Colombia - Reforming the decentralization law : incentives for an effective delivery of services (Vol. 1 of 2) : Main report

Report No. 15298-CO Colombia Reforming the Decentralization Law: Incentives for an Effective Delivery of Services (In Two Volumes) Volume l: The Main Report April 29, 1996 Country Operations Ccuntry Department III Latin America and the Caribbean Region u .1) Document of the World Bank   Colombia: Decentralization TABLE OF CONTENTS ACRONYMS AND ABBREVIATIONS ............................................................ iv EXECUTIVE SUMMARY .............................................................v RESUMEN EJECUTIVO ............................................................ viii 1. THE MAIN STRUCTURAL PROBLEMS WITH THE CURRENT SYSTEM ...........1 2. THE GUIDING PRINCIPLES .............................................................5 3. THE DIFFERENT COMPONENTS ............................ .................................9 A. FINANCING EDUCATION SERVICES ...........................................................9 B. FINANCING HEALTH SERVICES ..................................... ........................ 12 C. FINANCING NFRASTRUCTURE ........................................... 14 D. GENERAL PURPOSE GRANTS ............................ ................................ 16 E. THF RIGHT ENVIRONMENT .............................................. 20 4. ADJUSTMENTS TO THE TRANSFER SYSTEM ................................................... 25 A.TTHE SITUADO FISCAL .......................................................... 25 B.THEPARTICIPACiONESMUNvICIPALES ................................................................. 29 C.OTHERTRANSFERS ........................................ ..................................... 34 TABLES TABLE 1: THE SYSTEM OF INTERGOVERNMENTAL TRANSFERS ......................2 TABLE 2: CRITERIA TO ADJUST THE TRANSFER SYSTEM ..................................6 BOXES BOX 1: INTERGOVERNMENTAL TRANSFERS ........................................................1 BOX 2: SOME SIMPLE RULES .............................................................5 BOX 3: DESIGNING TRANSFERS: AN EXAMPLE ..................................1................I 1 BOX 4: PROPERTY TAXES ............................................................ 22 ANNEXES I. DECENTRALIZATION IN THE EDUCATION SECTOR ......................................1 II. REFORMS IN THE HEALTH SECTOR ............................................................ 17 III. LOCAL RESOURCE MOBILIZATION ....................................... 22 IV. EARMARKING AND EXPENDITURE MANDATES . ......................................... 54 V. SIMULATIONS ............................................................. 58 BIBLIOGRAPHY ............................................................ 74 - ii-  Colombia: Decentralization The report was written by Ariel Fiszbein (task manager) and Richard Bird. Ernesto Cuadra and Donald Winkler contributed to the analysis of the education sector, Philip Musgrove and Patricio Marquez to the analysis of the health sector, and Thomas Tsiopou- los to the analysis of municipal and departmental finances. The Unidad de Desarrollo Territorial (UDT) in the DNP coordinated the important contributions made by several agencies in the Colombian Government including UDT itself, UDS-DNP, DAF-Ministry of Finance, Ministry of Education, Ministry of Health and the Superintendency of Health. The report benefited from comments by Jose Antonio Ocampo, Bill Dilinger, Ehtisham Ahmad, Paul Isenman, Ernesto May, Elsie Garfield, Juan Luis Londofno, Dan Morrow, and Norman Hicks. A draft of the report was discussed in a two-day workshop held in Bogota on February 9-10 with the participation of representatives from DNP, and the Ministries of Education, Finance and Health. The report incorporates comments received during that workshop. - iii -  Colombia: Decentralization ACRONYMS AND ABBREVIATIONS CONPES National Council for Economic and Social Policies (Consejo Na cional de Politica Econ6mica y Social CR Current Revenues DANE National Department of Statistics (Departamento Administrativo Nacional de Estadistica) DNP National Planning Department (Departamento Nacional de Planeacion) DRG Diagnosis-related Groups DRI Rural Development Fund (Desarrollo Rural Integrado) EPS Health Promoting Entities (Empresas Promotoras de Salud) ESS Health Solidarity Entities (Empresas Solidarias de Salud) FER Regional Educational Funds (Fondos Educativos Regionales) FIS Social Investment Fund (Fondo de Inversi6n Social) FIU Fund for Urban Infrastructure (Fondos para la Infraestructura Ur bana) FIV Fund for Road Infrastructure (Fondo de Inversi6n Vial) IGAC Agustin Codazzi Geographical Institute (Institutio Geografico Agustin Codazzi) [PS Health Providers (Instituciones Prestadoras de Servicios) JED Departmental Board for Teachers (Junta de Escalaf6n Departmental) JUDE Departmental Board for Education (Junta Departamental de Educa ci6n) JUMEs Municipal Board for Education (Junta Municipal de Educaci6n) MEN Ministry of Education (Ministerio de Educacion Nacional) MOH Ministry of Health (Ministerio de Salud) NBI Index of Basic Unsatisfied Needs (Indicador de Necesidades Basicas Insatisfechas) PAB Public health plan (Plan de Asistencia Basica) PEI Institutional Educational Project (Proyecto Institucional Educativo) POS Basic health package (Plan Obligatorio de Salud) SISBEN Targeting program (Sistema de Identificaci6n de Beneficiarios) SSS Regional Health Units (Servicios Seccionales de Salud) UDT Decentralizaton Unit in DNP (Unidad de Desarrollo Territorial) UNMATA Municipal unit for agricultural extension - iv -   Colombia: Decentralization EXECUTIVE SUMMARY 1. Decentralization in Colombia is well advanced, having begun over a decade ago. Nonetheless, the process has recently been the focus of renewed attention. The issues un- der discussion include the performance and capacity of municipios (an issue addressed in World Bank (1995a)), the extensive use of earmarking of intergovernmental transfers, and the fiscal and macroeconomic implications of the system of intergovernmental transfers. 2. Despite the importance of such issues, in many ways the basic question underlying the current debates concems the adequacy of existing institutional and financial arrangements to provide the incentives for an effective delivery of services in a decentralized environ- ment. This is the question discussed in this report. 3. Many possible reforms are being discussed with respect to transfers. Indeed, Congress is required to consider adjustments in Law 60 (which establishes the responsibilities of different levels of govermment in essential areas such as education, health, water and housing, and the rules goveming the system of intergovernmental transfers) in 1996. The challenge is to use this opportunity in a positive way in order to provide information and analyses that contribute to informed decisions by policy-makers. 4. Over the last eleven months, the World Bank has been collaborating with the National Planning Department (DNP) and other national agencies (including the Direcci6n de Apoyo Fiscal of the Ministry of Finance and the Ministries of Education and Health) in a revision of the institutional and legal framework ruling decentralization with the purpose of recommending adjustments. This report summarizes the findings emerging from that work and makes recommendations oriented to making decentralization work. 5. The report, however, does not constitute a formal proposal for the reform of Law 60, let alone of the broader set of intergovernmental fiscal issues currently under study in Co- lombia. Rather, it should be seen as an input to the discussion process among all inter- ested parties in Colombia. The recommendations made in the report attempt to avoid, to the extent possible, conflicts with constitutional rules and minimize necessary legal changes. For example, the report takes as given the size of the transfers (as established by the Constitution and Law 60). The underlying logic of the report is to consider changes in the implementation, rather than the principles, of the existing framework for decentraliza- tion in order to make it more conducive to an effective delivery of services by local gov- ernments. The question of the fiscal sustainability of the system, especially of the increases mandated by the Constitution, is not discussed in this report. The recommendations in the report regarding the structure of the transfer system should be valid, within certain limits, regardless of the size of the transfers. 6. This document analyzes the problems of the system of intergovernmental transfers by, first identifying objectives, next the instruments that are being used, and finally evaluating the extent to which there is an adequate assignment of instruments to objectives. 7. One basic objective, indeed perhaps the principal one, of the transfer system derived from the National Constitution and Law 60 is to guarantee the provision of minimum service levels of education and health to the population. Given the important externalities - v -  Colombia: Decentralization associated with those services and their special status, the national level has in effect as- sumed the responsibility of financing the provision of those services. The main guiding principle that we propose should be followed is that money should flow to the entities (even if they are not part of the public sector) that are actually responsible for the provi- sion of these minimum service levels. This leads us to recommend that the cost associated with the provision of minimum service levels in education and health should be financed through a specific purpose transfer calculated on the basis of costs and number of benefi- ciaries. Given the institutional setting that prevails in these two sectors, more than one transfer will be needed for each sector. The main difference from the existing system is the establishment of a direct link between the guaranteed service and its financing that is not a function of complex formulae that attempt to measure poverty, fiscal capacity/effort, or other variables that do not represent the cost of providing the service. 8. The Colombian system assumes the national level will partially finance capital expendi- tures whenever it is necessary to expand coverage of critical services such as education, health or potable water, or in areas where there are large externalities as in the case of sewerage or roads. Other types of investment projects can be financed entirely by the beneficiaries either through taxes and betterment levies, and through borrowing in the credit markets. In a decentralized system, in principle, subnational governments identify infrastructure needs and execute projects. The national level only provides incentives to orient resource towards projects with externalities or those important for national goals. This leads us to recommend that national financing of lumpy capital investment projects undertaken by departments and municipios be done through matching grants and not through an automatic transfer. 9. The system of intergovernmental transfers would be incomplete without some type of general purpose grant which replaces the confusing use of redistributive measures in the Situado and Participaciones. Such grants enable all subnational governments to operate at some minimum level and provide other services besides health and education. At the same time they partially compensate for the large differences in fiscal capacity that exist among municipios, and hence place all municipios on a more level 'playing field' with re- spect to the equal (per service unit) capitation grant proposed for basic health and educa- tion. The key aspects of these grants involve the fact that the use of funds thus provided would be at the discretion of the municipio (and thus would exclude any type of earmark- ing) and, especially, that they would be distributed according to needs and fiscal capacity. It should be emphasized that if this grant should become too small the viability of the en- tire system may become suspect. 10. The natural way to implement the revised system would be to have three separate transfers. The amounts to be allocated to the education and health transfers would be de- termined based on the cost of providing the services, and not as a given percentage of cur- rent revenues. The amount allocated to the general purpose grant would be calculated either so as to level the fiscal capacity of all municipios to a given level, or as a percentage of GDP. However, the report argues that the main elements of the revised system can be accommodated within the existing constitutional framework which considers one integral transfer calculated as a percentage of current revenues. - vi -  Colombia: Decentralization 11. Adjustments are required in addition to those recommended for the system of inter- governmental transfers. Changes are needed to give subnational governments more flexibility to use resources, eliminating all earmarking (and mandates) that do not respond to very clearly established needs. To a great extent, the difficulties of implementing Law 60 and decentralization in general are associated with the prevailing confusion regarding the responsibilities of different entities. This has created some important institutional bot- tlenecks that must be removed in order to improve the incentives for an effective delivery of services. While this matter is not the focus of the report, analyses and recommenda- tions are included in the annexes to the main text. 12. Section 1 of the report identifies the main problems with the current system of inter- governmental transfers. Section 2 presents what we believe should be the guiding princi- ples for a reformed system. Section 3 discusses in depth the principal components of the system. Finally, Section 4 discusses some of the critical options and decisions faced by reformers in their effort to redesign the main transfers. Several annexes analyze specific issues in more detail and serve as background to the proposals and recommendations made in the main text. - vii -   Colombia: Decentralization RESUMEN EJECUTIVO 1. El proceso de descentralizaci6n en Colombia ha avanzado mucho desde que se inici6 hace mis de diez afios. Con todo, en los uiltimos tiempos ha vuelto a ser objeto de espe- cial atenci6n. Las cuestiones que se estan debatiendo son el desempefilo y la capacidad de los municipios (tema abordado en un trabajo del Banco Mundial (1995a)), el amplio uso del destino especifico para las transferencias intergubemamentales y las repercusiones fis- cales y macroecon6micas del sistema de transferencias. 2. A pesar de la importancia de dichos temas, en gran medida el interrogante basico de los debates que se estan sosteniendo en la actualidad se refiere a la pertinencia de los sis- temas institucionales y financieros existentes para proporcionar incentivos para la presta- ci6n eficaz de servicios en un medio descentralizado. Este es el tema de este informe. 3. En la actualidad se estan estudiando diversas posibles reformas en el campo de la des- centralizaci6n. En 1996 el Congreso debe considerar la modificaci6n de la Ley 60 (en la que se establecen las responsabilidades de los distintos niveles de gobierno en esferas esenciales como educaci6n, salud, abastecimiento de agua y vivienda, y las normas por las que se rige el sistema de transferencias intergubernamentales). El desafio que se encara consiste en aprovechar esta oportunidad de manera positiva a fin de proporcionar infor- maci6n y analisis que ayuden a los responsables de las politicas a adoptar decisiones bien fundamentadas. 4. A lo largo de los uiltimos once meses, y en respuesta a un pedido por parte del Gobier- no Nacional, el Banco Mundial ha estado colaborando con el Departamento Nacional de Planificaci6n (DNP) y otros organismos nacionales (incluida la Direcci6n de Apoyo Fiscal del Ministerio de Hacienda y Credito Puiblico, y los Ministerios de Educaci6n y Salud) en la revisi6n del marco institucional y juridico relativo a la descentralizaci6n con el prop6sito de recomendar ajustes. En este informe se resumen las conclusiones que surgen de esa labor y se formulan recomendaciones orientadas a lograr que la descentralizaci6n di bue- nos resultados. 5. Empero, el informe no es una propuesta formal de reforma de la Ley 60, mucho menos del amplio espectro de aspectos relacionados a las finanzas intergubernamentales que se estan discutiendo en Colombia. Debe considerarse mis bien como un aporte al proceso de estudio por las partes interesadas que esta teniendo lugar en Colombia. Las recomenda- ciones formuladas en el informe tratan, en la medida de lo posible, de evitar conflictos con las normas constitucionales y reducir al minimo los cambios juridicos necesarios. Por ejemplo, el informe toma como dada la magnitud de las transferencias (establecidas en la Constituci6n y la Ley 60). La 16gica en que se basa es contemplar cambios en la aplica- ci6n. mas que en los principios. del marco existente para la descentralizaci6n a fin de que sea mas propicio para que los gobiernos locales presten servicios eficaces. El tema de la sustentabilidad fiscal del sistema, en particular del aumento en transferencias establecido por la Constituci6n, no es tratado en el informe. Las recomendaciones con respecto a la estructura del sistema de transferencias que se hacen en el informe son validas, dentro de ciertos limites, independientemente del nivel de las transferencias - viii -  Colombia: Decentralization 6. En este documento se analizan los problemas del sistema de transferencias interguber- namentales identificando primero los objetivos y despues los instrumentos que se estAn usando y, por ultimo, evaluando la medida en que se han asignado los instrumentos ade- cuados para la consecuci6n de los objetivos. 7. La meta basica, y tal vez la principal, del sistema de transferencia derivado de la Constituci6n Nacional y la Ley 60 es garantizar la prestaci6n de niveles minimos de servi- cios de educaci6n y salud a la poblaci6n. Habida cuenta de las externalidades importantes vinculadas a esos servicios y su condici6n especial, el nivel nacional ha de hecho asumido la responsabilidad de financiar su prestaci6n. El principio rector que proponemos que se siga es encauzar dinero hacia las entidades (aunque no sean del sector piublico) que sean efectivamente responsables de la prestaci6n de estos niveles minimos de servicios. Esto nos lleva a recomendar que el costo vinculado con la prestaci6n de niveles minimos de servicios de educaci6n y salud debe financiarse mediante una transferencia para fines es- pecificos calculada sobre la base de los costos y el numero de beneficiarios. Teniendo en cuenta el marco institucional que prevalece en estos dos sectores, se necesitara mas de una transferencia para cada sector. La principal diferencia con el sistema vigente es la creaci6n de un vinculo directo entre el servicio garantizado y su financiamiento, que no sea fuinci6n de f6rmulas complicadas para tratar de medir la pobreza, la capacidad y el esfuerzo fiscal u otras variables que no representan el costo de prestaci6n de los servicios. 8. El sistema colombiano supone que el nivel nacional de gobierno financiarA parcialmen- te los gastos de capital cuando sea necesario ampliar la cobertura de servicios criticos de, por ejemplo, educaci6n, salud y abastecimiento de agua potable, o en esferas de grandes externalidades, como el alcantarillado o las carreteras. Otros tipos de proyectos de inver- si6n pueden ser financiados integramente por los beneficiarios, ya sea mediante impuestos de valorizaci6n y otros, o la obtenci6n de fondos en los mercados de credito. En princi- pio, en un sistema descentralizado, los gobiernos subnacionales determinan cuales son las necesidades de infraestructura y ejecutan los proyectos. El nivel nacional de gobierno s6lo proporciona incentivos encaminados a orientar los recursos hacia proyectos con externali- dades o que revistan importancia para objetivos nacionales. Esto nos lleva a recomendar que el financiamiento de proyectos de inversiones indivisibles de capital que sean llevados a cabo por los departamentos y municipios se efectue mediante donaciones de contraparti- da y no de una transferencia automatica. 9. El sistema de transferencias intergubernamentales quedaria incompleto sin algun tipo de transferencia de prop6sitos generales en reemplazo de la confusa aplicaci6n de las me- didas de redistribuci6n que hoy se incluyep. en el Situado Fiscal y las Participaciones Mu- nicipales. Dichas transferencias permiten a todos los gobiernos subnacionales funcionar a un nivel minimo y prestar otros servicios ademas de los de salud y educaci6n. Al mismo tiempo compensan parcialmente las grandes diferencias de capacidad fiscal que existen entre los municipios, y ayudan a ubicar a todos los municipios en mejores condiciones con respecto a las transferencias propuestas para educaci6n y salud. Los aspectos fundamenta- les de estas transferencias son el hecho de que el uso de los fondos proporcionados de esa manera quedaria librado a los municipios (por lo que excluiria cualquier tipo de asignaci6n especifica) y, concretamente, dichos recursos se distribuirian de acuerdo con las necesida- - Ix -  Colombia: Decentralization des y la capacidad fiscal. Debe enfatizarse que si el volumen de esta transferencia fuese demasiado pequenlo, la viabilidad del sistema como un todo estaria en duda. 10. La manera natural de implementar el sistema recomendado seria a traves de tres transferencias automaticas separadas. Los montos asignados a las transferencias para educaci6n y salud serian determinadas en base al costo de proveer dichos servicios y no como un porcentaje de los ingresos corrientes. El monto de la transferencia de prop6sito general seria calculada de modo tal de igualar la capacidad de todos los municipios hasta cierto nivel, o como un porcentaje del PIB. Sin embargo, el informe argumenta que los principales elementos del sistema propuesto pueden ser tambien implementados dentro el esquema constitucional vigente que considera una transferencia integral calculada como porcentaje de los ingresos corrientes. 11. Es preciso hacer otros ajustes ademas de los recomendados en el sistema de transfe- rencias. Es preciso introducir cambios que doten a los gobiemos subnacionales de mas flexibilidad para el uso de los recursos, eliminando las asignaciones especificas (y los man- datos) que no se ajusten a necesidades establecidas con gran precisi6n. En gran medida, las dificultades para aplicar la Ley 60 y poner en practica la descentralizaci6n en general se vinculan con la confusi6n vigente con respecto a las responsabilidades de las distintas enti- dades. Esto ha creado cuellos de botella institucionales importantes que deben eliminarse a fin de mejorar los incentivos para la prestaci6n eficaz de servicios. Si bien esta cuesti6n no se trata especialmente en este informe, en los anexos se incluyen analisis y recomenda- ciones al respecto. 12. En la Secci6n I del informe se identifican los principales problemas del actual sistema de transferencias intergubemamentales. En la Seccion 2 se presenta lo que consideramos que deberian ser los principios rectores del nuevo sistema tras la reforma. En la Secci6n 3 se analizan a fondo los principales componentes del sistema. Por uiltimo, en la Secci6n 4 se tratan algunas de las opciones y decisiones criticas que encaran los reformadores en su labor de reformulaci6n de las principales transferencias. En varios anexos se analizan problemas especificos de manera mas detallada; en ellos se apoyan las propuestas y reco- mendaciones contenidas en el cuerpo principal del informe.   Colombia: Decentralization 1. THE MAIN STRUCTURAL PROBLEMS WITH THE CURRENT SYSTEM 1.1 The system of intergovernmental transfers in Colombia has three main compo- nents: the Situado Fiscal, the Participaciones AMunicipales, and the National Cofinancing System (SNC). The first two components are recognized in the National Constitution and regulated by Law 60/93, while the SNC was created by Decree 2132/92 (see Box 1). Questions have been raised about the fiscal sustainability of the system, especially of the increases mandated by the Constitution. This topic, which in itself could be the subject of a complete study, is not discussed in this report, which focuses on issues related to the structure of the system of intergovernmental transfers. 1.2 The design and functioning of this system is of critical importance for the eco- nomic and social development of the country. First, it influences the effectiveness of the provision of those services that have been decentralized. The transfers generate incentives that can be more or less conducive to the maintenance and expansion of coverage and quality of such services as education, health or water. Second, it could have an important impact on the efficiency of public expenditures both from the point of view of allocative and production efficiency. 1.3 One key problem with the current system is that it focuses too much on the en- l I: Iirv tases tities receiving the transfer at the expense of The-.:. S ::: Fis:l c- ttt a- - erc- the services or activities these transfers are oc.urrnt.evnue(R)..a...re .t.. supposed to finance. T hus, in many cases the paments- ad distts to basic purpose that justifies a transfer of re- pr.im an -sed eui :i sources (i.e. a subsidy to pay for health serv- he.t-L 60- stablisesi ts, size. (4 o ices of poor families) remains in the back- GDP or.24.%ofC.: in19. 9 a the......f.or&- ground with respect to the discussion regard- ts by .i.t it i. .it . bt....... ing which level of government receives the ...P.....Mni.p..so... transfer. ..p n of C. trn. .to -u.k.W- Wznd -.rn. gr s kdier.. to. F. .. .. . . i 1.4 This document analyzes the problems .....en Th.e C