislative Decree 143-2009. The to a smaller reduction, and thus their share of total in- Decree stipulates that the central government should frastructure spending has actually increased, from 77 to transfer a portion of tax revenue to the municipalities, 83 percent. With the exception of 2011, the main share setting the mandated level at 7 percent in 2010, 8 per- of subsidies to public enterprises have been for current cent in 2011, 9 percent in 2012, 10 percent in 2013, and spending, of which energy subsidies to consumers have 11 percent in 2014 and thereafter. Although it specifies constituted about 90 percent in 2009–2011 and 58 per- that municipalities should dedicate 69 percent of these cent in 2012. transfers to investment projects, in practice many allo- cate most of the transfers to cover operational expens- The central government has increased its current ex- es. While it is perhaps less important whether it is the penditures and capital transfers at the expense of central government or the municipalities who carry out capital investment. The current expenditures of the investment spending, evidence from the municipalities budgetary central government increased from 17.5 per- suggests that the transfer resources are often used for re- cent of GDP in 2008 to 18.1 percent in 2012 due to ris- current spending. Therefore, the decline in capital invest- ing interest payments, transfers, and payments for goods ment may, in fact, be larger than previously indicated. and services. At the same time, a government decrease in capital spending was accompanied by a particularly sharp The decline in capital investment by the budgetary reduction in capital investment—from 2.8 percent of GDP central government would not be a matter of con- in 2009 to 1.4 percent in 2012—to provide for larger capi- cern if the level of investment by the public sector tal transfers to other levels of government. These trans- remained stable. However, in recent years public sector fers, which increased from 2.1 percent of GDP in 2010 to gross fixed capital formation, including capital spending 3.1 percent in 2012 (from 11.1 percent to 13.8 percent of for the central and subnational governments, SOEs, and total expenditures), were not always applied toward capi- other non-financial public sector entities, has declined tal investment, but rather were often used to finance cur- to a decade-low of 3.3–4.0 percent of GDP in 2010–2011, rent expenditures. One example is the Bono 10 Mil pro- compared to 5–6 percent in 2003–2004. The question gram involving cash transfers to 350,000 poor families in then arises on whether this decrease in capital invest- Honduras on the condition that they send their children ment is well informed given the significant gaps in in- Figure I.8: Consolidated Central Government Expenditures in Honduras and Countries with High GDP per Capita Growth (% of GDP and % of Total Expenditures) 10 30 Honduras Public Exphenditure Review 2008-2012 8 25 20 6 15 4 10 2 5 0 0 Honduras Chile Korea Malaysia Thailand Honduras Chile Korea Malaysia Thailand General Public Services Education Health Infrastructure Source: SEFIN; IMF GFS (1986) for Chile, Korea, Malaysia, and Thailand. Note: Data covers 2008–2012 for Honduras; 1984–1997 for Chile; 1970–2004 for Korea; 1976–1997 for Malaysia; and 1987–1996 for Thailand. Chosen years for these countries correspond to periods of sustained GDP per capita growth. Figure I.9: Composition of Consolidated Central Government Expenditures (% of GDP) on Infrastructure on Public Order & Safety 10 2 8 1.5 6 1 4 0.5 2 0 0 2008 2009 2010 2011 2012 2008 2009 2010 2011 2012 Subsidies to SOEs Other Infrastructure Safety Public Order Source: SEFIN. frastructure access and the efficiency of resource use in The public wage bill accounts for the bulk of the in- some sectors analyzed in this report. crease in total public expenditures over the past de- cade. Public sector wages in Honduras grew from an The twin challenge of declining investment levels and average of 12 percent of GDP in 2002–2008 to nearly 15 poor efficiency in public sector investment efforts is percent in 2009–2011, placing Honduras well above the even more apparent when gross public fixed capital regional average. The rapid growth in budgetary central formation in Honduras is compared to other countries government wages, which peaked at 10.9 percent of GDP in Central America as well as countries outside the in 2009, accounted for more than 70 percent of the overall region that succeeded in achieving rapid economic increase in the public wage bill. Efforts by the govern- growth. This comparison shows that the level of pub- ment to contain salary expenditures by linking wage in- lic sector investment in Honduras is only slightly higher creases for public sector workers to inflation rather than than in Chile but significantly lower than in Korea, Malay- the minimum wage in 2010 have proven effective.5 As sia, and Thailand during periods of high GDP per capita a result of this initiative, between 2009 and 2012 salary growth rates. It also illustrates that, in the recent years, Honduras has significantly lagged behind Nicaragua, Pan- 5  Tax and salary reforms were implemented through the Ley de Fortalecimiento de Ingresos, Equidad Social y Racionalización del Gasto ama, and Paraguay in terms of levels of gross public fixed Público, approved on March 28, 2010, and by Decree 224-2010, pub- capital formation. lished on October 29, 2010. Table I.3: Budgetary Central Government Expenditure Structure (2008 - 2012) Description % of Total % of GDP Honduras Public Exphenditure Review 2008-2012 2008 2009 2010 2011 2012 2008 2009 2010 2011 2012 Total expenditures 100.0 100.0 100.0 100.0 100.0 22.4 23.1 21.5 21.6 22.8 Current expenditures 78.4 80.7 83.3 78.6 79.5 17.5 18.6 17.9 17.0 18.1 Consumption 53.5 60.0 61.0 55.7 54.3 12.0 13.8 13.1 12.0 12.4 Salaries 41.8 47.0 49.5 44.5 42.6 9.3 10.9 10.7 9.6 9.7 Goods and Services 11.7 12.9 11.5 11.2 11.6 2.6 3.0 2.5 2.4 2.7 Interest 2.7 3.2 4.5 6.3 7.9 0.6 0.7 1.0 1.4 1.8 Transfers 22.1 17.5 17.8 16.7 17.3 4.9 4.1 3.8 3.6 3.9 Capital Expenditures 21.9 22.2 17.4 21.4 19.9 4.9 5.1 3.7 4.6 4.5 Investment 10.8 12.0 7.7 7.7 6.2 2.4 2.8 1.7 1.7 1.4 Transfers 11.1 10.2 9.7 13.7 13.8 2.5 2.4 2.1 3.0 3.1 Net Lending −0.3 −2.9 −0.6 0.0 0.6 −0.1 −0.7 −0.1 0.0 0.1 Source: SEFIN. spending for the entire public sector fell by 1.8 percent of in interest payments for domestic debt, which represent- GDP while the wage bill of the budgetary central govern- ed 1.4 percent of GDP in 2012 (or 78 percent of total inter- ment declined by nearly 1.2 percent (dropping below the est payments). 10 percent of GDP mark). Overall, the domestic debt of the central government Growing salary expenditures in education and health, grew from 2.8 percent of GDP in 2007 to an estimated triggered mainly by salary increases, had the most sig- 15.2 percent in 2012. The government’s acquisition of nificant impact on the wage bill. On average, these two a large amount of new domestic debt to finance current sectors accounted for around 70 percent of the budgetary expenditures has led to a rise in financing costs. Yet a sub- central government wage bill during the 2008–2012 pe- stantial portion of the debt accumulation resulted from riod—education accounted for 52 percent, while health public sector liabilities added during the 2006–2009 peri- accounted for 18 percent. Since in 1980, employees in od (equivalent to 4.7 percent of GDP), which included the both the education and health sectors have benefitted state’s pension contributions (1.8 percent of GDP), pend- from special employment regimes that allow their salaries ing payments for goods and services (1.7 percent of GDP), to grow at a faster rate than the rest of the public sector. and transfers to municipalities (0.8 percent of GDP). The As a result, the increase in the education and health wage new debt is expensive, both because of its short-term na- bill represents 70 percent of the total increase in salaries ture and because of a lack of depth in the domestic bond between 2008 and 2009, with the education sector alone market, where interest rates average more than 9 percent accounting for 46 percent. and at times exceed 15 percent. Interest payments have increased significantly over Subsidy spending declined in recent years, from 1.6 the last three years, mainly as a result of growing do- percent of GDP in 2008 to 0.4 percent in 2012. De- mestic debt. Between 2002 and 2008, interest payments creases in international fuel prices and tighter eligibil- declined from 1.6 to 0.6 percent of GDP, with interest pay- ity requirements have contributed to declines in subsidy ments on external debt falling from 1.4 to 0.3 percent of spending. Honduras has three main subsidy programs: GDP. This reduction was the direct result of debt allevia- fuel, electricity, and urban transport. It eliminated a tion through the HIPC and MDRI multilateral initiatives. fourth program—a taxi subsidy—in 2010. Fuel subsidies By 2012, however, interest payments rose sharply, to 1.88 were introduced in 2007 to mitigate the adverse impact percent of GDP. The jump was driven by a drastic increase on consumers of a spike in international fuel prices. How- Figure I.10: Gross Public Fixed Capital Formation Figure I.11: Composition of Budgetary Central (% of GDP), 2000–2011 Government Capital Transfers (% of GDP), 2009–2012 7 3.5 Honduras Public Exphenditure Review 2008-2012 6 3 5 2.5 4 2 3 1.5 2 1 1 0.5 0 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2009 2010 2011 2012 Others Bono 10 Mil Decentralized and Source: IMF World Economic Outlook (WEO). Public Enterprises Local Governments Deconcentrated Institutions Figure I.12: Central Government Capital Expenditures Figure I.13: Central Government Capital Expenditures (% of GDP)* (% of Total) 7 6.4 30 6 24.4 25 20.6 21.1 5 4.56 20 17.7 3.8 4 2.9 15 13.2 2.9 3 10 2 1 5 0 0 Honduras Chile Korea Malaysia Thailand Honduras Chile Korea Malaysia Thailand * years for averages noted below. Source: SEFIN; IMF GFS (1986) for Chile, Korea, Malaysia, and Thailand. Note: Budgetary central government data covers 2008–2012 for Honduras; consolidated central government data cover 1984–1997 for Chile; 1970–2004 for Korea; 1976–1997 for Malaysia; and 1987–1996 for Thailand. Figure I.14: Gross Fixed Capital Formation (% of GDP) Figure I.15: Gross Public Fixed Capital Formation (% of GDP) 35 14 30 12 25 10 20 8 15 6 10 4 5 2 0 0 Honduras Chile Korea Malaysia Thailand Honduras El Salvador Guatemala Nicaragua Panama Paraguay Gross public xed capital formation Gross private xed capital formation 2009-2011 2005-2008 2001-2004 Source: IMF WEO. Note: Data covers 2008–2012 for Honduras; 1984–1997 for Chile; 1980–2004 for Korea; 1980–1997 for Malaysia; and 1987–1996 for Thailand. Source: IMF WEO. Figure I.16: Budgetary Central Government Salary Figure I.17: Total Central Government Payment Arrears Expenditure by Sector (Millions of Lempiras) 14 35,000 12 Honduras Public Exphenditure Review 2008-2012 30,000 25,000 10 20,000 8 15,000 6 4.8 10,000 4 2.1 2.8 2.5 1.7 5,000 2 0 0 2008 2009 2010 2011 2012 2008 2009 2010 2011 2012 Ministry of Education Ministry of Health Rest Arrears (percentage of actual total expenditure) Arrears (percentage of GDP) Source: SEFIN. Source: SEFIN. ever, the subsidy costs quickly skyrocketed as fuel prices (Secretaría de Finanzas or SEFIN) estimates the share of continued to increase, surpassing 1 percent of GDP in predetermined spending in the central government bud- 2008. Authorities eliminated the fuel subsidy in 2009 and, get at around 85 percent, which exceeds (by more than in 2010, made it only applicable to liquefied petroleum 20 percentage points) the rigidity of the average budget gas (LPG) for household use. Urban transport subsidies in other Latin American countries with large entitlement for Tegucigalpa and Comayagüela are paid directly to programs (e.g. Brazil). The main drivers of such elevated service providers in exchange for charging a fixed fare of expenditure levels are: rising interest payments on do- L3.5 to regular users and providing free transportation to mestic debt, escalating salaries of public employees, par- seniors and the disabled. The cost of these subsidies has ticularly in the education and health sectors, and trans- been limited to 0.1–0.2 percent of GDP.6 Finally, the gov- fers to municipalities following the 2009 Decentralization ernment gradually began reducing the coverage of elec- Law. In the absence of new financing sources, legally- tricity subsidies in 2008 until their elimination in 2012. mandated increases in both salaries and intergovernmen- tal transfers have become sticky expenditures that are politically difficult to reverse. These factors significantly B. Managing the Deterioration of the Fiscal Stance constrain government efforts to improve expenditure pri- oritization, address societal needs, and still execute the The growing fiscal deficit in Honduras stems from in- budget effectively over the course of a fiscal year. efficient public financial management, pre-electoral spending increases, and pressures from sector inter- The bulk of payment arrears stems from rising ex- est groups. Corrective measures focusing on tax reform penditures that are driven by only a few categories of and fiscal consolidation in post-election years tend to ad- spending, which indicates that the government has dress short-term fiscal imbalances rather than the reduc- been unable to pay for budgeted commitments. Ar- tion of long-term fiscal vulnerabilities. The reversal of the rears represented an average 2.8 percent of GDP over the 2010–2011 fiscal adjustment has brought about a signifi- 2008–2012 period. Although, at 2.5 percent of GDP, 2012 cant deterioration in fiscal discipline. This situation raises arrears were below the 2009 peak of 4.8 percent, they in- concerns about the fiscal sustainability of current govern- creased from the 2.1 percent level registered at the begin- ment spending. ning of the period. Transfers to municipalities and other decentralized institutions, teacher and civil servant allow- Rising expenditures have increased the rigidity of the ances, as well as social security contributions comprised central government’s budget. The Ministry of Finance 41.6 percent of total arrears in 2009, rising to 45.5 percent in 2011 (Figure I-18). Payment arrears for public debt and 6  The subsidy began as a daily amount of L1,075 to each operator; goods and services were also sizeable, but from 2009 to in March 2012, the amount was increased to L1,325. 2011 they showed declining shares of the total govern- Figure I.18: Composition of Central Government Payment Arrears, 2009 and 2011 2009 2011 Honduras Public Exphenditure Review 2008-2012 Salaries & wages, incl. allowances & social security contributions Salaries & wages, incl. allowances & social security contributions Other Transfers Goods & services Debt service Other Transfers Goods & services Debt service Source: SEFIN. Note: In 2009, the share of other payment arrears was comprised solely of arrears in capital investment spending. ment payment arrears—from 21.2 percent to 7.9 percent When compared to the original approved budgets, re- and from 27.2 percent to 8.8 percent, respectively. How- cent total expenditure outturn figures suggest an in- ever, the expanding arrears category that gives reason for creased deviation in the implementation of budgeted concern is capital spending. Although the disaggregated expenditures. In 2008, before the financial and econom- data on arrears for 2009 includes capital spending as a ic crisis, the government managed to release funds with separate category, while the 2011 data does not, we can, little deviation from the original total budget by cutting however infer that the 2011 category “other arrears” in- expenditures across sectors. However, in recent years, cludes capital spending, which increased two-fold. Even revenue shortfalls and growing spending pressures have if total arrears have been brought under control (Figure left it unable to achieve the same result. In 2011, both I17), the risk of a continued escalation looms large be- the composition of expenditure outturn and the aggre- cause the main contributors to government arrears are gate expenditure outturn deteriorated again, resulting in the fastest growing categories of spending. lower PEFA scores of B and C+, respectively8 (Figure I-19). The changes in the composition of spending suggest that the central government is not delivering public services C. The Immediate Causes of Weak Public Financial as originally planned (Honduras PEFA Assessment, 2012). Management An array of dysfunctions in the public financial man- The significant deterioration of the fiscal stance in agement system contributes to the poor performance recent years, coupled with the low quality of public of budget execution. The system is hampered by unre- spending, warrant a closer look at government institu- alistic budget planning, lack of commitment controls, and tions and processes that influence public expenditure weak cash management. Budget implementation is af- management.7 This section focuses on the institutional fected by internal factors within SEFIN and line ministries arrangements that impact aggregate fiscal discipline and as well as undue political interference from Congress, budgetary spending efficiency. It is not intended to pro- the Executive branch, and sector interests pushing for in- vide a comprehensive review of public expenditure man- creased shares in the total budget. agement practices, but rather to highlight areas that need improvement and careful progress monitoring given their The budget preparation process undermines the credi- connection to sectoral performance. bility of the annual budget. Budgets are routinely revised over the course of the year. The annual budget totals are con- 8  Guidance on the methodology for calculating aggregate expen- 7  Institutions are viewed as sets of rules, both formal and informal, diture outturn and composition expenditure outturn is available on roles, and information flows that pertain to planning, allocation, and the PEFA website: http://www.pefa.org/en/content/methodologi- spending of public resources. cal-guidance-and-practical-tools. Figure I.19: Aggregate and Composition Expenditure Outturn Aggregate Primary Expenditure Outturn (%) Composition of Expenditure Outturn (%) 15 15 Honduras Public Exphenditure Review 2008-2012 10 10 5 5 0 0 2008 2009 2010 2011 2008 2009 2010 2011 Source: Honduras PEFA Assessment, 2012. tinuously increased during preparation and further revised to support aggregate fiscal discipline and introduced ceil- upward during implementation (Figure I20). However, the ings for line ministries to limit their claims on the com- budget actuals, which represent the government’s resource mon resource pool. These spending limits aim to produce availability to fund expenditures at the budget execution predictable resource envelopes to help line ministries stage, tend to be much smaller. Such poor resource predict- plan and manage expenditures more effectively. The effi- ability encourages ministries, agencies, and other partici- cacy of such measures crucially depends on the credibility pants of the budget process to give insufficient weight to of expenditure and revenue totals as well as on the as- the budget formulation and treat it as a pro-forma exercise. sumption that sector ceilings will not be revised except The programming of salaries and operational expenditures to accommodate unforeseen expenditures exceeding the tends to dominate the budget preparation of line ministries amount of a contingency reserve. If total expenditures and agencies, while the budgeting of the programs that are continuously adjusted over the course of a year—first they implement is a residual of total allocated resources. upward to accommodate new or increased spending and then downward to reconcile with actual resources—then The central government continuously underestimates sector ceilings will inevitably prove ineffective. actual expenditures. Although SEFIN’s revenue-fore- casting capacity is quite robust, the central government Ineffective expenditure ceilings undermine strate- continuously underestimates actual expenditures. To ac- gic prioritization of government spending and per- commodate a lower spending level than required by exist- petuate budget incrementalism. The previous bud- ing policy commitments, the proposed and subsequently get continues forward as the basis for the ensuing one, approved budget often does not include certain catego- which then relies on incremental additions to spending ries of spending mandated by law or policy decisions. For programs derived from formalistic analysis of legal ex- instance, the 2013 budget does not provide funds for the penditure requirements. Meanwhile, spending priorities increased teacher wages approved in 2012 following na- receive little consideration in relation to emerging eco- tional strikes, which the Ministry of Education conserva- nomic and social needs. In the end, the back-and-forth tively estimates at an additional L800 million. In another process of budget proposals and revisions between SEFIN illustrative example, Congress received a proposed 2012 and line ministries becomes merely a perfunctory deter- budget with an estimated fiscal deficit of 3.5 percent of mination of budget allocations before the budget is sub- GDP, while SEFIN presented a different macroeconomic mitted to Congress. framework with a projected fiscal deficit of 4.5 percent of GDP and a larger allocation for public investments. The budget process encounters considerable spend- ing pressures at both the approval and execution Furthermore, the sector expenditure ceilings are not stages. These budget demands often undermine expen- enforced. Honduras adopted a top-down budget model diture ceilings of ministries and thus impact the compo- Figure I.20: Total Budget at Different Stages of the Figure I.21: Total Budget Execution (%) Budget Cycle (Millions of Lempiras) 100 180 160 Honduras Public Exphenditure Review 2008-2012 80 140 120 60 100 80 40 60 40 20 20 0 0 2008 2009 2010 2011 2012 2008 2009 2010 2011 2012 Prepared by Institutions Approved by Council of Minitries Accrued expenses to modi ed budget Actuals to modi ed budget Approved by Congress Modi ed Accrued Expenses Actuals* Source: SEFIN. Source: SEFIN. Table I.4: Upward Adjustment of Aggregate Budget Ceilings Budget Adoption (Millions of Lempiras) Budget Adoption (% Change) Years Original Ceilings Approved by Approved by Modified Approved Approved Modified Change between (proposed budget Council of Congress Budget by Council by Congress Budget Original Ceilings by ministries/ Ministers of Ministers and Modified spending agencies) Budget 2008 101,634.6 111,374.6 111,161.3 119,509.4 9.6 - 0.2 7.5 17.6 2009 119,220.2 112,939.7 112,939.7 125,577.1 - 5.3 0.0 11.2 5.3 2010 117,255.4 121,990.2 121,990.2 127,732.5 4.0 0.0 4.7 8.9 2011 128,489.5 133,288.4 133,288.4 144,501.1 3.7 0.0 8.4 12.5 2012 136,449.1 144,744.5 145,022.0 164,461.1 6.1 0.2 13.4 20.5 Source: SEFIN. sition of spending. The first upward adjustment occurs resource availability over the course of a fiscal year. The at the approval stage, when the Council of Ministers re- budget incorporates unrecorded commitments as expen- quests changes during budget formulation that Congress ditures either later in the current year or, in the case of subsequently reviews and eventually approves. However, multi-year contracts, in subsequent years whenever the the main upward adjustments result from programming government receives invoices and associated payments changes due to revised budget estimates, new projects, or materialize. Unrecorded commitments are difficult to shifting priorities at the budget implementation stage. A quantify because such commitments are not reflected in major source of rising expenditures are the various sector accounting records. The analysis of approved and modi- interests pushing for increases, either by influencing high- fied expenditures by economic classification (Table I5) level policymakers or by exerting public pressure (e.g. shows that, except in 2010, the highest increase for an teacher and police strikes). As shown in Table I-4, the total approved budget occurred for capital spending, which expenditure hikes ranged from 4.7 to 13.4 percent of the is primarily due to increased disbursements of external original proposed budget during the 2008–2012 period. loans in recent years.9 This analysis also suggests that a Sizeable expenditures also originate from unrecorded 9  Any supplier and / or contractor of the State at the time of re- commitments. Such commitments include purchase or- ceipt of the purchase order must request a copy of a pre-credit com- mitment registered in the Integrated Financial Management System ders issued and contracts signed without verification of (SIAFI), to ensure that the contracting institution has the funds avail- Table I.5: Current and Capital Expenditures—Approved and Modified Budget Current Expenditures Capital Expenditures Approved Modified Change Approved Modified Change Honduras Public Exphenditure Review 2008-2012 (Millions of Lempiras) (Millions of Lempiras) (%) (Millions of Lempiras) (Millions of Lempiras) (%) 2008 95,263.00 100,030.70 5.0 15,902 19,480.10 22.5 2009 98,235.50 104,359.50 6.2 14,704 21,217.60 44.3 2010 107,280.70 110,098.80 2.6 14,709 17,633.70 19.9 2011 119,875.90 123,653.80 3.2 13,412 20,847.30 55.4 2012 128,475.80 140,114.40 9.1 16,546 24,346.60 47.1 Source: SEFIN. Table I.6: Additional and Modified Budget by Functional Classification (% of Total) Additional Budget Modified Budget 2008 2009 2010 2011 2012 2008 2009 2010 2011 2012 General Public Services 29.8 38.0 19.3 21.9 63.8 30.0 28.8 26.5 30.2 32.9 Education 12.9 1.8 0.0 0.5 0.0 20.2 23.4 23.4 23.9 25.2 Health 3.7 6.0 1.2 6.9 5.0 6.4 7.8 7.9 7.2 7.0 Social Security & Welfare 2.1 0.3 7.3 10.8 6.2 5.1 5.0 5.6 5.4 6.9 Infrastructure 19.9 28.1 36.2 27.4 12.0 25.5 25.0 26.1 24.3 23.8 of which SOEs 12.6 0.7 1.6 9.5 4.0 19.4 17.1 19.5 18.5 19.4 Environment 6.8 0.8 19.5 6.2 0.3 1.1 0.6 1.4 1.2 0.8 Agriculture 10.8 1.1 5.1 7.9 2.9 2.2 1.9 2.1 2.2 1.9 Public Order & Safety 5.5 4.5 7.1 8.2 4.7 3.8 3.9 4.0 4.0 3.8 Defense 5.5 4.0 1.2 2.8 1.6 3.3 3.3 3.7 3.7 3.5 Other 3.0 1.5 3.1 2.4 3.4 3.9 2.8 2.4 3.4 2.4 Source: SEFIN. significant share of unrecorded commitments is likely to 738–943 percent increases (for the Electoral Supreme comprise capital investment. The growth of current ex- Court and the Financial Services of the Central Adminis- penditures contributed only marginally to higher levels of tration, respectively) to 25 percent decreases (the Nation- total spending. al Program for Social Rehabilitation). However increases in education and infrastructure have been declining since The distribution of budget modifications among sec- 2010 and 2011, respectively. The drastically-reduced in- tors based on functional classification shows that the creases to educational spending are explained by a gov- general public services category has benefited the ernment decision to freeze salary increases for teachers, most from spending increases, followed by infrastruc- which in earlier years crowded out capital expenditures. ture (primarily subsidies to SOEs) and education. Real- The pattern for overall social spending is uneven, as rela- location of expenditures across budget heads varied sig- tively low shares for health, social security, and welfare nificantly. In the 2012 budget, reallocations ranged from have increased. able to honor the commitments made. Otherwise the government, Cash rationing due to resource constraints has led to through the Ministry of Finance, will not accept claims for payment further distortions in spending across sectors at the of debts that do not have the respective budgetary support. Officials expense of capital investments. Actual spending by who violate this article shall be jointly liable financially to honor all the debts generated by not having appropriate credit reserves. line ministries and agencies has significant deviations Figure I.22: Budget Execution by Economic Classification (%) 100 100 Honduras Public Exphenditure Review 2008-2012 80 80 60 60 40 40 20 20 0 0 2008 2009 2010 2011 2012 2008 2009 2010 2011 2012 Current Expenditure Current Expenditure Capital Expenditure Capital Expenditure Accrued Exp. To Modi ed Budget Actuals to Modi ed Budget Accrued Exp. To Modi ed Budget Actuals to Modi ed Budget Source: SEFIN. from the approved modified budget. Execution rates of financed infrastructure and social spending (except edu- government spending on a cash basis hovered around cation) are the areas that have faced cuts in fund releases, 70–82 percent for the 2008–2012 period, indicating while the most protected sectors include general public the government’s real fiscal capacity set by SEFIN. The services, environment, and agriculture. In fact, the edu- government’s approach to defining the budget execu- cation sector, which historically has received the highest tion rate as the ratio of accrued expenses to the modi- share of the budget, had expenditure execution rates of fied budget, without considering the ratio of actual nearly 90 percent until 2010. payments to modified budget, is misleading for future budget preparation. Such a definition assumes a greater The absence of mechanisms for commitment con- fiscal capacity of the state and thus results in setting an trol exacerbates cash rationing, adds rigidity to the over-optimistic total resource envelope in the following budget, and poses a substantial fiscal risk. Although year. The budget execution rates for current expendi- stipulated in the Organic Budget Law, no ex-ante control tures have exceeded those of capital spending—72–83 exists at the commitment stage and no mechanisms are percent and 61–78 percent, respectively. This suggests provided to program and control expenditures before that cash rationing is one of the main factors that further they are generated through scheduling and allocation compromises capital investments.10 Moreover, the dif- of commitment quotas. A number of contracts without ference between execution rates on the accrual versus pre-assigned budget allotments have produced arrears cash basis is larger for capital spending, which implies payable in future fiscal years in violation of the Organic that it constitutes a larger share of payment arrears that Law. In 2009, in-year expenditures that were not in- are most likely to reflect unrecorded capital-investment cluded in the appropriated budget reached 4.7 percent commitments. of GDP. This absence of commitment control also gives discretion to the Executive to add new projects and ex- In recent years, the pattern of execution rates across sec- penditures without allocating funding, which further ex- tors has skewed expenditure priorities in favor of core pands payment arrears. In 2008, 2.6 percent of the funds general public services and away from health, infrastruc- committed were not paid, while only 77.4 percent of the ture, and social security and welfare. Thus, domestically modified budget was paid (Figure I-23). Cash rationing often has adverse implications for capital spending be- 10  A study examining constraints to budget execution in the in- cause delays in payments to service providers can slow or frastructure sector in Indonesia (World Bank, 2012) finds that low stall the work and lead to higher procurement costs and execution rates for capital spending can also be caused by delays in budget preparation, appointment of project managers, and procure- inefficient management.11 ment. However, compared to Indonesia where capital expenditure execution rates were 80 percent and above in 2009–2011, there is significant room for improvement in capital spending execution 11  “Impunidad, el verdadero problema presupuestario en Hondu- rates in Honduras. ras,” FOSDEH, 2012. Table I.7: Budget Execution by Functional Classification (%) Actuals to Modified Budget Accrued Expenses to Modified Budget 2008 2009 2010 2011 2012 2008 2009 2010 2011 2012 Honduras Public Exphenditure Review 2008-2012 General Public Services 70.2 59.7 71.9 74.0 84.0 72.1 74.2 75.6 79.0 90.4 Education 91.6 85.5 91.9 88.2 79.9 92.5 87.1 93.1 94.2 94.6 Health 85.3 82.9 89.8 92.4 74.2 93.8 95.4 94.5 96.1 92.9 Social Security & Welfare 88.6 28.2 36.9 73.1 66.7 88.7 30.0 41.7 74.7 69.9 Infrastructure 74.2 75.9 73.0 84.3 68.9 75.9 83.1 77.9 90.6 83.6 of which SOEs 71.6 78.0 73.3 85.5 70.0 72.8 82.3 77.7 91.7 83.2 Environment 61.8 57.2 90.2 85.8 88.5 95.8 73.6 91.9 87.5 91.1 Agriculture 73.8 71.1 72.2 79.5 87.8 75.8 75.9 74.5 85.4 93.2 Public Order & Safety 93.7 85.4 90.1 92.2 83.5 98.8 93.8 96.5 97.3 97.0 Defense 93.1 77.4 85.3 93.2 72.1 95.2 83.0 85.8 95.4 80.1 Other 35.6 31.5 89.8 75.6 79.0 41.1 36.6 90.6 76.1 80.9 Source: SEFIN. Table I.8: SIAFI Coverage (Millions of Lempiras) 2008 2009 2010 2011 2012 Government total payments 92,529.2 88,328.2 99,137.1 118,687.8 127,260.9 Payments out of SIAFI made by centralized institutions 2,961.6 3,068.5 3,503.8 3,712.2 3,768.4 Payments out of SIAFI made by decentralized institutions 18,652.9 13,454.2 18,475.1 24,329.6 31,788.9 Total of payments out of SIAFI 21,614.5 16,522.7 21,978.9 28,041.8 35,557.3 % from total government payments out of SIAFI 23% 19% 22% 24% 28% Source: SEFIN. The implementation of the Integrated Financial Man- directors of the Public Credit, Budget, and Planning Unit agement System (SIAFI) remains incomplete. SIAFI is (UPEG)—monitors cash flows and decides on the release bypassed at the commitment control stage and used only of the quarterly allotment based only on availability of at the payment stage when commitments, invoices, and funding in a given quarter, which leads to cash rationing. payment obligations are recorded. Furthermore, the pro- curement information system is not linked to SIAFI, which At present, only 63 percent of government institutions makes it difficult to keep track of multi-year commitments connect to SIAFI, while approximately 28 percent of under infrastructure investment contracts. There is no total payments are made outside of the system. SIAFI, practice of preparing schedules of annual payments for which covers all line ministries but not all public agencies, multi-year contracts. Until recently, the system of pro- comprises the development and integration of modules graming and allocation of commitment quotas established on: (i) budget, (ii) accounting, (iii) treasury, (iv) human re- in the Budget Law, although referred as commitments, in sources, (v) administration of goods, (vi) travel manage- practice consisted of quarterly planning and allocation ment, (vii) revolving funds (fondos rotatorios), and (viii) of cash for payments, broken down by month. There is management of externally financed projects (unidades no programming or control at the moment of expendi- ejecutoras de proyectos con recursos externos or UEPEX). ture commitment, which, according to the provisions of The initially-designed modules for travel management Article 31 of the Budget Law, is the time of entering into and revolving funds have never been implemented, while a legal contract with third parties. In practice, the Cash those for human resources, administration of goods, and Flow Committee—comprised of the National Treasury and UEPEX have only partially been implemented.    This in- Figure I.23: Arrears and Actual Payments The absence of robust institutional mechanisms for (% of Modified Budget) maintaining fiscal discipline and achieving expendi- ture efficiency lies at the heart of public expenditure management concerns. SEFIN plays a pivotal role in fis- Honduras Public Exphenditure Review 2008-2012 cal discipline that ranges from setting budgetary ceilings to controlling cash release during budget execution. The strengthening of SEFIN’s role was meant to ensure fiscal discipline. However, the recent hikes in expenditures, which were particularly pronounced in pre-election years and led to rising fiscal deficits, show the limits of such institutional aims. High-level political influences are also reducing SEFIN’s authorizing power because the 2008 2009 2010 2011 2012 main drivers of increased expenditures, such as growing domestic interest payments, transfers to municipalities, Actuals Accrued not paid Overestimation and salaries, originate outside of its control in Congress Source: SEFIN. and the Executive. Honduras lacks formal institutional mechanisms backed by political will that ensure coher- complete implementation of SIAFI has led to the creation ence in public expenditure management, such as effec- of specific systems for the needs of all public entities, pro- tive and formal commitment controls and assessment ducing duplication and inconsistency in information gen- of the pertinence and affordability of various sector de- eration due to the lack of an integrated single system for mands over the medium term. In addition, its mandate all public entities. overlaps to a large extent with the newly established Ministry of Planning, which could undermine SEFIN ef- A weak capacity to track expenditures and monitor forts if the government does not clearly define their re- performance undermines aggregate fiscal discipline. spective responsibilities and introduce effective coordi- No detailed reporting or feedback exists on the perfor- nation mechanisms. Thus, despite its rather significant mance of programs, projects, and policies that can inform technical capacity to manage public finances—including line ministries in their budget decisions or help counter- staff skills, resources, and information technology—the act persistent interference from the Executive. Program recent expenditure hikes reveal its limited capability to performance reviews are minimal and audits are largely facilitate aggregate fiscal discipline and allocative effi- financial. These factors make scrutiny and implementa- ciency of public spending. tion of in-year modifications difficult to manage. In the short to medium term, Honduras faces major challenges in implementing and sustaining fiscal con- D. The Deeper Causes of Weak Public Financial solidation while at the same time undertaking equally Management important quality fiscal adjustments. The lack of al- locative spending efficiency can potentially translate into Dysfunctions at the budget formulation and imple- inefficient consolidation whenever political expedience mentation stages, which have led to fiscal profligacy, determines sector spending cuts with little consideration are symptomatic of deeper institutional problems. for economic and social expenditure priorities. While The prevailing informality of institutional arrangements tough policy choices to reduce budget rigidity are neces- and practices, combined with a limited track record in sary, enhancing public financial management practices is implementing new policies, have undermined overall fundamental in order to achieve a sustained quality fiscal government fiscal performance. The people of Honduras adjustment. have one of the lowest trust levels for the State in Latin America. Poor quality of public services and the ineffec- Dysfunctions in budget planning and execution un- tive and inefficient use of public resources are damaging dermine aggregate fiscal discipline and efficient allo- public confidence. Improving the quality of public ser- cation of resources. The primary constraint arises from vices should gradually enhance citizen confidence in the SEFIN’s nonbinding expenditure ceilings, which are rou- trustworthiness of their government. tinely breached. Sector ministries request significantly Figure I.24: Results of the Citizens’ Perception Survey Citizens’ Trust in the State Citizens’ Perception of Tax Evasion 60 7 Honduras Public Exphenditure Review 2008-2012 50 6 5 40 4 30 3 20 2 10 1 0 0 El Salvador Costa Rica Guatemala Nicaragua Panama Paraguay Latin America Costa Rica El Salvador Guatemala Nicaragua Panama Paraguay Latin America Honduras Honduras Source: Latinobarómetro, 1996–2011. Perception of trust in government, Source: Latinobarómetro, 1998–2011. Answer to question: In a scale of 1 to 10, rank 0–100%, being 100% more trust. how justi able is it for you to evade taxes? Being 10 totally justi able. E ciency of the State Satisfaction with Public Services 7 3 6 5 2 4 3 1 2 1 0 0 El Salvador Costa Rica El Salvador Costa Rica Guatemala Honduras Guatemala Nicaragua Nicaragua Panama Paraguay Latin America Panama Paraguay Latin America Honduras Source: Latinobarómetro, 2011. Citizens’ perception of e ciency of the State, Source: Latinobarómetro, 2009–2011. Citizens’ satisfaction with public services rank 0–10, being 10 very e cient. (central and municipal), rank 0–11, being 11 very satis ed. higher expenditure envelopes over numerous negotia- also leading to overspending at the sectoral level, grow- tion rounds that extend into budget execution. As a re- ing arrears, and widening fiscal deficits. The government sult of the many budget revisions, cash rationing has be- recently adopted de jure laws to impose commitment come endemic, while the level and composition of actual control that have yet to become operational. Although resource allocations frequently deviates significantly from the measures authorized SEFIN to refuse approval for in- the approved budget. To make expenditure limits cred- year budget modifications made without a correspond- ible and enable effective cash management, SEFIN needs ing budget allocation, whether it actually manages to to base these ceilings on a realistic deficit target and then exercise this power will prove a determining factor in enforce them at both the budget preparation and imple- whether commitment control efforts succeed. mentation stages. In order to enforce expenditure ceilings and commit- The absence of de facto mechanisms for commitment ment control effectively, SEFIN needs an authorizing control is exemplified by the discretion of the Execu- environment backed by political will at the highest tive to add new projects and programs without allo- level. Without the strong and sustained support of politi- cated funding. Pressures from sector interest groups are cal leadership, SEFIN efforts to enforce spending ceilings and commitment control will continue to lack credibility. management and overall performance, which in turn Merely adopting laws and implementing information sys- have led to lax fiscal behavior. The prevailing informal- tems is insufficient because both can be circumvented ity of institutional arrangements and practices, as well as and adapted to traditional informal practices. Effective a weak track record for implementing reforms, are major Honduras Public Exphenditure Review 2008-2012 management requires coordination among the planning obstacles to devising a more stable fiscal policy. The lack and the execution functions, a clear definition of roles, and of stability and continuity of civil service in key areas of a corresponding implementation capacity across entities. public financial management is another important con- It also requires a transformation from incomplete and ad- tributing factor. As a medium- to longer-term priority, hoc policymaking into a more thorough and detailed ap- public financial management requires enhanced coordi- proach, considering that policy activism with weak imple- nation in managing public finances across central finance mentation can generate unexpected outcomes. functions, sectors, and levels of government; effective capacity-building in the financial management of line The 2013 Budget Law addresses de jure some of the ministries and agencies; and the assurances that SEFIN dysfunctions of the public financial management will maintain a leading role throughout the process. system, although its enforcement capacity is yet un- tested.   Article 75 of the General Provisions includes a While the fiscal situation that Honduras faces today provision to prevent expenditure commitments without a is not a result of the decentralization process and cor- prior verification of budget availability. The government responding transfers to municipalities, a continued plans to develop an application in SIAFI that requires each push toward decentralization could pose significant spending agency to make necessary budgetary modifica- fiscal and political challenges. Growing fiscal deficits tions prior to execution, schedule the dates of receipt of all and weak public financial management practices con- goods and services (accrued expense), and arrange a cor- strain the ability of the central government to implement responding payment plan with suppliers and contractors.  and finance the decentralization process. The next chap- The agencies would send this information to the Treasury, ter looks at the decentralization efforts in detail. which would in turn use it for its cash plans and releases. SEFIN plans to develop a new SIAFI module that requires spending agencies to develop an annual fiscal plan for all monthly expenses paid with national Treasury funds.  Its goal is to develop an annual overview of cash payments, greater predictability of cash releases, and an accurate es- timate of payment arrears.  SEFIN has also assigned staff (pre-interventores) to the Ministries of Education, Public Works, Health, and Transportation and Housing to exert formal control over expense-related documents. Over- sight and sanctions against violators will occur under the administrative procedure of the Superior Court of Audi- tors (TSC). The enhanced monitoring of budget execution will help inform proposed changes in allocation. The Min- istry of Planning and External Cooperation (SEPLAN) will conduct monthly monitoring of public expenditures us- ing information related to physical progress and financial execution of investment at the national, regional, and municipal levels. This review will utilize data from SIAFI and from the Monitoring Platform of the Annual Operat- ing Plan and Budget of SEPLAN. Political support is required to address the deeper in- stitutional issues that underlie poor public financial Honduras Public Exphenditure Review 2008-2012 Fiscal Decentralization: Building Credible Reforms Honduras has announced an ambitious process of Current decentralization aspirations would pose sig- fiscal devolution that aims to increase considerably nificant fiscal and political challenges. Growing fiscal the municipal share of total expenditure. The govern- deficits and weak public financial management practices ment pledged in 2010 to increase the share of central constrain the ability of the central government to imple- revenue transferred to local governments from 5 to 20 ment and finance the decentralization process. The cen- percent by 2017 and to 40 percent by 2038. Since the tral government has complied with the mandated rev- passing of the 1990 Law of Municipalities that initiated enue transfer percentage only twice in the last decade the decentralization process, the government has at- (and only three times in the more than two decades of tempted numerous reforms seeking to reinforce the pro- decentralization). Politically, mandated transfers tend to cess. However, these reforms have only been partially be difficult and costly to revert. implemented. A multitude of management, administra- tive capacity, and accountability problems remain both Improving transfer predictability is an important pre- at the central and subnational level.12 requisite for promoting effective local management of resources. If the government announces a new increase in transfers but cannot effectively comply with them, the new decentralization policy can unrealistically elevate ex- 12  César Vargas (2011) provides insights into the political econ- pectations. The reform could then lead to an unintended omy behind the decision to push for decentralization in Honduras. He suggests, for example, that during the 2009 crisis municipalities increase in local recurrent expenditure and higher levels played an important role in supporting the new president, who in of debt because municipalities use budgeted transfers to turn promised an increase of transfers from 5 to 11 percent by 2014. A political economy analysis of the decentralization process, how- determine how much they can spend on personnel and ever, is beyond the scope of this report. to back short-term loans. The plans for deepening fiscal decentralization are This chapter looks at the decentralization13 process not informed by an assessment of the comparative in Honduras from three complementary angles. First, advantage of municipalities in delivering social and it reviews the features and evolution of the intergovern- infrastructure services. For sectors in which there are mental framework and previous attempts to strengthen Honduras Public Exphenditure Review 2008-2012 clear scale economies and spillovers, devolution to the the role of municipalities. Second, the chapter examines municipalities is not advisable. The present policy dis- the effects of transfer unpredictability and shortfalls, and cussion has focused solely on the distribution of finan- the institutional weaknesses of municipalities. Third, it cial resources without considering what functions local presents selected sectoral outcomes and existing evi- governments can carry out effectively. The assignment dence from community-level interventions to illustrate of resources has to be consistent with the distribution of the potential benefits and hurdles of localizing service responsibilities (“funding follows function”). provision. Devolving more responsibilities to all municipalities is unlikely to help improve service delivery or increase A. Intergovernmental Framework capital formation. The large majority of municipalities and Local Conditions lack the capacity and scale (with an average population size of 27,000) to efficiently finance and manage complex The current intergovernmental fiscal framework cre- services. Furthermore, currently most municipalities do ates positive and negative incentives for municipali- not comply with the central government’s investment tar- ties. On the one hand, a considerable number of mu- gets and spend most of the transfers on personnel and nicipalities finance a sizable share of their expenditure other recurrent expenditure. Further devolving expen- through the collection of their own revenue and have, diture responsibilities can lead to a reduction in public over time, increased their tax mobilization. While debt capital expenditure because it is unlikely that municipali- levels are high and unsustainable for 25 municipalities, ties will be able to effectively absorb and scale up invest- the majority are demonstrating some level of fiscal re- ments in the medium term. To ensure a higher quality sponsibility by not assuming new debt beyond the legal of decentralized capital spending requires sound public limit. On the other hand, most municipalities are not investment management systems and adequate moni- complying with the earmarks on transfers, in particular toring by and coordination with the central government. the share assigned to capital expenditure. In addition, Moreover, the gains expected from decentralization in due to the unpredictability and lateness of transfers, terms of greater accountability as well as increased pro- many municipalities regularly resort to short-term loans duction and allocative efficiency are not likely to be real- from private banks to finance their recurrent spending ized in a context of weak institutional capacity. obligations. These loans, which generally imply high in- terest rates that can divert sizable resources from produc- The present transfer system only partially accommo- tive spending, are backed with budgeted transfers from dates for the heterogeneity in size, population, and the central government. economic development of municipalities. Half of the transfer pool is allocated in equal shares to all munici- palities and the rest according to population and poverty 13  The design of an intergovernmental framework involves a shares. The result is that per capita transfers are many- number of interrelated dimensions, including political, administra- tive, and fiscal. Political decentralization assigns decision-making fold higher in municipalities with small populations than authority to subnational governments that are directly elected by in urban centers. Most small municipalities have few the residents of their jurisdiction (Manor, 2000). Fiscal decentraliza- tion requires that subnational governments be granted decision- sources of revenue besides transfers, but some are rich making power over taxation and expenditure policies. Administra- and able to mobilize sizable resources. Presently, the tive decentralization guides subnational governments to assume authority over policies and public officials in their jurisdiction (Eaton distortions are not problematic because the overall vol- and Schroeder, 2010). Different degrees of administrative authority ume of transfers to the lowest tier is not large. Channel- result in different models of decentralization. Deconcentration gives subnational governments or regional units of line ministries the ing more resources through that system, however, would responsibility for implementing policies, but they have little discre- magnify distortions, as the fixed transfer per municipality tion over their design or the allocation of resources. In a delegation would be several times larger. Asymmetric approaches model, subnational governments are granted partial responsibilities and authority, while the central government retains significant deci- could help factor in the heterogeneity in the municipali- sion-making power. With devolution, subnational governments are ties’ capacity and size. granted complete authority over functions and resources without any oversight from national officials. While the fiscal and administrative frameworks pres- Municipalities vary considerably in size and popula- ent problems, the implementation process poses the tion density. Over half of the population lives in the eight greatest challenge to municipal governance. Before most-developed urban municipalities, while the rest lives embarking on new decentralization efforts, special at- in municipalities of 70,000 inhabitants or less (the average Honduras Public Exphenditure Review 2008-2012 tention should be given to increasing the municipalities’ population is below 20,000 and the lowest is 1,100). Also, capacity to operate within the existing framework and municipal surface areas vary considerably, from 17 to manage the current transfer levels. Improving transfer 8,063 sq. kilometers (the average municipal surface area is predictability is an important prerequisite for promot- 376 sq. kilometers), while population densities vary from ing effective local management of resources. Enhancing 2.5 to 800.6 inhabitants per sq. kilometer. Although the vertical and horizontal coordination in public investment poverty headcount is high across all municipalities (66.2 management is another priority because different levels percent of the population in 2011), the average poverty of government and the various entities involved jointly rate is higher in rural areas than in cities—71.6 versus 59.8 provide much of the infrastructure. percent, respectively. All these factors impact both the municipal costs for providing services and infrastructure i. Heterogeneity in Size and Capacity and the administrative requirements necessary for effec- tive service delivery. Honduras is a unitary country with two main levels of government, central and local. Territorially, the country The vast majority of municipalities lack the necessary is organized into 18 departments and divided into 298 capacity to effectively manage resources and have municipalities. Departments operate with limited func- made limited progress since decentralization began. tions under governors directly appointed by the Presi- Municipalities exhibit wide disparities in development dent. The 1982 Constitution recognizes municipalities and institutional strength. The Secretariat of Interior and as autonomous political and administrative entities with Population divides municipalities into four categories their own treasuries and entitlement to revenue gener- based on several indicators: the human development in- ated in their jurisdiction. In practice, they could only ex- dex, the level of urbanization, institutional capacity, the ercise their autonomy after the 1990 Law of Municipalities resource base, and access to basic services such as tele- gave them fiscal resources along with several of their own phone, piped water, and power. The categories include: sources of revenue. A (High Capacity), B (Intermediate Capacity), C (Poor Ca- pacity), and D (Very Poor Capacity). In 2010, the govern- While municipalities have political and administra- ment classified 8 percent of municipalities as developed, tive autonomy, horizontal and vertical accountability 11 percent as intermediate, and the remaining 81 percent remains weak, especially in small municipalities. A were categorized as poor or very poor despite numerous direct vote elects a Mayor and a Council to govern; the initiatives for institution building at the local level. number of Council members varies within a range of four to ten, depending on municipal size. Although the Coun- Municipalities have varying capacities to mobilize cil is expected to provide a balance of power and repre- their own resources and access external financing. sent a diversity of interests, in practice mayors wield con- Type A and B municipalities have a broader economic siderable influence (USAID, 2004). base and a better capacity for tax collection. Their sub- Table II.1: Average Population and Human Development Index per Category of Municipality, 2010 Category Number HDI Income per Population Surface Area Population Capita (sq. km) Density Average Minimum Maximum A 23 0.72 4,307 174,582 14,651 1,126,533 733.9 272.4 B 32 0.68 2,981 32,545 2,772 112,909 555.8 114.5 C 107 0.65 2,360 15,720 1,352 78,797 408.6 72.7 D 136 0.60 1,729 9,611 1,163 31,538 247.6 70.9 Source: SEFIN. stantially higher resource base provides them with great- ment, regulation of public spaces, local infrastructure, and er fiscal space to invest in social infrastructure, manage other local services to the lowest tier of government. Mu- the unpredictability of transfers, and meet service needs nicipalities are also expected to assume responsibilities and citizen demands. Conversely, Type C and D munici- related to municipal police and public lighting services. Honduras Public Exphenditure Review 2008-2012 palities lack the capacity to mobilize resources and re- Since it came into effect, the law’s functional assignments main primarily dependent on vertical transfers. have remained unchanged, even when Decree 143/2009 mandated that the share of central revenue transferred ii. Municipal Responsibilities Focused on Local Services to municipalities be more than double by 2013. Table II2 provides a list of essential public services and the role of The functions assigned to municipalities have re- different levels of government in their delivery. mained limited to the few local services mentioned in the 1990 Law of Municipalities. The law assigns munici- The central government is still responsible for the palities certain functions related to solid waste manage- management and financing of the main social services Table II.2: Functional and Expenditure Assignment across Levels of Government in Honduras Functions National Local Observations Security X Municipalities assume some expenses to support the Ministry of Security (creating the Munici- pal Police, providing infrastructure facilities, etc.). Health X Municipalities assume some expenses for maintenance, guards, and the purchase of medicines and other items in support of the Ministry of Health. Education X Municipalities assume some maintenance expenses for schools, guards, purchase of school sup- plies, temporary teachers, etc., in support of the Ministry of Education. Social Programs X X Municipalities assume some responsibilities in caring for vulnerable groups as well as some expenses to support relevant institutions. Housing and Urban X X Municipalities assume some expenses to support the Ministry of Public Works, Transportation, Planning and Housing (SOPTRAVI) for housing and counterparts in housing programs (community land). Municipalities are assigned the responsibility of urban planning. Solid Waste Collec- X Municipalities have the sole responsibility to provide all related services. tion and Treatment Roadways and X X Municipalities are responsible for opening and maintaining the urban and secondary roadways. Urban Streets Transport X The Ministry of Public Works, Transportation, and Housing (SOPTRAVI) has responsibility for op- erating permits and local route openings. Potable Water X Municipalities participate in programs of investment in water and sanitation through central- ized and decentralized entities such as: Honduran Social Investment Fund (FHIS); National Rural Development Program (PRONADER); Ministry of Finance (SEFIN); National Autonomous Aque- duct and Sewer Service (SANAA), etc. Municipalities also offer technical assistance and regulate the sector. In Article No. 67, the Water Law (Decree 181/2009) states that the municipalities will issue wa- ter rights, permits, and licenses under established regulations for the following consumption: 1) household use; 2) industrial, artisanal, and micro and small enterprises; 3) artisanal and sports fishing; 4) local tourism; 5) irrigation systems up to 10 hectares in size; 6) agro-cattle whose isolated consumption does not exceed 0.06 liters per second; and 7) legally recognized water administration boards. Article 86 of this Law indicates that the water authority will establish the tariff framework and its review, and, depending on the case, by the municipalities after the calculations are assessed and reviewed by the respective regulating entities. Electricity X X Municipalities manage the construction of new distribution networks and the promotion of ecological protection and reforestation. Municipalities are directed to use resources from the Poverty Reduction Strategy in part to bring power projects to their communities. Municipalities also participate in the control and promotion of commercial and industrial activities, with a high involvement in the policies for electrical energy generation and transmission. Source: SEFIN. Table II.3: Expenditure Assignment across Levels of Government in Central America Social Services Transportation Other Services Utility Services Honduras Public Exphenditure Review 2008-2012 Primary and Preschool Education Ports and Navigable Waterways Drinking Water and Sewerage Public Order and Safety Electric Power Supply Urban Transportation Interurban Highways Secondary Education Oil and gas Pipelines Telecommunications Nutrition Programs Waste Collection Urban Highways Fire Protection Social Welfare Public Health Universities Irrigation Railroads Hospitals Housing Airports Heating Police Costa Rica C C C C C C C C C C C C C C C C C L C C El Salvador C C C C C C C C C C C C C C C C C L C C Guatemala C C C C C C C C L C C C C C C C,L L C C Honduras C C C C C C C C C, L C C C C C L C, L L C C Panama C C C C C,I C C C C C C C C C C I,L C C Source: World Bank, 2006; Law of Municipalities, 1990. Note: C = Central, I = Intermediary, and L = Local. such as education and health. The central government iii. Features and Evolution of the Transfer System still manages the provision of education services through deconcentrated structures. Municipalities are involved in Reform processes to date have focused almost ex- early education and are responsible for the maintenance clusively on the design of the transfer system, with and rehabilitation of schools. Some degree of local par- less attention paid to the capacity of the municipali- ticipation now takes place through parent-teacher asso- ties to invest or to raise their own revenue. While the ciations, but schools maintain direct links to the Ministry intergovernmental fiscal system of Honduras is simple of Education rather than to municipalities. Health centers and transparent, compliance problems critically ham- operate at the local level, but with lines of accountability per effective local fiscal management. Delays, unpre- to central agencies.14 The central government is currently dictability, and shortfalls in transfers limit the ability of considering a move towards a deconcentrated model of municipal governments to deliver services and make services in rural areas through a memorandum of under- investments. This unreliability presents particularly se- standing between a service provider and the Ministry of rious problems for small municipalities that have little Health. In the first round, this model would serve 52 mu- resources of their own. nicipalities in the C and D categories. The transfer system allocates a high share of resources The distribution of responsibilities across levels of on a per unit basis and by population. Municipalities government in Honduras is very similar to that of benefit from having tax revenue that accrues to them other unitary countries in the region. Health and edu- directly, but large disparities exist in the economic base cation remain central government functions, as in most of these jurisdictions. The overall intergovernmental sys- countries of the region. The role of municipalities is large- tem does not fully factor in their ability to collect these ly focused on local services and small-scale infrastructure resources. Although other countries give block grants to provision (Table II3). local governments for basic operating expenses, these transfers are either adjusted to the size of the municipal- ity or comprise smaller shares of the total resource pool. 14  The Ministry of Health has: seven national hospitals, six re- For example, Guatemala allocates part of its transfers as gional hospitals, and 16 area hospitals with a total of 4,000 beds; 57 maternal child clinics; 380 health centers with doctors and dentists; a fixed amount per municipality, but the share is consid- 1,017 rural health centers; and 4 emergency periphery clinics. erably smaller and the formula gives greater weight to a number of other equalizing indicators. Beyond Latin extent that size and poverty incidence are correlated, the America, Ghana and Uganda have per unit subsidies, but transfer could also have an equalizing effect. Conversely, they are much smaller in relative terms. large municipalities benefit from the use of population size to distribute two-thirds of the resource pool and have Honduras Public Exphenditure Review 2008-2012 The current transfer system established by the 1990 the strongest economic and institutional capacity. Law of Municipalities has been amended twice. Ini- tially, the central government was obligated to trans- Municipalities also received revenue from local taxes. fer 5 percent of tax revenue to the municipalities (Table The main categories include: the Personal Income Tax, the II4). Authorities computed the transfer amount using a Industry, Commerce, and Service Tax, and the Extraction formula that distributed 40 percent of resources equally and Exploitation of Livestock Resources Tax. In the case and the remaining 60 percent in proportion to the num- of personal income and sales taxes, municipalities collect ber of inhabitants in each municipality. The law limited these taxes directly but share the base with the central the resources available for administrative or operational government. The size of the base for these main tax cate- expenditures to 15 percent of the transfers and assigned gories highly correlates with population density, meaning the remaining funds to maintenance and additions to the that small municipalities must rely on the block transfer as social infrastructure. their almost-exclusive source of revenue. The original 1990 formula introduced two sources of In 2005, a new allocation formula increased the block horizontal imbalance or disparities. A block grant that transfer to half the share of the resource pool and provided equal portions for all municipalities favored introduced poverty and fiscal effort as new criteria. small municipalities disproportionally. The per capita Decree 200/2005 increased the unitary transfer to 50 per- transfers received by a very small municipality could be cent of the shared resources, gave poverty a 20 percent 1,000 times greater than that of an urban center. Yet to the weight, and assigned fiscal efficacy a 10 percent weight. Table II.4: Reforms to Allocation Formula (Law of Municipalities’ Article 91) 1990 Law of Municipalities Decree 200/2005 Decree 143/2009 Amount 5 percent of central tax revenue 5 percent of central tax revenue 7 percent of central tax revenue and expenditure in 2010, 8 percent in 2011, 9 percent in 2012, 10 percent in 2013, and 11 percent in 2014 Allocation 40 percent of transfers allocated in 50 percent of transfers allocated in equal parts 50 percent of transfers allocated in equal parts Formula equal parts 30 percent according to share of population 20 percent according to share of population 60 percent according to share of 10 percent according to share of population 30 percent according to share of population population living under the poverty line living under the poverty line 10 percent according to level of fiscal efficacy Uses 10 percent earmarked for administra- 10 percent earmarked for administrative 15 percent earmarked for administrative tive expenditure expenditure expenditure 5 percent earmarked for operation 15 percent earmarked for operation and 13 percent earmarked for operation and mainte- and maintenance of social infrastruc- maintenance of social infrastructure (including nance of social infrastructure (including salaries ture (including salaries of teachers salaries of teachers and nurses) of teachers and nurses) and nurses) In municipalities with budgets below L500,000, In municipalities with budgets below L500,000, In municipalities with budgets below the two percentages above can be doubled the two percentages above can be doubled L500,000, the two percentages above The remainder is allocated to public investment 1 percent earmarked for child and youth protec- can be doubled (75 percent) tion programs and projects The remainder is allocated to public 2 percent earmarked for social development and investment (85 percent) gender-based violence prevention The remainder is allocated to public investment (69 percent) Special 1 percent earmarked to audits of the TSC and for 1 percent earmarked to audits of the TSC and for Provisions strengthening internal controls strengthening internal controls Source: Law of Municipalities and Vargas, 2011. By reducing the weight of the population share, the re- ria, the transfer system no longer rewards municipalities’ form aimed to correct some of the horizontal inequities efforts to raise their own revenue. from the previous formula. The revised Article 91 also ear- marked transfers to a number of specific uses but allowed The current transfer system favors small municipali- Honduras Public Exphenditure Review 2008-2012 municipalities with a budget smaller than L500,000 to ties and those that have a lower level of economic double both the share of the administrative costs and the and institutional development. When considered on amount invested in the operation and maintenance of so- a per capita basis, the fixed per unit transfer or aliquot cial infrastructure (which includes the salaries of teachers (alícuota) increases geometrically as the size of the mu- and nurses). Over 85 percent of municipalities had met nicipalities’ population decreases, as shown in Figure the revenue efficacy criteria by 2008, while the rest saw II1. The transfer amount per capita received in munici- their transfers cut by 10 percentage points. palities of 3,000–5,000 inhabitants is eight times larger than that of a municipality with more than 300,000 In 2009, the central government revised the formula people (Figure II2). The differences between large and again, reducing the weight of the share of population, small municipalities increase as population decreases, increasing the weight of the share of poverty, and reaching an 18:1 ratio between the least and the most eliminating the reward for fiscal efficacy. It also intro- populated groups. duced earmarks for child protection and gender-based vi- olence prevention programs. The new formula increased Transfers per capita are negatively associated with the equalization effect of the transfer by prioritizing the population and income, but there are wide disparities poverty share over population. However, by continuing to in the transfers received by localities of similar levels provide a fixed amount to each municipality regardless of of human development. Figure II3 and Figure II4 illus- wealth or size, it not only generated horizontal inequali- trate the variation. Equally, there is no clear relationship ties but even encouraged excessive spending among the between transfers and revenue mobilization capacity or small municipalities that do not require 15 to 30 percent the overall level of municipal development, which corre- in operating costs. By eliminating the fiscal efficacy crite- lates highly with income (Table II5). Transfer Formula Transfer = ( 0.50 * P ) + ( 0.20 * P ) + ( 0.30 * P ) 298 x/X Population x/X Population under the Poverty Line P is the pool of funds to be allocated to all municipalities. x/X represents the share of each factor that is present in a given municipality. Figure II.1: Aliquot per Municipality per Capita, 2009 Figure II.2: Total Transfers per Capita by Population Range, 2009 >300,000 >300,000 50,001-300,000 50,001-300,000 Population range 10,001-50,000 Population range 10,001-50,000 5,001-10,000 5,001-10,000 3,001-5,000 3,001-5,000 1,501-3,000 1,501-3,000 <1,500 <1,500 0 500 1000 1500 2000 2500 0 1000 2000 3000 4000 5000 Lempira Lempira Source: Vargas, 2011. Source: SINEMUN, 2012. Although transfers have some degree of conditional- palities. The policy initiative was not successful because ity, the central government lacks the capacity to effec- the decisions remained centralized and the municipalities tively monitor the allocation of funds to ensure their were not given the necessary resources to implement the proper use as prescribed by the Law of Municipalities. projects (Calix, 2012). Honduras Public Exphenditure Review 2008-2012 Without monitoring or enforcement mechanisms, the government cannot verify that municipalities are spend- Efforts and programs designed to support local capac- ing resources in the intended manner. Municipalities thus ity-building have had limited success. The most recent have greater autonomy to use resources at their own dis- is the Decentralization and Municipal Development Pro- cretion given that the enforcement of rules is unlikely. In gram (PRODDEL), which was established to support the addition to the general administrative weakness of mu- implementation of the PRS by consolidating and deepen- nicipalities, the central government has limited capacity ing the process of decentralization and local accountabili- to monitor and promote coordination across levels. ty. The program provided support to the central agencies working with municipalities to improve their capacity to iv. Recurrent Attempts at Deepening Decentralization deliver services and local financial management. Along with changing the transfer formula, succes- The inconsistent implementation of existing commit- sive governments have repeatedly attempted to ments due the limited fiscal space and shifting priorities strengthen the role of municipalities. Development of the central government remains a major obstacle to partners have been important supporters of these efforts. deepening the decentralization process. The 2011 es- Nonetheless, many of the pledges in support of decen- tablishment of a national security tax (Box 1) provides an tralization have not translated into actions or results. Fur- example of the perils involved in creating a law without a thermore, a multitude of management, administrative ca- careful implementation design. pacity, and accountability problems are still present both at the central and subnational level. B. Municipal Finance in Numbers Most attempts to delegate additional responsibili- ties and resource management to municipalities have The most relevant trends and fiscal indicators are pre- failed due to the resistance by central line agencies sented in this section. It is important to note that the and the weak local capacity to absorb functions. One analysis is limited due to lack of complete and reconciled example is the 2001 Poverty Reduction Strategy (PRS) data on subnational finances. Honduras has established adopted as part of the debt forgiveness process, under a National Municipal Information System (SINEMUN) that which the government delegated major responsibilities is maintained by the Ministry of Interior and Population for project and program implementation to the munici- and integrates the account statements from all 298 mu- Figure II.3: Transfers per Capita and Population, 2010 Figure II.4: Transfers and Human Development Index, 2010 5000 Mercedes de Oriente 5000 Mercedes de Oriente 4000 Humuya 4000 Humuya San Miguelito 3000 3000 San Juan 2000 2000 1000 Puerto Cortés 1000 Tegucigalpa San Pedro Sula 0 0 6 8 10 12 14 50 60 70 80 Population (In) Index of Human Development Source: INE and SINEMUN, 2012. Source: INE and SINEMUN, 2012. Box 1: The National Security Tax—Implementation Challenges and Its Effects on Municipalities In Honduras, local governments have designed and implemented innovative, although often extra- Honduras Public Exphenditure Review 2008-2012 legal, approaches to citizen protection. Due to the high social demand for improved security services, many municipalities assumed an expanded role in safety matters. Acting beyond their traditional jurisdic- tion and legal responsibilities, several mayors imposed municipal security taxes, which in some cases raised significant revenues for the municipalities. Some municipalities, such as Puerto Cortés, developed success- ful mechanisms to raise security taxes and manage them transparently through a trust fund monitored by local authorities, the private sector, and civil society organizations. In response, the central government designed a unified national system for security tax collection. In July 2011, Congress passed the Ley de Seguridad Poblacional, which defined the security tax rate and its application. The central government prohibited the collection of the municipal security taxes starting in January 2012 and, in return, promised to redistribute the resources collected from the implementation of a new national security tax. For a temporary period of five years, the security tax would apply to financial transactions and specific sectors, including mobile phones, food and beverages, mining, and casinos. The Executive Revenue Directorate (Dirección Ejecutiva de Ingresos, DEI) within SEFIN would assume responsi- bility for its collection and transfer the revenue to a trust fund that a committee of representatives from the government, business sector, and civil society would administer. The new revenue system for public security was implemented slowly. In January 2012, no regulation existed yet for the financial transactions tax. Derived from efforts to protect individuals and make firms bear the costs of the tax, the regulations were so complex that banks did not understand how to withhold the tax properly. With this uncertainty lasting well into mid-2012, revenues from the financial transactions tax slowed, while the central government failed to fill the gap caused by the municipalities’ loss of their own security tax revenue. Moreover, the new law required that the committee of representatives agree on the process for using the earmarked revenues. Although the Law established the committee, it has not yet reached an agreement on a common strategy. The revenues collected up until April 2013 (L1.3 billion) had not yet been utilized. Several other municipalities that had previously collected security taxes, committed revenue, and begun implementing programs have since lost their capacity to respond to public security needs. In response to the slow implementation of the new system, some municipalities have created mechanisms for obtaining resources under a new tax rate called a “governability rate” or through the private sector. Source: Authors’ summary of the National Security Tax Law. nicipalities. However, there are discrepancies with the in- Strategic Plan made decentralization a core goal, with the formation municipalities provide to the Superior Tribunal aim of giving municipalities 40 percent of the national of Accounts (Tribunal Supremo de Cuentas) and the data revenue by 2038. Figure II5 illustrates the goals and inter- on transfers that SEFIN compiles. The discrepancies are mediate milestones guiding the current efforts for greater highlighted when relevant. fiscal devolution. i. Transfer Shortfalls: A Source of Allocative Distortions Compliance with transfer targets, however, has been partial and inconsistent. These goals contrast starkly By 2013, mandated transfers in Honduras had dou- with the actual transfers.15 During the first decade that bled, relative to the beginning of the decade. Dur- the Law of Municipalities was in force (1990–2000), the ing the 2009 political crisis, the Executive issued Decree central government transferred the full 5 percent man- 200, which changed the transfer allocation formula and mandated their increase to 11 percent of central revenue 15  According to the Honduran Association of Municipalities (AM- HON), the government has a pending debt of L860 million with the by 2014. Subsequently, the 2010 Country Vision and the lowest tier of government. Figure II.5: Planned Transfer as Share of Central Figure II.6: Actual Transfer as Share of Central Revenue (%), 2002–2038 Revenue (%), 2002–2011 45 45 Country Decree Honduras Public Exphenditure Review 2008-2012 40 Vision 40 143/2009 35 35 30 30 25 25 Decree 20 143/2009 20 15 15 10 10 5 5 0 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2017 2022 2038 Source: Vargas, 2011; Law of Municipalities; and Strategic Plan, 2010. Source: SEFIN, 2012. dated allocation to local governments only once, in 1992. ii. Municipal Revenue: Untapped Sources Since 2002, transfers have slowly increased as a share of and Bright Spots central government revenue, but they still fall below the target. Only in 2006 and 2008 did transfers meet the tar- Municipal revenue as a share of government reve- get (Figure II6). nue declined from 2001 to 2009, although the trend reversed in 2010. On average, municipal revenue rep- Transfers fall short of budgeted amounts with varying resented less than 2 percent of GDP for the 2002–2010 rates of execution among municipalities. On average, period, about 2 percentage points below the ratio of mu- between 2008 and 2012 the ratio of actual transfers to the nicipal expenditure to GDP (Figure II9). The trend in the amount budgeted was 85 percent (Figure II7). Yet this av- data reported by the IMF Government Finance Statistics erage masks variation across units. In 2010, while most (GFS) is similar despite differences in 2010 (Figure II10). municipalities received about 70 percent of the budgeted transfer, approximately one-sixth of municipalities re- Municipal current revenue16 represented, on average, ceived 50 percent or less (Figure II8). 58 percent of total municipal revenue for the 2002– 2010 period. Despite the decrease in current revenue as Transfer shortfalls and delays distort allocation deci- a share of total revenues from 2005 to 2007, which plum- sions. Municipalities typically determine the shares of meted to 36 percent, it started to increase again in 2009. recurrent and capital spending on the basis of budgeted Current revenue takes a larger share of total revenues transfers and what they forecast to be their own-source now than it did at the beginning of the period. Whereas revenue. They tend to front-load recurrent expenditure in 2002 it accounted for 53 percent of the total, in 2010 it and delay capital investments until they receive transfers represented 58 percent. from the central government. Consequently, shortfalls in transfers are often translated into cuts to capital spend- Tax revenue as a share of municipal revenue appears ing. When there are transfer delays, it is common for to be declining. Municipalities do collect some local municipalities with few resources of their own to resort taxes,17 but, whereas in 2002 tax revenue represented to costly short-term commercial loans to finance the op- eration of basic services. Improving transfer predictability 16  Current revenue comprises tax revenue and other non-tax is an important prerequisite for promoting effective local revenue. Non-tax revenue refers to current revenue such as fines, management of resources. Before embarking on new de- surcharges, interest for late payments, and revenues recovered by administrative and judicial procedures. Capital revenue comprises centralization efforts, special attention should be given to loans, revenue from asset sales, transfers, subsidies, capital gains, increasing the capacity of municipal governments to op- and other capital income such as bonds, legacies, donations, and balance resources. erate within the existing framework and manage current 17  Funding for municipalities comes from a variety of sources that transfer levels. include both tax and non-tax revenue. These include taxes that can- Figure II.7: Budgeted Transfers versus Actuals, Figure II.8: Transfer Outturn (%), 2010 2009–2012 5000 100 100 Honduras Public Exphenditure Review 2008-2012 4000 80 80 million Lempira 3000 60 percentage 60 2000 40 1000 20 40 0 0 20 2008 2009 2010 2011 2012 0 50 100 150 200 Modi ed Budget Actuals Actuals as % of Modi ed Budget Frequency Source: SEFIN, 2013. Source: SEFIN, 2013. Figure II.9: Municipal Revenue as Share of Total Figure II.10: Local Revenue as Share of Total Government Revenue and GDP (%), 2000–2010 Government Revenue (%), 2003–2010 16 16 14 14 12 12 10 10 8 8 6 6 4 4 2 2 0 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2003 2004 2005 2006 2007 2008 2009 2010 Municipal revenue as share of total government revenue Local Revenue as Share of Total Government Revenue Municipal revenue as share of GDP Source: SINEMUN, 2012. Source: IMF GFS, 2013. Figure II.11: Municipal Revenue Total and Detailed Figure II.12: Municipal Revenue Total and General Composition (Millions of Lempiras), 2002–2010 Composition (Millions of Lempiras), 2003–2009 6000 6500 5000 5500 4000 4500 3000 3500 2500 2000 1500 1000 500 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 -500 Service Fees ICS Tax Property Tax Personal Tax 2003 2004 2005 2006 2007 2008 2009 Pecuniary Tax Extraction Tax Transfers Taxes Other revenue Grants Source: SINEMUN, 2013. Source: IMF GFS, 2013. Figure II.13: Tax Revenue per Capita and Population, Figure II.14: Tax Revenue per Capita and Transfer Share 2010 of Municipal Revenue, 2010 4000 San Buenaventura 4000 San Buenaventura Honduras Public Exphenditure Review 2008-2012 3000 Utila 3000 Utila Roatán Roatán 2000 2000 Santa Ana Santa Ana José Santos Guardiola San Pedro Sula José Santos Guardiola 1000 La Unión Las Vegas 1000 La Unión Tegucigalpa 0 0 6 8 10 12 14 0 20 40 60 80 100 Population (In) Transfer Share of Municipal Revenue Source: INE and SINEMUN, 2012. Source: INE and SINEMUN, 2012. 46 percent of total municipal revenue, in 2010 it only ac- Although tax revenue per capita is positively associ- counted for 29 percent. Although marked by fluctuations ated with population, some small municipalities are throughout the period, tax revenue has increased in ab- collecting substantial tax revenue, indicating that size solute terms. The decrease in the share of tax revenue on it is not the only determining factor. The more depen- total municipal revenue is therefore explained by a faster dent municipalities are on transfers, the smaller their own- increase in other sources of revenue, especially transfers. source revenue per capita is. Size and income per capita are closely associated. Both economic development and The composition of tax revenue remained relatively size affect the revenue base and the level of transfer that stable throughout the 2002–2010 period, despite a given municipality receives. Yet this relationship is not some fluctuations in 2007 and 2009. The fee on ser- deterministic, and positive outliers exist. Some munici- vices accounted for the largest share of municipal tax palities such as José Santos Guardiola, La Unión, Las Ve- revenue, followed by the Tax on Industry, Commerce, and gas, Roatán, San Buena Ventura, Santa Ana, and Utila have Services, contributing to total tax revenue an average of high rates of tax revenue per capita, despite their small 43 and 30 percent, respectively, for the 2002–2010 period. size and high transfer amount per capita. While discrepancies exist between the data reported by SINEMUN and that of the IMF GFS, the figures show a simi- The significance of tax revenue versus that of trans- lar trend (see Figure II-11 and Figure II12). fers varies significantly across different types of mu- nicipalities. Between 2002 and 2010, tax revenue aver- A wide disparity in revenue collection performance aged approximately half of the municipal revenue of the still exists. In 2010, the 23 municipalities classified as most developed group, while the share for municipalities category A collected 80 percent of all local revenue, with very poor capacity was below 10 percent. During while the remaining 20 percent was mobilized by the 275 the same period, the more developed municipalities only municipalities in the other three categories. Their per derived 16.4 percent of their revenue from transfers, while capita tax revenue was also more than seven times that in the least developed municipalities transfers accounted of stagnant or poor capacity municipalities. This trend is for 73 percent of total revenues. consistent with the main source of income from indus- tries that are likely to be concentrated in a small number iii. Municipal Expenditure: Giving Precedence of municipalities. to Recurrent Expenditure not be modified by municipalities (i.e. real estate tax, personal in- Since 2002, municipal expenditure has increased come tax, and business tax) as well as taxes for services and contribu- both in absolute and relative terms. Subnational ex- tions imposed at the discretion of the municipality. Non-tax revenue includes fines, surcharges, transfers from the central government, penditures have gone from L2.6 billion in 2002 to L11.2 and others. billion in 2010 (as per data from SINEMUN illustrated in Table II.5: Tax Revenue and Transfer Shares of Municipal Revenue by Categories of Municipalities (%), 2002–2010 Category 2002 2003 2004 2005 2006 2007 2008 2009 2010 Tax Revenue Share of Municipal Revenue (%) Honduras Public Exphenditure Review 2008-2012 A 59.6 52.4 53.0 52.0 52.5 42.8 48.7 44.5 37.4 B 36.7 38.5 36.1 34.4 42.9 34.8 36.0 33.0 29.0 C 29.4 23.5 23.9 17.3 22.6 17.8 19.2 18.4 16.7 D 13.8 8.4 11.1 11.8 10.2 6.0 8.2 7.9 7.4 Transfer Share of Municipal Revenue (%) A 17.6 12.2 15.3 16.0 14.9 18.7 21.9 18.3 12.8 B 46.0 33.9 34.1 41.1 27.3 31.2 35.7 37.2 26.9 C 60.0 57.1 58.5 69.0 55.0 55.4 60.6 58.3 50.2 D 73.8 69.9 72.0 79.0 73.5 69.2 78.3 71.7 69.7 Source: SINEMUN, 2012. Figure II19). The share of subnational expenditure also gone from representing 41 percent of municipal expendi- rose from 10 percent of total government expenditure in ture in 2003 to 27 percent in 2010, although in absolute 2002 (equivalent to 2 percent of GDP) to 14 percent (or 3.4 terms it increased by L1.2 billion. Conversely, the service percent of GDP) in 2011 (Figure II15). This average is com- of debt was 5 percentage points higher, going from 14 to parable to that of other unitary countries in the region 19.4 percent in 2010. Once again, there are discrepancies (World Bank, 2012). As illustrated in Figure II-15, the share between the SINEMUN and IMF GFS data (Figure II-19 and of municipal expenditure reached its highest level in 2007 Figure II20), but it is important to note that the latter does and has since declined. In per capita terms, municipal ex- not include capital expenditure. penditure has increased at a faster rate than central gov- ernment expenditure. However, discrepancies do exist An examination of expenditure composition sug- between the data reported by SINEMUN (Figure II-15) and gests that most municipalities are not complying the data that SEFIN reports to the IMF GFS (Figure II16). with the earmarks established for transfers in Article 91. They are required to invest 54 to 69 percent of the Expenditure per capita is inversely related to the pop- transfers they receive from the central government and ulation but has a positive association with transfers. can only spend 15 to 30 percent in administrative costs, Small municipalities have higher spending per capita and depending on the size of the municipality. There is an higher transfer amounts per capita (Figure II17, Figure inverse relationship between the ratio of recurrent ex- II18). This suggests that transfers may be driving expen- penditure to total municipal expenditure and the level diture increases. of transfer dependence (Figure II21). The data also sug- gest that municipalities are choosing to use the majority Recurrent expenditure accounts for approximately of their own resources to finance personnel and other half of municipal expenditure, with personnel outlay recurrent expenses (Figure II22). Notably, larger munici- as one of the main subcategories. In the 2002–2010 palities spend more per capita on recurrent expenditure period, recurrent expenditure represented, on average, than smaller ones. 45.6 percent of total municipal expenditure (Figure II-19). Wages accounted for an average 58 percent of recurrent iv. Municipal Borrowing: Covering Transfer Delays expenditure and 28 percent of total municipal expendi- ture for the same period. In absolute terms, personnel The Law of Municipalities allows municipalities to se- expenditure increased from L720 million in 2002 to 3.3 cure public loans, carry out financial operations with billion in 2010. national and foreign institutions, and issue bonds to finance investments. In each case, they must comply Capital investment expenditure has shown a decreas- with the legal requirements and adhere to an established ing trend since 2003. Expenditure on capital assets has borrowing limit of 20 percent of the municipality’s annual Figure II.15: Municipal Expenditure as a Share of Total Figure II.16: Local Expenditure Total and as a Share Government Expenditure and GDP (%), 2002–2011 of Total Government Expenditure (%), 2002–2010 20 20 Honduras Public Exphenditure Review 2008-2012 15 15 10 10 5 5 0 0 2002 2003 2004 2005 2006 2007* 2008* 2009* 2010* 2011* 2003 2004 2005 2006 2007 2008 2009 2010 Municipal Expenditure as Share of Total Government Expenditure Local Expenditure as Share of Total Government Expenditure Municipal Expenditure as Share of GNP Source: SINEMUN and SEFIN, 2012. Note: * Non-adjusted data. Source: IMF GFS, 2013. Figure II.17: Municipal Expenditure per Capita Figure II.18: Municipal Expenditure per Capita and Population, 2010 and Transfer Amount per Capita, 2010 6000 6000 Mercedes San Miguelito Puerto Cortés Puerto Cortés San Miguelito de Oriente Mercedes de Oriente Humuya Cabañas San Juan Humuya 4000 Cabañas 4000 San Buenaventura San Buenaventura Amapala Amapala Fraternidad Yauyupe Tauyupe Fraternidad 2000 2000 0 0 6 8 10 12 14 0 1000 2000 3000 4000 5000 Population (In) Transfer per Capita Source: SINIMUM, 2012. Source: SINIMUM, 2012. Figure II.19: Municipal Expenditure Total and Figure II.20: Municipal Expenditure Total and Composition, 2002–2010 (%) Composition, 2003–2009 (%) 100 12 100 10 80 10 80 8 Billions Lempira Billions Lempira 60 60 6 40 40 4 20 20 2 0 0 0 0 2002 2003 2004 2005 2006 2007* 2008* 2009* 2010* 2003 2004 2005 2006 2007 2008 2009 Recurrent Expenditure Interest Total Expenditure Compensation of employees Use of goods and services Expense Capital Goods Capital Transfers Personnel Expenditure Interest Other Expense Source: SINEMUN, 2012. Source: IMF GFS, 2012. Figure II.21: Recurrent Expenditure as a Share of Expenditure Figure II.22: Recurrent Expenditure per Capita and Transfers as a Share of Municipal Revenue, 2010 and Transfer Amount per Capita, 2010 100 6000 Mercedes Honduras Public Exphenditure Review 2008-2012 Puerto Cortés San Miguelito de Oriente 80 4000 Cabañas San Juan Humuya 60 Fraternidad San Buenaventura Amapala Tauyupe 40 2000 20 0 Earmark 0 0 20 40 60 80 100 0 1000 2000 3000 4000 5000 Transfer Share of Municipal Revenue Transfer per Capita Source: SINEMUN, 2012. Source: SINEMUN, 2012. ordinary revenue. Municipalities cannot acquire external revenue and 48 of own-source revenue. The limited expe- debt without SEFIN authorization and they must not ex- riences in issuing bonds have not been successful due to ceed the administration period. poor design and implementation. For example, the mu- nicipality of San Pedro Sula recorded the first municipal The law grants municipalities broad financial auton- bond issue in 1997, but, as a consequence of a series of omy, including the right to secure loans, but only a evaluation and implementation errors, the security led to few have the capacity to access credit and financing. a financial imbalance for the municipality (USAID, 2004). The practices of subsidized money, donations, and debt Table II6: Borrowing per Capita and Debt Service Share of payments from the central government have hampered Municipal Expenditure, 2002–2010 efforts to develop a system of municipal financing. Mu- nicipalities often obtain short-term loans from domestic High levels of indebtedness are a source of concern for commercial banks guaranteed with budgeted transfers a number of municipalities. In 2010, approximately 28 from the central government. The financing cost of such municipalities took new loans in excess of the 20 percent loans results in inefficient spending focused on recurrent of revenue ceiling (Figure II14) whereas 20 spent more expenditure and short-term needs rather than on medi- than one-fifth of their expenditure on debt service (Figure um- to long-term capital investments. II23). While information about cumulative levels of mu- nicipal debt was not available, indications suggest that Between 2002 and 2010, borrowing increased in ab- the levels are high for some municipalities. At least 25 solute terms for all categories of municipalities, but municipalities have far exceeded their repayment capac- decreased as a share of municipal revenue and ex- ity and exhibit dangerous levels of debt (Vargas, 2011). penditure. In per capita terms, both borrowing and debt service have increased for all categories of municipalities v. Local Public Investment: Increasing but Falling Short (Table II6). Borrowing represented 14 percent of munici- of Targets pal revenue in 2002, compared with 8 percent in 2010. Service of debt has decreased as a share of municipal ex- Municipalities are responsible for a sizable part of penditure and appears to have stabilized at 5 percent. public investment and additions to public fixed capi- tal, but many constraints for effective capital spend- Municipal borrowing is concentrated on municipali- ing remain. Public investment at the subnational level is ties with a greater capacity for savings. In 2010, the constrained by: (i) the low levels of management capacity debt service represented on average 8 percent of total in most municipalities; (ii) limited coordination in invest- municipal revenue. The municipalities in Category A (less ment projects across levels of governments and among than 8 percent) accounted for 81 percent of new loans neighboring municipalities; (iii) unpredictability and late- and their debt service was on average 17 percent of total ness of transfers; and (iv) limited access to credit. Table II.6: Borrowing per Capita and Debt Service Share of Municipal Expenditure, 2002–2010 Category 2002 2003 2004 2005 2006 2007 2008 2009 2010 New Borrowing per Capita Honduras Public Exphenditure Review 2008-2012 A 114.8 197.8 164.6 181.1 191.8 195.2 142.2 212.2 225.2 B 25.5 50.0 35.8 50.7 32.6 49.2 102.7 165.4 162.8 C 24.9 41.1 39.2 13.4 72.7 93.9 75.4 12.9 54.3 D 41.9 35.1 61.2 6.1 36.9 59.6 68.5 45.8 45.4 Debt Service per Capita as Share of Municipal Expenditure A 16.6 14.1 14.0 15.6 16.5 10.5 10.5 10.5 17.2 B 8.9 7.9 10.8 7.9 6.1 5.7 5.7 5.7 9.3 C 4.9 7.7 8.8 7.5 5.7 6.6 6.6 6.6 4.8 D 10.1 9.7 11.7 10.4 4.5 3.7 3.7 3.7 4.1 Source: Authors’ calculations based on data from SINEMUN and SEFIN. Figure II.23: Borrowing as Share of Municipal Current Figure II.24: Debt Service as Share of Municipal Revenue (%), 2010 Expenditure (%) , 2010 200 80 San Lorenzo San Lorenzo Puerto Cortés San Pedro 150 60 Sula Choluteca Jesus de Otoro Puerto Cortés Tela 100 40 Ocotepeque Morolica Nacaome La Trinidad Tomalá 50 Morolica El Triunfo 20 San Jorge Legal Limit 0 0 0 8 10 12 14 0 8 10 12 14 Population (In) Population (In) Source: SINEMUN and INE, 2012. Source: SINEMUN and INE, 2012. Fiscal constraints and volatile economic conditions (Figure II26).18 Compared to other Latin American coun- have negatively affected the overall public investment tries as well as LMI countries, Honduras ranks at the bot- level. Capital spending and public fixed capital forma- tom in terms of the Logistics Performance Index (LPI) for tion as a share of GDP have been relatively low and vari- Infrastructure, which measures the quality of trade- and able from year to year. The average share of fixed capital transport-related infrastructure.19 formation was 4.5 percent of GDP during the 2000–2010 period (Figure II25), a level considerably lower than that Against this backdrop, municipal fixed capital forma- of 1990s. This level is below the regional average for the tion has been steadily increasing, and it represents a same period of time—5.7 percent—and that of develop- growing share of total public capital formation. Be- ing and lower middle income (LMI) countries. tween 2003 and 2009, municipal capital formation qua- drupled (Figure II27). During the same period, additions The rate of public capital formation remains insuf- to fixed capital by municipalities went from representing ficient to cover the current infrastructure gap. While 17 percent of total public additions to 30 percent (Figure some infrastructure services such as access to water and II28). However, those additions are likely to be heavily con- sanitation have slowly improved, the overall quality of in- frastructure remains low relative to the regional average 18  WEF, 2012. 19  World Development Indicators, 2012. Figure II.25: Public Gross Fixed Capital Formation Figure II.26: Access to Improved Water and Sanitation as a Share of GDP (%), 2000–2010 Facilities (% of Total Population), 1990–2010 10 90 9 Honduras Public Exphenditure Review 2008-2012 8 80 7 6 5 70 4 3 60 2 1 0 50 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 1990 1995 2000 2005 2010 Improved sanitation facilities (% of population with access) LAC Honduras LMI Improved water source (% of population with access) Source: IMF WEO, 2013. Source: World Development Indicators, 2012. Figure II.27: Municipal Net Public Fixed Capital Formation Figure II.28: Municipal Net Public Fixed Capital and Transfer Assets (Millions of Lempiras), 2002–2010 Formation as a Share of Total (%), 2002–2010 6000 40 5000 35 4000 30 3000 25 2000 1000 20 0 15 2003 2004 2005 2006 2007 2008 2009 2002 2004 2006 2008 2010 Net Fixed Capital Formation Total Expenditure Source: IMF GFS, 2013. Source: IMF GFS, 2013. centrated in large urban centers. The horizontal distribu- outcomes are contingent on the overall level of institu- tion of fixed public capital formation cannot be examined tional capacity and the degree of accountability. Without through the available data. Furthermore, it is not possible these factors, decentralization can instead shift resourc- to determine if the additions to public fixed capital at the es from needed infrastructure investments to recurrent municipal level are the result of local investments or rather spending, as it appears to be happening in Honduras. assets produced by central agencies or development part- ners and then transferred to local governments. On average, small municipalities invest higher shares in capital assets than large municipalities with greater The transfer system has been designed with the aim resources of their own, but a large number of mu- of encouraging public investment and prioritizing nicipalities do not meet the investment targets. The maintenance of capital assets and social infrastruc- greater the ratio of transfer to municipal revenue is, the ture at the local level. Decentralizing public investment larger the share of expenditure in capital assets (Figure responsibilities was expected to increase its allocative II29). Municipality size seems to be inversely associated and production efficiency. By placing decision-making at with expenditure in capital assets (Figure II30). These pat- the government level with the most direct contact with terns suggest that larger municipalities are using their citizens, municipalities could obtain and better leverage own resources to finance social and recurrent spending. more accurate information about local needs. However, They may be also be benefitting from economies of scale. Figure II.29: Expenditure in Capital Assets as a Share Figure II.30: Expenditure in Capital Assets as a Share of Municipal Expenditure and Transfer Amount as a of Municipal Expenditure and Population, 2010 Share of Municipal Revenue, 2010 Earmark Honduras Public Exphenditure Review 2008-2012 80 80 60 60 40 40 20 20 0 0 0 20 40 40 60 100 6 8 10 12 14 Transfer as Share of Municipal Revenue Population (In) Source: SINEMUN and SEFIN, 2012. Source: SINEMUN and SEFIN, 2012. Figure II.31: Expenditure in Capital Assets per Capita Figure II.32: Expenditure in Capital Assets per Capita and Infrastructure Gap, 2010 and Population Density, 2010 4000 4000 Mercedes de Oriente Mercedes de Oriente Utila Utila 3000 Humuya San Juan 3000 Humuya San Buenaventura Fraternidad San Juan San Buenaventura Cabañas Fraternidad San Miguelito San Miguelito 2000 San Juan Guarita 2000 San Juan Guarita Amapala La Libertad La Libertad Amapala Yauyupe Yauyupe 1000 1000 0 0 10 20 30 40 50 0 200 400 600 800 1000 Percentage of Household without Access to Improved Sanitation (2001 Census) Population Density Source: SINEMUN, 2012. Source: SINEMUN, 2012. Small municipalities appear to be complying more closely As population density increases, expenditure in capi- with the earmarks established by the Law of Municipali- tal assets per capita decreases, highlighting cost dif- ties, though not uniformly. ferences across municipalities. Economies of scale favor large urban localities (Figure II32). Currently, the Expenditure in capital assets is positively associated transfer formula compensates indirectly for differences with the infrastructure gaps, which suggests some in size but does not fully consider population density or responsiveness to needs, although with considerable cost differences. However, the transfer formula may be variation across localities. Municipalities that have a encouraging the few small and wealthier municipalities lower percentage of households with access to improved to overinvest, as it appears to be the case in Utila. sanitation or running water are generally investing a greater amount of resources than those that are better endowed (Figure II31). This could be linked to municipali- C. Impact in Service Delivery: ties having better information about local needs and tar- The Case of Water and Sanitation get resources accordingly. One outlier is the municipality of Utila, which invests a significantly higher amount per Due to a lack of concise data at the municipal level, capita than other municipalities exhibiting similar levels decentralization results were analyzed only for the of infrastructure needs. water and sanitation sector. In the health and educa- Table II.7: Service Providers—Types and Population Served Service provider type Description # of service Population served % of total providers (thousands) population served Honduras Public Exphenditure Review 2008-2012 SANAA National water and sewerage utility (where service has not yet 1 20 1,550 20% been decentralized) Municipality (directly) A municipal department that provides services directly, usually 4,7 21 490 6% in municipalities with smaller urban populations Autonomous municipal Public utility with independent governance structure and budget 12 510 7% provider 22 predominantly in mid-size cities Private concession A private firm providing service within a specific geographical 1 720 9% area under government agreement, with the sole example exist- ing in San Pedro Sula Patronatos and urban A local development committee or community- based organiza- 300– 450 6% water boards tion operating in urban areas, sometimes with bulk water from 50,0 23 another provider, for example SANAA; prevalent in many cities Urban dwellers without No service from organized service provider (self-provision) N/A 200 3% organized water services Rural water boards A community-based organization operating in rural areas as the 5,000 24 2,900 38% quasi-exclusive model in rural areas Rural dwellers without No service from organized service provider (self-provision) N/A 770 10% organized water services Source: Authors’ calculations. tion sectors, community-driven models present better portant trends of employees per 1,000 connections by results than centralized service delivery (Box 2). However, different management model. While SANAA has more community-driven projects and decentralized municipal than 6 employees per 1,000 connections, the concession- service delivery are distinct models and should not be ary water utilities and water boards have just above 2 em- used interchangeably. ployees for the same number. The Ordinary Least Squares (OLS) analysis confirms that the trend is statistically sig- The methods for delivering water and sanitation ser- nificant within the sample. vices differ considerably among municipalities. In ru- ral areas, water boards mainly facilitate service provision. The independent service providers show better per- In urban areas, the municipality or dedicated autonomous formance in financial and operational efficiency. municipal providers deliver these services. The national Tariff effectiveness measured by the ratio of average water and sewerage utility (SANAA) is also an important tariff over average cost is significantly higher among service provider that serves over 1.5 million people in the water boards and municipal companies at 2.62 and capital city of Tegucigalpa. 1.38, respectively, compared to 1.23 and 1.21 by SANAA and the municipality, respectively. Service providers The more autonomous service providers are more ef- further removed from political decision-making (and ficient. Operating efficiency differs considerably among the corresponding possibility of receiving operating different management models. Figure 33 shows the im- subsidies) appear to have more liberty and less political pressure to keep tariffs below cost recovery level (Ta- 20  Currently serving 13 cities and 2 rural communities (based on “Ac- ble II8). Community-based water boards fare better in ueductos Operados por SANAA—propuesta tentativa de entrega”). service continuity than SANAA and other management 21  As registered by ERSAPS. The actual number is likely to be higher. 22  Even within this category, there are several different models, models. SANAA has the highest rate of water metering ranging from a decentralized unit within municipal administration of all the management models, but it is not statistically with budget and managerial autonomy, to a commercial company with (partial) municipal ownership. significant. Municipal direct service and water boards 23 Estimated. have very low disinfection rate compared to SANAA and 24  CONASA, 2011. Figure II.33: Employees per 1,000 Connections Based Figure II.34: Service Provider Performance on Type of Service Provider by Management Model 25 7 100 Increased Autonomy 2 Honduras Public Exphenditure Review 2008-2012 6 Number / % of models observed 80 3 5 4 60 7 18 3 6 11 40 2 20 1 4 0 1 0 SANAA Municipality Municipal Water Private SANAA Municipality Municipal Water Private Company Board Concessionaire Company Board Concessionaire Poor Medium Good Source: Authors’ calculations based on 2010 ERSAPS data. Source: Authors’ calculations based on 2010 ERSAPS data. Table II.8: Selected Characteristics of Distinct Service Providers Service provider Tariff effectiveness Water treated (%) Metering level Continuity Index SANAA 1.23 0.63 11% 1.4 Municipality 1.21 0.05 2% 1.8 Municipal Company 1.38 0.87 10% 1.6 Water Board 2.62 0.07 N/A 2.5 Source: Authors’ calculations based on 2010 ERSAPS data. municipal companies, perhaps due to the low level of However, decentralization alone is unlikely to solve all professionalization of services in these two models. the problems in the sector. Whether it is decentralized to autonomous service providers or local government, The overall performance of autonomous service pro- the decentralization process of the sector should couple viders such as municipal companies or water boards is appropriate management models and capacity-building better than SANAA or direct municipal service provi- with autonomy and incentive structures in order to reap sion. A performance index, based on indicators covering the potential benefits of sector reform. financial, operational, and service quality aspects, divides service providers into low, medium, and good perfor- Evidence from the health and education sectors simi- mance utilities. The results summarized in Figure II34 sug- larly point to beneficial impacts on performance from gest that municipal companies and water boards perform the more direct engagement of clients in the imple- better than SANAA and municipalities. mentation of programs. A number of implementation examples involving deconcentrated service delivery The results suggest that decentralization and commu- models and community-driven development interven- nity-based models have had some positive impacts on tions show sizable gains compared to purely centralized sector performance. Most of the autonomous munici- models (Box 2). These examples could inform reforms at pal companies, which were created more recently when the local level and create linkages between community- the decentralization process began, perform better than level interventions and municipal governments. SANAA. Community-based water boards appear to work quite well, providing a small portion of the clientele with decent quality of service while operating within their available financial resources. 25  The models are ranked by increasing level of autonomy from political influence. Box 2: Deconcentrated Service Delivery Models and Community-driven Development Interventions Communities have assumed service delivery through particular projects in a number of sectors such as education, transportation, and rural development. Many are financed by the Honduran Social In- vestment Fund (FHIS), a unit of the Presidency of the Republic that funnels investment towards municipal Honduras Public Exphenditure Review 2008-2012 projects. In the education sector, community-based schools present better results than traditional schools. The Honduran Program for Community Schools (PROHECO) is a model in which the responsibility for regu- lating and financing basic education and schooling as well as coverage in rural areas of the country remains centralized under the Secretariat of Education. However, the functions related with the contracting and administrative supervision of personnel are decentralized and transferred to the association of parents who act in a legal capacity. PROHECO was originally financed by the World Bank and today is integrated under the Secretariat of Education. PROHECO schools show evidence of greater efficiency compared to traditional schools and they have a greater impact on student achievement. Evidence shows that these schools: (i) have lower drop-out rates and grade failure instances; (ii) earn better test scores in Spanish, Mathematics, and Science; (iii) are open more frequently; (iv) have teachers that work more hours even though their salaries are lower than traditional schools; (v) use existing capacity more effectively; and (vi) have better school infrastructure and more resources than traditional schools. In addition, the government successfully devolved some of its functions to school councils. Each school council (called Asociación Educativa Comunitaria or AECOS) is responsible for hiring and firing teachers, managing funds, monitoring teacher performance and absentee- ism, and construction and maintenance of schools. PROHECO increased its coverage significantly over the last 13 years. It started with 648 teachers serving 16,501 students in 506 schools. In 2012, the program had 6,468 teachers serving 144,249 students in 3,100 schools. In the health sector, community-driven models improved service, but the system remains very cen- tralized. Decentralized primary health care (PHC) has increased its coverage and shows evidence of provid- ing better quality services than the Ministry of Health (Secretaría de Salud or SS) facilities. The innovative health coverage extension strategy was introduced in 2005 to implement community-based models in ar- eas that meet a set of socio-economic criteria, including a designated poverty level, access to health and education services, and certain health indicators. This program provides a basic package of health services oriented toward prevention, health promotion, and basic curative care with special attention to young chil- dren and reproductive-age women. By 2012, this service delivery model had increased to 28 management entities overseeing 269 facilities in 13 departments, which covered at least 850,000 people or 10.3 percent of the Honduran population. Evidence exists that these models provide better quality services than the SS facilities (Table 6). Users are also more likely to return to a decentralized facility. In terms of results, reduced incidences of maternal death have been reported in areas where a decentralized facility operates, such as the one managed by a consortium of municipalities of Copán Ruinas, Santa Rita, and Cabañas. Deconcentrated facilities not only performed better than SS facilities, they also had lower unit costs. A 2009 study found that, in general, total unit costs in decentralized facilities for 11 types of services (in- cluding family planning, prenatal care, pneumonia and acute respiratory infection treatment, and births) were lower than total unit costs in centralized/SS facilities. However, some variations exist depending on the type of service: (i) SS facilities tend to spend more on personnel even though they have fewer staff than decentralized units, partly because they pay higher salaries—doctors and auxiliary nurses working in SS facilities earn 25 percent and 19 percent more, respectively, than those in decentralized facilities; (ii) decen- tralized facilities pay 40 percent more for their basic medicine stocks than centralized ones because they purchase medicines at the local level to facilitate timely delivery without the benefits of bulk purchases and lower unit costs. However, the health administrative system is not harmonized. The 2010 public expenditure survey (PETS) noted that institutional and financial arrangements vary depending on the particular governance and management model of each facility. This situation contributes to diversity in reporting arrangements with lines of authority not always clear to facility managers. For instance, some public ambulatory facilities managers, including managers of rural health centers (CESARs) and maternal-infant clinics (CMIs), report to SS central offices, while most facilities report to SS departmental offices and some to municipal offices. The PETS also underscored that weak data recording prevented the accurate tracking of resource flows. No mechanism exists to systematically record and consolidate all sources of funding according to health facility. Table II.9 Characteristic Decentralized/Alternative Facility SS Facility Adherence to Protocols:* % of Facilities % of Facilities Honduras Public Exphenditure Review 2008-2012 Pneumonia Treatment 81.5 69.5 Diarrhea Treatment 69.4 37.8 Functioning Medical Equipment** 83 55 Clean Bathrooms** 50 0 Users Received Needed Medicines** 91.6% of Users Surveyed 68.1% of Users Surveyed Users Waited Less than 1 hour** 50% of Users Surveyed 39% of Users Surveyed Source: * SS, MEASURE Evaluation/PRODIM Consultores (2009); **García Prado and Peña (2010). D. Policy Recommendations be reflected in process design. Only a small set of munici- palities (approximately 23) have substantive administra- The essential preconditions for meaningful devo- tive capacity to assume greater responsibilities. lution include increased capacity at the municipal- ity level and stronger public financial management Vertical Balance: Potential accountability gains could practices at the central level. Although decentraliza- result from reducing dependence on transfers and pro- tion plays an important role in the public agenda, the viding local governments with additional opportunities central government should avoid accelerating its imple- to generate own-source revenue, especially in the richer mentation until it recovers control of its own finances and more densely populated municipalities. Two major and improves the flow of transfers, which is often ir- untapped potential revenue sources are the property tax regular or delayed. This unpredictability has had a det- and motor vehicle taxes. In highly urbanized municipali- rimental impact on municipalities. Under the current ties, particularly the A and some B categories, increased circumstances, deepening decentralization could raise taxes have the potential to both raise local budget re- expectations that cannot be met and lead to unintend- sources and improve resource allocation in the economy. ed consequences such as increased local borrowing. Subnational Revenue Effort: The government might The following policy issues and recommendations consider re-introducing an incentive for revenue mobi- are offered for consideration by the central govern- lization. Between 2002 and 2009, the collection of local ment: taxes increased substantially when the transfer formula rewarded municipalities for their revenue mobilization Focus on implementation: Intergovernmental and efforts. Additional incentives should come with capacity- subnational governance reform should focus on imple- building support, investments in improved cadasters, and menting existing rules and strengthening and stream- other tools to better assess the revenue potential of mu- lining existing institutions, rather than on passing a nicipalities. Using an asymmetric approach, additional new legal framework and additional rules. New reforms taxes could also be devolved to more developed subna- should not only be informed by evidence, but also un- tional entities. derpinned by sound assumptions about fiscal space and expected gains in terms of service delivery and ac- Equalization formula: An equalization formula seeks to countability. eliminate the gap between revenue and the cost of pro- viding a minimum level of services and local revenue. Asymmetrical decentralization: Subnational govern- The current formula considers poverty rates to assess ments are heterogeneous in resources and capacity and needs and uses population to assess the revenue base. may best be served by an asymmetric approach to de- However, half of the resource pool is allocated on a per- centralization. Heterogeneity across municipalities in municipality basis, which introduces disparities given the revenue base, income per capita, spending challenges, large heterogeneity of local governments. A more refined and administrative capacity are central issues that should formula should consider not only the level of tax effort but more accurately the economic potential and needs of various municipalities. While limitations still exist on the availability of data, enough information is currently col- lected on income per capita, human development, and Honduras Public Exphenditure Review 2008-2012 regional GDP that could complement the existing mea- sures included in the formula. Public Service Delivery: A rethinking of the expen- diture assignments and the public service delivery re- gime is necessary if the commitment to expenditure decentralization is to deepen. The structural fixes needed to reach this goal include removal of con- straints on local decision-making, greater clarity in ex- penditure assignments, and containment of fiscal dis- parities to acceptable levels. However, these reforms are conditional on strengthening the capacity of sub- national governments to upgrade the quality of their service delivery. E. Appendix Figure II.35: Number of Municipalities by Population Range, 2010 Honduras Public Exphenditure Review 2008-2012 >300,000 50,001-300,000 10,001-50,000 Population range 5,001-10,000 3001-5000 1,501-3,000 <1500 0 30 60 90 120 150 Number of municipalities Source: SINEMUN and INE, 2012. Table II.10: Correlation Coefficients Population Surface area Population under Transfer Amount Expenditure per Recurrent Expenditure (sq. km) the Poverty Line per Capita Capita per Capita Surface area (sq. km) 0.18*** 1.00 Population under the 0.66*** 0.20*** 1.00 Poverty Line Transfer Amount per -0.02 -0.01 0.17*** 1.00 Capita Expenditure per Capita -0.01 -0.03 0.00 0.01 1.00 Recurrent Expenditure 0.09*** 0.13*** 0.03 -0.04 0.04*** 1.00 per Capita Expenditure in Capital -0.22*** -0.15*** -0.14*** 0.05* -0.04 -0.80*** Assets per Capita *** Significant at the 0.01 level (2-tailed) ** Significant at the 0.05 level (2-tailed) * Significant at the 0.10 level (2-tailed) Table II.11: Multivariate Regression Analysis of Vertical Transfer Received (Fixed Effects) Dependent Variable Vertical Transfer Received Constant 3,608,753.0 Honduras Public Exphenditure Review 2008-2012 (359,451.4)*** Population 26.0 (5.9)*** Population under the Line of Poverty 104.8 (9.2) Population Density (per sq. km) -8458.1 (2540.1)*** N 1549 N Groups 184 R-sq 0.44 F= 3.51 Prob>F= 0 *** Significant at the 0.01 level (2-tailed) ** Significant at the 0.05 level (2-tailed) * Significant at the 0.10 level (2-tailed) Table II.12: Multivariate Regression Analysis of Municipal Expenditure (Fixed Effects) Dependent Variable Total Expenditure (Lempiras) Recurrent Expenditure (Lempiras) Expenditure Constant -14,800,000 -5,945,732 (4,011,222) (2,280,779)*** Vertical Transfer Received -1.1 -1.0 (0.3)*** (0.2)*** Tax Revenue -0.3 -0.2 (0.1)*** (0.0)*** Population 1,104.3 540.6 (64.4)*** (36.6)*** Population under the Line of Poverty 459.1 284.7 (100.3)*** (57.0)*** Population Density (per sq. km) 80,438.9 34,975.2 (26,766.2)*** (15,219.2)** N 1549 1549 N Groups 184 284 R-sq 0.73 0.69 F= 3.45 2.86 Prob>F= 0 0 Source: Authors’ calculations based on 2010 ERSAPS data. 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% Afghanistan Afghanistan Dominican Rep. Jamaica Congo, Rep. Lesotho Swaziland Congo, Rep. Azerbaijan Swaziland Lesotho Panama Panama Tunisia Mauritius Iran, Islamic Rep. Costa Rica Nicaragua Armenia Dominican Rep. Iran, Islamic Rep. Azerbaijan Jordan Costa Rica Nicaragua Trinidad and Tobago Paraguay Thailand El Salvador Mauritius Morocco Armenia Honduras Jordan Indonesia El Salvador Bosnia and Herzegovina Morocco Malasya Honduras Turkey Paraguay Chile Turkey Croatia Indonesia Source: World Bank Decentralization Indicators, 2012, based on GFS data. Source: World Bank Decentralization Indicators, 2012, based on GFS data. Macedonia, FYR Cape Verde Cape Verde Bosnia and Herzegovina Mongolia Chile Zimbabwe Croatia Thailand Macedonia, FYR Serbia Malasya Bulgaria Mongolia Slovak Rep. Serbia Ecuador Zimbabwe Slovenia Georgia Georgia Bulgaria Albania Slovenia Hungary Slovak Rep. Estonia Mexico Moldova Romania Romania Albania Czech Rep. Moldova Latvia Czech Rep. Lithuania Latvia Ukraine Peru Peru Hungary Colombia Lithuania Poland Estonia Belarus Ukraine Mexico Poland Korea Rep. Colombia South Africa Tajikistan Kazakhstan Belarus Figure II.37: Subnational Revenue Share of Total Public Revenue (%) in Developing Countries, 2010 Argentina Kazakhstan Brazil Russian Fed. Russian Fed. Korea Rep. India South Africa China India Figure II.36: Subnational Expenditure Share of Total Public Expenditures (%) in Developing Countries, 2010 Argentina Brazil Honduras Public Exphenditure Review 2008-2012 Map 1: Transfer Amount per Capita (Lempiras), 2010 Honduras Public Exphenditure Review 2008-2012 Map 2: Recurrent Expenditure per Capita (Lempiras), 2010 References Honduras Public Exphenditure Review 2008-2012 Acevedo, Javier, Sergio Mejía, and Alex Matamoros. 2004. López, Javier, and José Vicente Castro. 2007. Ajuste fiscal y Construyendo ciudadania: Los presupuestos locales, un sistema de transferencias incompleto. Mimeo. espacio que garantiza la efectividad de la participación Ocampo, José Antonio. 2004. Policies and Institutions ciudadana. Fundación Arias para la Paz y el Progreso Hu- for Local Economic Development: The Case of Honduras. mano, San José, Costa Rica. United Nations Biletska, Nataliya and Anand Rajaram. 2007. Public Expen- Parker, Andew, and Rodrigo Serrano. 2000. Promoting diture and Economic Growth: Insights from a Review of Good Local Governance through Social Funds and Decen- Evidence from Six Growth Success Stories. A Background tralization. Social Protection Discussion Paper 0022. World Paper for the Fiscal Policy for Growth and Human Develop- Bank, Washington, DC. ment Project. World Bank. Parker, Andrew. 2001. Promoting Good Governance with de Mello, Luiz. 2012. “Fiscal Decentralization and Public In- Social Funds and Decentralization. World Bank, Washing- vestment.” In: Brosio, Giorgio, and Juan Pablo Jiménez, eds. ton, DC. Decentralization and Reform in Latin America: Improving In- tergovernmental Relations, London: Edward Elgar. Sano, Yoshiko. 2009. Water Management Decentralization in Rural Honduras. Masters dissertation, Oregon State Uni- Di Gropello, Emanuela, and Jeffrey H. Marshall. 2011. “De- versity. centralization and Educational Performance: Evidence from the PROHECO Community School Program in Rural USAID. 2004. Trends in Decentralization, Municipal Honduras.” Education Economics 19 (0): 161-180. Strengthening, and Citizen Participation in Central Ameri- ca, 1995-2003. Report Number No. LAG-A-00-98-00060-00. Dickson, Eric. 2006. Management Models of Water and USAID, Washington, DC. Sanitation: Approaches to Decentralization in Honduras. Urban Poverty and Environment Report series 4. Interna- Vargas, César. 2011. Descentralización de Honduras: de tional Development Research Centre, Ottawa. un enfoque de reparto a un enfoque de desarrollo. Tegu- cigalpa: PROADES. Available at: http://www.seip-utd.hn/ Eaton, Kent and Larry Schroeder. 2010. “Measuring Decen- publicacion-libro-de-descentralizacion.html tralization,” in E. Connerley, K. Eaton, and P. Smoke, eds. Making Decentralization Work: Democracy, Development, Viñuela, Lorena. 2013. “Decentralized Public Investment: and Security, Boulder: Lynne Reinner Publishers (2010): Evidence and Trends,” in Martínez-Vázquez, Jorge and Jo- 167-190. nas Frank, Decentralization and Infrastructure, ed., Lon- don: Edward Elgar (forthcoming). Espitia Avilez, Gonzalo. 2001. El estado de la Descentral- ización Fiscal en Honduras. BID-GTZ. World Bank. 2006. Qualitative Decentralization Indicators. World Bank, Washington, DC. Espitia Avilez, Gonzalo. 2004. Decentralización Fiscal en Centroamérica, Estudio Comparativo. GTZ-CONFEDELCA, World Bank (2008). The Growth Report: Strategies for Sus- Tegucigalpa. tained Growth and Inclusive Development. Falleti, Tulia. 2010. Decentralization and Subnational Poli- World Bank. 2012. Decentralization Indicators. World Bank, tics in Latin America. Cambridge University Press. Washington, DC. Available at: http://www.worldbank.org/ publicfinance/decentralization Fiedler, J. L. and J. Suazo. 2002. “Ministry of Health user fees, equity and decentralization: lessons from Honduras.” World Bank, 2012. Identifying the Constraints to Budget Health Policy Plan 17 (4): 362-77. Execution in the Infrastructure Sector: DIPA Tracking Study. Gómez Sabaini, Juan C., and Maximiliano Geffern. 2006. World Economic Forum. 2012. Global Competitiveness Re- Honduras: el papel de los municipios en el combate a la port. pobreza. Series Medio Ambiente y Desarrollo No. 128. CE- PAL, Santiago de Chile. Annex Honduras Public Exphenditure Review 2008-2012 Figure A.1: Comparison with regional average (Latin America and Caribbean - lower bar) Voice and Accountability Political Stability / Absence of Violence Government E ectiveness Regulatory Quality Rule of Law Control of Corruption 0 25 50 75 100 Country´s Percentile Rank (0-100) Figure A.2: Comparison with income category average (Lower middle income - lower bar) Voice and Accountability Political Stability / Absence of Violence Government E ectiveness Regulatory Quality Rule of Law Control of Corruption 0 25 50 75 100 Country´s Percentile Rank (0-100) Source: Kaufmann D., A. Kraay, and M. Mastruzzi (2010), The Worldwide Governance Indicators: Methodology and Analytical Issues. Note: The Worldwide Governance Indicators (WGI) are a research dataset summarizing the views on the quality of governance provided by a large number of enterprise, citizen and expert survey respondents in industrial and developing countries. These data are gathered from a number of survey institutes, think tanks, no-governmental organizations, international organizations, and private sector rms. The WGI do not re ect the o cial views of the World Bank, its Executive Directors, or the countries they represent. The WGI are not use by the World Bank Group to allocate resources. Honduras at a glance Key Development Indicators Honduras Latin Lower America middle & Carib. income (2011) Population, mid-year (millions) 7,8 589 2.533 Surface area (thousand sq. km) 112 20.394 20.842 Population growth (%) 2,0 1,1 1,6 Urban population (% of total population) 52 79 39 GNI (Atlas method, US$ billions) 15,4 5.050 4.488 75-79 GNI per capita (Atlas method, US$) 1.980 8.574 1.772 60-64 GNI per capita (PPP, international $) 3.820 11.582 3.837 45-49 GDP growth (%) 3,6 4,7 5,5 30-34 GDP per capita growth (%) 1,6 3,6 3,9 15-19 (most recent estimate, 2005–2011) 0-4 Poverty headcount ratio at $1.25 a day (PPP, %) 18 6 30,2 8 6 4 2 0 2 4 6 8 percent of total population Poverty headcount ratio at $2.00 a day (PPP, %) 30 12 59,5 Life expectancy at birth (years) 73 74 66 60 Infant mortality (per 1,000 live births) 18 16 46 50 Child malnutrition (% of children under 5) 9 3 24 40 Adult literacy, male (% of ages 15 and older) 85 92 80 30 20 Adult literacy, female (% of ages 15 and older) 85 91 62 10 Gross primary enrollment, male (% of age group) 114 118 106 0 Gross primary enrollment, female (% of age group) 114 114 102 8 1990 1995 2000 2011 Honduras Latin America & the Caribbean Access to an improved water source (% of population) 87 94 87 Access to improved sanitation facilities (% of population) 77 79 47 Net Aid Flows 1980 1990 2000 2011 (US$ millions) 8 6 Net ODA and official aid 102 448 448 574 4 Top 3 donors (in 2010): 2 0 United States 19 215 110 103 -2 Spain 0 6 35 69 -4 -6 European Union Institutions 5 10 18 58 95 05 Aid (% of GNI) 4,2 16,0 6,4 3,9 GDP GDP per capita Aid per capita (US$) 28 92 72 76 Note: Figures in italics are for years other than those specified. .. indicates data are not available. Development Economics, Development Data Group (DECDG). Long-Term Economic Trends 1980 1990 2000 2011 Consumer prices (annual % change) .. 23,3 11,0 2,9 GDP implicit deflator (annual % change) 13,2 21,2 30,8 9,3 Honduras Public Exphenditure Review 2008-2012 Exchange rate (annual average, local per US$) 2,0 4,1 15,0 18,9 Terms of trade index (2000 = 100) .. 129 100 85 1980– 1990– 2000–11 90 2000 (average annual growth %) Population, mid-year (millions) 3,6 4,9 6,2 7,8 3,0 2,4 2,0 GDP (US$ millions) 2.566 3.049 7.106 17.427 2,7 3,2 4,4 (% of GDP) Agriculture 23,7 22,4 15,9 14,5 2,7 2,2 3,1 Industry 24,3 26,4 32,5 27,3 3,3 3,6 3,4 Manufacturing 15,0 16,3 22,7 19,2 3,7 4,0 3,8 Services 52,0 51,2 51,7 58,1 2,5 3,8 5,6 Household final consumption expenditure 69,4 66,8 70,8 77,7 2,6 3,0 4,3 General gov't final consumption expenditure 12,7 12,9 13,4 16,5 3,3 2,0 5,6 Gross capital formation 24,8 23,0 28,3 27,0 3,0 7,8 3,0 Exports of goods and services 37,2 37,2 54,0 47,6 1,1 1,6 3,7 Imports of goods and services 44,1 39,9 66,4 68,8 1,2 3,8 3,4 Gross savings .. .. 21,2 17,9 Note: Figures in italics are for years other than those specified. .. indicates data are not available. Development Economics, Development Data Group (DECDG). Balance of Payments and Trade 2000 2011 (US$ millions) Total merchandise exports (fob) 1.297 3.897 Voice and accountability Total merchandise imports (cif) 2.863 8.658 Political stability Net trade in goods and services -831 -3.606 Regulatory quality Current account balance -508 -1.503 Rule of law as a % of GDP -7,2 -8,6 Control of corruption Personal transfers and 0 25 50 75 100 compensation of employees (receipts) 484 2.811 2011 Country´s percentile rank (0-100) 2000 higher values imply better ratings Reserves, including gold 1.319 2.821 Central Government Finance Technology and Infrastructure 2000 2011 (% of GDP) Paved roads (% of total) 20,4 .. Current revenue (including grants) 15,1 23,2 Fixed line and mobile phone Honduras Public Exphenditure Review 2008-2012 Tax revenue 13,8 16,2 subscribers (per 100 people) 7 112 Current expenditure 14,8 21,0 High technology exports Overall surplus/deficit -5,2 -2,8 (% of manufactured exports) 0,4 1,3 Highest marginal tax rate (%) Individual .. .. Environment Corporate 25 30 Agricultural land (% of land area) 26 29 External Debt and Resource Flows Forest area (% of land area) 57,1 45,3 (US$ millions) Terrestrial protected areas (% of land area) 18,2 18,2 Total debt outstanding and disbursed 5.478 4.642 Freshwater resources per capita (cu. meters) 14.810 12.371 Total debt service 397 1.008 Freshwater withdrawal (% of internal re- 1,2 1,2 sources) Debt relief (HIPC, MDRI) 814 1.884 CO2 emissions per capita (mt) 0,81 1,0 Total debt (% of GDP) 77,1 26,6 GDP per unit of energy use Total debt service (% of exports) 9,9 12,2 (2005 PPP $ per kg of oil equivalent) 6,0 5,9 Foreign direct investment (net inflows) 382 1.043 Energy use per capita (kg of oil equivalent) 481 601 Portfolio equity (net inflows) 0 0 Short-term, 268 IBRD, 0 World Bank Group portfolio 2000 2011 IDA, 732 (US$ millions) Private, 1.061 IMF, 215 IBRD Total debt outstanding and disbursed 151 0 Disbursements 0 0 Bilateral, 766 Other Principal repayments 27 0 multilateral, 1.600 Interest payments 15 0 IDA Private Sector Development 2000 2011 Total debt outstanding and disbursed 838 732 Disbursements 38 135 Time required to start a business (days) – 14 Total debt service 8 7 Cost to start a business (% of GNI per capita) – 46,7 IFC (fiscal year) Time required to register property (days) – 23 Total disbursed and outstanding portfolio 42 192 of which IFC own account 27 192 Ranked as a major constraint to business 2000 2011 Disbursements for IFC own account 9 0 (% of managers surveyed who agreed) Portfolio sales, prepayments and Corruption .. .. repayments for IFC own account 26 19 Access to/cost of financing .. .. MIGA Gross exposure 16 6 Stock market capitalization (% of GDP) 8,8 .. New guarantees 0 0 Bank capital to asset ratio (%) 8,8 11,0 Millennium Development Goals Honduras With selected targets to achieve between 1990 and 2015 (estimate closest to date shown, +/- 2 years) Honduras Public Exphenditure Review 2008-2012 Goal 1: halve the rates for extreme poverty and malnutrition 1990 1995 2000 2011 Poverty headcount ratio at $1.25 a day (PPP, % of population) 46,9 20,6 18,0 17,9 Poverty headcount ratio at national poverty line (% of population) 74,8 65,8 70,0 66,2 Share of income or consumption to the poorest qunitile (%) 2,8 3,1 2,7 2,0 Prevalence of malnutrition (% of children under 5) 15,8 19,2 12,5 8,6 Goal 2: ensure that children are able to complete primary schooling Primary school enrollment (net, %) 89 90 89 97 Primary completion rate (% of relevant age group) 65 71 .. 101 Secondary school enrollment (gross, %) 33 33 .. 74 Youth literacy rate (% of people ages 15-24) .. .. 89 95 Goal 3: eliminate gender disparity in education and empower women Ratio of girls to boys in primary and secondary education (%) 104 .. .. 107 Women employed in the nonagricultural sector (% of nonagricultural employment) 41 43 42 42 Proportion of seats held by women in national parliament (%) 10 .. 9 20 Goal 4: reduce under-5 mortality by two-thirds Under-5 mortality rate (per 1,000) 55 44 35 21 Infant mortality rate (per 1,000 live births) 43 36 29 18 Measles immunization (proportion of one-year olds immunized, %) 90 89 98 99 Goal 5: reduce maternal mortality by three-fourths Maternal mortality ratio (modeled estimate, per 100,000 live births) 220 180 160 100 Births attended by skilled health staff (% of total) 47 55 56 67 Contraceptive prevalence (% of women ages 15-49) 47 50 62 65 Goal 6: halt and begin to reverse the spread of HIV/AIDS and other major diseases Prevalence of HIV (% of population ages 15-49) .. .. .. .. Incidence of tuberculosis (per 100,000 people) 125 125 116 43 Tuberculosis case detection rate (%, all forms) 60 71 89 96 Millennium Development Goals Honduras With selected targets to achieve between 1990 and 2015 (estimate closest to date shown, +/- 2 years) Goal 7: halve the proportion of people without sustainable access to basic needs Honduras Public Exphenditure Review 2008-2012 Access to an improved water source (% of population) 76 79 82 87 Access to improved sanitation facilities (% of population) 50 57 64 77 Forest area (% of total land area) 72,7 .. 57,1 45,3 Terrestrial protected areas (% of land area) 13,6 15,0 18,2 18,2 CO2 emissions (metric tons per capita) 0,5 0,7 0,8 1,0 GDP per unit of energy use (constant 2005 PPP $ per kg of oil equivalent) 5,5 5,5 6,0 5,9 Goal 8: develop a global partnership for development Telephone mainlines (per 100 people) 1,8 2,9 4,8 7,9 Mobile phone subscribers (per 100 people) 0,0 0,0 2,5 104,0 Internet users (per 100 people) 0,0 0,0 1,2 15,9 Households with a computer (%) .. .. 3,7 12,9 Note: Figures in italics are for years other than those specified. .. indicates data are not available. – indicates observation is not applicable. Development Economics, Development Data Group (DECDG). 125 100 160 140 100 75 120 75 100 50 80 50 60 25 40 25 20 0 0 0 2000 2005 2010 1990 1995 2000 2011 2000 2005 2010 Primary net enrollment ratio Honduras Latin America & the Caribbean Fixed + mobile subscribers Internet users Ratio of girls to boys in primary & secondary education Honduras Public Exphenditure Review 2008-2012 The World Bank 1818 H Street, NW, Washington, DC 20433, USA. www.worldbank.org