Lessons from International Experience on Fiscal Decentralization for Regional Governments Volume II: Overview and Case Studies: Colombia, Indonesia, Poland and Spain prepared under the World Bank Partnership RAS for Chile1 1The Overview Chapter was written by Ewa Korczyc and Natalia Campora under the supervision of Jorge Martinez-Vazquez. Case studies were authored by country experts: Colombia - Juan Gonzalo Zapata; Indonesia - Gabriele (Gabe) Ferrazzi; Poland- Pawel Swianiewicz; and Spain - Santiago Lago-Peñas. The work benefitted from comments by Pedro L. Rodriguez. 1 Foreword: The World Bank has been requested to accompany the Chilean government in conducting a comprehensive analysis of international experience with fiscal decentralization processes for regional level of subnational governments. Such analysis will provide lessons and recommendations that can assist the Ministry of Finance in designing and implementing the fiscal decentralization framework for regional governments in Chile. This work is aligned with the Government’s decentralization strategy which places the regionalization process, in particular democratically elected self-governing regions with resources and responsibilities, at the center of the reform effort. This work is very timely, as Chile - confronted with social unrest towards the end of 2019 - will hold a referendum regarding the current constitution and is embarking on a national debate to redefine the fundamental principles of state organization. The study is comprised of two volumes. Volume I provides a systematic review of the relevant international experience in fiscal decentralization for the mid-level of government. The work focuses primarily on the aspects of fiscal decentralization from the fiscal discretion angle, specifically expenditure responsibilities, revenue assignments, transfers and subnational borrowing. This work is supplemented with a brief overview of selected aspects of fiscal accountability related to the budget process, audit and control; as well as transparency and citizen engagement. Volume II includes in-depth case studies with descriptions and analyses of decentralization processes in four selected countries: Colombia, Indonesia, Poland and Spain. This volume is prefaced by an overview that summarizes the main issues and draws relevant lessons from the experience of these four countries. The case studies for the report were selected jointly with the Government of Chile. The following criteria were applied in selecting the case study countries: • “unitary constitution” (rather than a federal structure); • income per capita similar to that of Chile; • decentralization experiences and insights are relevant and interesting in terms of both achievements and failures; and • existence of relevant “in-house” experience at the World Bank. All case studies follow a similar structure – they provide a brief history of the decentralization process in the respective country and describe major adjustments to the decentralization model. Next, they provide a detailed analysis of the expenditure assignments and sources of funding (revenue assignments) for the intermediate level of subnational governments. Each report also includes a discussion about subnational borrowing and fiscal rules. A separate section is dedicated to issues related to accountability, budget process and control mechanism. Each report ends with a summary of the key lessons and their applicability to Chile. 2 Overview Decentralization process and its drivers The case studies show that the process of decentralization is closely linked to specific circumstances of the individual countries. The decentralization process in Indonesia is known in the literature as “big bang”, as most of the decentralization reforms were completed within a short period of two years. In 1998, the effects of the Asian Financial Crisis and the ensuing political turmoil provided Indonesia with the fertile ground for decentralization reforms, which were generally driven by long-standing grievances regarding limited democracy, concentration of wealth and political centralization. Without the benefit of a longer period of preparation, Indonesia enacted several laws and the central government suddenly transferred political authorities, responsibilities and financial resources to the districts and villages, but much less so to the intermediate level, the provinces. In contrast, the process of decentralization in Colombia has been evolving for decades. The early decentralization efforts in Colombia date back to the mid-19th century but it was only in the 1980s that Colombia finally achieved fiscal, political and administrative decentralization by assigning fiscal resources to the municipalities, allowing the election of mayors, and by transferring functions from the central government to the subnational governments. For Colombia, in contrast to the case of Indonesia, the decentralization process was gradual and comprehensive, as the policies were oriented simultaneously to decentralize functions, resources and political responsibilities. Similarly, the drivers behind the decentralization process varied considerably among the four countries. In the case of Poland, the process of fiscal decentralization was part of a broader reform package, as the country moved from a central planning system to a market economy. The prospect of the European Union accession and the important role of subnational governments in other European Union countries were important factors supporting decentralization reforms in Poland. In fact, Poland -as well as Colombia- has had a long history of decentralization, but it was not until the 1990s, when comprehensive decentralization reforms were launched. Political events also prompted decentralization in Spain, where a phase of transition towards democracy coincided with death of General Franco in 1975. By 1977, the first Autonomous Communities (Regional Governments) were established, and the decentralization process was officially launched. The new Constitution of 1978 incorporated three governmental levels: Central Administration; the regions called Autonomous Communities; and the Local Corporations, which include the municipalities and the provincial councils. Expenditure assignments The experiences of case study countries show that most sectoral competencies are overlapping, although central government tends to play a leading role in regulatory framework, while implementation is devolved to subnational governments. For example, in Colombia, the 1991 Constitution granted the central government responsibilities related to the design of sectoral policies, while their implementation was delegated to the subnational level. Initially the financing of these policies was provided with resources from the National Budget; however, the system evolved with the introduction of a system of intergovernmental transfer systems and the strengthening of own revenue sources for subnational governments. In Poland, regional governments have just a few specific service delivery functions that are related to sectoral competencies, and their primary role is with regards to the programming of regional development policies, especially in the context of the significant European Union structural funds. Regional governments also play a much more pronounced role in public investment than other layers of subnational governments, which helps to exploit existing economies of scale. The case of Spain shows much more emphasis on delegating functions to the intermediate level of subnational governments. Regional governments in Spain share many regulatory functions with the central government, especially with respect to territorial management, culture, environment, R & D policy, transport infrastructure, etc. And there some 3 hierarchical relationship in the sense that regional governments can decide on the organization of municipalities and provinces within their regional territory. Among case studies, there are examples of both uniform and asymmetric decentralization models. For example, in the three-tier structure of subnational governments in Poland, each tier has its specific functions defined in the legal framework. All subnational governments belonging to the same subnational tier, have the same set of responsibilities, hence there is no asymmetric decentralization. In contrast, in Indonesia, Colombia and Spain, the process of decentralization was asymmetric with regards to political, fiscal and administrative/management function. For example, 5 regions in Indonesia are granted a stronger role in selected aspects of management or economic policy. In Colombia, political asymmetry in decentralization is present through the special rights of indigenous peoples to manage their own territory or special metropolitan status of Bogota. Special administrative and management functions have also been granted to six other metropolitan areas. Finally, in Spain, at the regional level, decentralization is asymmetric, with two distinct regimes: the common regime (15 ACs) and the “foral” regime (Basque Country and Navarra), with this latter characterized by an almost complete spending and revenue autonomy. In addition, ACs of the common regime each have their specific statute, allowing for some distinctive features, especially since the 2000s (several statutes have been reformed case by case). In some of the countries the decentralized subnational governments coexist with the deconcentrated central government units. Poland’s case is interesting as at the regional level there are two parallel structures operating to a large extent independently of each other: the regional governor/administrator (wojewoda) who is appointed by the Prime Minister and who is the head of the state administration in the region; and the elected regional government, headed by the Regional Marshal (marszalek). While at the beginning of such dual system there were some conflicts over competencies, the model has been successfully implemented over last twenty years and currently there are no conflicts between the two structures even when the subnational authorities represent different political views than the central administration holding office. However, in the case of Indonesia, the dual role of provincial governor as a regional head and a representative of the Central Government for the purpose of supervising the district/city governments, has not been well understood or fully utilized leading to complications with public accountability. In the case of Spain, there are territorialized agencies of the central administration and a “delegate” of the central government in each region, but the regional administrations (CAs) function entirely in a separate and autonomous way. Revenue sources In most cases analyzed subnational tax autonomy is rather limited. The subnational regional governments in Poland have virtually no own revenues, except for small revenues from fees and property constituting less than 5 percent of regional budgets. Even though the legislation in Spain allows that the Autonomous Communities create their own taxes within the regional borders, in practice there are many impediments and as a result the role of own taxes in the revenue structure of regional governments at 2 percent is rather insignificant. However, the so called “ceded” central taxes, including 50 percent of the central personal income tax and others, and for which regional governments can select rates and tax base, provide much higher levels of revenue autonomy. In the case of the Basque Country and Navarra e autonomy is still much higher, as these regional governments benefit from all taxes except import duties, payroll taxes, VAT and excise duties. In Colombia, the central government retained the main sources of public revenues including the income tax, the value-added tax and taxes on international trade, and subnational governments have limited taxing autonomy and little discretion over tax rates and bases. Several taxes are earmarked for specific use defined by the law. Taxes represent around 20 percent of revenues for departments and include receipts from the excise taxes (beer, tobacco, liquor), vehicle tax, register tax and gasoline tax. 4 Regional governments seem to receive quite substantial resources from tax sharing arrangements with the central government. While there are different types of tax sharing arrangements, it seems that regional governments in Spain have somewhat larger fiscal leeway whereby they participate in sharing of major taxes (PIT, VAT, excise) and also have autonomy over the wealth tax, inheritance and gift tax, tax on capital transfers, gambling tax, vehicle excise and hydrocarbons retail sales tax. In Poland, tax sharing arrangements include PIT and CIT, the latter being particularly problematic both in terms of its sharing formula and high volatility leading to large horizontal disparities among regions and cyclical swings in their revenue base. Different transfers play an important role in subnational financing and help achieve important objectives within the fiscal decentralization process. • In Poland, the general grant (non-earmarked) constitutes the most important grant. It is made up of several shares, including the education share, the equalization share, the balancing share and the regional share. The education share is by far the largest, and it aims at covering educational expenses, including teacher’s salaries, but it is not earmarked. The equalization share is allocated to all SNGs with below-average tax capacities. The regional share is calculated for each region based on several criteria (demography, unemployment, GDP per capital, roads, etc.). Earmarked transfers include specific transfers for central government delegated tasks (e.g. social spending), capital expenditure, etc. • In Spain, Basque Country and Navarra do not receive transfers from the central government but, on the contrary, they transfer funds to central government to participate to national general expenses. Other ACs (in the common regime) receive general unconditional equalization grants and conditional grants. There are two main funds: (i) the Guarantee of Basic Public Services Fund, non-earmarked, which is intended to ensure equal funding for basic public services (health care, education and social services), and (ii) the Global Sufficiency Fund, which is based on the assessment of the fiscal gap between expenditure needs and fiscal capacity. In addition, regions receive conditional grants aimed at fostering regional development, such as the compensation fund. • In Colombia, the General System of Transfers (Sistema General de Participaciones, SGP) is the most important central government transfer. SGP funds are earmarked for the most part, and their distribution between key sectors (education, health and water supply and sewerage) is determined by law according to a formula based on a combination of population coverage, social equity and efficiency criteria. • In Indonesia, transfers from the central government consist of three major components: the general allocation fund, the revenue-sharing fund and the specific purpose fund. The general allocation fund (Dana Alokasi Umum, DAU) is an equalization transfer system aiming at reducing fiscal imbalances between sub-national governments. Transfers are formula-based, consisting of a base allocation (equal to the amount of spending on personal) and a fiscal gap allocation (which can be positive or negative). The shared revenue fund (Dana Bagi Hasil, DBH) is sub-divided in two categories. The DBH from taxes is a shared tax system, based on receipts from the personal income tax, etc. The DBH from natural resources is based on revenues derived from forestry, mining, oil, etc. Last, the special allocation fund (Dana Alokasi Khusus, DAK) is a transfer system to fund responsibilities that are considered as national priorities and are mostly earmarked capital transfers. 5 Subnational borrowing and fiscal rules The case studies underscore the need for ensuring fiscal stability at the subnational level by putting in place borrowing rules. In Colombia, the subnational entities suffered significant financial problems. During the decade of the 1990s, many departments and large cities became excessively indebted and could not meet their payment obligations. That prompted the national government to bail out those subnational governments. To avoid the situation from re-occurring, a robust regulatory framework was issued imposing borrowing restrictions on subnational governments, in order to encourage fiscal sustainability. In Indonesia, subnational borrowing prior to the decentralization reforms used to be significant. The central government sought to clean up the defaulting loans early in the reform period. Nowadays, regional governments can borrow but are limited by regulations and capacity to design and manage large capital projects. The subnational governments in Poland have a right to take loans or issue bonds to borrow in the financial market without asking for central government prior permission, although, they are subject to regulations and ex-post controls. If the sub-national entities have problems for servicing their debt, they need to approve a recovery program. It often includes fiscal adjustment programs in exchange for a low-interest loan from the central administration. Subnational borrowing is often embedded in the subnational fiscal rules and supervised through independent bodies or the central government: • In Spain, the subnational debt has significantly increased after the 2008 global financial crisis. This situation led to profound reforms regarding fiscal rules. A constitutional reform was adopted in 2011 to underpin the fiscal consolidation targets for all Spanish administrations from 2020 onward. Moreover, an Organic Law on Budgetary Stability and Financial Sustainability passed in 2012 introduced a structural balanced budget rule and debt ceilings for all levels of governments, as well as expenditure rules for SNGs. The regional Liquidity Fund (FLA) was also temporarily established in 2012 to support heavily indebted regions with difficulties borrowing in open markets. Finally, an independent authority for Fiscal Responsibility was set up to monitor and report on compliance of all levels of government. • In Poland, the 2011 Public Finance Act stipulated to balance local current budgets and strengthened debt limitations requiring that the sum of loan instalments and interest payments must not exceed 15% of total debt. From 2014 onwards, the mode of calculation for debt ratios is changed in order to reduce subnational debt. • In Indonesia, sub-national debt must not exceed 75% of the previous year’s budget revenues. Sub- national governments are not allowed to borrow internationally; those with outstanding arrears on government loans are prohibited from borrowing. A cumulative lending limit for subnational loans is set by the ministry of finance. • In the case of Colombia, recourse to borrowing is regulated by strict prudential rules, established in 1997 (“Traffic Light Law” or Ley de Semáforo); in addition, in 2000 and in 2003 the Law on Fiscal Transparency and Responsibility curbed the growing subnational government indebtedness observed during the 1990’s. In addition, since 2003, departments and large municipalities must obtain satisfactory credit ratings from international rating agencies before they can borrow. 6 Challenges and lessons The case studies show that the decentralization process is a continuous effort that has to be carefully sequenced and managed. Challenges are multiple: • The decentralization process is unlikely to succeed if it is only partial -political decentralization. If these reforms are not accompanied by fiscal and administrative decentralization the overall outcome can in fact have detrimental effects on the quality and effectiveness of governance in the country. • Accountability becomes complicated and diffused when there are overlapping competencies or when these responsibilities are not well-defined, which also increases the likelihood of political conflicts. • Politically it is much easier to decentralize expenditure responsibilities than revenue sources. However, insufficient revenue decentralization and financial autonomy is a significant threat to accountability and the overall efficiency of the decentralized system. • Local fiscal accountability and citizens’ participation on subnational governments’ affairs represent a difficult challenge in the decentralization process of any country, but they are key to successful decentralization. Case studies also offer many lessons to other countries embarking on decentralization reforms: • While decentralization processes are very country specific, they need to be informed by fiscal federalism principles and the ample international experience on what has worked and what has not work in other countries. • Decentralization is not an effortless and simple process, on the contrary, it requires a significant planning effort and political leadership in its execution. • Supporting newly created subnational governments is a core responsibility of the central government in the evolved decentralized intergovernmental architecture. • Macroeconomic and fiscal stability are core objectives that the fiscal decentralization architecture must support and respect. Subnational governments, even when they have autonomy in their fiscal matters, must still be subject to the rules and regulations that are necessary to ensure economic stability. What can Chile learn from the case studies? Articulate a vision. Decentralization reforms are too complicated to be planned and implemented on ad hoc basis and only with a partial view. While they can be spread over years, it is important to have a clear vision or roadmap in mind. All stakeholders must get involved in the process. Chile is not in a rush, thereby, all kind of stakeholders should get involved in the process to converge as much as possible on goals, concepts, terminology and key options. Good planning is essential. For expenditure responsibilities, it is important to identify and asses which is the lowest organizational level of government at which functions can be done efficiently and successfully. It will be necessary to find a balance between the subsidiarity principle and economies of scale. Then, a timely and relevant capacity development plan must be prepared that recognizes critical and urgent transition issues, but which also incorporates a longer-term perspective. 7 Create a dynamic and flexible system. The architecture and typology of functions scheme must provide a dynamic and creative system. Chile should provide certainty to the subnational level in order to undertake the new functions but must guarantee that there is room for changes and improvements, of course, under the central government’s guidance and control, when needed. Focus on establishing a comprehensive role for the intermediate level of government. Overriding political considerations shapes the choice of intermediary roles and mechanisms. Chile may have the option (with only one level in play) to create from the beginning an integral but at the same time transparent intermediate level of government. Design an efficient accountability mechanism. Lack of accountability mechanisms may generate a slow progress on service improvement and innovation. Lack of accountability will also lead to the inefficient and unresponsive provision of public services, and ultimate to the waste of public resources. Chile has the chance to design good mechanisms (especially in terms of the use of the information) which are not burdensome and involve all stakeholders more meaningfully. Match revenues to needs. To ensure accountability, revenue should be matched as closely as possible to expenditure needs that is: “finance follows function.” In addition, tax instruments intended to further specific policy goals should be assigned to the tier of government which is in charge of providing that service. Manage increases on aggregate national expenditures. While in theory decentralization could be considered neutral in terms of expenditures, in practice Chile should be prepared for an increase in overall national public expenditures. This is specially the case if decentralization implies funding underprovided key public services in lagging areas. Chile needs to find its own boundaries and priorities in implementing a fiscal decentralization plan at the intermediate level of government. 8 Fiscal decentralization of regional governments – the case of Colombia prepared under the World Bank TA RAS for Chile1 1 The note for Colombia was written by Juan Gonzalo Zapata 1 Índice Contexto y breve revisión histórica ........................................................................................................... 3 Descentralización en Colombia .................................................................................................................. 5 Competencias de departamentos y municipios ........................................................................................ 9 Transferencias Intergubernamentales Generales de la nación ............................................................. 19 Sistema General de Participaciones ........................................................................................................ 19 Sistema General de Regalías ................................................................................................................... 22 Deuda y sostenibilidad fiscal .................................................................................................................... 26 Balance de la Descentralización en Colombia: Ajustes sobre la marcha ............................................. 29 2 Contexto y breve revisión histórica Desde mediados del siglo XIX los niveles intermedios de gobierno en Colombia han sido importantes en términos fiscales y han tenido autotomía administrativa y política. Esto se explica porque, si bien este país ha sido marcadamente centralista, tuvo un periodo largo en donde, inspirado en ideas liberales, se abrió el camino para la expedición de tres constituciones de corte federal. Las constituciones de 1853, 1858 y 1863, en un proceso lento pero persistente, crearon en total 9 estados. Estos estados se crearon a partir de las anteriores provincias heredadas del largo periodo colonial. En la década de los años 50 del siglo XIX las 42 provincias se agruparon y se les dio autonomía plena en los ámbitos políticos, económicos y sociales.2 Ya para la década de los años sesenta, en el marco de la Constitución de Rionegro de 1863, que funda los Estados Unidos de Colombia, los anteriores estados se convierten en estados soberanos en donde cada uno tenía su asamblea legislativa, tenía su ejército, creaba sus propios impuestos y priorizaba el gasto público que ejecutaba, entre muchas funciones más. Era un modelo federal muy cercano a una confederación. La autonomía de estos estados era muy amplia; por ejemplo, algunos de ellos deciden abolir la pena de muerte con una pena máxima de 10 años. Por esto, las relaciones con el gobierno nacional eran complejas pues cada estado tenía un voto para la elección presidencial y gran parte del poder se ejercía desde el Congreso, en donde los estados tenían una alta representación. Además, cada estado nombraba su presidente, tenía asamblea legislativa y en algunos de ellos nombraban sus mandatarios locales por voto calificado. Si bien no es del caso mencionar las duras confrontaciones entre conservadores y liberales, es necesario recordar que la agenda pública estaba signada por el papel de la religión, por la reforma educativa, por la distribución de algunos impuestos nacionales con los mencionados estados; por ejemplo, la renta de la producción y comercialización de las salinas, y por la difícil puesta en marcha de proyectos nacionales. El mejor ejemplo de esto último fueron los intentos de construir ferrocarriles y otros numerosos esfuerzos adelantados tanto por los estados soberanos como por el gobierno nacional. De otro lado, la independencia de España no cambió la estructura tributaria existente. La mayoría de los impuestos al consumo se mantuvieron e inclusive buena parte de los monopolios coloniales aún estaban en la lista de los gravámenes departamentales vigentes. Esto es bien importante, puesto que el monopolio del aguardiente, los ingresos por juegos de suerte y azar y del degüello mayor y menor, entre otros, han sido importantes para las finanzas departamentales desde que estos fueron creados. A pesar de todos los cambios políticos producto de las guerras civiles y nuevas constituciones, estos impuestos se han mantenido en el tiempo. Es más, a finales de siglo XIX cuando los estados soberanos se degradan y se convierten en departamentos, varios de sus tributos mencionados los incorpora el gobierno nacional y su peso dentro del total de ingresos fue importante. 2 Zapata, G. J.G. Reforma radical en el Estado de Santander, 1850-1885. Bogotá: Universidad del Rosario, 2015. 3 Aunque es conocido que este periodo federal termina con la expedición de la constitución de 1886, que era muy centralista, buena parte del desarrollo institucional que se adelantó durante el periodo federal se mantuvo en los años siguientes. Los estados soberanos pasaron a ser departamentos, los gobernadores eran designados por el gobierno nacional, y estos a su vez nombraban a los alcaldes. Si bien la autonomía política del modelo federal murió, se mantuvo la misma estructura de las secretarías (ministerios) sectoriales (salud, educación, vías, etc.). Por supuesto, los ejércitos de los estados se acabaron y las asambleas no tenían funciones legislativas mayores, ya que no expedían leyes sino ordenanzas con un alcance mucho más limitado. Es el momento de mencionar que el primer esfuerzo de descentralización en Colombia fue a mediados del siglo XIX. Se dio durante el gobierno de José Hilario López, cuando en 1850 su secretario Manuel Murillo Toro descentralizó en favor de las provincias varios de los impuestos nacionales. Esto con el fin de agilizar su cobro y darles más autonomía y poder a los mandatarios territoriales. Desde lo fiscal, se buscaba además que el gobierno nacional se concentrara en el cobro de pocos impuestos y, que las provincias pusieran en marcha el cobro del impuesto directo. Por explicarlo en forma resumida, este impuesto era una combinación de impuesto de renta y de impuesto a la propiedad de acuerdo con parámetros actuales. Por ejemplo, en el estado de Santander, la autoridad entregaba una declaración de bienes a la persona llamada riqueza gravada, y a partir de esta declaración se le aplicaba una tarifa cercana al 1% anual. Era un sistema de cobro elaborado pues la persona podía solicitar una revisión de la declaración de su riqueza gravada y existía una autoridad encargada de dicha revisión. Además, las alcaldías, llamadas en Santander distritos, decidían la tarifa a aplicar sobre la riqueza gravada, ver Zapata (2016).3 En este impuesto muchos de los líderes del partido liberal que apoyaban el modelo federal, centraron muchas de las esperanzas de autonomía fiscal de los gobiernos federales. Más adelante, durante el gobierno del general Rafael Reyes a comienzos del siglo XX, varios departamentos se escindieron, con lo cual se crearon nuevos departamentos. Se llegó así, a un total de 15 y casi al mismo tiempo se crearon las intendencias y comisarias para los territorios nacionales poco poblados y alejados del centro. Estos fueron administrados desde el nivel central por una entidad pública y un departamento administrativo nacional. Ya durante el siglo XX los departamentos encontraron en el monopolio de los licores y en sus loterías sus principales fuentes de recursos. Luego, la explotación del chance y el impuesto a los vehículos automotores, entre otros tributos tomaron relevancia. Se ha criticado que estos impuestos al ocio son regresivos, que tienen cuño colonial y que sus bases gravables son poco dinámicas. Este tema, sin embargo, por lo pronto lo dejaremos de lado. De lo anterior queda en claro que, desde el siglo XIX, los departamentos son importantes en la vida fiscal y financiera del sector público colombiano. También que estos siempre han tenido fuentes propias de recursos y han tenido claridad en las funciones que les competían. A pesar de lo anterior, a lo largo del siglo XX los departamentos pierden relevancia al interior de la vida 3 Zapata, J. G. (2015). Caps. 2 y 3. Ibid. 4 pública por varias razones. Primero, sus fuentes de recursos propias perdieron fuerza; segundo, los impuestos nacionales crecieron notablemente y desplazaron en términos relativos a los impuestos territoriales. Tercero, la descentralización, que comenzó en la década de los años noventa del siglo XX, inicialmente se concentró en los municipios y el marco legal que se desarrolló favoreció su financiamiento, tanto sus recursos propios como transferidos. Descentralización en Colombia En los años ochenta del siglo XX, una serie larga de paros cívicos eran la expresión de las regiones. La mayoría de estos paros no tenían reivindicaciones políticas y eran más sociales, en especial por la falta de servicios públicos. Acueductos, interconexión eléctrica y vías eran los principales temas (Maldonado: 2012).4 En este sentido la descentralización fue en gran parte, la respuesta política a las demandas que comenzaron en las regiones. Las demandas venían desde abajo pues se identificaban como malos o inoperantes a los alcaldes elegidos directamente por los gobernadores. El descredito de la vida política local era muy alto. Sin embargo, con la posterior elección popular de alcaldes los paros cívicos disminuyeron rápidamente. De otra parte, el proceso de descentralización tuvo antecedentes técnicos importantes en los años 80, antes de los cambios normativos centrales. El informe Bird- Wiesner, una misión de expertos nacionales e internacionales presentó sus resultados en 1979, demostró que el 84.6% de los ingresos del Estado estaban concentrados en el nivel central, mientras que solo 9.7% en los departamentos, intendencias y comisarias.5 Con este muy bajo nivel de gasto territorial era muy difícil responder a las demandas locales. Una de las recomendaciones del informe fue de fortalecer las fuentes de recursos territoriales. Es así como se expiden varias leyes que fortalecen los fiscos municipales; la Ley 14 de 1983 mejoró el recaudo de impuestos como el de timbre nacional, licores, cigarrillos. Esta ley lo que hizo fue aumentar las tasas y las bases gravables. Mas adelante en 1995, se crea la sobretasa a la gasolina y los municipios eran los encargados de decidir su tarifa; sin embargo, la legislación posterior volvió obligatorio el cobro del 20% en todos los municipios. También en 1998 con la expedición de la ley 488 se creó el impuesto a los vehículos automotores que reemplazó al de Circulación y Transito, el de timbre sobre vehículos automotores y el impuesto unificado de vehículos que había creado Bogotá. Con esto se aumentó la base gravable y es el ministerio del transporte el que decide el precio de los vehículos nuevos y usados (base gravable) para su cobro. Su recaudo se destina mayoritariamente a los departamentos. A su vez, más adelante, con la creación del Impuesto al Valor Agregado (IVA), que reemplazó al anterior impuesto a las ventas, aumentó al mismo tiempo la transferencia en favor de los municipios. Debe mencionarse que un artículo transitorio de la Constitución Nacional obliga la expedición de un nuevo estatuto tributario territorial; sin embargo, después de 28 años este estatuto aún no se ha 4 Maldonado, A. Los límites de la descentralización territorial: el caso de Colombia 1991- 2008. Tesis de Doctorado. Universidad Complutense. Madrid. 2011. 5 DNP (1981). Las finanzas intergubernamentales en Colombia: informe de la Misión Wiesner-Bird de finanzas intergubernamentales. Bogotá. 5 expedido. El gobierno nacional ha utilizado las reformas tributarias nacionales para mejorar, poco a poco, el recaudo de los ingresos corrientes o propios de departamentos y municipios al incluir siempre algunos artículos al respecto. Por ejemplo, la ley 1943 incluyó un aumento importante en la tarifa del impuesto a los cigarrillos. Esta tarifa se triplica en un periodo de sólo 3 años. Una de las últimas reformas tributarias nacionales, la mencionada ley 1943 de 2018, creo la Comisión de Estudio del Sistema Tributario Territorial. Esta comisión se reúne desde el mes de junio y se espera que sus recomendaciones estén para mediados del 2020. Estas recomendaciones, no obligan su adopción o la presentación de un proyecto de ley, pero es la primera vez que una comisión de este tipo se concentra en analizar los ingresos territoriales. Un paso importante para la descentralización política se dio con el Acto Legislativo 1 de 1986 que estableció la elección de alcaldes, anteriormente eran funcionarios de libre nombramiento y remoción por parte de los gobernadores. Esta decisión contribuyó a ampliar la democracia representativa en el país. Para este mismo año hubo un cambio importante en la distribución de funciones de las entidades territoriales, especialmente de los municipios. La Ley 12 de 1986, al mismo tiempo que aumentó la participación de los municipios del recién creado impuesto al valor agregado (IVA), se les asignó más competencias en la administración de servicios en el sector salud, educación, desarrollo agropecuario, desarrollo urbano y saneamiento básico. Este salto no fue exento de críticas pues no se crearon planes o mecanismos claros de implementación que ayudaran a los territorios a asumir estas funciones. Adicionalmente, se criticó que no se diferenciaran las funciones de acuerdo con criterios como el número de habitantes, nivel de ruralidad o generación de recursos propios, entre muchos más. La mayoría de las funciones descentralizadas pasaron del nivel nacional al municipal, en especial en el sector salud. En esta coyuntura no se entregaron funciones a los departamentos. El proceso de descentralización se volvió más integral con la expedición de la Constitución Nacional de 1991. Esta constitución permitió que los departamentos lograran desarrollarse como entidades autónomas en lo político, lo administrativo y lo fiscal en forma paralela a como se hizo con los municipios. Se decidió que las autoridades departamentales se eligieran por voto popular. Hay que tener presente que durante los debates de la Asamblea Constituyente se expresaron dos líneas sobre la elección popular de gobernadores. Los que querían que la descentralización fuera únicamente municipal y los que querían un tercer nivel de gobierno fuerte. Los principales reparos para extender la descentralización a los departamentos eran fiscales; se entendía que no se disponía de los recursos necesarios para financiar tres niveles de gobierno. Sin embargo, las realidades políticas, los departamentos siempre han tenido presencia en la vida nacional, permitieron que esta decisión no tuviera fuerte oposición en la Asamblea Constituyente. Como veremos, esta decisión explica el gran aumento de las transferencias intergubernamentales en favor de departamentos y municipios. La elección de gobernadores por voto popular se entendió como un complemento de la elección municipal. Además, las anteriores intendencias y comisarías se volvieron departamentos con iguales funciones y responsabilidades. Se llegó así a un total de 32 departamentos, con lo cual en 6 Colombia se formalizó un proceso de descentralización con tres niveles de gobierno fuertes. Falleti argumenta que en Colombia se dio muy fácil la descentralización política pero que la cesión de funciones y recursos fue mucho más lenta.6 Esto generó problemas en las relaciones entre el gobierno nacional y los mandatarios de los territorios.7 El proceso de descentralización fue complejo pues el gobierno nacional cerró varias entidades descentralizadas nacionales para seguidamente ceder las funciones a los territorios. En especial se debe mencionar al sector agua potable y saneamiento en donde una entidad era la encargada de construir los acueductos. Esta entidad, el Instituto de Fomento Municipal (INSFOPAL) tenía muy malos resultados. Hay ejemplo en el sector agrícola y en la construcción de vías, etc. El proceso de cesión de funciones y recursos tuvo con la Constitución Nacional de 1991 un salto cualitativo pues la descentralización fue uno de los temas ganadores con la nueva carta fundamental. Este cambio constitucional hizo que el aumento de las transferencias intergubernamentales, junto con el de las regalías por la explotación de recursos naturales no renovables, tuvieran en cuenta a los dos niveles de gobierno territoriales. La distribución de estos recursos favoreció a los gobiernos subnacionales. De esta manera se reconoció el gasto social como prioritario, así como la destinación condicionada de los recursos de inversión en sectores como salud, educación y saneamiento básico. En pocas palabras los territorios y la descentralización fueron los ganadores con el nuevo marco normativo producto de la CN de 1991. Se desarrolló igualmente una nueva institucionalidad por parte del gobierno nacional encargada de administrar los recursos transferidos, de hacer seguimiento de la planeación en los territorios e intervenir mediante convenios de desempeño en las entidades territoriales con problemas financieros críticos, ley 358 junto con la ley 550 de 1999, que es una ley de quiebras que tiene su capítulo para el sector público. Esta ley permite reestructurar sus pasivos. Cuando un gobierno territorial, departamento o municipio, tiene problemas financieros puede acogerse a cualquiera de estas dos leyes. Para esto debe firmar un convenio de desempeño o un acuerdo de reestructuración con la Dirección de Apoyo Fiscal (DAF) del Ministerio de Hacienda. Estos convenios y acuerdos son ortodoxos, puesto que obligan a que el gobierno subnacional mejore sus ingresos, controle sus gastos de funcionamiento o generales, limite la inversión en infraestructura y, cierre o reorganice sus empresas de servicios públicos o cualquier entidad de su jurisdicción. Las metas son anuales y se hace un seguimiento de ellas permanente. A su vez la firma y el cumplimiento de los convenios de desempeño les permite acceder a recursos frescos de los bancos para refinanciarse. Sin la firma de los convenios de desempeño no es posible acceder al crédito, ver Caja 1. Caja 1 - Reestructuración de Pasivos y Convenios de Desempeño de los Territorios Los departamentos y municipios se encuentran en programas de Saneamiento Fiscal y Financiero, siempre y cuando tengan suscritos convenios o planes de desempeño en el marco de la ley 358 de 6 Falletti, T. A sequential theory of decentralization and its effects on the intergovernmental balance of power: latin american cases in comparative perspective. Working Paper # 314. Kellog Institute. July 2004. 7 En este trabajo cuando se utiliza la palabra territorios se debe entender que son al tiempo los departamentos y municipios. Así los mandatarios territoriales son los alcaldes y gobernadores. 7 1997 o suscriban acuerdos de reestructuración de pasivos, ley 550 de 1999. Estos programas de saneamiento deben contar con el concepto favorable del Ministerio de Hacienda y Crédito, ley 617 de 2000, y es la dirección de Apoyo Fiscal (DAF) del Ministerio de Hacienda la encargada de cumplir con la reestructuración o con los convenios. Veamos un resumen de estas dos figuras de ajuste financiero para auxiliar a los territorios con problemas financieros. Reestructuración de pasivos de las entidades territoriales - Ley 550 de 1999 La Ley 550 de 1999 establece un régimen que facilita la reactivación empresarial y la reestructuración financiera de los entes territoriales. Específicamente en el Título V “De la reestructuración de pasivos de las entidades territoriales” regula su aplicación a las entidades territoriales y descentralizadas del nivel territorial. Objetivos de la Ley Instrumentos de intervención estatal La negociación y celebración de acuerdos de Restablecer la capacidad de pago de las reestructuración de pasivos que restablezcan la entidades de manera que puedan atender capacidad de pago de las entidades territoriales o adecuadamente sus obligaciones. de las entidades descentralizadas para que puedan atender sus obligaciones. Procurar una óptima estructura La normalización de los pasivos pensionales. administrativa financiera y contable Propender porque las empresas y sus La concertación al interior de cada entidad trabajadores acuerden condiciones territorial de condiciones laborales que faciliten su especiales y en materia laboral reactivación y viabilidad. La negociación de deudas contraídas con cualquier Facilitar la garantía y el pago de los clase de personas privadas, mixtas o públicas. pasivos pensionales a cargo de las Incluyen las deudas parafiscales distintas de las entidades del orden territorial. previstas en el régimen seguridad social, así como las deudas fiscales. El Ministerio de Hacienda y Crédito Público, a través de la Dirección de Apoyo Fiscal, es el competente para realizar los procesos de reestructuración de pasivos de las entidades descentralizadas del nivel territorial. Para esto se firma un Acuerdo de reestructuración con la entidad territorial (Art. 5 de la Ley 550). Este convenio tiene efecto vinculante, tanto para la entidad como para la totalidad de los acreedores. El articulo especifica las condiciones para acogerse al acuerdo de reestructuración y el artículo 7 establece el orden de prioridad para programar los gastos corrientes de la entidad territorial, conforme con al acuerdo de reestructuración. Estas prioridades son un su orden: a) Mesadas pensionales b) Servicios personales c) Transferencias de nómina d) Gastos generales e) Otras transferencias f) Intereses de deuda g) Amortizaciones de deuda h) Financiación del déficit de vigencias anteriores i) Inversión Desde el 2000 hasta mayo de 2019 se han ejecutado 112 Acuerdos de Reestructuración de Pasivos en las Entidades Territoriales, de los cuales 75 se terminaron por haberse cumplido en forma anticipada o en el plazo establecido, 24 se encuentran en ejecución, 6 se terminaron por 8 incumplimiento, 2 se encuentran en negociación y 5 no se suscribieron porque no se obtuvo la votación requerida, fracaso la negociación o por no tener facultades del concejo municipal. Convenios de Desempeño Estos planes o convenios de desempeño son programas de ajuste fiscal, financiero y administrativo que buscan restablecer la solidez económica y financiera de la entidad territorial. Buscan garantizar la capacidad de pago de largo plazo y el mejoramiento de los indicadores de endeudamiento de las entidades territoriales (artículo 7 – ley 358). Estos planes de desempeño incluyen medidas de racionalización del gasto y el fortalecimiento de los ingresos propios. También obligan a tomar decisiones como el cierre o la mejora de la gestión de las entidades descentralizadas del departamento o municipio. Los planes son ortodoxos, pue se hace un seguimiento periódico de las metas acordadas. Por esto las entidades territoriales deberán enviar trimestralmente la información de la evolución de los planes de desempeño a la DAF. Esta dirección evalúa el cumplimiento de dichos planes. Las entidades que incumplan esta medida quedarán sujetas a las sanciones pertinentes. Estos planes son de largo plazo y los mismos se mantienen vigentes hasta que la entidad territorial registre un nivel de intereses/ahorro operacional menor o igual al 40%. Sin el cumplimiento de los planes de desempeño, los departamentos y municipios no pueden acceder a los recursos del crédito. A continuación, se presentará el desarrollo que ha tenido este proceso en la definición de las competencias asignadas a los departamentos y municipios; el comportamiento de las finanzas públicas en los últimos 20 años; el funcionamiento de los sistemas de transferencias y, el desempeño fiscal y al endeudamiento de las entidades territoriales. Como veremos, la legislación que se expidió abarcó a los dos gobiernos subnacionales. Competencias de departamentos y municipios De acuerdo con lo señalado con anterioridad, la asignación de competencias a los departamentos y municipios tuvo cambios importantes durante la década de los años 80. En general, la mayoría de las funciones son compartidas entre los tres niveles de gobierno, en especial en los sectores de salud y educación en donde se movilizan los mayores recursos y las nóminas públicas son muy grandes; por ejemplo, en Colombia hay más de 300 mil maestros públicos. La tabla 1 nos muestra la competencia de los gobiernos por sectores del gasto. La lógica que hay al compartir responsabilidades es que el gobierno nacional es el encargado de diseñar las políticas sectoriales y los gobiernos subnacionales los encargados de ejecutarlas. Sin embargo, dado que en Colombia se demoró 20 años la expedición de la Ley Orgánica de Ordenamiento Territorial (LOOT), estipulada en la CN de 1991, lo que se dio fue un proceso en donde las leyes sectoriales, al tiempo que distribuían las funciones y recursos, creaban su propio ordenamiento territorial sectorial. Tabla 1 – Funciones y responsabilidades por sectores del Gasto Función Responsabilidad de Responsabilidad Financiación política y control de provisión 9 Cotizaciones al Sistema General de Seguridad Social. Seguridad social N N, D, M Presupuesto General de la Nacional – GNC. Sistema General de Educación N, D, M N, D, M Participaciones Recursos propios territorios Sistema General de Salud N, D, M N, D, M Participaciones Recursos propios territorios Presupuesto General de la Asistencia social N, D, M N, D, M Nacional Presupuesto General de la Seguridad N, M N, M Nación, Recursos propios territorios Sistema General de Agua y N, M N, M Participaciones alcantarillado Recursos propios territorios Sistema General de Participaciones, Presupuesto Medio ambiente N, D, M N, D, M General de la Nación, Cooperación Internacional Recursos Presupuesto General de la N N, D, M naturales Nación Ferrocarriles y Presupuesto General de la N N, M aeropuertos Nación Presupuesto General de la Carreteras N, D, M N, D, M Nación, Recursos propios territorios De otra parte, es necesario señalar que con la elección de los alcaldes y gobernadores en Colombia cambió la lógica de la planeación territorial y la presupuestación. Antes, estos niveles de gobierno poco participaban en la gestión de los recursos públicos pues esta se hacía desde el nivel nacional. Pero con la CN de 1991 se decidió que la nación se encargara del diseño de las políticas sectoriales, mientras que su implementación se debía descentralizar en los gobiernos subnacionales. Inicialmente la financiación de estas políticas era con recursos del Presupuesto Nacional; sin embargo, la creación de los sistemas transferencias intergubernamentales aumentó los recursos para varios sectores. Debe también quedar claro que en Colombia los gobiernos subnacionales tienen fuentes propias de recursos. En forma complementaria, se crearon Superintendencias y Comisiones de regulación y vigilancia que vigilan el cumplimiento de la ley y hacen los ajustes normativos para cada sector, ver Caja 2. Desde la misma lógica, el proceso de planeación y presupuestación quedó articulado. La presentación por ley de programas de gobierno de los candidatos a alcaldes y gobernadores, al momento de registrar su candidatura, los hace responsables de cumplir el programa de gobierno. Así, estos deben obligatoriamente diseñar sus planes de desarrollo y sus planes plurianuales de 10 inversión para garantizar el cumplimiento de estos programas. Estos presupuestos plurianuales de inversión los deben ejecutar en sus presupuestos anuales. Si esto no se cumple pueden inclusive ser revocados de su mandato; esto último ha ocurrido en algunos casos, ver tabla 2.8 Caja 2 – Superintendencias y Comisiones de Regulación en Colombia Comisiones de regulación La ley 142 de 1994, creó las comisiones de regulación con el fin de desarrollar la intervención estatal necesaria, tanto en el ámbito económico, como en el social. Así, estas se encargan por el correcto desarrollo de la competencia y por los derechos de los usuarios. De esta forma, las comisiones son unidades administrativas especiales que cuentan con independencia en términos administrativos, técnicos y patrimoniales. Sin embargo, cada una de ellas se encuentra adscrita al ministerio que corresponda a la actividad sobre la que intervienen. La delegación de las funciones de las comisiones de regulación la realiza el presidente de la república, en ejercicio de su poder, durante el acto de creación. Las principales Comisiones de regulación en Colombia son: La Comisión de Regulación de las Comunicaciones (CRC), la Comisión de Agua Potable y Saneamiento Básico (CRA), la Comisión de Regulación en Salud (CRES) y la Comisión de Regulación de Energía y Gas (CREG). Superintendencias Las superintendencias se crean con la Constitución Nacional de 1991, donde se creó la figura dentro de la rama ejecutiva. Las funciones de las superintendencias, en términos generales, están definidas por el artículo 66 de la ley 498 de 1998. Estas realizan labores de inspección y vigilancia según la ley o según lo ordene el presidente de la república. Las superintendencias se encuentran adscritas a la rama de poder ejecutiva y su función es inspeccionar y vigilar determinadas actividades económicas y de servicios. Estas funciones se definen por medio de la actividad que cubran y pueden llevar a cabo labores de auditoria, vigilancia e incluso impartir sanciones y penalidades a quienes no cumplan con lo estipulado en la ley. Así, estos organismos realizan funciones de control que resultan claves para Colombia. En momento hay 10 superintendencias: notariado y registro, servicios públicos domiciliarios, industria y comercio, salud, economía solidaria, puertos y transporte, financiera, subsidio familiar, puertos y transporte y, seguridad privada. Por su parte, la tabla 2 nos muestra un resumen de las principales leyes expedidas en el marco de la descentralización. En primer lugar, queda en claro que fue un proceso acelerado en donde se expidieron numerosas leyes que cambiaron en forma radical la gestión de los recursos públicos y la relación entre niveles de gobierno. Segundo, vemos como la planeación en los territorios se vuelven mucho más compleja, ya que queda relacionada con el proceso político, en específico el electoral. Esto incluye la expedición de planes de ordenamiento territorial por parte de los municipios. Estos planes son los que definen el uso del suelo en los municipios. Tercero, aparece una profusa legislación sobre finanzas y responsabilidad fiscal. Estas leyes limitan el acceso a la deuda, definen limites a los gastos de funcionamiento y obligan a una planeación financiera de largo plazo. 8Velásquez, F. y González, E. ¿Qué ha pasado con la participación ciudadana en Colombia? CIDER – Uniandes y Fundación Corona. Bogotá. 2003. 11 Tabla 2- Marco normativo de la gestión territorial Temas Norma Objeto Ley 136 de Establece las reglas y criterios para la organización y funcionamiento de los 1994 municipios, de los concejos y del personero. Ley 1368 de Establece el valor de los honorarios por cada sesión a que asistan los 2009 concejales, conforme a las categorías de los distritos y los municipios. Modifica y adiciona la Ley 136 de 1994. Esta ley tiene modifica el régimen Organización municipal, dentro de la autonomía de los municipios. Es un instrumento de y Ley 1551 de gestión para cumplir sus competencias y funciones que modifica los criterios 2012 funcionamien de categorización y los requisitos para la creación de los municipios. to Ley 1617 de Contiene las disposiciones que conforman el Estatuto Político, 2013 Administrativo y Fiscal de los distritos. Dicta las normas orgánicas para dotar a las Áreas Metropolitanas de un Ley 1625 de régimen político, administrativo y fiscal, que, dentro de la autonomía 2013 reconocida por la Constitución Política y la ley, sirva de instrumento de gestión para cumplir con sus funciones. Por medio de la cual se dictan normas orgánicas en materia de recursos y Ley 715 de competencias de conformidad con los artículos 151, 288, 356, 357 (Acto 2001 Legislativo 01 de 2001) de la Constitución Política y se dictan otras Distribución, disposiciones. competencias Modifica la estructura del Sistema General de Participaciones (SGP). Ajusta y recursos las competencias de las entidades territoriales en materia de agua potable y Ley 1176 de 2007 saneamiento, y los criterios para la distribución y destinación de estos recursos. Establece los procedimientos y mecanismos para la elaboración, aprobación, ejecución, seguimiento, evaluación y control de los planes de desarrollo, así Ley 152 de como la regulación de los demás aspectos contemplados por el artículo 342 1994 de la C. N. En general, precisa las normas constitucionales que se refieren al plan de desarrollo y la planeación en el estado. Establece los mecanismos que permiten al municipio, en ejercicio de su autonomía, promover el ordenamiento de su territorio, el uso equitativo y Ley 388 de racional del suelo, la preservación y defensa del patrimonio ecológico y Planeación y 1997 cultural. También define en su ámbito territorial la prevención de desastres Ordenamient en asentamientos de alto riesgo. o Territorial Establece mecanismos de integración, coordinación y armonización de las Ley 614 de diferentes entidades competentes en materia de ordenamiento del territorio, 2000 para la implementación de los planes de ordenamiento territorial. Ley Orgánica de Ordenamiento Territorial (LOOT). Dicta las normas orgánicas para la organización política administrativa del territorio Ley 1454 de colombiano. Limita el ejercicio de la actividad legislativa en materia de 2011 normas y disposiciones de carácter orgánico relativas a la organización político-administrativa del Estado en el territorio. Establece los principios rectores del ordenamiento. Regula la programación, elaboración, presentación, aprobación, modificación Decreto 111 y ejecución del presupuesto, así como la capacidad de contratación y define de 1996 el gasto público social. Por el cual se regulan los planes regionales ambientales de las Corporaciones Decreto Presupuesto y Autónomas Regionales y de las de Desarrollo Sostenible y su armonización 1865 responsabilid con la gestión ambiental territorial (Decreto 1865). ad fiscal Ley 549 de Establece que, con el fin de asegurar la estabilidad económica del Estado, las 1999 entidades territoriales deberán cubrir en la forma prevista en esta ley, el valor 12 de los pasivos pensionales a su cargo, en los plazos y en los porcentajes que señale el Gobierno Nacional. Por la cual se establece un régimen que promueva y facilite la reactivación empresarial y la restructuración de los entes territoriales para asegurar la Ley 550 de función social de las empresas y lograr el desarrollo armónico de las regiones. 1999 Se dictan disposiciones para armonizar el régimen legal vigente con las normas de esta ley. Establece los requisitos para la creación de municipios, reglas para la Ley 617 de transparencia de la gestión departamental, municipal y distrital, y, fija los 2000 criterios para la categorización presupuestal de los departamentos. Establece normas orgánicas de presupuesto para la transparencia fiscal, la Ley 819 de estabilidad macroeconómica, la disciplina fiscal y el endeudamiento 2003 territorial, entre otras disposiciones Ley 1483 de Establece los criterios y requisitos para la aprobación de vigencias futuras 2011 excepcionales en las entidades territoriales. Fuente: autor – varias leyes Asimismo, la nueva legislación sobre ordenamiento territorial determinó que, además de los departamentos y municipios, hay nuevas maneras de organizar el territorio y sus funciones. En el año 2011, la ley de orgánica de ordenamiento territorial (LOOT), establece nuevas formas de organización político-administrativa para mejorar la planeación, gestión y administración. Hay así, nuevas formas de asociación de entidades territoriales como: asociaciones municipales, regiones departamentales, áreas metropolitanas, provincias y distritos especiales. Cada una de ellas tiene definidas las condiciones mínimas para su conformación y funcionamiento. En general, las áreas metropolitanas se han desarrollado, pero las demás figuras de la LOOT han tenido un desarrollo discreto. Asimismo, recientemente se utilizaron los contratos plan, una figura administrativa utilizada por los franceses, como una forma asociativa entre los entes territoriales, el gobierno nacional y los privados que, permite sumar esfuerzos para planear y financiar en conjunto proyectos estratégicos de un territorio o región específica. La Caja 3 nos explica cómo funcionan las áreas metropolitanas y los distritos y la Caja 4 los contratos plan. Caja 3 - Áreas Metropolitanas y Distritos en Colombia El proceso de descentralización de Colombia que inició en la década de 1980 y que se acentuó con la Constitución del 1991 dio paso a una serie de cambios institucionales que modificaron las estrategias de integración, planeación y desarrollo regional. Entre otros temas, se abrió paso la formación de áreas metropolitanas y Distritos. Áreas Metropolitanas La Constitución de 1991, por medio del artículo 319, contempla la formación de áreas metropolitanas e indica que se puede hacer y cómo formarlas. Posteriormente las leyes 128 de 1994, la Ley Orgánica de Ordenamiento Territorial (LOOT), la ley 1454 de 2011 y la ley 1625 de 2103 precisan las reglas del juego para su creación, sus funciones y límites entre muchos temas más. Esta última ley expide el régimen de áreas metropolitanas y define, su objeto, naturaleza, competencias y funciones, la relación de los municipios que las integran, su alcance y hasta los órganos de dirección y administración. En la actualidad, en Colombia existen seis áreas metropolitanas constituidas y reconocidas ante la ley. Estas son: 1)El área metropolitana del Valle de Aburrá. Conformada por: Medellín, Bello, Barbosa, Copacabana, La Estrella, Girardota, Itagüí, Caldas, Sabaneta y Envigado; 2) El área metropolitana de Bucaramanga. Conformada por Bucaramanga, Floridablanca, Piedecuesta y Girón; 3) El área metropolitana de Barranquilla. Conformada por Barranquilla, Puerto Colombia, Soledad, Malambo y Galapa; 4) El área 13 metropolitana de Cúcuta. Conformada por: Cúcuta, Villar del Rosario, Los Patios y El Zulia; 5) El área metropolitana de Centro Occidente. Conformada por Pereira y Dosquebradas y, 6) El área metropolitana de Valledupar. Conformada por: Valledupar, Agustín Codazzi, La Paz, Manaure Balcón del Cesar y San Diego. Las áreas metropolitanas pueden ser integradas por municipios de un mismo departamento, municipios de diferentes departamentos que limiten entre sí y, por municipios que no limitan entre sí, pero que conservan una relación histórica en términos económicos y/o sociales. Con estas áreas se integran las políticas públicas de los municipios y se logra una mejor coordinación en términos económicos, políticos y de gobernanza territorial. Esto es muy positivo si se considera el hecho de que los municipios siguen conservando su autonomía. Dado que cada municipio suele llevar a cabo su propio Plan de Ordenamiento Territorial (POT) sin tener en cuenta las dificultades o los impactos causados a los municipios vecinos. El área metropolitana soluciona esto por medio de la integración de la gobernanza territorial en un solo POT. Se optimizan de una mejor manera los recursos, la capacidad de gestión aumenta y se reducen las brechas económicas entre las regiones. Se pueden llevar a cabo proyectos regionales. Distritos9 Los distritos son una forma alternativa de organización y funcionamiento de los municipios. Solo pueden ser distritos aquellos que cumplen una serie de características. Los distritos tienen un régimen legal, administrativo, político y económico propio. En Colombia, hay varias categorías de distritos y esta guarda relación con la actividad económica básica. Hay distritos turísticos, industriales, portuarios y culturales, entre otros. También, existe la categoría de Distrito Capital, que se refiere a la ciudad capital del país, Bogotá. Es preciso señalar que la ley 1617 de 2013, por la cual se expide el régimen para los distritos especiales, define los estatutos políticos, administrativos y fiscales de los distritos. En ella, se discute la creación, el funcionamiento y los límites de los distritos, además de su organización política, administrativa y de carácter territorial. Actualmente, Colombia cuenta con los siguientes distritos: 1) Bogotá: Distrito Capital. 2) Barrancabermeja: Distrito especial, portuario, industrial, turístico y biodiverso. 3) Barranquilla: Distrito especial, industrial y portuario. 4) Buenaventura: Distrito especial, industrial, portuario, biodiverso y ecoturístico. 5) Cartagena: Distrito turístico y cultural. 6) Mompox: Distrito especial, turístico y cultural. 7) Riohacha: Distrito especial, turístico y cultural. 8) Santa Marta: Distrito turístico, cultural e histórico. 9) Turbo: Distrito portuario, logístico, turístico, industrial y comercial. Y, 10) Cali: Distrito especial, deportivo, empresarial y de servicios. 9Gómez, L. (2017). Así nacieron las áreas metropolitanas. Revista Semana. Documento en línea, recuperado de: https://www.semana.com/contenidos-editoriales/valle-de-aburra-sin-fronteras/articulo/historia-del-nacimiento-de-las-areas- metropolitanas-de-colombia/544873. Ley 1625 de 2013. Diario Oficial de la República de Colombia. Bogotá, Colombia, 29 de abril de 2013. Ley 1617 de 2013. Diario Oficial de la República de Colombia. Bogotá, Colombia, 05 de febrero de 2013. Zapata, O. (2017). Las provincias y las áreas metropolitanas en Colombia: Potencias para la integración y el desarrollo regional. Revista Políticas Públicas. Vol. 10(2). 14 Caja 4 - Los Contratos Plan en Colombia La 1454 de 2011, ley Orgánica de Ordenamiento Territorial (LOOT), entre otros objetivos, busca concertar esfuerzos estatales para la planeación integral del desarrollo territorial con visión de largo plazo. Permite que entidades u organismos del nivel nacional puedan firmar con entidades u organismos del nivel territorial esquemas asociativos territoriales. Estos se conocen como los Contratos Plan (CP). Mas adelante, se utiliza en los planes de desarrollo nacionales de 2010-2014 y 2014-2018 de la administración Santos. En mayo de 2016 los bautizó como Contratos Plan para la Paz. Estos son una herramienta de gestión de la inversión pública que permite focalización y convergencia de los diferentes niveles de gobierno en la reconstrucción de los territorios donde la violencia tuvo alto impacto. En estos contratos plan inclusive pueden participar los privados. Los Contratos Plan son un acuerdo de voluntades entre los diferentes niveles de gobierno. Estos identifican los temas estratégicos para el desarrollo integral y mancomunado de una población y territorio en enfoque de largo plazo. Lo más novedoso es permite la ejecución concertada de programas y proyectos de inversión, la prestación de servicios públicos y el ejercicio de competencias y funciones que de manera conjunta realicen los territorios. Debe quedar en claro que no es una nueva fuente de financiamiento para los territorios; cada participante debe aportar sus recursos con las fuentes actualmente disponibles. No obstante, sí implican mayor disponibilidad de recursos para los sectores y programas objeto del Contrato Plan ya que, buena parte de los recursos provienen del Presupuesto Nacional. Se estima que la Nación contribuya con un aporte máximo de hasta un 60% de las inversiones, al tiempo que las entidades territoriales deberán concurrir con el 40% restante. Inicialmente se hicieron 7 CP en su primera fase: Atrato Gran Darién, Boyacá, Santander, Sur del Tolima, Cauca, Arauca y Nariño. Una evaluación de estos siete contratos dejo en evidencia que lo mejor no es hacer múltiples proyectos, sino más bien concentrarse en iniciativas capaces de transformar la vida de las comunidades. Es decir, proyectos de alto impacto. Por esta razón el Documento Conpes 3822 DNP permitió aumentar en 10 el número de Contratos Plan y modificó la gestión de los recursos. Se creó así un Fondo Regional para los Contratos Plan en el Plan Nacional de Desarrollo 2014-2018 Todos por un nuevo país. Los Contratos Plan no pueden sustituir, ni entrar en competencia con los demás instrumentos de priorización y gestión de la inversión; por lo contrario, deben ser complementarios. Su importancia es concentrar en pocos años inversiones sociales y en mejora de la infraestructura de regiones o zonas consideradas clave o estratégicas y que por diferentes razones no han tenido la atención del estado que merecen. Estas son regiones con altos niveles de pobreza, malos indicadores sociales (salud, educación, saneamiento básico, etc.), alejadas geográficamente o con un gran atraso en su infraestructura. Se estima que una región en donde lleguen al tiempo y de manera coordinada inversiones de los tres niveles de gobierno y del sector privado se logran sinergias definitivas. Finalmente, se debe mencionar que el DNP es el encargado de la gestión y seguimiento de estos contratos. Para esto se debe garantizar que las inversiones deberán ser verificables y, debe tener metas de gestión y que son monitoreadas. También se busca mejorar la gestión de algunos territorios y para esto se incluyen en las entidades territoriales institucionales. En el actual gobierno los Contratos Plan continúan, pero su nuevo nombre es Pactos Territoriales. Muy en la línea del nombre del actual Plan de Desarrollo Nacional Pacto por Colombia, pacto por la equidad. Fuente: autor 15 Finanzas territoriales En Colombia los impuestos solo se pueden crear mediante leyes y estas son solo nacionales y expedidas por el Congreso de la República; así, los territorios, sólo pueden ejercer su autonomía en el marco de dichas leyes. En general, estas les permiten fijar bases gravables, definir las tarifas y la gestión de la recaudación, entre varias funciones más. Debe quedar en claro que los gobiernos subnacionales no tienen iniciativa tributaria; esto es, no pueden crear impuestos. Las tablas 3 y 4 nos muestran la lista de los impuestos departamentales y municipales vigentes. Estos son los principales impuestos de cada nivel de gobierno y la mayoría de ellos son solo para departamentos y municipios; sin embargo, en algunos casos como el de Bogotá, que tiene su propio estatuto orgánico, se le permite cobrar al mismo tiempo algunos impuestos departamentales o participar de ellos. Al analiza la lista de cada nivel de gobierno se encuentra que los impuestos o gravámenes municipales son más numerosos y tienen bases gravables más fuertes como los son la propiedad inmueble y la actividad económica local. Mientras que los departamentos tienen menos impuestos y estos se encuentran concentrados en lo que se conoce en Colombia como los impuestos al ocio: cerveza, licores, juegos de suerte y azar, cigarrillos, que además concentran en promedio más del 50% de sus ingresos propios. Como veremos estas características explican la diferencia en el nivel del recaudo de departamentos y municipios. Tabla 3- Impuestos departamentales Registro y anotación Sobretasa a la gasolina Impuesto al consumo de licores Estampillas Impuesto al consumo de cerveza Impuesto al rodamiento de vehículos Cigarrillos y tabaco Juegos de suerte y azar Degüello de ganado mayor Tabla 4- Impuestos municipales Predial unificado Sobretasa bomberil Sobretasa ambiental Degüello de ganado menor Industria y comercio Alumbrado público Avisos y tableros Sobretasa consumo de gasolina Publicidad exterior visual Circulación y transito Delineación y urbanismo Telefonía Espectáculos públicos Registro de marcas y herretes Rifas municipales Impuesto de pesas y medidas Contribución a contratos de obra pública De otra parte, al analizar el desempeño del recaudo de los ingresos de los territorios desde el 2000 en adelante los ingresos municipales crecen mucho más rápido. En términos de recaudo departamental vemos como el impuesto a la cerveza es el más importante, le sigue en importancia el de licores, registro y anotación, tal y como se muestra en la tabla 5 y Grafica 1. Estos cuatro impuestos explican en promedio el 60% de los ingresos propios. Finalmente se encuentra que los 16 ingresos de recaudo bajo, al consolidarse, han aumentado en los últimos años. El crecimiento real de los ingresos departamentales es bajo y muy similar al PIB por lo que estos ingresos representan en los últimos años 1% del PIB. Un nivel muy inferior al de los ingresos propios o corrientes de los municipios que es del 2.8% del PIB en el 2018. Pero más importante aún, esta relación ha crecido en forma sostenida desde el 2000 al pasar de 1.9% a 2.8% en el 2018, ver tabla 5 Gráfica 3. Tabla 5 – Ingresos propios departamentales y municipales en Colombia (2000 – 2018) INGRESOS MUNICIPALES ($2018 MILES DE MILLONES) Cuenta 2000 2003 2006 2009 2012 2015 2018 1. Ingresos Corrientes 8.075 11.013 15.882 16.916 21.076 26.293 27.430 1.1. Ingresos Tributarios 6.713 8.523 12.129 14.477 17.233 22.745 24.165 1.1.1. Predial 2.318 2.783 3.609 4.284 5.234 7.368 8.387 1.1.2. Industria y Comercio 2.666 3.587 4.987 5.953 7.231 8.485 8.795 1.1.3. Sobretasa a la gasolina 1.053 1.381 1.631 1.499 1.457 1.593 1.587 1.1.4. Otros 676 772 1.902 2.741 3.312 5.299 5.396 1.2. Ingresos No Tributarios 1.362 2.489 3.753 2.440 3.843 3.548 3.265 Ingresos Corrientes % PIB 1,9% 2,3% 2,5% 2,4% 2,5% 2,8% 2,8% Ingresos Tributarios % PIB 1,6% 1,7% 1,9% 2,1% 2,0% 2,4% 2,5% Ingresos Tributarios % Ingresos Corrientes 83,1% 77,4% 76,4% 85,6% 81,8% 86,5% 88,1% INGRESOS DEPARTAMENTALES ($2018 MILES DE MILLONES) Cuenta 2000 2003 2006 2009 2012 2015 2018 1. Ingresos Corrientes 4.859 8.910 5.757 3.944 8.492 9.451 10.127 1.1. Ingresos Tributarios 3.704 7.404 4.535 3.436 6.726 7.819 7.613 1.1.1 Cerveza 1.163 2.289 1.319 941 1.998 2.221 2.170 1.1.2 Licores 997 1.586 931 844 1.289 1.131 789 1.1.3 Cigarrillos y Tabaco 501 627 460 386 571 502 928 1.1.4 Registro y Anotación 382 844 463 309 884 1.124 1.108 1.1.5 Vehículos automotores 307 607 386 286 557 613 796 1.1.6 Sobretasa a la Gasolina 266 526 374 273 378 432 432 1.1.7 Otros 89 923 602 396 1.050 1.796 1.388 1.2. Ingresos No Tributarios 1.155 1.506 1.222 508 1.765 1.632 2.514 Ingresos Corrientes % PIB 1,1% 1,8% 0,9% 0,6% 1,0% 1,0% 1,0% Ingresos Tributarios % PIB 0,9% 1,5% 0,7% 0,5% 0,8% 0,8% 0,8% Ingresos Tributarios % Ingresos Corrientes 76% 83% 79% 87% 79% 83% 75% Impuestos al ocio % Ingresos Corrientes 55% 51% 47% 55% 45% 41% 38% Fuente: Cálculos autor Gráfica 1- Composición del ingreso tributario departamental 17 10,000,000 9,000,000 8,000,000 7,000,000 6,000,000 5,000,000 4,000,000 3,000,000 2,000,000 1,000,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Cerveza Licores Cigarrillos Registro y anotación Vehiculos automotores Sobretasa a la gasolina Otros Fuente: Estimaciones autor Asimismo, al analizar la composición de los ingresos departamentales vemos como los propios o corrientes explican el 30% del total. La Gráfica 2 nos muestra así, que son las transferencias intergubernamentales y las regalías las fuentes de recursos con mayor peso. Gráfica 2 – Composición de los ingresos departamentales 100% 90% 80% 70% Regalías 60% Transferencias 50% 40% Cofinanciacion 30% No tributarios 20% Tributarios 10% 0% 2006 2013 2000 2001 2002 2003 2004 2005 2007 2008 2009 2010 2011 2012 2014 2015 2016 2017 Fuente: Estimaciones autor Grafica 3- Contribución de los ingresos departamentales y municipales al PIB Nacional 18 3.0% 2.5% 2.0% Porcentajes (%) 1.5% 1.0% 0.5% 0.0% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Municipal Departamental Fuente. Estimaciones autor Transferencias Intergubernamentales Generales de la nación Los ingresos totales de municipios y departamentos cuentan con una alta participación de transferencias nacionales; estas se dividen en transferencias generales y en regalías. Actualmente ambas funcionan como sistema, aunque en el pasado se distribuían a partir de un porcentaje fijo del presupuesto general de la nación. A continuación, se presentará la distribución y asignación actual del presupuesto en cada uno de los dos sistemas. Sistema General de Participaciones Antes de 1993 existía el situado fiscal para los departamentos. Este era un porcentaje de los ingresos corrientes de la Nación que se distribuía en su mayoría por población (85%) y el 15% como una alícuota igual para todos los departamentos. Estos recursos financiaban mayoritariamente las nóminas de los sectores de salud y educación. Igualmente, se habían creado participaciones de los municipios del anterior impuesto a las ventas y posteriormente del impuesto al valor agregado. La CN de 1991 decidió unir las dos transferencias mencionadas y aumentar el monto de las mismas, en un proceso gradual, se aumentaron hasta llegar al 42.5% de los ingresos corrientes de la nación en el año 2002. La ley 60 de 1993 distribuyó estas transferencias, pero es solo con la expedición de la ley 715 que se crea el sistema general de participaciones (SGP). Este se divide en 3 grandes partidas con destinación sectorial: salud (24%), educación (56%) y agua y saneamiento básico (5%). Estas se distribuyen entre departamentos y municipios. Además, se creó una nueva participación llamada propósito general que corresponde al 11% de las transferencias destinada a municipios y distritos. Estos recursos favorecen principalmente a los municipios de menos de 25.000 habitantes y son de libre asignación y de libre inversión. La gráfica 4 nos muestra 19 la distribución de los recursos del SGP entre asignaciones especiales y sectores y la tabla 6 detalla los criterios utilizados para su distribución a departamentos y municipios.10 Gráfica 4 – Distribución del SGP entre sectores y asignaciones especiales Fuente: Departamento Nacional de Planeación 10Fedesarrollo. 2017. APOYO AL DNP EN LA PROPUESTA DE AJUSTE A LA DISTRIBUCIÓN Y DESTINACIÓN DE LA PARTICIPACIÓN DE PROPÓSITO GENERAL Y LAS ASIGNACIONES ESPECIALES DEL SISTEMA GENERAL DE PARTICIPACIONES SGP. Bogotá. 20 Tabla 6 - Criterios para la Distribución de los recursos del SGP Asignaciones sectoriales – 96% Asignaciones Especiales – 4% Prestación de servicios (Entidades 100% según población Educación Territoriales Certificadas*) Resguardos indígena de cada resguardo (58.5%) Calidad Matrícula y calidad indígenas como proporción de la gratuidad (distritos, municipios y (0.52%) población total que habita áreas no municipalizadas) en resguardos Régimen subsidiado (87%) – (municipios y distritos). Salud Salud pública (10%) – (dep/tos Alimentación Equidad (95%) (24.5%) (45%) y municipios (55%) escolar Eficiencia (5%) Prestación de servicios y subsidio a (0.52%) (Distritos y municipios) la oferta (3%) – (departamentos, distritos y municipios certificados) Déficit de cobertura (35%) 100% en proporción de los Agua Población atendida y balance del Municipios kilómetros de ribera potable y esquema solidario (30%) ribereños del compartida con el Río saneamiento Esfuerzo ampliación cobertura (5%) Río Grande de la Magdalena básico Pobreza (20%) Magdalena por cada municipio (5.4%) Eficiencia gestión sectorial (10 %) (0.08%) (municipios y departamentos) ▪ Municipios menores de 25.000 En proporción al pasivo habitantes (17%): pensional (30%) Pobreza relativa (60%) Fondo de En proporción a la Propósito Población (40%) pensiones participación en las 12/12 general ▪ Todos los municipios (83%): territoriales – PG vigencia anterior (11.6%) Pobreza relativa (40%) FONPET diferentes a AE (30%) Población (40%) (2.9%) NBI (30%) Eficiencia fiscal (10%) Población (10%) Eficiencia administrativa (10%) Fuente: Departamento Nacional de Planeación Vemos como la ley 715 además creó unas asignaciones especiales, que son el 4% del total del SGP que se destinan así: resguardos indígenas (0.52%), alimentación escolar (0.5%), municipios ribereños (0.08%) y para un fondo de pensiones territoriales FONPET (2.9%). Estos recursos tienen condicionado su uso. Los de los resguardos indígenas se distribuyen según la cantidad de habitantes y solamente pueden ser invertidos en vivienda, desarrollo, salud y educación. La alimentación escolar se asigna de acuerdo con la matrícula oficial de los establecimientos estatales y debe gastarse en compra de alimentos, contratación de personal, transporte, dotación y contratación de terceros. Los recursos de los 111 municipios ribereños del Río Magdalena deben ejecutarse en proyectos de reforestación y revegetalización y en reducir y controlar la erosión del suelo. A su vez, los recursos destinados al FONPET se asignan según el pasivo pensional y solo pueden usarse en el pago de pensiones. Debe mencionarse también que para el gobierno nacional no fue posible cumplir con el monto de las transferencias definido por la nueva constitución. Esto es por razones fiscales pues era muy alta la participación de estas transferencias en los ingresos corrientes de la Nación. Esto obligó la expedición de dos Actos Legislativos que reformaron el monto transferido. Los Actos Legislativos 21 01 del 2001 y el 4 de 2007 modificaron los artículos 356 y 357 de la Constitución Política, estos establecieron periodos y tasas de crecimiento anual del SGP por periodos de ocho años. Por ejemplo, en los años 2008 y 2009 el monto SGP creció con la tasa de inflación más un 4% de crecimiento real y, a partir del año 2010 estas transferencias crecieron con la inflación más una tasa de crecimiento real del 3%. En resumen, se decidió que las transferencias crecen en términos reales. Con esto se evitó que las mismas sean un porcentaje fijo de los ingresos corrientes de la Nación. Recientemente desde el Ministerio de Hacienda se destacó la importancia del SGP en el Presupuesto General de la Nación. En el último año los recursos del SGP crecen a una tasa del 12,2% y su aumento es de $4,5 billones. Esta transferencia intergubernamental pesa más de 4 puntos del PIB. Por último, debe mencionarse que hay en discusión un proyecto de ley que reforma la distribución del SGP; se busca reducir la asignación al FONPET y aumentar los recursos destinados a los resguardos indígenas y a la educación. Sistema General de Regalías El cobro de regalías por la explotación de los recursos naturales en Colombia existe desde hace 30 años. Inicialmente, estas se dividían en regalías directas, entre los departamentos y municipios productores, y las regalías indirectas, que eran administradas por el Fondo Nacional de Regalías. Por su diseño, las regalías se concentraban en pocos territorios productores, unos 4 departamentos y 25 municipios recibían un monto muy alto de regalías, en pocas palabras, tenían una bonanza fiscal. La mayoría de estas regalías se giraban directamente a departamentos y municipios y no existía un mayor control del gasto. Para superar estos problemas, se aprobó la ley 1530 del 2012 que crea el Sistema General de Regalías (SGR) y modifica en forma radical la distribución y el acceso a estos recursos por parte de los departamentos y municipios. En resumen, se disminuyeron en mucho las regalías directas y estas se distribuyen con criterios de población y pobreza que hace que lleguen a todos los departamentos y municipios. Además, se crea un sistema de distribución de los recursos en 5 fondos, 2 de ahorro y 3 de inversión, a los cuales sólo se pueden acceder a ellos con la presentación de proyectos de inversión. La aprobación de estos proyectos se hace en unos Órganos Colegiados de Administración Decisión (OCADs) en el cual tienen asiento los gobiernos territoriales y el gobierno nacional. Hay OCADs municipales, departamentales y regionales. Algunos han sido eficientes, pero en otros hay una gran cantidad de proyectos no aprobados o pendiente de decisión. Con esto se hace muy lenta la ejecución de los recursos. El nuevo sistema de fondos del Sistema General de Regalías (SGR) creó los siguientes fondos: 22 1. Fondo de Desarrollo Regional para mejorar la competitividad de la economía. Se distribuyen los recursos por los criterios de población y pobreza. Este último medido por el índice de necesidades básicas insatisfechas. 2. Fondo de Compensación Regional. Se distribuye en un 60% para los departamentos y el 40% restante para municipios y, los criterios para la asignación son también los de población y pobreza. Se incluye igualmente el desempleo. Esta es la fórmula: 3. Fondo de Ciencia, Tecnología e Innovación- FCTel; los departamentos participan en proporción al Fondo de Desarrollo Regional y Fondo de Compensación Regional. 4. Fondo de Ahorro para el pago de pensiones públicas de los territorios (FONPET). Se distribuye entre las entidades territoriales de acuerdo con el monto de los pasivos pensionales de cada departamento o municipio. 5. Fondo de Ahorro y Estabilización. Este se estima como un valor residual derivado del porcentaje que obtienen los departamentos en los fondos de desarrollo regional, compensación regional, FONPET y Ciencia, Tecnología e Innovación. 23 Además, recientemente fue creado un fondo destinado a la financiación de los acuerdos de Paz con las FARC. Debe mencionarse que, hay en discusión una reforma de la ley 1530 que busca aumentar nuevamente las regalías directas a los territorios en donde se explotan los recursos naturales no renovables. También, se propone eliminar los OCADs pues los mandatarios territoriales quieren tener autonomía en la decisión del gasto de estos recursos. Es muy probable que esta reforma se apruebe, pero no veo posible que las regalías directas vuelvan a los niveles de antes. Gráfica 5 - Distribución entre Fondos de las Regalías del SGR Fuente: Fedesarrollo 2018 Gráfica 6 - Distribucion de las regalías antes y despues de la creación del Sistema General de Regalias Antes del SGR 24 Después del SGR 25 Deuda y sostenibilidad fiscal Debe mencionarse que el proceso de descentralización no estuvo ausente de problemas financieros y que, durante la década la de los años noventa, una buena parte de los 32 departamentos y de las grandes ciudades se endeudó y no pudieron cumplir con el pago de la deuda. Esto obligó a que el gobierno nacional rescatara a algunos gobiernos subnacionales respaldando estas deudas (bail outs) y logrando que las mismas se refinanciaran. Como consecuencia, se expidió un marco normativo rígido que hace casi muy restrictivo el endeudamiento de los territorios y que propende ante todo por finanzas sanas y sostenibles en el largo plazo. El primer avance normativo en el camino de una mayor sostenibilidad fiscal fue la Ley 358 de 1997 que reglamenta el acceso al endeudamiento de las entidades territoriales. Se crea un conocido sistema de semáforos para los departamentos y municipios. Para lo anterior se tiene en cuenta dos criterios, el de liquidez que estima la relación entre intereses de la deuda y el ahorro operacional y define que este no debe superar el 40%. Este indicador no incluye las amortizaciones de la deuda efectuadas en el periodo. El segundo criterio es de solvencia que fija el 60% como el límite de la relación del saldo de la deuda y los ingresos corrientes. A partir de estos dos criterios, los territorios que cumplan con las dos condiciones pueden endeudarse libremente (semáforo verde). Si solo se cumple un criterio pueden endeudarse con previa consulta de la Dirección de Asuntos Fiscales (DAF) del Ministerio de Hacienda (semáforo amarillo). Por último, si no se cumple con ninguno de los criterios, no pueden endeudarse (semáforo rojo). La misma ley permite que los territorios con problemas fiscales y financieros, no solo de deuda, adelantar un plan de desempeño con la Dirección de Apoyo Fiscal (DAF) del ministerio de Hacienda. Estos convenios fijan metas ortodoxas de ajuste del gasto y aumento de los ingresos. Ha sido una política utilizada con buenos resultados. Posteriormente la legislación avanzó con la Ley 617 del 2000 en varias dimensiones. Primero clasifico a los departamentos y municipios según su población e ingresos propios con el fin de controlar la creación de entidades territoriales. Segundo, reglamentó que el funcionamiento debe financiarse con los recursos propios, esto con el objetivo de un mayor control sobre el gasto. Para esto se definió un límite del gasto de funcionamiento cómo % de sus recursos propios. De acuerdo con la clasificación departamental o municipal; este límite se encuentra entre el 50% y el 70% para los departamentos y hasta el 80% para los municipios. Asimismo, en el año 2003 la ley 819 hace obligatoria la expedición de los marcos fiscales de mediano plazo para municipios y departamentos. Se busca garantizar la sostenibilidad de la deuda. Los marcos fiscales deben asegurar que las entidades territoriales generan superávit primario, lo que les permite pagar los intereses de la deuda o financiar parte de su inversión con sus ingresos corrientes. La ley igualmente analiza si el saldo de la deuda es muy alto o bajo de acuerdo estos ingresos corrientes. Además, para cada vigencia debe presentarse un plan financiero específico para asegurar el control y ejecución del presupuesto articulado a las metas de planeación. 26 Adicionalmente la institucionalidad del gobierno nacional creó herramientas para fomentar la responsabilidad fiscal. El Departamento Nacional de Planeación creó el índice de desempeño fiscal que califica el desempeño de los departamentos y municipios, de acuerdo con las siguientes variables: Tabla 7- Indicador de Desempeño Fiscal – Criterios de medición Variable Composición (%) Autofinanciación del funcionamiento Gasto funcionamiento/ Ingresos corrientes. Magnitud de la deuda Servicio de la deuda / ingresos disponibles Dependencia de las transferencias Transferencias + Regalías / ingresos totales Dependencia de los recursos propios Ingresos tributarios / ingresos corrientes Magnitud de la inversión Inversión / gasto total Capacidad de ahorro Ahorro corriente / ingresos corrientes De esta manera, los municipios y departamentos son calificados todos los años y pueden situarse en una de las siguientes categorías de desempeño: deterioro, riesgo, vulnerable, sostenibles y solvente. Este índice ha posicionado el desempeño fiscal como un logro para las administraciones de departamentos y municipios y además ha permitido un control nacional a lo largo del tiempo de las finanzas de los territorios. A modo de ejemplo, la Tabla 8 muestra la calificación de todos los departamentos del desempeño fiscal del año 2017. Debe mencionarse, que estas calificaciones han mejorado a lo largo de los últimos 10 años y además hay convergencia en los resultados. Cada vez hay menos dispersión entre los mejores y peores calificados. 27 Tabla 8 DESEMPEÑO FISCAL DE LOS DEPARTAMENTOS 2017 Porcentaje de Porcentaje de Porcentaje de Porcentaje del gasto Indicador de Posición ingresos corrientes Respaldo de la ingresos que ingresos ctes que Capacidad de Rango Departamento total destinado a desempeño 2017 a nivel destinados a deuda 2/ corresponden a corresponden a ahorro 6/ Clasificación inversión 5/ Fiscal 7/ nacional funcionamiento 1/ transferencias 3/ recursos propios 4/ VALLE 48,1 0,23 34,46 87,89 77,63 56,58 81,23 1 L MAGDALENA 36,1 0,00 55,71 96,45 87,59 54,24 80,43 2 L NARIÑO 38,7 0,00 56,26 90,55 90,24 53,64 79,69 3 S ATLANTICO 29,9 7,47 35,88 71,82 82,15 66,21 79,47 4 S RISARALDA 41,6 2,08 53,44 80,19 83,83 64,77 78,88 5 S CUNDINAMARCA 43 15,43 36,36 82,65 75,85 55,87 77,10 6 S META 52,8 5,32 67,36 92,02 86,19 54,89 76,74 7 S ANTIOQUIA 40,4 7,44 33,13 79,79 70,30 49,29 76,47 8 S BOYACA 51,4 3,92 62,69 89,22 86,37 49,41 76,40 9 S SUCRE 54,5 2,58 79,07 95,14 90,06 53,26 76,14 10 S SANTANDER 37,5 7,51 45,53 90,60 76,54 42,46 76,09 11 S CORDOBA 46,8 6,16 66,46 95,82 89,22 40,28 75,45 12 S NORTE SANTANDER 44,9 4,21 66,50 92,71 85,18 41,92 74,85 13 S QUINDIO 51,8 5,01 47,43 82,69 73,72 44,10 74,68 14 S CAQUETA 59,5 0,00 65,05 80,26 90,77 41,09 74,51 15 S CALDAS 57,2 7,09 53,69 85,24 79,32 42,76 74,42 16 S HUILA 59,2 2,65 62,06 85,55 83,37 38,80 73,83 17 S CESAR 52,6 9,53 74,43 90,99 90,02 42,39 73,24 18 S TOLIMA 54,3 9,24 55,79 79,72 84,01 36,09 72,46 19 S CAUCA 35,8 3,95 77,47 73,08 91,04 46,39 71,52 20 S BOLIVAR 66,4 1,22 59,08 78,83 83,53 46,82 70,59 21 S GUAVIARE 61,9 1,28 69,75 52,88 86,94 46,74 69,26 22 V CASANARE 68,7 11,17 77,33 77,96 87,20 35,38 68,68 23 V GUAINIA 41,3 0,00 75,36 35,37 89,67 58,76 68,07 24 V VAUPES 42,4 0,00 72,03 28,87 88,46 54,52 66,64 25 V GUAJIRA 70,6 41,81 79,05 95,15 88,60 25,83 64,45 26 V SAN ANDRES 34,5 0,00 51,00 18,34 61,82 52,03 63,53 27 V PUTUMAYO 60,4 0,00 82,24 55,42 87,95 19,07 63,37 28 V CHOCO 81,6 2,91 77,45 95,69 87,65 10,14 62,41 29 V VICHADA 66,5 0,75 77,34 26,16 86,97 38,33 62,23 30 V AMAZONAS 70,2 0,00 75,45 37,09 85,59 26,15 62,12 31 V ARAUCA 92 2,19 80,99 68,47 91,34 37,19 56,75 32 R Fuente: Dirección de Desarrollo Territorial Sostenible - DNP 28 Balance de la Descentralización en Colombia: Ajustes sobre la marcha En Colombia se dieron grandes cambios en materia fiscal y política en favor de las autoridades territoriales. Con la Constitución Nacional de 1991 se dio una redistribución del poder político entre niveles de gobierno. Este cambio inclusive afectó la estructura de los partidos políticos. Esto fue importante puesto que el gasto territorial disminuyó desde la década de los años sesenta y por casi 20 años, la gestión de lo público estuvo en manos del gobierno nacional. La descentralización, en un país de regiones, era inevitable. Así, tras estos cambios políticos, se fortaleció el proceso de descentralización en Colombia, que había comenzado en la década de los años ochenta. La gran decisión fue que se coexistieran tres niveles de gobierno con igual importancia. Si en un comienzo la descentralización fue municipal, rápidamente se aceptó que los departamentos son igual de importantes. Esta decisión fue acertada pues los departamentos tienen una gran tradición política, las gobernaciones siempre han sido importantes y la elección popular las fortaleció. Asimismo, la nueva constitución modificó los procesos de planeación y presupuestación y los articuló con la vida política. Los programas de gobierno de los políticos son la base para la planeación y presupuestación de los recursos públicos. Al mismo tiempo, se reconoció la importancia de la participación política y ciudadana en la planeación y ejecución de los recursos públicos. Este proceso de planeación – presupuestación detallado y compulsivo también contribuyó a atar las metas políticas con la responsabilidad fiscal de nuevos mandatarios. No se puede presupuestar lo que no está financiado. Sin embargo, los primeros años de la descentralización, 90´s, generaron una crisis fiscal que obligó a la expedición de una legislación ortodoxa de responsabilidad fiscal con el objetivo de tener finanzas sanas en el largo plazo. En resumen, el gasto y el endeudamiento territorial se elevaron a niveles no sostenibles en el largo plazo. El gobierno nacional demostró tener capacidad para ajustar el marco normativo y la gestión para hacer un seguimiento y control de las finanzas de los territorios. En pocos años se crearon nuevas instituciones públicas, direcciones en ministerios y el DNP, que controlan a los territorios e impulsa la buena gestión territorial. Lo anterior se complementó con un marco normativo estricto alrededor del control del endeudamiento y un buen desempeño fiscal. Para ello, también se modificó la distribución de los sistemas de transferencias con el objetivo de garantizar un mejor reparto y una inversión alta en los sectores sociales. De otra parte, al estudiar los ingresos propios de los territorios se encontró que estos han crecido en forma sostenida desde el 2000 en adelante. Sin embargo, los departamentos han tenido un crecimiento más discreto, sobre todo si compara con los municipios. Esto se explica porque las bases las bases gravables departamentales son inelásticas y el número de impuestos es mucho menor. Igualmente, los sistemas de transferencias aumentaron notablemente los ingresos para los territorios. Vimos cómo fue necesario reformar la C. N. de 1991 mediante dos Actos Legislativos 29 para ralentizar el crecimiento anual de las transferencias; sin embargo, estas todavía crecen en términos reales. Asimismo, un proceso similar se tuvo con las regalías producto de la explotación de los recursos naturales no renovables. La creación del SGR buscó una mejor distribución de estos recursos entre los territorios y un mayor control de las inversiones financiadas con estos recursos. Por último, vemos como el ajuste de estos sistemas es periódico y en la actual coyuntura es probable que en ambos sistemas de transferencias ajusten los criterios de distribución de sus recursos y la manera como deben invertirse. Un balance después de más de 30 de iniciada la descentralización nos muestra que Colombia tiene departamentos fuertes en lo político, estables en lo fiscal y algunos con gran capacidad para desarrollar políticas sociales. Sin embargo, no hay una homogeneidad. Hay aún departamentos débiles en lo institucional y financiero; en particular los que son grandes en territorio y poco poblados. Finalmente, la fortaleza de esta compleja transición del poder político y recursos entre niveles gobierno, se explica por la capacidad de ajuste sobre la marcha. La experiencia colombiana demuestra la necesidad de tener flexibilidad para corregir errores y retomar el rumbo en el corto plazo. Estos ajustes, sin embargo, son cada vez más difíciles puesto que los gobernadores y alcaldes son cada vez más fuertes y tienen gran peso político. No negocian su autonomía. La experiencia reciente de Colombia sobre la gestión de los gobiernos intermedios deja varias enseñanzas: • En primer lugar, las autoridades de Chile deben tener en claro que al momento de tener gobernadores electos por voto popular la relación de poder se modifica en mucho. Esto significa que los nuevos gobernadores, serán mas fuertes y por lo tanto sus demandas financieras por diversos motivos serán mayores. Entre estos se encuentra la necesidad de cumplir con sus propuestas de campaña que los obliga a ejecutar gasto social o proyectos de infraestructura. En este contexto veo difícil que se pueda mantener el modelo actual de cogobierno con la Nación. La no entrega de mayor autonomía y mayores recursos puede ser un limitante para el desarrollo político de los gobernadores. • Segundo, Es necesario que en Chile el ordenamiento territorial tenga nuevas posibilidades como las áreas metropolitanas, distritos, asociaciones de municipios o departamentos, regiones administrativas de planeación e inclusive contratos plan. Un amplio abanico de posibilidades permite monitorear a lo largo del tiempo cuales de estas nuevas figuras fueron útiles o no. Esto significa que la estructura rígida de gobierno nacional y subnacionales es limitada. Hay muchos problemas y retos regionales que necesitan el concurso de autoridades de varios niveles de gobierno e inclusive del sector privado. • Tercero, la experiencia muestra que es inevitable disponer de mayores recursos para financiar a un estado que descentraliza sus funciones y competencias entre varios niveles de gobierno. En Colombia esto significó un aumento importante de las transferencias intergubernamentales a través de dos sistemas: el SGP y SGR. Esto puede ser una buena decisión siempre y cuando la administración de estos recursos sea eficiente y transparente. En general, estos sistemas de transferencias financian muy poco gasto corriente y se concentran en inversiones en infraestructura o gasto social. En Colombia, las nóminas de los sectores de salud y educación 30 se financian en gran parte con estos recursos. Asimismo, esto implica que hay que definir para cada sector del gasto la distribución de recursos y funciones entre niveles de gobierno. • Cuarto, preocupa que no estén definidas fuentes de recursos estables para los gobiernos intermedios en Chile. Como vimos en Colombia estos gobiernos dependen en gran medida de los impuestos al ocio como los impuestos a las loterías, licores, cerveza, cigarrillos y juegos de suerte y azar. Estas son fuentes poco dinámicas, pero son una garantía de ingresos en el largo plazo. Estos recursos son los que han garantizado el funcionamiento de los departamentos desde mucho tiempo atrás. Chile debe identificar posibles fuentes de recursos que puedan ser cedidas y / o compartidas ente niveles de gobierno. En todo caso es mejor que los departamentos administren sus propios impuestos o gravámenes, de lo contrario su fortaleza política se ve comprometida. • Quinto, la experiencia del proceso de descentralización colombiano nos muestra que los problemas no son solo de distribución de funciones y recursos entre gobiernos. Se necesita igualmente desarrollar una institucionalidad que garantice tener los espacios coordinación entre el gobierno nacional y los subnacionales. En Colombia hay direcciones en el DNP y del ministerio de Hacienda encargadas de esta función. Entre estas tareas se han convertido en muy importantes los sistemas de información y de ejecución presupuestales y contables de obligatorio reporte periódico. Es mejor manera de hacer seguimiento a la gestión de los territorios. • Sexto, es necesario ajustar la normatividad a las nuevas condiciones creadas por los cambios políticos. Es claro, la mayor autonomía no significa bajo compromiso en materia financiera. Hay que expedir leyes de responsabilidad fiscal o capítulos para los territorios en las leyes existentes. Se les debe imponer límites al gasto de funcionamiento, controlar el crecimiento de las nóminas y por supuesto el acceso al endeudamiento. Por ejemplo, los gobiernos subnacionales que no generen superávit corriente no pueden endeudarse. Colombia y otros países tienen este tipo de leyes con muy buenos resultados. • Séptimo y último, al que tener presente que estos procesos de descentralización política, administrativa y financiera son mucho más complejos de los que inicialmente se estima. Demandan mucho tiempo, requieren de una visión estratégica, se debe generar un mayor volumen de recursos y distribuirlo apropiadamente entre niveles de gobierno, rediseñar el ordenamiento territorial y el reparto de funciones recursos y de pensar en nuevas posibilidades de organización en el territorio. Además, hay que tener la capacidad de ajustar sobre la marcha. Muy buena parte de la nueva legislación y de la reorganización institucional en Colombia, se dio a la luz de los problemas que surgieron con el proceso de descentralización ya en ejecución 31 Bibliografía: Acosta O. y Bird, R. El dilema de la Descentralización en Colombia. En Bases para una Reforma Tributaria Estructural en Colombia. Banco de la República – Fedesarrollo. Bogotá. 2005. Fedesarrollo. 2017. Apoyo al DNP en la propuesta de ajuste a la distribución y destinación de la participación de propósito general y las asignaciones especiales del sistema general de participaciones SGP. Bogotá. Daughters, R. Harper, L. Reformas de descentralización fiscal y política. En Lora, Eduardo (editor). El estado de las reformas del Estado en América Latina. BID. Washington. 2007. Dirección de Apoyo Fiscal –DAF- 10 años de Transformación Fiscal Territorial en Colombia 1998 - 2008. DAF- Ministerio de Hacienda. Bogotá. Mayo de 2009. DNP (2002). Evaluación de la descentralización municipal en Colombia: balance de una década. Bogotá. Duque, N. (2007). El departamento en la descentralización político-administrativa en Colombia. Bogotá: Escuela Superior de Administración Pública. Econometría Consultores. (2014). Evaluación de operación del sistema general de participaciones. Bogotá Estupiñan, L. (2012). Desequilibrios territoriales. Bogotá: Universidad del Rosario. Falletti, Tulia. A sequential theory of decentralization and its effects on the intergovernmental balance of power: Latin American cases in comparative perspective. Working Paper # 314. Kellog Institute. July 2004. Federación Colombiana de Municipios (2009). Tributación municipal para la equidad y el desarrollo. Memorias Foro nacional de autoridades locales sept. 2008. GTZ- Federación Colombiana de Municipios. Bogotá. Maldonado, A. Los límites de la descentralización territorial: el caso de Colombia 1991- 2008. Tesis de Doctorado. Universidad Complutense. Madrid. 2011. Núñez, J. Zapata, J. 2018. Evaluación de Impacto del Sistema General de Regalías. Fedesarrollo – evaluación para el DNP. Bogotá. Pening, J. P. El modelo de descentralización en Colombia. Capítulo de tesis doctorado en Desarrollo. Universidad de Lovaina. PGR. Guía básica de tributos municipales. Bogotá. 2018. Piza, J. R. (2008). El régimen fiscal en la Constitución de 1991. Revista Derecho del Estado, núm. 21, diciembre de 2008. Universidad Externado. Bogotá. Porto, A., Pineda, C. R. y Eguino, H. (2018). Descentralización y autonomía fiscal subnacional en América Latina: Panorama comparado de Brasil, Colombia, México y Perú. BID. Restrepo, D. (2014). Procesos de Descentralización en Bolivia y Colombia, 1980 – 2005 Una propuesta de economía política espacial comparada. Universidad Nacional de Colombia - Facultad de Ciencias Humanas - Departamento de Historia. Tesis de Grado. Bogotá. 32 Sánchez, A. & Herrera, D. Revisión histórica del proceso de descentralización administrativa y sus restos en Colombia, de 1985 a 2016. Universidad Santo Tomas. Tesis. Bogotá. 2016. Sánchez, F., & Zenteno, J. (2010). Descentralización y desempeño fiscal subnacional - el caso de Colombia. En Descentralización y sostenibilidad fiscal subnacional: los casos de Colombia y Perú (págs. 47-97). BID. Valencia-Tello, D; Karam De Chueiri, V. Antecedentes de la descentralización territorial en Colombia. (2014); 23(1):171-194. DOI: 10.5294/DIKA.2014.23.1.7 Velásquez, F. y González, E. ¿Qué ha pasado con la participación ciudadana en Colombia? CIDER – Uniandes y Fundación Corona. Bogotá. 2003. Ver Perry et. al. (2009) Propuestas para el fortalecimiento Fiscal de los Departamentos. Fedesarrollo - Federación Nacional de Departamentos. Bogotá. Zambrano Pantoja, F. El contexto histórico de la descentralización territorial en Colombia. Revista Nómadas núm. 3. Universidad Central. Bogotá. 1995. Zapata, G. J.G. Reforma radical en el Estado de Santander, 1850-1885. Bogotá: Universidad del Rosario, 2015. 33 Fiscal decentralization of regional governments – the case of Indonesia prepared under the World Bank TA RAS for Chile1 1 The Indonesia note was written by Gabriele (Gabe) Ferrazzi 1 Contents Introduction ................................................................................................................................................... 1 Chapter 1. Indonesia and its regional government at a glance ...................................................................... 2 Chapter 2. Indonesia’s Big Bang decentralization ........................................................................................ 4 Chapter 3. Indonesia’s choices in functional assignment ............................................................................. 8 Chapter 4. Functions of regional government ............................................................................................. 12 Chapter 5. Regional government revenues ................................................................................................. 13 Chapter 6. Regional government borrowing ............................................................................................... 20 Chapter 7. Regional government budget process........................................................................................ 21 Chapter 8. Effectiveness of regional government spending........................................................................ 23 Chapter 9. Improvement and accountability in regional government ......................................................... 25 Chapter 10. Lessons and implications for other countries .......................................................................... 28 References ................................................................................................................................................... 30 Appendix 1: Comparative view of Indonesia and Chile ............................................................................. 32 Appendix 2: Example of education function assignment in Law 23/2014 on Regional Government ........ 33 Appendix 3: Changes in regional taxes over time/legislation in Indonesia ................................................ 34 Appendix 4: Essential features of decentralization modalities ................................................................... 35 i Introduction2 1. Spurred by political necessity, in 1999 Indonesia embarked on what has become known as its “big bang” decentralization (Hofman and Kaiser 2004, Lamont and Imansyah 2012, Decentralization Support Facility 2012, Nasution, 2016). From a heavily centralized country it quickly became one of the world’s most decentralized countries, with the two levels of regional government (provinces and district/city levels) becoming responsible for about half of total public expenditure. 2. Indonesia’s approach to decentralization holds lessons for other countries contemplating decentralization reforms. In some respects, Indonesia’s radical reforms were a success, surprising advisors of international finance institutions that called for a much more cautious approach. Aspects of the design and the political will and pressure for decentralization figured prominently in the reforms’ success. In other respects, the Big Bang was disruptive and has required considerable clean-up and retrofitting over the years. As might be expected, it is still a work in progress. Both the successes and pitfalls can be instructive to other countries. 3. Chile’s policy makers in particular can learn from Indonesia’s experience. The countries have some similarities (see Appendix 1 for a general statistical comparison), with the scale of the Indonesian district/city regional government approximating the scale of the regional governments in Chile. Although the province in Indonesia is considerably larger in scale, the efforts to suppress and then enlarge its intermediary role also offers lessons useful to the evolving regional government of Chile. The large variation in Chile’s regions mirrors that of Indonesia, making the latter’s attempts to bridge income and capacity disparities in the context of decentralization very relevant. Additionally, Indonesia’s attempt to build greater accountability to improve service delivery holds lessons that are relevant to regions regardless of their scale. 4. The relevance of the Indonesian experience for Chile is best analyzed via the essential features of decentralization and subnational governance. These essential features relate to choices made in functional assignment (architecture and modalities), own revenues, intergovernmental transfers, public financial management practices, and accountability mechanisms. Moreover, the lessons are more meaningful if they bridge to the specific circumstances and challenges of Chile’s situation. An effort is made in the case study to achieve this bridging, but the benefit of a wide-ranging and participatory dialogue within Chile is necessary to extract maximum benefit from any international experience. In line with these observations, the case study is organized to highlight some features of Indonesia and its regional government that can help in understanding the context and assessing relevance (Chapter 1). It then provides the essential changes made in the decentralization reforms of Indonesia (Chapter 2), followed by a deeper examination of choices made in functional assignment (Chapter 3) and the resulting distribution of functions (Chapter 4). Chapter 5 explain the regional revenues composition, followed by its borrowing powers/limitations (Chapter 6). The vertically integrated budget process is covered in Chapter 7. Evidence is offered in Chapter 8 for the effectiveness of regional government spending, and regional government performance is viewed more broadly through the lens of improvement and accountability incentives in Chapter 9. In the final Chapter 10, some lessons are drawn relevant to countries contemplating decentralization and specific suggestions are offered for Chilean policy-makers. 2 The author is grateful for the opportunities provided by the World Bank to become familiar with the Chilean context and policy information needs. The video session with Ministry of Finance officials in Chile was particularly useful. The sharing of recent information on the reforms in Indonesia provided some inspiration and data for the issues raised here. 1 Chapter 1. Indonesia and its regional government at a glance 5. Indonesia is a large and sprawling archipelagic middle-income country that is making strides in development and governance reforms. Indonesia’ 265 million inhabitants are 85% Moslem and is ethnically diverse, particularly in its outer islands. It is urbanizing quickly and maintaining a reasonable growth rate of over 5%; however not enough to fully absorb its graduates in the formal economy. Its official unemployment rate (Table 1) obscures the high level of informal economy and underemployment. Since embarking on reforms in 1999 it has made strides towards improving subnational governance. Table 1: Indonesia at a glance Population and Geography Economic and social data • Area: 1,904,569 km2 • GDP: US$ 1,022 billion (2018) • Population: ~ 265 million • GDP per capita: USD 3,871 (2018) • Density: 140 inhabitants/km2 • Real GDP growth: 5-6% since 2005 • Urban population: 55% of national population • Unemployment rate: 5.34% (2018) • Capital city: Jakarta (10 million; almost 4% of • Public debt: 30% of GDP (2018) national population) • Human Dev Index: 0.694 6. Subnational government in Indonesia has two levels of regional government and below these village governments. Both regional levels have intermediary roles. The larger regions are the provinces (34 units) and the smaller (second level of) regions comprise 416 districts and 98 cities. This second level of regions has roots in the Dutch colonial period. Provinces have come into their own in the independence period but in a tense relationship to the central government that factored into the political decision to favor the districts/cities in the eventual reforms of 1999. The population estimates for the Indonesian regions (Table 2) indicate that district/city regions are similar in scale to the Chilean regional government tier. However, both regional levels in Indonesia have an intermediary role for governments below them that can provide insights into the evolving intermediary role of Chilean regions toward the municipal level. The intermediary role of the district in relation to the villages is gaining in importance as a result of the additional resources and expanded role of villages in recent years (Box 1). Table 2: Subnational governments (elected) in Indonesia – with Chile comparison Average pop. Chile (est.) SNG level Head of administration Number (est.) Province Governor 34 7.9 million Region: District/City Bupati (Regent)/Mayor 416/98 500, 000 1,125,000 Village Village Head 75,244 1,800 Municipalities: 52,000 Source: Approximate calculations by author Box 1: The district links higher level government and village government Village government comprises village administration (village, head, secretary, other officials in charge of public service and financial affairs) and a village council. The village head is elected, and the village council 2 must be established through democratic means. The reforms of law 6/2014 greatly increased village revenue, with funds directly obtained from the national state budget, and also via the district budget. Villages may also raise their own revenue, for example through the exploitation of village land and enterprises. The expenditures of villages have focused on small infrastructure, but they are slowly expanding to encompass services, such as local health centers and early childhood education centers. Districts play a vital role in financing villages, through passing on a portion of their own transfers from the central government and revenues from selected local taxes and charges. The district, though sub-district offices, also guides village governance and development, hosting the annual cross-village development forums for instance. District politicians and planners also seek village input to expenditures managed at the district level that have a high impact on villagers, for example the location of health centers and schools. Districts also perform the financial administration that is needed for the central government to determine village allocations. The village head is expected to submit an annual financial report to the council and to the district government on the implementation of village governance. District government further influences village governance through its right to approve and terminate the appointment of village heads and council members. District heads must also evaluate draft village-level regulations and settle elections disputes. 7. Provincial and district/city level governments are diverse and have disparate development outcomes. Indonesian regions are diverse in their ethnic composition and natural endowments, with considerable income disparity per capita as a result. District and city regions in Indonesia have as much as a multiple of 50 between the lowest and highest incomes per capita, whereas the disparity in Chile is substantial but at most reaches a multiple of five times in GDP terms (OECD, 2018). Regional disparity has been shrinking in Chile in recent years whereas it shows no reduction in Indonesia. 8. Regional government in Indonesia sits within a larger national administration that is heavy on regulation but weak in governance. In contrast, the Chilean bureaucracy is reasonably efficient and clean. For example, on the corruption perception index Indonesia ranks 89 as compared to Chile at 27 (Transparency International, 2019). Indonesia, however, is making strides in conducting internal audit and prosecuting malpractice, reaching the regional level in recent years, placing many regional heads under investigation and in prison. 9. The resources afforded to provinces and districts/cities in Indonesia are considerable, with a high dependence on transfers revealed by the very low average own source revenue, especially for districts/cities (13%). The typical provincial and district/city government political and administrative structures and resources are described in Table 3. Provinces are more self-sufficient than districts/cities in terms of own revenues, reaching around over 50% of their budget on average. Table 3: Structures and resources of regional government Province District/city Elected Council of 35-100 members (directly Council of 20-50 members (directly political body elected) elected) Regional head (Governor) is directly elected Regional head (Bupati) is directly elected (since 2005); does not sit on the Council (since 2005); does not sit on the Council Bureaucracy 27-39 organizations of the provincial 27-49 organizations of the district/city government government Ave. 16,700 employees (full civil servants 2,000 - 10,000 employees (full civil and contract workers) servants and contract workers) Budget Annual budgets in $1-5 billion range Annual budgets in $50-500 million range Own source Own source revenue: Ave. 54.4% (2017) Own source revenue: 2%-78% (average revenues 13%) 3 Lower level About 15 districts/cities within province 50-300 villages/wards within the governments boundaries district/city boundaries Source: Ministry of Finance and Ministry of Home Affairs online data bases; Law 23/2014 on Regional Government Chapter 2. Indonesia’s Big Bang decentralization 10. Indonesia has a long history of centralization. Early kingdoms in Indonesia (e.g., Mataram) were very centralistic and patrimonial in their governance. In the colonial period, the Dutch rulers made use of this existing structure, ruling through Indonesian “regents” (equivalent to today’s districts). The Netherlands imposed federalism as a last-ditch effort to maintain its role post World War II, but the newly independent country soon after reverted to a unitary and centralized system. Under Soekarno’s “guided democracy” some early steps and experimentation in decentralization took place. This was followed by Soeharto’s New Order, characterized by centralism and neo-patrimonialism. His regime expanded deconcentration structures in the regions, but promised regional autonomy was never delivered. The most visible effort was a 1994 pilot project launched in 26 districts, but the set of functions transferred were limited and were not accompanied by corresponding resources. This pilot was overtaken by events; in 1998 Indonesia suffered the effects of the Asian Financial Crisis. The ensuing political turmoil, with regime change at the center, provided fertile ground for decentralization reforms. 11. Indonesia’s decentralization was shaped by the political tumult arising from the Asia Financial Crisis, and was driven by long-standing grievances regarding limited democracy, concentration of wealth and political centralization. The decentralization reforms were essentially regional elite driven, especially from resource-rich regions, and raised concerns about national disintegration. Hence the response was a political accommodation that was radical in content (deep devolution) but targeted to the district/city level regions to safeguard national integrity and the unitary state.3 A tentative discussion about federalism at this time was brief and truncated (Ferrazzi, 2000). Provincial governments, as autonomous governments, were given only a weak intermediary role toward lower levels of government. The Provincial Governor dual role was maintained, in relation to supervision of the district/city governments but this dual role was not clearly delineated at this time. 12. Decentralization reforms were radical and swift (earning the label Big Bang), raising risks that were managed in part by opting for a relatively clean transition from deconcentration to devolution. The 1999 laws on regional government and fiscal balance were drafted in a rushed and closed fashion by a handful of officials and selected academics (dubbed the Team of Seven) and were implemented fully by 2001. The functions transferred through Law 22/1999 on regional government (and its Government Regulation 25/2000 providing the detailed list) were extensive and comprehensive (sector wise). Resources somewhat consistent with the functions were part of the reform. Over 2 million central government staff were shifted to regional government status. Law 25/1999 on regional finances gave the regions 26% of the total net revenue of government (after revenue sharing) in a 90/10 split between the districts/cities and provinces.4 The dominant transfer 3Accommodations were made in the case of the provinces of Papua and Aceh, which were given a special form of autonomy to dampen separatist sentiments in 2001 and 2005 respectively. In 2002, under international pressure, the province of East Timor was allowed independence. 4 For 2019 this is reset slightly higher at 28.7% of net revenues, and with a slightly adjusted split as well, with provincial government receiving 14.1% and district/city governments 85.9%. The latter split presumably takes into account the recent recentralization of some functions from district/city government to provincial government. 4 mechanism was a block fund (General allocation Fund - DAU). Figure 1 shows how this radical reform was made organizationally manageable – by the relative simplicity of shifting from deconcentration to devolution structures in the regions. About 20,000 organizational units were transferred, along with staff and assets/documents (Fengler and Hoffman, 2009). But the public was hardly aware of the massive reorganization as only the office placards and staff uniforms indicated any change. The only central government units that remained in the regions were those that discharged central government functions (e.g., finance offices of treasury, national audit bodies, special technical/training units of sector agencies). Figure 1: Organizational restructuring in the regions: shift from deconcentration to devolution Source: Author 13. Indonesia adhered to “resources follows functions” in a hurried analytical process that was rather closed but included sector ministries. The approach to deciding functional assignment is in the process of being systematized in Asian decentralization efforts (Ferrazzi and Rohdewohld, 2017). Indonesia was able to adopt a rather hurried and closed process that nonetheless resulted in a meaningful reform; this distinguishes it from other countries that have applied more rigor in the method, but with less evident results in terms of the transfer of meaningful functions. The process had some rationality; it adopted and to some extent used criteria for assigning functions (Box 2). But the quick design was facilitated by the decision to undertake a visible and deep form of devolution; shifting central government units in the regions to the corresponding level of regional government. Hence the assessment of non-financial resources required to accompany the shift was straightforward; the shift naturally included the staff and assets in those organizations. The finances calculation was more difficult but used as an approximation the budgets that were associated with those organizations. There was not a perfectly clean division; a few functions being carried out in the shifted organizations were really functions that the central government wished to keep, so some disentangling was required, but this was minimal. It was out of this approximation that the government came to the aggregate pool for the transfer (26% of net revenue), which at that time was largely the general block transfer (DAU). Box 2: Criteria formally adopted in the assignment of functions in 1999-2014 reforms The criteria adopted in Indonesia are generally referred to as “principles” and are applied only to those functions (sectors and sub-sectors) that are to be further subdivided because they are to be shared (the law uses the term “concurrent” in this respect, although the usage is not in line with the international understanding 5 of concurrent, where the division of functions process has ended and still two or more levels of government are empowered to carry out the very same function). The principles were culled and adapted from international experience, with development partners playing a large role in exposing Indonesian officials to the relevant literature. The principles, or criteria as they are generally referred to internationally, were discussed quite deeply within the ministry leading the decentralization policy (The Ministry for Administrative Reform). Stakeholders or experts were not brought into the discussion. As explained in the Law 23/2014, the principles are as follows: Accountability: related to the proximity of the regional government to the impact of the function. Efficiency: related to the ability to extract the highest utility from the function. Externalities: related to the scope of the impact arising from the function Strategic national interest: related to maintaining the integrity and unity of the state. The principles were shared informally with the sector ministries, and several were given support in applying them. However, there was little checking on how thoroughly and consistently sector ministries applied the principles. One of the challenges in applying the principles is the way they were adapted; the above definitions veer somewhat from those adopted internationally (see Ferrazzi and Rohdewohld, 2017). For instance, externality is generally viewed as the impact felt by other jurisdictions in the discharge of a function. Notably missing is the principle/criteria of economies of scale. Source: Author’s elaboration. 14. While the larger political reforms ensured more contested regional government elections, decentralization lacked sufficient complementary mechanisms to ensure greater accountability to citizens and better overall governance. This stems in part from the fact that decentralization was political accommodation of regional elite (at least initially) and did not arise from any groundswell of citizen demand for improved service delivery. Decentralization reforms reinforced electoral reforms (local politicians had larger decisions to make and more resources) but the motivation for these related reforms had more to do with the regional elite wanting their turn to enjoy the benefits of development and what comes with power than a desire to improve governance. For this reason, decentralization also served to enlarge the patronage system, although it also destabilized it for a while, upsetting the informal market for promotions, contracts etc. 15. Resistance to decentralization did make itself felt, largely by sector ministries fighting a rear- guard battle, to retain funds and “their” laws/regulations. While the initial pressure was felt by all politicians to come through with deep reforms, some senior officials in certain ministries were not enthusiastic, or rightfully were concerned about some aspects of the reforms. As some predicted, the communication/reporting channels were disrupted as district/city regional heads (dubbed “little kings”) thumbed their nose at the central government with the weight of favourable constitutional and legal principles (ignoring some legislative and regulatory detail on their obligations). The resistance from sector ministries took the form of directives to regional government that referred to legal instruments that were not yet amended to conform to the new regional government framework.5 There was also some resistance from other stakeholders, with provincial governors concerned that their role as representatives of the central government was not sufficiently robust. Teachers also were wary of being managed by district/city governments. Some of these concerns dissipated over time, while others grew to fuel the 2004 and 2014 revisions of the regional government framework. 5These directives often clashed with those of the Ministry of Home Affairs, the institution charged with providing general guidance to subnational government and thus the source of many of the directives that sought to operationalize the new regional government law and government regulations. 6 16. Ministry/central agency turf battles also marred the development of the framework and consistency on the ground. The most serious turf battle was between the Ministry of Finance and the Ministry of Home Affairs. Guiding the fractious district/city governments called for a united front in the coordinating national agencies, but this took a long time to evolve, particularly on financial management rules and reporting. Over time the Ministry of Home Affairs (MoHA) and the Ministry of Finance (MoF) came to a better working relationship, with reduced overlap (e.g., on accounting packages required) and points of tension. MoHA focused more on budgeting and planning priorities, and MoF on intergovernmental transfers and financial reporting. Both of these institutions deferred to the National Development Planning Agency (Bappenas) on capital investment. This understanding was complicated, however, by the actions of the Ministry for State Reform, when, in 2009, it began to introduce its own requirements on subnational government service delivery and performance reporting. Even today, regional government is saddled with excessive and fragmented reporting requirements emanating from various national level agencies. These have had the effect of nearly paralyzing some regional government officials, who at the same time must contend with a more aggressive internal/external audit regime and Anti-corruption Agency (KPK). 17. The inherent risks of decentralization were managed through several policy decisions pertaining to the transition period and longer-term safeguards. Some of the safeguards were built into the system from the start, in an anticipatory fashion. Others were ad hoc fixes that were generated as implementation problems became more evident (Box 3). Box 3: Safeguards put in place to manage decentralization risk • The radical but simple model of decentralization (deconcentration to devolution) simplified the transfer of functions and resources, and reduced civil service resistance (people were not rotated to remote areas of the country). • The use of the historical cost of services (under the deconcentration stream) to approximate the new expenditure needs of regional government ensured budget affordability and aggregate fiscal discipline for the consolidated government sector. • Regional government borrowing was banned except through the central government (CG) and limits were set on borrowing by individual regional governments (RGs). • A hold harmless provision placated RGs that would lose out in the new block formula; this was particularly important for regions that received the bulk of the natural resource revenue sharing. This policy explains in part why decentralization was not revenue neutral. • CG continued to pay former central civil servants for a transitional period of five months, deducting the wage bill from the DAU transfers to the regions. • CG “shortchanged” RG in the transfer computation (which served to maintain some fiscal discipline) by disbursing transfers to the regions based on budgeted amounts rather than actual (until 2005).6 Even with this fudge, decentralization was found to be fiscally expansionary. • Minimum service standards were made explicit and placed in regulations (starting in 2003) to emphasize the basic service delivery obligations of RGs. Source: Author’s elaboration. 18. The deficiencies of the original decentralization design were in part corrected with two rounds of revisions. The 2004 revision allowed regional heads to be directly elected and provided a positive 6The approach may revert to budgeted amounts for 2019 as the central government contends that regional government is not well able to adjust to actual revenues and would do better to project their budgets based on budget estimates, giving more predictability. 7 list to replace the black box (badly formulated general competence) for district/city functions. The 2014 revisions recentralized senior secondary education and natural resource management from district/city to provincial government level (Figure 2).7 The latter changes in particular failed to adequately involve stakeholders, making the change rather unstable. Figure 2: Timeline of decentralization reforms in Indonesia Source: Author Chapter 3. Indonesia’s choices in functional assignment 19. Indonesia gave general competence powers to the district/city government level initially, but later reverted to a positive list architecture, embedded in an omnibus regulation (and more recently in law). The lack of certainty in the district/city government “black box” of the 1999 reforms hampered all actors; even district/city officials were not comfortable with it. Since the 2004 revision of the framework regional government law (Law 32/2004) a positive list has covered all levels (central, provincial and district/cities). The list has always been uniform across districts/cities. Only at provincial level was there any asymmetry, with five of the 34 provinces holding special autonomy/status. Indonesia has adopted the typology of obligatory versus discretionary functions but did not fully adopt the international meaning of these terms. Concurrent functions in Indonesia indicate which functions can be shared with regional government. These are further categorized as obligatory (with performance benchmarks applied, such as minimum service standards in the case of basic services) and optional, where less uniformity is expected. Optional functions are simply a term used to describe functions that are largely economic or cultural, and therefore need to be taken up in a flexible way that reflects local endowments and characteristics (as opposed to uniform service standards that must be met across all jurisdictions). Figure 3 captures the typology described above, including the “general government” functions of the President; these are undertaken by the Provincial Governor (and to some extent by the district/city regional heads) as deconcentrated tasks. Despite some quirks – in concurrent and optional definitions for instance - Indonesia has a reasonably well- developed regulatory approach to decentralization modalities (devolution, delegation, deconcentration) that largely accords with international definitions. 7 See Law 23/2014 on Regional Government. 8 Figure 3: Architecture and typology of governmental functions in Indonesia Source: Author, derived from Law 23/2014 on Regional Government 20. Provinces were given a weak intermediary role, and districts were confirmed in their strong intermediary role toward villages. The provincial support and supervision of district/city government was poorly delineated, and undermined by the notion that there was to be no hierarchy between the provincial and district/city levels, 8 resulting in the complex arrangement where hierarchy was achieved only through the dual role played by the Provincial Governor (Figure 4). 8 This non-hierarchical arrangement is rather unusual internationally. Intermediary level regions in federal systems generally have jurisdiction (e.g., Canada and United states) or significant influence over lower level subnational governments (e.g., India and Pakistan). Even countries with unitary systems tend to empower intermediary regional levels with respect to lower levels of government (e.g., the region in Italy with respect to the provinces and communes, and the province in the Philippines with respect to cities and municipalities). Allowing regional government this role can reduce the felt need of central government to create positions in regional government with a marked dual role, serving both the reginal government and the central government. It will then tend to establish its own structures separate from the regional government, to discharge deconcentrated tasks (e.g., the commissioner and prefect in Italy’s region and province respectively; the deconcentrated offices of the Department of Interior and Local Government in its own regional divisions). While there is little research focused on the comparative advantages of these constructions, it can be argued that a separate construction of central government in the regions provides greater clarity, and thus accountability, in regional government and central government roles. 9 Figure 4: Relationship between the two levels of regional government Source: Author’s elaboration. 21. Only in the case of the five special status/autonomy provinces was there asymmetric decentralization (Box 4) that included a stronger provincial role and a stronger hierarchy over component administrative/territorial units. In the same reforms, and revisions made over the years, the role of the provincial governor was developed to be the instrument for supervision of the district/city governments (see para. 15). The latter governments’ support and supervision role toward the village government was strengthened; the village law of 1974 was dropped in 1999 and relevant provisions were subsumed/amended under the Law 22/1999 on Regional Government. The district provides special funding to the villages and coordinates them through its deconcentrated Sub-district offices (Kecamatan) led by the sub-district head (Camat). Box 4: Five cases of extended or new asymmetric decentralization at provincial level • Jakarta: “Special Status” (city is given provincial status) due to its national importance as the capital. It gains the functions of both the province and district/city levels as there are no further subdivisions of the capital into elected governments. • Yogyakarta province “Special Status”: Sultan of Yogyakarta as unelected Governor and given additional land use rights. • Aceh province “Special law on its governance”: self-regarded as own nation (not fully colonized); given religious, economic (e.g. greater revenue cut of oil and gas) and other functions not given to other provinces. • Papua and Papua Barat provinces “Special Autonomy”: Papuans contest their entry into Indonesia; given traditional leadership, economic (special autonomy fund) and other functions not given to other provinces. Source: Author’s elaboration. 22. The deconcentration role of provincial governor is focused on the supervision of the district/city governments, but must unavoidably involve the autonomous provincial units in this supervision work (Figure 5). The dual role of the provincial governor was problematic from the start and has been slow to take effect. The need for this structure arises out of the choice of not making the provincial government (Council and Regional Head) formally hierarchical in relation to district/city government. 10 The provincial governor is instead asked to not only be the regional head, but to also add the role of representative of the Central Government, for the purpose of supervising the district/city government. Unfortunately, this role has not been well understood or fully utilized, in part because it was to be supplied with central government units at provincial level to implement deconcentrated tasks given to the Governor, but financial tightness on the part of the Ministry of Finance kept this additional bureaucracy from being realized. In recent framework revisions, the central government has settled for enlisting the provincial bureaucracy to also implement the Governor’s tasks deconcentrated from the central government. This compromise, a unified structure serving two masters (the region – with a regional council, and the central government), raises complications in accountability. 23. The functions architecture has lacked a workable mechanism for adjusting functions and harmonizing the legal framework. The legal placement/approach chosen (omnibus regulation and then law in 2014) worked against a harmonized and flexible legal framework. In the fractured political context where coalition governments resulted in near party control over sets of national institutions, it has been difficult to attain harmonization across legal streams, particularly between the RG legal stream (dominated by the Ministry of Home Affairs), the sector ministries, and the Ministry of Finance and National Development Planning agency (Bappenas). For many years, sector ministries held on to “their” laws and regulations, even when these were in conflict with the law/regulations governing RG. Figure 5: Dual role for provincial governor through deconcentration Source: Author, derived from Law 23/2014 on Regional Government 11 Chapter 4. Functions of regional government 24. The functions transferred in the 1999 reforms cover the entire governmental spectrum (social services, economic, environment, cultural, governance) and are listed in considerable detail in the appendix of Law 23/2014 on Regional Government; Figure 6 shows a summarized list of functions of all subnational levels of government.9 Flowing from the criteria adopted in functional assignment (Box 1), the provincial government undertakes infrastructure and services that cut across boundaries of district/city governments; were they to be managed at this lower regional level the spill-overs would be large to make the functions efficient. As mentioned above, in practice the provincial level also undertakes coordination, guidance and supervision functions toward the district/city governments that formally rest with the central government and in principle are exercised by the provincial governor (wearing their hat as representative of the central government). Figure 6: Summarized functions of subnational government Province District/city Village Basic services focus: Larger scale Small infrastructure/services e.g., roads, water, agriculture, health infrastructure and and resources: services: clinics and small e.g., referral hospitals, hospitals, e.g., vilage roads, senior secondary and elementary and street lighting, universities/teacher junior secondary community halls, training, vocational schools, public health clinics, early schools, forestry and safety, waste childhood mining management, management, social education. provincial roads services. Source: Author 25. The tasks of the Provincial Governor are weighty with respect to its ongoing supervision of district/city government, as well as largely ceremonial or episodic in the case of general government functions that come from the President (Figure 7). The involvement of the provincial government has been mentioned for the supervision functions. In principle, the Provincial Governor should be able to undertake the general government functions without leaning too much on the provincial government units. The Governor has the possibility of enlisting the assistance of the next level of regions (district/city heads; the Bupati and Mayor respectively). Moreover, the city/district heads have the possibility of being assisted in turn by the sub-district heads that are part of their administration (their own deconcentration channels). 9 See Appendix 2 for a detailed example from the education sector; the entire appendix is too large to append. When freshly issued in the version found in Regulation 38/2007, it was referred to as Buku Bantal, or Pillow Book. 12 Figure 7: Summarized functions of the provincial governor through deconcentration Deconcentration from Deconcentration from Central President (General Government Government functions) Coordination guidance, Tasks related to: and supervision of district/city government National unity Any task given by national State ideology organization (coordinated Resiliency via Minister of Home Affairs) Social conflict Source: Author 26. This architecture of functional assignment has resulted in the transfer of functions that are weighty, and has also generated some further challenges: • A list construction in rule focused bureaucracy reduces regional government initiative and innovation; • Uniformity of functions across districts/cities does not take into account different size and capacities; • Mega-urban regions (especially in Java) feel they do not have the power needed to address the special challenges they face (e.g., floods, traffic). • The devolution and deconcentration fusion has led to some confusion. Chapter 5. Regional government revenues 27. Decentralization in Indonesia was largely on the expenditure side, leading to a large vertical fiscal imbalance. Intergovernmental transfers have therefore continued to be necessary to close the fiscal gap, with districts/cities depending on transfers for more than 80% of their revenues and provinces 54% (Table 4), much higher than the Chilean municipal reliance on transfers of around 40% of their budgets. Box 5 provides an overview of Indonesia’s major transfers to regional government following the decentralization reforms and subsequent adjustment in the framework. Table 5 shows the relative sizes of the transfers (including those for villages) from central government. Table 4: Own revenues of regional governments compared to other sources (2015) Regional level Own revenues Revenue sharing Central Govt. Others transfers Province 54.7% NA 24.8% 20.4% District/city 22.4%* 0.08% 65.4% 0.04% * For 2015 district/city own revenues were higher than usual; for most years the amount is under 20% of the realized budget Source: Adapted from Ministry of Home Affairs, 2015. 13 Box 5: Major intergovernmental transfers for regional government • DAU (General Allocation Fund) is a block grant provided to provincial governments and district/city governments. • DAK (Special Allocation Fund) is an earmarked grant to finance special activities of national importance. DAK consists of DAK Fisik, mainly for financing capital expenditures, and DAK Non Fisik, mainly to provide additional financing for the operational cost of services, e.g., school and health center operational assistance (BOS and BOK). • DBH (Revenue Sharing Fund) is a part of national revenue that is shared with provinces and districts/cities. These include personal income tax, natural resource revenues (oil, gas, other mining, forestry, fishery), and tobacco excise. • Other types of transfers, including the Special Autonomy Fund for Aceh and Papua, the special fund for Yogyakarta, and regional incentive grants (DID) distributed to local governments according to their rank in performance ratings by the Ministry of Finance. 28. Revenue sources favored the provinces, but a decade after the initial reforms, the property tax and land transfer tax were decentralized to the districts/cities . Appendix 3 tracks the changes in regional government tax assignment from pre-reform to the latest changes in 2009. The reform era revenue assignment (Table 6) largely excluded sources that are large or very buoyant, such as income tax and value added tax, which were retained by the central government. However, a small portion of the personal income tax was shared with the regions (8% for provinces and 12% for district/cities). Natural resources revenue sharing also became a substantial source for some districts (Figure 8). A very modest effort was made to share the wealth of well-endowed districts with neighboring districts. While the size of natural resource revenue sharing is relatively small overall (5.6% of all transfers - see table 5), the concentration in the few districts of derivation, and their neighbors in the same province, adds significantly to horizontal inequality. Table 5: Composition of all transfers from central government 2014-2016 (IDR Trillion) World Bank 2018. Indonesia – Public Expenditure and Financial Accountability (PEFA) Report 2017 14 Table 6: Regional taxes assignment at time of major reforms* Responsibility Distribution of revenue Base Rate Adm CG P D/C Vehicle tax CG, P P P 0 30 70 Vehicle transfer tax CG, P P P 0 30 70 Fuel tax CG, P P P 0 90 10 Exploration tax of surface and CG, P P P 0 100 0 underground water Hotel tax CG, D/C D/C D/C 0 0 100 Restaurants taxes CG, D/C D/C D/C 0 0 100 Entertainment taxes CG, D/C D/C D/C 0 0 100 Advertisement taxes CG, D/C D/C D/C 0 0 100 Street Lighting taxes CG, D/C D/C D/C 0 0 100 Exploration tax of mines CG, D/C D/C D/C 0 0 100 Parking tax CG, D/C D/C D/C 0 0 100 CG=Central Government; P=Province; D/C=District/City * This chart does not show the property tax and property transfer tax that was initiated in 2011 (see Appendix 3) Source: Adapted from Simanjuntak and Mahi,2003 as reported in Nursini 2012. 29. Horizontal imbalances were reinforced in the reforms; fiscal revenues vary greatly across Indonesia’s district/city governments. World Bank figures indicate that in 2017 total fiscal revenues10 per capita in high-revenue districts were about 30 times higher than in low-revenue districts (Figure 9). The districts with the highest fiscal transfers per capita are found in eastern Indonesia (Figure 10). 10Total fiscal revenues here are defined to comprise total transfers from the central government and from provinces plus distri cts’ own-source revenues. 15 Figure 8: Natural Resources Revenue Sharing (%) between all levels Source: Law 33/2004 as reported in Agustina et al. 2012. Figure 9: Indonesia’s regional inequality in per capita revenues compared to other countries Source: OECD, 2018. 16 Figure 10: Spatial view of fiscal transfers per capita in Indonesia Source: World Bank, 2019 30. This spatial pattern in fiscal revenues arises because Indonesia’s transfer system tends to equalize places rather than people.11 The regions with the highest proportion of the poor are in the eastern part of Indonesia, but these regions have very small overall populations and a very small proportion of the all the poor people in the country. Transfers per capita to these regions tend to be higher per capita because the DAU formula emphasizes place through the focus on geographic size and reimbursing the regional wage bill, which tends to be disproportionately high relative to population in eastern Indonesia – up to six times higher (Figure 11). Population is just one of several adjustment factors in the DAU formula, allowing districts/cities with low population (but high poverty rates) to receive disproportionately high transfers (Figure 12). Because of the population concentration in Java/Sumatra/Bali, the majority of poor people are actually found in the highly populated parts of Indonesia – the relatively better off regions.12 Transfers are far from proportional to the number of people who lack access to basic services (as a measure of unmet service delivery needs). Yet, despite decades of higher per capita resource allocations, there is a little to suggest a convergence of lagging to regions to national infrastructure and service standards, suggesting other measures are needed in addition to financing. 11 Indonesia has considerable income inequality (Gini coefficient of 0.395). 12 Over one quarter of Indonesia’s poor population lives in the 100 districts with the lowest poverty rate. 17 Figure 11: Variance in number of civil servants per capita across districts World Bank, 2019 Figure 12: District population versus median total revenue per capita, FY2017 Source: COFIS; World Bank staff estimates based on SIKD-MoF data. Note: District per capita revenues comprise local tax and non-tax revenues, as well as transfers (DAU and DAK). 31. The DAU (General allocation grant) formula has been adjusted over time, with the most recent reform proposals from the World Bank recommending removing the place-based starting point. Figure 13 shows the current DAU formula. The Basic Allocation, tied to the wage bill, is about half of the value in the calculation, the other half being the fiscal gap calculation. There is a desire to make the DAU more focused on regions with higher numbers of poor people, less susceptible to encouraging a bloated wage bill, and some concern around the possible discouragement of own revenue effort in the fiscal gap component. While there is evidence that the wage bill concern is valid, the own source revenue disincentive appears to be overplayed as own source revenues have also risen as DAU amounts 18 have risen. There is some discussion of also making the DAU formula notionally reflect the basic cost of service delivery (e.g. as conditioned by minimum service standards) but this has been long discussed and avoided due to data challenges. 32. The natural resources income sharing favors producer and neighbor regions and the distribution does not appreciably equalize across regions. This revenue sharing aimed to garner greater political buy in for “keeping the country together” after the era of centralization came to an end. Some of the regions with the greatest endowment of resources were among those with lagging infrastructure and public service delivery (Agustina et al., 2012). Natural resources revenue sharing was a more significant portion of the regional government budget in early years of decentralization (about 25% in 2008 compared to about 6% in 2016) but its concentration in a few districts makes it very important to the receiving regions. The “hold harmless provision” implemented for several years after the introduction of the general transfer (DAU) was premised on safeguarding these regions from losing on their DAU, due to having a high share of natural resource revenue. The political sensitivity of giving with one hand and taking with the other – in regions feeling historically neglected and plundered – necessitated this accommodation. Figure 13: Formula of the DAU (General Allocation Grant) 33. DAK, the main specific-purpose grant, has grown considerably and tends to be “inequitable” as it is proposal based (since 2016), favoring districts/cities with high capacity. The DAK initially focused on a few sectors (e.g., health, education, roads), but over time grew to cover practically all sectors. An allocation formula was used for many years (although it was open to manipulation), but since 2016 the allocation has been made on the basis of regional government proposals that are guided by sector guidelines. 19 Chapter 6. Regional government borrowing 34. Regional governments can borrow but are limited by regulations and capacity. More restrictive rules during the initial period of decentralization have been relaxed somewhat in recent years but the amounts are still small. Over the period 2014 – 2018, about 50 regional government have borrowed a total of IDR 8.58 trillion (about $600 million) for roads, bridges, markets, and terminals (infrastructure with the potential to generate revenues). This is a small fraction of the national budget ($173 billion) and total government borrowing (0.85% as at 2017). In addition to the regulatory limitations on borrowing, regional government is also hampered by limited capacity to design and manage large capital projects. 35. Borrowing prior to the decentralization reforms was significant (via public sector mechanisms), but poor repayment and lack of an adequate borrowing framework have led to a decline relative to infrastructure investment needs. The central government sought to clean up the defaulting loans early in the reform period. This was a large task as over 50% of the loans (to central government agencies/donor mechanisms) were in default; the main obstacle to repayment was not poor finances but rather unwillingness to repay and central government acquiescence (Lewis, 2003). Following the decentralization reforms, there has been an attempt to avoid moral hazard in regional government borrowing. The central government has stressed that it does not intend to back up the loans taken by regional governments from other lenders. More limited loan funds and stringent safeguards emanating from the Ministry of Finance13 have curbed borrowing, to the degree that their safeguard provisions are not really being tested.14 The main existing safeguards for borrowing, including for bonds, are outlined in Box 6. The focus is on encouraging regions to borrow responsibly, while making more borrowing options available to them. But as this calls for considerable regional government capacity building and adjustment in lending institutions, the new approach has been slow to gain traction. There is no sanction on the books for not borrowing optimally. Box 6: Safeguards for borrowing by regional government • Total of debt outstanding (new debt and existing debt) cannot exceed 75% of total regional government revenues in the previous fiscal year. • DSCR (debt service coverage ratio) cannot exceed 2.5 every single year. • Maximum limit of cumulative deficit in national budget (APBN) and regional budget (APBD) is 3% of current GDP. For regional government bonds: • Bonds should be for revenue generating projects (self-liquidating); • Regional parliament's approval is needed for issuing each bond; • Satisfactory regional government audit rating and • Macroeconomic stability are required. • There is a review by the Securities Commission and Ministry of Finance 13Regional governments must report on the status of their obligations to both the Ministry of Finance and the Ministry of Home Affairs on a semester basis, see Government Regulation 30/2011. 14In fact, over the last decade or so regional government has tended to not spend part of their budgets and provide the excess funds to the rest of the economy through the purchase of certificates of deposit (SBI) in commercial banks (Lewis and Oosterman, 2009). 20 Chapter 7. Regional government budget process 36. The vertical coordination for the budget and planning system is elaborate but rigid and impractical. The central government seeks to guide regions on an annual basis through the budget process. The Ministry of Finance (and the National Development Planning Agency – Bappenas, particularly after changes made in 2017) sets the budget framework and guidance. Central and regional government long to short term plans, and budgets, are expected to be synchronized and harmonized, through annual planning/budgeting directives, coordination meetings (particularly on DAK allocations) and review of draft regional budgets (Figure 14). This idealistic framework is difficult to operationalize in practice. It absorbs a great deal of energy without resulting in good plans in many cases. It is important to note that national government reviews have focused on proposals of regional government for the use of the special allocation grant (DAK). The central government also reviews the overall budget of the regional governments (in the case of district/city budgets this is done by the provincial governor on behalf of the central government), but these reviews tend to be limited to procedural norms rather than whether the expenditures conform with priorities laid out nationally (e.g., on SDGs or minimum service standards). Regional budgets are often amended as a result of these reviews, but not substantially. 37. The proposal approach to the DAK is onerous on all concerned (Figure 15) although there are efforts now to make better use of online templates to lessen the burden on regional government and reviewers. Even so, for district/city government proposals, four different actors are involved in the review; sector ministries, Bappenas, Ministry of Finance, and the provincial governor/provincial government. Figure 14: Indonesia’s nationally integrated regional planning and budgeting Source: National Statistical Agency (BPS), 2014. 21 38. Historically, the central government has been late in providing information to the regional government on DAK allocations, making it difficult for regional governments to budget with confidence. Ideally, the DAK information should flow during the May-July period, when regional planners are conferring with regional councils. This approach has very recently been placed in the revision of the planning system (Government Regulation 17/2017). The initial estimate of the DAK pool and allocations to regions – even if announced in May-July - is subject to additional confirmation once the national parliament has its budget review much later in the national budget review cycle (Figure 16). The impracticality of the regional government framework on the requirement of coordination is further compounded by the lack of suitable assessment tools of the national and provincial governor/governments for the review of plans and budgets. In the short time available, only the most glaring deviations from the rules can be caught. There is little guidance on how the district/city governments can reflect national imperatives while also exercising their autonomy, in the effort to close minimum service standards gaps or in pursuing the targets in the Sustainable Development Goals. Figure 15: The budget cycle for proposal based DAK (Special Allocation Grant) Source: Ministry of Finance, 2018 (translated by author) 22 Figure 16: Indonesia’s regional government budget cycle Source: Ministry of Home Affairs, 2018 (adapted by author) Chapter 8. Effectiveness of regional government spending 39. Access to basic services has seen some improvement since decentralization, but far from the pace of growth in regional government spending. The World Bank estimates that real per capita spending increases of 258% between 1994 and 2017 has only led to 33% improvement in “access to basic services”15 on average, with only a weak correlation between spending and access improvement (see Figure 17 for the district level data between 2008 and 2017). 40. Decentralization has led to some convergence in spending across regions. Where regions faced large gaps in public investments, some change in the composition of spending can be noted, contributing to a slight convergence in public service delivery across districts. A study found that a one standard deviation lower coverage of health care centers prior to decentralization, resulted in about 14% higher investments in the health care sector after decentralization. The same effects were about 7% for schooling and 6% for physical infrastructure (Kis-Katosab and Sjahrir, 2017). Drawing more detailed conclusions about the links between spending and outcomes is hampered by the lack of reliable and timely data, at a low-enough level of geographic disaggregation. Another challenges is the 15 As a simple aggregate measure of “average access to basic services”, the average of five indicators that reflect access to lo cally provided basic services is used: (i) net enrollment rate for junior high school; (ii) net enrollment rate for senior high school; (iii) access to protected water; (iv) access to protected sanitation; and (v) proportion of births attended by a skilled health worker. An average score of 100 means that all households in the relevant district have access to protected water and sanitation, all births were attended by a skilled worker, and everyone in the relevant age groups are enrolled in junior or secondary school; a score of 0 means that no households have access to protected water or sanitation, no births were attended by a skilled worker, and no child in the relevant age groups are enrolled in junior or secondary school. 23 lack of feedback loops to local government, to take informed actions on the basis of data that is collected and reported upwards to central government. Figure 17: Changes in service access over changes in district spending per capita, 2008-2017 Source: World Bank, 2019 (drawing from Susenas and SIKD data) Inefficiency in spending can be attributed to several causes, including weak incentives to perform and capacity constraints. High levels of transfers do not foster public scrutiny of budgets and their implementation. The transfers lack performance incentives and are not well matched to sector needs (only recently have efforts been made to experiment with incentives). Access to information by citizens on regional government performance is lacking. Culturally, citizens have not asked government for much and are easily contented with some or any effort to provide services. World Bank surveys of service satisfaction have shown high satisfaction regardless of objective indicators. 41. Fiscal decentralization fueled the proliferation of regional government at both levels, and particularly at district/city level where most fiscal transfers flowed. This expansion of jurisdictions (Table 7) since 2000 has required more administration, largely in the district capitals (building, staff, equipment). Some districts had to shift their capital to another urban settlement in the district. This upheaval has distracted the regions from the task of improving service delivery. Local political aspirations figure prominently in regional splitting, led by desire to gain more central transfers, manage local resources, and reinforce ethnic and traditional identities (Fitrani et al., 2005). This proliferation has not appreciably improved service delivery in those new regions. In a study by Lewis, there was no effect on education, and about a 1% decline in household access to protected water and sanitation in the second year as compared to original districts (Lewis, 2017). While services have not deteriorated significantly, there has been a large opportunity cost in foregone improvements. 24 Table 7: Growth in number of regional governments in Indonesia, 1980 –2019 Government level 1980 1990 2000 2010 2019 Province 27 27 26 33 34 District 246 241 268 399 416 City 54 55 73 98 98 Note: Decentralization took full effect by 2001, making 2000 the baseline for regional units Sources: Adapted from Nasution, and Wikipedia 2019. Chapter 9. Improvement and accountability in regional government 42. Service improvement and accountability to citizens is hampered by cultural norms; citizens recognize that government is ineffective but are not greatly upset by it. In line with the previously mentioned high user satisfaction scores, Indonesians are quite confident in their government. Figure 18 shows the high confidence of Indonesians in their national government; data from the Asian Barometer suggests that confidence in regional government is at least as high (Asian Barometer, 2006). These sentiments are held despite perceptions that government is not effective, including in the provision of services (Figure 19). The Global Corruption Barometer reveals that Indonesians are pleased with their government’s efforts to eradicate corruption, their favorable view perhaps fueled by the visible efforts of the Anti-Corruption Agency (KPK) established in 2002. In 2018, the KPK charged its 100th regional government head as a suspect in corruption activity. The patient attitudes of citizens on governance issues serves to reduce pressure on government to act in an urgent way on the shortcomings of regional government processes and their results. Figure 18: Confidence in national government (2017) Source: Statistics Indonesia; Gallup World Poll, reported in OECD, 2018. 43. Accountability between levels of government has been slow to take hold and hampered by poor reporting from regional government, and insufficient capacity. The internal audit function of regional government has not developed rapidly due to insufficient capacity development of internal auditors and rules that compromise their independence. External audits by the Supreme Audit Body (BPK) took some time to reach all regional governments. Audit reports have improved substantially since the early years of decentralization; now 76% of regional government receive a “satisfactory” opinion. However, the 25 parliament or justice system follow-up to audit findings are still insufficient. In terms of the national guidance for policy making, planning and budgeting, the vertical reviews of draft plans and budgets (by the Provincial Governor for district/city governments and by the Ministry of Home Affairs for the provincial government) are bereft of adequate tools to assess the submissions aside from a check on legality. The power of the executive branch to overturn regional regulation is debatable in view of constitutional provisions on judicial review. Some regional government regulations that clearly conflict with the constitution and legislation (e.g., on religious/human rights matters) do not draw a proper response. However, over time, the central government has been able to improve the situation in some areas; addressing regional government regulations on mining and forestry concessions (these were proliferating and overlapping) and reducing burdensome taxes and charges that contravened central government regulations. Central ministries are still not able to fully assess regional government progress against minimum service standards and the SDGs, in part because the disrupted vertical reporting relationships have yet to be fully reestablished. Figure 19: Views of Indonesians on government effectiveness (2010, 2016)16 Source: OECD, 2018. 44. Citizen’s participation and government information on its performance are still in early stages of development. The Freedom of Information Act (FOIA) No. 14/2008 guarantees public access to information from all public bodies in Indonesia, including regional government. Regional government is mandated to make public some documents (plans and budgets and accountability reports), and the FOIA pertains to other information. The provincial government is required to set up an Information Commission, and all regional governments need to appoint an Information and Documentation Officer and issue regional regulations to manage the access process. This law has not yet been operationalized in the spirit it was intended (Faisol, 2010). As regional government reports on performance submitted to central government are weak, it is not surprising that information on government decisions and performance to the public is even less useful. Innovative efforts to inform and involve the public are being pioneered in some regions. For instance, leaders 16The Government Effectiveness Indicator compiled by the World Bank. It is a composite indicator ranging from -2.5 to 2.5 (highest score). It reflects perceptions of the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government's commitment to such policies. A higher value indicates more effective government. 26 are holding public meetings or radio shows to reach the public. The planning system requires a bottom-up approach that includes citizen participation in an annual development meeting (in villages and regions), but this has been rather pro-forma in most localities or captured by local elites. More genuine participatory budgeting is being attempted in a few regions, facilitated by NGOs (Kota Kita, 2017). Indonesia has joined the Open Budgets Portal of the World Bank, and has received a score of 62 (2012) placing it in the “substantial” openness category, but with the finding that its public participation could be improved 45. Innovation has been slow to take hold after decentralization but is picking up momentum and is incentivized by recognition events. As new leaders emerge, from the CSO sector or the private sector, regional governments are showing some innovation in public service delivery – beyond the efforts of the central government to introduce new approaches (like e-procurement). One visible effort is the provincial effort to establish Innovation Hubs, beginning with East Java 17 which has now spread to two other provinces. Strongly supported by KemenPan R&B (Ministry of State Apparatus and Bureaucratic Reform) and development partners (e.g., GIZ), the provinces currently involved have put competitions and technical support to good use in fostering district/city level innovations. The spread (replication) of these innovations is less evident or incentivized. Replication does not have the same cachet as innovation and is not rewarded as much. Innovation is often hamstrung by regional officials’ concerns about overstepping their roles/functions; the rigid positive list of functional assignments and auditing practices underlie this apprehension. 46. Minimum service standards (MSS), a safeguard born in the decentralization reform design, are slowly being incorporated by regional governments in their planning and budgeting. MSS were designed to give district/city government some parameters in the initial general competence architecture of functions (Ferrazzi, 2005) (Box 7). The challenge in setting normative service standards is to focus these on the highest priority services and ensure that regional government will have the resources to achieve them within a reasonable timeframe. A proliferation of minimum service standards, as seen in the period 2005-2013, has led to a loss of credibility that they can be achieved and that they are priority, losing their power to motivate improvement. Since 2014, an effort has been made to refocus these on basic services. Even attaining a streamlined set of standards has been a challenge in Indonesia, particularly with the persistent vertical and horizontal fiscal imbalances. Even so, many regional governments can now boast of having met most of these standards (pertaining to basic services). They have done so by generally increasing spending on these services relative to other expenditures or finding innovative ways of delivering the services. At national level there is continuing exploration on how fiscal transfers and other levers can be used to further spur their achievement across all regions. Box 7: Concept of Minimum Service Standards as applied in Indonesia • Nationally set standards applicable to all regional governments. • Focused on basic service delivery (health, education, water provision etc.). • Introduced in 2001 to ensure basic services would be delivered by regional government (at that time districts/cities had a general competence functional assignment architecture). • Have been supported in some fashion through the Special Allocation Grant (DAK). • Some regional governments have struggled to meet them, due to lack of funds and capacity to cost them and to plan and budget accordingly. 17 See the East Java platform at http://jipp.jatimprov.go.id/ 27 • Central Government does not have complete/reliable data on achievement. • The concept is slowly becoming internalized in public discourse. Chapter 10. Lessons and implications for other countries 47. Indonesia’s experience has several lessons that can serve any country contemplating decentralization: 1) Deep forms of decentralization (devolution/Big Bang) have some advantages, and the risks entailed can be managed. Such reforms generally need to be demanded and seen as a political necessity. 2) Matching resources to functions is essential to the achievement of success by subnational governments. A clean shift of modalities (deconcentration to devolution) can facilitate this matching and simplify reform. 3) There is value in keeping reforms simple, particularly in limiting asymmetry in the assignment of functions; it may be better to help some weaker regions to manage a uniform assignment than to deal with numerous regions with a variable range of functions. 4) Functional assignment should be understood as setting roles across levels and deciding on the overall architecture/typologies. It should result in both certainty and sufficient room for discretion/initiative to realize the promises of decentralization. 5) Decentralization design needs to address and balance all major building blocks of decentralization, including appropriate planning, reporting, and accountability systems. 6) The success in fulfilling new roles requires substantial time and capacity development efforts, particularly where there is considerable variation in capacity. 7) Decentralization can lock in or even exacerbate regional disparities. Reducing these is arduous and politically charged work, with difficult trade-offs in addressing historical grievances versus current needs (e.g., place versus people trade-off in Indonesia). 8) The quest for increased efficiency and improved governance at the regional government level has no simple policy solution, but rather calls for the collection of data to confidently report on spending trends and development results, the communication of such data to create political incentives to improve, sustained pressure from citizens, experimentation, and learning from experience. 9) Decentralization reforms may have a big bang moment, but they are generally an ongoing process that can add or reduce local autonomy and build systems. It is useful to have consensus building forums/processes to review progress and develop corrective measures. 48. In particular for Chile, the following cautions and suggestions might be especially relevant: 1) Indonesia was rushed, but Chile has more time, so do involve stakeholders (all levels of government and others) to converge on goals, concepts, terminology and key options, e.g., scope and mechanisms of intermediary roles; balance of expenditure and revenue assignments; policy on regional disparities. 2) Indonesia did a few things right on the architecture and typology of functions but has not provided for a dynamic and creative system; Chile should ensure there is CG control, certainty and safety in RGs taking up functions, but also room for RG to take the initiative and be creative. 3) Overriding political considerations shaped Indonesia’s choice of intermediary roles and mechanisms. Chile may have the option (with only one level in play) to make the intermediary role more fulsome and simplified from the start. 28 4) The slow progress on service improvement and innovation in Indonesia points to a lack of accountability mechanisms, with clumsy and burdensome retrofitting thereafter. Chile has a chance to design mechanisms (especially on information use) that are not burdensome and involve stakeholders more meaningfully. 5) While in principle decentralization can be considered revenue neutral, in practice Chile should be prepared for an increased overall national expenditure. This may be desirable, if it puts more funds into key services and lagging areas but has to be managed. Indonesia had some safeguards in this respect that were helpful for its context; Chile may need to find its own brakes and guard rails. 6) If Chile is interested in a more rigorous approach to minimum service standards – as it transfers service functions to regional governments – it would do well to limit these to the most essential services, ensure data on progress is collected and communicated to regions and the public, and to stress the attendant shift of resources from less-priority expenditures to these higher-priority services. 7) A careful assessment of transfer formulae should be undertaken, recognizing the trade-off between focusing on people (population numbers or service users) versus places (relative indices e.g., poverty rates across certain regions). This entails discussions of national values (including the need to address historical placed-based injustices), and an understanding that views on equity can be highly political and often incommensurate in nature. It should also be noted that increased revenue alone will likely not address the service gaps of lagging regions, which will also require other support (e.g. capacity injections) to catch up to national averages. 8) Prepare a timely and relevant capacity development plan that recognizes critical and urgent transition issues and also has a longer-term perspective. 29 References Agustina, Cut Dian, Ahmad, Ehtisham, Nugroho, Dhanie and Siagian Herbert. 2012. Political economy of natural resource revenue sharing in Indonesia. World Bank, LSE Asia, and University of Bonn. Asian Barometer. 2006. The 2006 Asian barometer survey; Wave 2. Asian Development Bank. 2019 (forthcoming). Indonesia: Country Assistance Program Evaluation. Manila. Boex, Jameson, and Jorge Martinez-Vazquez. 2007. Designing intergovernmental equalization transfers with imperfect data: Concepts, practices, and lessons. In Fiscal Equalization, 291-343. Springer. Decentralization Support Facility. 2012. Proceedings of a conference on Alternative Visions for Decentralization in Indonesia. Jakarta: Decentralization Support Facility. Faisol, A. et al 2010. Fulfilling to Right to Information: Baseline Assessment on Access to Information in East Nusa Tenggara, Indonesia. Article 19, TIFA, Bappeda NTT, AusAID. http://www.article19.org/data/files/pdfs/publications/fulfilling-the-right-to-information.pdf Fengler, W. and Hoffman, B. 2009. Managing Indonesia’s Rapid Decentralization: Achievements and Challenges, in Bahl, R.W. and Ichimura, S. Decentralization Policies in Asian Development. Hackensack, N.J.: World Scientific. Ferrazzi, G. (2005). Obligatory Functions and Minimum Service Standards for Indonesian Regional Government: Searching for a Model. Public Administration and Development, 25 (3): 227-238. Ferrazzi, G. (2000). Using the “F” Word: Federalism in the Decentralization Discourse in Indonesia, Publius - The Journal of Federalism, 30 (2), (Spring). Ferrazzi, G. and Rohdewohld R. 2017. Emerging practices in Intergovernmental Functional Assignment. Routledge: London. Feruglio, F. and Rifai, A. 2017. Participatory budgeting in Indonesia: past, present and future, Making All Voices Count. Practice Paper, Brighton: IDS Fitrani, Fitria, Hofman, Bert, and Kaiser, Kai. 2005. Unity in diversity? The creation of new local governments in a decentralising Indonesia, Bulletin of Indonesian Economic Studies, 41 (1): 57-79. Hofman, B., Kaiser, K. 2004. The making of the big bang and its aftermath: A political economy perspective’, in Alm, J., Martinez-Vazquez, J. and Indrawati, S. eds. Reforming intergovernmental fiscal relations and the rebuilding of Indonesia. Cheltenham: Edward Elgar. Krisztina Kis-Katosab and Bambang Suharnoko Sjahrir. 2017. The impact of fiscal and political decentralization on local public investment in Indonesia. Journal of Comparative Economics, 45(2): 344- 365. Lamont, J., Imansyah, H., eds. 2012. Ten Years After the Big Bang: The Future of Fiscal Decentralization in Indonesia. Jakarta: University of Indonesia Press. Nursini. 2012. Implementation of Fiscal Decentralization in Indonesia: 2001-2009, Jurnal Ekonomi Pembangunan, Volume 13, Nomor 1, Juni, hlm.1-17. 30 Lewis, Blane D. 2017. "Local government spending and service delivery in Indonesia: the perverse effects of substantial fiscal resources." Regional Studies. 51 (11):1695-1707. Lewis, Blane. 2017. Does local government proliferation improve public service delivery? Evidence from Indonesia. Journal of Urban Affairs, 39(8): 1047–1065. Lewis, Blane. (2003). Local Government Borrowing and Repayment in Indonesia: Does Fiscal Capacity Matter? World Development. 31: 1047-1063. Lewis, Blane D, Neil McCulloch, and Audrey Sacks. 2016. "Measuring local government service delivery performance: Challenges and (partial) solutions in Indonesia." Journal of International Development. 28 (5):808-817. Lewis, B. and Smoke, P. 2017. Intergovernmental fiscal transfers and local incentives and responses: the case of Indonesia. Fiscal Studies 38(1):111-139. Lewis B. and Oosterman, A. 2009. Impact of Decentralization on Subnational Government Fiscal Slack in Indonesia. Public Budgeting & Finance, Summer. 29(2), 27-47. Ministry of Home Affairs (Kementerian Dalam Negeri). 2016. Pemutakhiran Data Pajak Daerah dan Retribusi Daerah. Nasution, A. 2016. Working Paper Series GOVERNMENT DECENTRALIZATION PROGRAM IN INDONESIA. Asian Development Bank Institute. No. 601 October. Manila. OECD. 2018. Regions and Cities at a Glance 2018 – CHILE. Paris: OECD Publishing. https://www.oecd.org/cfe/CHILE-Regions-and-Cities-2018.pdf OECD. 2018. Economic Surveys: Indonesia. Paris: OECD Publishing http://www.oecd.org/eco/surveys/Indonesia-2018-OECD-economic-survey-overview.pdf OECD. 2016. Subnational governments around the world. Structure and finance. A first contribution to the Global Observatory on Local Finances. Paris: OECD Publishing. Pakaya, J. S. 2017. Redesign of Judicial Review System of Regional Regulations. Jurnal Legislasi Indonesia, 14 (1) - Maret: 91 – 98. World Bank. 2017. Indonesia Subnational Public Expenditure Review. Jakarta. 31 Appendix 1: Comparative view of Indonesia and Chile (https://countryeconomy.com/countries/compare/chile/indonesia) Annual GDP 2018 298,172M.$ 1,022,450M.$ 2018 GDP per capita 2018 16,245$ 3,871$ 2018 Debt (%GDP) 2018 25.56% 29.20% 2018 Debt Per Capita 2018 4,152$ 1,130$ 2018 Deficit (%GDP) 2018 -1.52% -1.75% 2018 Education Expenditure (%Bud.) 2016 21.18% 20.50% 2015 Gov. Health Exp. (%Bud.) 2016 19.74% 8.31% 2016 Expenditure (%GDP) 2018 25.21% 16.64% 2018 Expenditure Per Capita 2018 4,095$ 644$ 2018 Education Expenditure P.C 2016 737$ 121$ 2015 Gov. Health Exp. P.C. 2017 742$ 50$ 2016 Corruption Index 2018 67 38 2018 Competitiveness Ranking 2018 33º 45º 2018 Fragile States Index 2018 40.7 72.3 2018 Innovation Ranking 2018 47º 85º 2018 Density 2017 24 138 2018 Birth Rate 2016 13.28‰ 18.99‰ 2016 Fertility Rate 2016 1.77 2.36 2016 Population 2017 18,355,000 264,162,000 2018 HDI 2017 0.843 0.694 2017 Gender Gap Ranking 2018 54º 85º 2018 Life expectancy 2016 79.52 69.19 2016 32 Appendix 2: Example of education function assignment in Law 23/2014 on Regional Government NO SUB-FUNCTION CENTRAL PROVINCIAL DISTRICT/CITY 1. Educational a. GOVERNMENT Establishment of national a. Management REGION of a. Management REGION of Management education standards. secondary education. primary education. b. Management of higher b. Management of b. Management of early education. special education. childhood education and non-formal education. 2. Curriculum Establishment of the national Establishment of local Establishment of local content curriculum for secondary content curriculum for curriculum for primary education, education, primary education, secondary education and early childhood education, and non- early childhood education, and local special education. formal education. non-formal education. 3. Accreditation Accreditations for higher -- -- education, secondary education, primary education, early childhood education, and non-formal education. 4. Educators and a. Control of the educators’ Transfers of educators and Transfers of educators and Educational formation, educators’ educational staff across educational staff within a district/city Staff transfers, and educators’ district/city regions within 1 region. career development. (one) provincial region. b. Transfers of educators and educational staff across provincial regions. 5. Educational Licenses a. Issuance of licenses for a. Issuance of licenses a. Issuance of licenses for private universities for secondary primary education organized organized by the public. education organized by the public. b. Issuance of operating by the public. b. Issuance of licenses for early licenses for foreign b. Issuance of licenses childhood education organized educational units. for special education by the public and non-formal organized by the education organized by the public. public. 6. Language and Fostering of Indonesian Study of language and Study of language and literature Literature language and literature. literature whose speakers are whose speakers are present within spread across district/city the district/city region. regions within 1 (one) provincial region. 33 Appendix 3: Changes in regional taxes over time/legislation in Indonesia As reported in Smoke P. and Sugana R. 2012. Subnational Own Source Revenues and Shared Taxes In Indonesia: Taking Stock And Looking Forward. International Conference Alternative Visions For Decentralization In Indonesia 12 – 13 March, Borobudur Hotel, Jakarta. 34 Appendix 4: Essential features of decentralization modalities Deconcentration Delegation/Agency Devolution Tasks Ministerial decree and Law, regulation, Constitution, law and related Instrument circular. government decree, or regulations. ministerial decree/circular. From ministry, “delegated” to From representative body From the “state”, or central Source and its own dispersed branches or or ministry/agency to SNG level representative body to receiver of field offices (but also special or special agency. SNG. authority case of delegation to SNG officials having a dual role) From ministry to its branches From the assigning entity Receiving level (through directly (does not show in to the SNG/special assigned revenues or block SNG budget). agency. In cases, funded or conditional grants). Funding from broader transfers that are deemed to be sufficient to cover delegated tasks. Branch staff is central-level SNG/ special agency have SNG have own staff, but civil servants, part of the own staff, but operate operate under a national ministry establishment. Their under a national frame. frame; considerable duties may include May also use seconded discretion in hiring, firing, Staffing coordinating with SNG. staff of central size of establishment etc. government. May also use seconded staff In the case of a dual role by an of central government - official, the official is part of treated essentially as SNG. the SNG executive arm. Discretion for Branches are structured by the SNG/ special agency can SNG can shape their units Structuring ministry, though often shape their units within a within a national frame, and Internal approved at cabinet or higher national frame, and handle handle functions in units of level. tasks in/within units of their choosing. Organization their choosing. Variable but usually limited Considerably constrained High degree of discretion, by ministry regulations, by policy, procedures and but may be limited Implementation procedures, standards and standards set by assigning somewhat by national Discretion instructions. May be entity; some discretion on standards considerable ad hoc guidance. implementation in some cases. 35 From branch/official to Primarily to the assigning Primarily to citizens of ministry head-quarters. entity, but also to the SNG receiving level, through the Reporting/ representative body and SNG representative body; citizens. vertical accountability Accountability remains and in principle is more pronounced in early stages of decentralization Source: In Ferrazzi and Rohdewohld, 2017 (adapted from GTZ 2009) 36 Fiscal decentralization of regional governments – the case of Poland prepared under the World Bank TA RAS for Chile1 1The note was written by Paweł Swianiewicz, Department of Local Development and Policy, Faculty of Geogr aphy and Regional Studies at University of Warsaw with contributions from Ewa Korczyc (Economist, World Bank). 1 Contents Country context............................................................................................................................................. 3 Brief history of decentralization in Poland ................................................................................................... 3 Prior to independence in 1918 .................................................................................................................. 3 Since 1918 until the modern wave of decentralization ............................................................................. 4 Modern wave of decentralization 1990 and 1998 reforms ........................................................................ 5 Current decentralization model ..................................................................................................................... 8 Description of the administrative structure and the key characteristics .................................................... 8 Transition process ................................................................................................................................... 11 Functions of regional governments ............................................................................................................. 12 Revenue sources of regional governments.................................................................................................. 15 Own revenues and tax sharing ................................................................................................................ 15 Transfers to regional budgets .................................................................................................................. 17 Subnational borrowing ................................................................................................................................ 23 Budget process – the role of the MEF......................................................................................................... 25 Supervision, transparency and accountability ............................................................................................. 26 Financial supervision .............................................................................................................................. 26 Transparency and participatory mechanisms .......................................................................................... 27 Other country-specific observations ........................................................................................................... 28 Role of regions in regional development policies ................................................................................... 28 Lessons learnt and insights for other countries ........................................................................................... 34 References:.................................................................................................................................................. 36 Further readings in English: ........................................................................................................................ 37 2 Country context With a population of about 38 million and GDP per capita of US$14,150 (2018), Poland has the largest economy in Central Europe. Since joining the EU in 2004, the country’s ambitions have been marked by the desire to rapidly catch up with the core of the EU in terms of economic growth and living standards. Indeed, Poland’s economic performance in the past few years has been quite impressive with average annual growth rate above 4 percent and record low unemployment rate at 6 percent. Public debt at 46 percent of GDP and fiscal deficit below 3 percent of GDP are considered sustainable for the medium-term. In terms of geography, Poland’s surface is 312 679km2, which makes it sixth largest country in Europe. Poland’s population at around 38.5 million is rapidly ageing thereby creating pressures for increased spending on pensions, healthcare and old-age care. Poland’s capital city - Warsaw - with around 1.8 million inhabitants is a home to around 4.5 percent of the population. Apart from Warsaw, Poland has another 4 cities with population above 0.5 million. The key facts and figures are presented below: Brief history of decentralization in Poland Prior to independence in 1918 Poland has a long history of decentralization. The history of Polish regions (województwo) as territorial units goes back to the end of the thirteenth century, when Poland was being unified after a period of division. During the sixteenth to the eighteenth century the Polish-Lithuanian Commonwealth was divided into big territorial units: 22 regions of the Polish Crown and 10 regions of the Grand Duchy of Lithuania. This division was quite stable for over 200 years. The regional governor (wojewoda) was a member of the Senate (upper house of the Parliament) and although he was nominated by the King, the dignity of the office derived from the fact that the function was allocated to the members of powerful aristocratic families in a given region. In the 17th and 18th centuries, the central power of the King was weakening while the regions were becoming increasingly centrifugal forces, eventually leading to the disintegration of the state. This negative historical experience has been still refereed to during contemporary discussions on decentralization. 3 Since 1918 until the modern wave of decentralization After regaining independence in 1918, the Second Republic (1918-1939) created a network of 16 relatively big regions. The governor (wojewoda) was appointed by the President of Poland and was predominantly a representative of the central government in the region. He was also the head of the regional administration and had strong supervisory powers in relation to local governments. However, contemporary regional attachment is not very strong, except of some small sub-regions, and the present territorial boundaries of the regions do not always respect historical borders. After the 2nd World War, there have been three main periods of the regional organization: • Between 1945 and 1975 Poland was divided into 17 regions plus 5 largest cities of regional status. The sub-national system was based on three tiers: a changing number of municipalities, over 300 counties (powiat), and 17 regions. • In 1975 that organization was replaced by 49 smaller units, which were more easily controlled by the center. The same reform abolished the county level (changing the system from three- to two- tier of sub-national authorities) and amalgamated small municipalities into nearly 2500 larger units. The system was highly centralized – the governor of the region was appointed by the center and was strongly vertically subordinated to the central government in Warsaw. Weak and undemocratically elected regional councils had very weak powers and played a façade role in the system. Such system was sustained until 1998. • Post 1998 territorial reform, which is described in the next section. The summary of the evolution of sub-national tiers in Poland after 1945 is summarized in the Table 1. Table 1. Development of the territorial organization in Poland after 1945 Upper Intermediate Municipal tier Notes (regional) tier (county) tier (gmina, (województwo) (powiat) gromada) 1945-1974 14-17 (plus five Over 300 Changing No democratic local government possible, largest cities) number (2-8 vertical structures of state administration thousand) 1974-1990 49 - Below 2500 No democratic local government possible, vertical structures of state administration 1990-1998 49 - Below 2500 Local self-government system on a municipal level. Deconcentrated state administration on the regional level. 1999 - present 16 315 + 66 cities Below 2500 Elected local self-governments on all three of county status tiers. Parallel state administration structures in regions 4 Modern wave of decentralization 1990 and 1998 reforms The 1990 Decentralization Reform The regional decentralization reform was proceeded by the reform of local governments in 1990. The 1990 reform is considered the most important mile-stone in decentralization of Poland: it introduced democratic local elections and the local (self-)government system in almost 2500 Polish municipalities. The main motivation for the reform came from the political turnover in Europe. Decentralization was considered one of the key priorities of the first non-communist government after the 1989. In September 1989, professor. Jerzy Regulski, an academic expert working on the concept of local government system during the previous decade, was appointed a minister responsible for the decentralization of the country. Subsequently, the Law on Municipal Governments was adopted on March 8, 1990, and the first democratic elections were held on May 27, 1990. The regional reform had to wait for the next political “window of opportunity” which opened at the end of 1997. The latter reform was not as far reaching as the creation of local governments. Also nowadays – as reflected in this report – the municipal tier is by far the most powerful sub-national tier in Poland, both in terms of spending functions, as well as in terms of fiscal autonomy. Nonetheless, the regional tier plays an increasingly important role in the planning of development policy and its implementation. The 1998 Regional Reform There were several motivations for the 1998 reform. It was clear that the structure of 49 small regions created in 1990 was dysfunctional and required substantial modifications. At the same time, it was quite clear that a new, elected regional government should be introduced together with the reform of territorial units, which was then expected to be implemented within the next two to three years (Regulski 2000). Since early 90s’, more than 20 versions of the new administrative division were drawn up by government experts as well as by academics (Kowalczyk 1993). The proposed number of regions varied from 6 to over 40. An even more important issue, though less prominent in popular discussions, was the question of the function of the future regions. Basically, three options were identified by the Government Committee (Mażewski 1992): • Functional region – powers of regions limited to narrowly-defined functions; domination of state administration over the elected representation; • Self-government model – but still within a unitary state system; • Autonomous region – closer to the model of a federal state2. Meantime, in 1997, Poland adopted a new Constitution. The position of the regions in this new Constitution was rather weak. The only tier which is mentioned and protected by the constitution is the municipality (gmina). The decision on the existence of other tiers of elected government was delegated to the Parliament (art. 164). After 1990, discussions on possible changes on a regional level were soon shadowed by a more alive debate on the introduction of an inter-mediate county (powiat) tier which could take-over those local functions, which were too complex for communal governments (Emilewicz & Wołek 2002). There was no strong bottom up pressure for the regional reforms, from any of the regions. The situation changed – quite unexpectedly – after the 1997 Parliamentary elections, when a new centre-right coalition brought a “second stage of decentralization reform” as one of the flagships of its reform programme. 2 The third option was particularly controversial, since some politicians argued that it could lead to the disintegration of Poland. However, the (left-wing) central government elected after the 1993 Parliamentary elections, effectively stopped any progress of the reform for the next four years. 5 This “second stage” was to include the introduction of two new tiers of elected government – county and regional. One may say that the origin of the debate and of the reform was more technocratic than reflecting bottom-up pressure. Necessary legislations were prepared very fast and – after a short but colourful battle with the Parliamentary opposition – the reform was adopted at the end of 1998, effective of 1 January 1999. The number of new regions created was 12 in the initial government proposal adopted by the Parliament. However, the new Law met with resistance, especially of political elites of cities which were afraid of losing their regional capitals’ status. As a consequence, the new Law was vetoed by the President, which led to furt her negotiations among the main political parties and the eventual number of 16 regions (Figure 1). The preparation of the reform started with appointment of the special ministry for the administrative reform. About half a year later, on 5 June 1998 the law on regional government was adopted. The first elections to the regional councils were held in October 1998. In November 1998 the law on financing sub-national governments was adopted (further modified in 2003), and on January 1st of 1999 was a day for the take-off of the new regions. The debates over the reform had three main dimensions. • First, it was related to functions regional governments would be responsible for. • Second, the relationship between regions and central government as well as between regions and lower tiers of territorial organization. • The third dimension was focused on the number of regions and their boundaries, and this was a dimension which was the hottest issues in popular public debates. The future integration with the European Union was an important issue in the discourse of 1998 regional reform (see Box 1). Proponents of the reform argued that only big regions could be competitive in integrating Poland into the European economy and that they could become stronger partners for their West European counterparts. The stronger version of the European argument was to put forward the argument that regional reform was a pre-condition for access to EU structural funds, though the argument was not strictly correct. And, as Hughes et. al. (2003) notice, the final shape of the reform should be seen as inherently endogenous development. 6 Figure 1. Changes to the map of the regions following the 1998 reform Source: Own documents Box 1: EU’s role in Poland’s decentralization reforms The idea of “Europe of the regions” has been widely discussed in academic literature since the late 1980s. The debate was triggered by the development of the EU regional policy and the strengthening of regional governments in several member States. The EU regional policy and its financial instruments enabled the regional actors to better represent their interests on a supranational level. Furthermore, the involvement of sub-national actors on the European level was expected to establish a closer relationship between the EU and its citizens (Crepaz, 2016). In the Central East European Countries (CEECs) that underwent a strong centralisation process during the communist regime, the regionalisation could start only after the fall of the Iron Curtain and was closely connected with the European integration process. At the outset of the transformation process in 1989, Poland was a highly centralised state. In contrast, the European Union underlined the importance of multi-level governance, and the role of sub-national authorities, by implementing the cohesion policy and its structural funds. The accession to the EU was the main goal of Poland’s foreign policy. However, its desire for EU-membership placed Poland and other CEECs in a relationship of “asymmetrical interdependence” with EU actors, so that they could demand some reforms and influence the polities, politics and policies of candidate states (Scherpereel, 2010, p. 45). Indeed, the power asymmetries between the EU and Poland might have suggested that the EU drove the regionalisation process. According to the Europeanisation thesis, EU has provided both negative and positive incentives for regionalisation in CEE. Candidate countries on the one hand were obliged to implement the acquis communautaire prior to accession, on the other hand they could gain pre-accession financial assistance as from the PHARE or SAPARD Programme (O’Dwyer, 2006). Nevertheless, the opportunity to put pressure should not be considered the same as an impact. As many researchers (i.a. O’Dwyer, 2006; Scherpereel, 2010) argue, the EU’s role in regionalisation in CEECs was secondary to the primary domestic dynamics. The decentralisation process offered political parties the possibility to tailor the regional administrative structures to their own aims and to 7 build party structures in regions: “the EU figured as a useful pretext for reforms that allowed domestic political forces, in particular governing party coalitions, to shape regional institutions in their own interests” (O’Dwyer, 2006, p. 222). In Poland the decentralisation was guaranteed in the Constitution of 1997 (Art. 15): “The territorial system of the Republic of Poland shall ensure the decentralisation of public power”. Source: Elżbieta Opiłowska, « Regionalisation in Poland: background, features and public perceptio n. A first appraisal », Belgeo [En ligne], 2 | 2019, mis en ligne le 24 mai 2019, consulté le 01 février 2020. URL : http://journals.openedition.org/belgeo/34254 ; DOI : 10.4000/belgeo.34254 How could the rationale of the reform and arguments used in the preceding debates be summarized? First of all, it has been widely presented as the second step of decentralization of the country, which was initiated by the 1990 reform which concentrated on the municipal tier. The 1998 reform did not affect the functions allocated to the municipal tier, which stayed untouched, but passed more functions to newly created elected governments at the county and regional levels. This has remained true even for those functions which were allocated to municipalities, but which might be more rationally allocates at the county level – such as waste management, for which most of single municipalities are too small to benefit from economies of scale, and its organization requires inter-municipal cooperation. But from the point of view of the regions the main argument was based on an assumption that 49, relatively small regions were too weak to play an important role in the preparation and implementation of regional development policies and to be capable to compete in this respect with larger regions of other European countries (e.g. with the regions of Germany, France, Italy or Spain). In this case economic competitiveness combined with an impact of expected further European integration were on the top of the policy arguments. Current decentralization model Description of the administrative structure and the key characteristics As a result of the reform process, there are currently three tiers of sub-national governments in Poland: - 2479 municipalities (gminy) with the mean population size around 15,500 (the median size of municipality is 7,500 and the smallest one has around 1,300 citizens) - 315 counties (powiaty) plus 66 cities of county status (performing functions allocated both to municipal and county level). The mean population size of a county is around 100,000, and the smallest is just over 20,000. - 16 regions (województwa) with an average population size of around 25. Million. Two (the smallest) regions have just about 1 million population (Opole and Lubuskie) while the largest (Mazovia, with Warsaw being its capital) has more than 5 million citizens. The number and size of units on each of the tiers is summarized in the Table 2. 8 Table 2. Present territorial organization of sub-national authorities in Poland – summary Municipalities County level Regions Rural counties Cities of county status Number 2479 315 66 16 Mean population size 15,500 90,375 151,000 2.39 million Median population size 7,500 72,000 117,000 2.13 million Minimal population size 1,300 20,300 36,500 0.99 million Maximum population size 1,764,000 381,600 1,764,000 5.38 million Source: Author based on data from Central Statistical Office. Poland has a relatively balanced settlement structure. The largest city (Warsaw) is inhabited by less than 5% of total country population, and the Warsaw metropolitan area has less than 10% of total population. There are four other cities over 500,000 residents, and none of regional capitals is smaller than 120,000. Also, differences in the level of economic development, although noticeable, are not extremely large – the richest region’s GDP per capita is less than two and half times higher from the least developed region3. There are two important characteristics regarding the current structure of subnational governments in Poland: (i) lack of hierarchical relationship between the tiers of subnational government and (ii) parallel administration structure at the regional level. All tiers of subnational governments elect their own authorities acting independently of each other and there is no vertical subordination between the tiers (de jure). As explained later in the report, part of sub-national revenues originates from inter-governmental transfers (see section 5), however the flow of transfers is always directly from the national budget to budgets of local jurisdictions of each of the tiers. De facto, there is one important exception to the rule of no hierarchical dependency among the tiers which is related to the role of regional governments in the context of strategic planning. The regional strategy is the main document which includes the most important development projects for a given region. Regional subnational authorities, in liaison with the central government and the European Commission prepare also the Regional Operating Programme (ROP) which is the main instrument for European Union regional policy co-financed from the EU structural and cohesion funds. It is mainly the regional government which decides upon the selection of individual projects to be financed from ROP. Several of them are implemented by municipal governments for which ROP is one of the main sources for development finance of local infrastructure. In this indirect way municipal authorities become somewhat dependent on regional authorities with respect to their development policies and priorities. An important feature of the regional tier is a dual structure of regional administration consisting of state (central) administration and regional self-government. These are two parallel structures operating to a large extent independently on each other: • The regional governor/administrator (wojewoda) who is appointed by the Prime Minister and who is the head of the state administration in the region; • The elected regional government headed by the Regional Marshal (marszalek). The regional government is composed of the council elected by residents in proportional elections, and the regional executive board appointed by regional councillors. The board consists of 3-5 members (regional politicians) headed by the Regional Marshall. The regional council has 30 members in 3 The ratio of GDP per capita in the richest and in the poorest NUTS-2 region in Poland is higher than in seven EU countries, but lower than in 11 other EU countries (not all EU countries are divided into NUTS-2 regions, some of them are too small). 9 regions with less than 2 million residents, and an additional 3 members for each 500,000 citizens. In practice, this means that the size of the council varies from 30 to 51. In terms of organization, the internal structure of regional administration is not strictly regulated, and it is not uniform across the country. Regional Boards headed by the Marshals have a discretion to arrange internal structures in the way which best suits local conditions. Also, the state governor has some amount of discretion to decide upon internal organization of the state administration within the region, so the structures may vary from one region to another. In terms of functions, those two structures of state and regional administrations operate in parallel, but they have to a large extent separate responsibilities. The overview of the main tasks is included in Figure 2 and elaborated further in the report. Figure 2. Main responsibilities of state governors and regional marshals Source: Author. The Governor/state administrator (wojewoda) is responsible for the following policy areas: - Coordination and implementation of central government policies in the region - The special role related to taking care of crisis situations, such as related to the natural hazards (e.g. floods, big fires etc.) - Legal supervision of elected sub-national governments. Every resolution of the regional or local council has to be sent to wojewoda, who checks its legality. If he (she) decides that the sub-national government decisions violate national law, it may abolish the decision. However local government has the right to appeal to the administrative court, which makes the final decision. The supervision concerns conformity with the national legislation only, but not assessment of the directions of sub- national governments’ policies. Governor has his budget (which is a separate section of central government budget) which he controls and may spend on functions related to his competences. The Marshal and the regional Council (i.e. the elected, decentralized regional government) are responsible for the preparation and implementation of regional development policies and the delivery of public services allocated to this tier (as specified in detail further in the report). This model of dual structures on the regional level has been successfully implemented over last twenty years in general. The first few years after the reform has brought some conflicts over competencies, since 10 governors had been accustomed to their larger role in regional policies under the previous system (before 1998 regional decentralization) and were not ready to accept giving up these roles to the new, elected bodies. However, in the following years those cases of competence conflicts have become rare and nowadays the division of functions between elected regional government and appointed regional governors is considered relatively clear. Transition process The introduction of the decentralized government structure at the regional level in 1998 changed the administrative structure of regional tier. The evolving role of the state and elected self-government administrations at the regional level may be illustrated by the changing size and composition of public employees hired by both types of regional authorities (see Table 3). In 1999 the number of employed in the regional office headed by state appointed governor/administrator has decreased considerably, as some of the functions were transferred from the state administration to democratically elected regional self- governments. Nonetheless, even a decade after the reform, the offices of the state governors/ administrators continued to be larger than in the elected self-governments. This was largely due to the following reasons: first, the state governors continued to play a substantial role in the supervision of sub-national jurisdictions, management of state policies implementation, dealing with crisis situation, and, second, for a limited time the management of EU funded regional development programmes continued to be coordinated through the state regional administration. In particular, in 2002-2005 the administration of state governor offices grew again due to the role in implementation of pre-accession and later (since 2004) EU structural funds. Until 2007, the Ministry of Regional Development was still a managing authority for the Integrated Programme of Regional Development (IROP), and the state governors played an important role in the implementation (e.g. managing financial operations of the Programme), and the IROP 2004-2006 programme was implemented until late 2008 (along with the N+2 rule of the EU budget implementation). Only after this period the level of employment in self-governing regional administration has clearly exceeded the number of working in the governor-led offices. The employment trends were also a result of a well-known inertia of administrative structures, which tend to continue for some time even if the rationale for their operation evaporates, as was in Poland in the case of regional state administration. Table 3. The size of employment in regional administration (,000 of employees) in Poland 1998 2000 2005 2010 2017 Governors’ administration 15.5 9.1 13.0 12.0 8.3 Administration of self-governing regions 0 4.0 5.5 12.5 15.0 Source: Author based on data from the Central Statistical Office. Box 2: Adjustments to the model – past and ongoing discussions Since 1998 reform, there have been no major changes in the model of regional government after its implementation. This does not mean that there are no debates about the possible changes of the existing system. The most far going has been the recent proposal of Civic Platform (Platforma Obywatelska) – one of the main opposition parties at the central level – which suggested to abolish the position of the regional governor and to transfer all its powers into the hands of the elected regional governments. However, the proposal has not explained how the legal supervision over decisions of sub-national governments should be exercised in the absence of governor’s position. In a less radical proposal, made some years ago, it was suggested that limited functions of the governor did not justify having 16 governors (one in each of the regions). A suggestion was made to reduce number of governors to 6-8, so each of them would play its roles for two or more regions, which would also allow to reduce employment in state administration. 11 On the other hand, the present governing party (Law and Justice – Prawo i Sprawiedliwość) advocates for the strengthening the current role of governor and his (her) larger involvement in shaping regional development policies. However, those debates have not materialized in any concrete reforms so far, and the system seems to be relatively stable at the moment. There are sometimes political discussions on creation of one or two additional regions, but it has never been turned into a more concrete plan or realistic proposal. The most important changes after 1999 included growing role of regional elected governments in forming and implementing Regional Operating Programmes (this process is discussed in detail further in this report). There were also relatively minor changes concerned sub-national finance (e.g. changing in regulations on sub-national borrowing, evolution of the horizontal equalization systems) – which are discussed later in this report. Source: Author’s elaboration Functions of regional governments In the three-tier structure of subnational governments in Poland, each tier has its specific functions defined in the legal framework. All subnational governments belonging to the same subnational tier, have the same responsibilities, hence there is no asymmetric decentralization in Poland. Despite 1998 reform, which introduced two new tiers of sub-national jurisdictions, the municipality (gmina) has remained the main sub- national tier, responsible for the most basic (and also most important) public services, including provision of primary education and early child-care, solid waste management, water provision, street lightening, social housing, municipal investment in roads and their maintenance. In contrast, counties are responsible for these public services, which exceed the capacity of municipalities, like for example secondary education, some health services, as well as county roads and several services related to social protection and active labour market policies. Finally, the regional subnational tier’s main functions have to do with regional planning and supporting long-term development (Figure 3). The functions of regional governments are specified in the Law on Regional Government ( Ustawa o samorządzie wojewódzkim) adopted in June 1998, as well as in the series of specific sectoral laws. The origin of the division of responsibilities was to a large extent a legacy of the past. Other government functions have been already decentralized to the local (municipal) level and the newly created regions generally took over most responsibilities from the 49 voivodships that existed prior to the reform. The most important addition to these functions, was regional development planning, that was related to the EU accession process. According to the legal framework, the most important regional services include: - Transport: o Maintenance and construction of regional roads o Operation of regional railway networks - Health care (regional hospitals). In this sector regional governments are responsible for the management of hospitals, necessary investments and maintenance of the buildings. However, financing current operation of the health care is part of the national health insurance system (national health fund contracts regional hospitals providing funds for their current expenditures) - Culture – museums, theatres and other cultural institutions of regional importance - Agriculture – amelioration of arable lands - Labour market policies – training for unemployed etc. Despite having sectoral service delivery functions, the main task of the regions is related to the programming regional development. Each region prepares its regional development strategy and is responsible for its 12 effective implementation using financial resources provided under the EU co-financed Regional Operating Programmes (See “Other country specific observations” section which includes a detailed description of the ROP and EU funds). Figure 3. Overview of the main functions of subnational governments in Poland Source: Own elaboration The role of different subnational tiers in service provision is reflected in the budgetary data. Altogether, the accumulated spending of sub-national budgets is slightly lower than the size of the national budget (roughly 45% of budget spending is sub-national and 55% of public budgets is spent by the central government). Looking only at the sub-national budgets, it is clear that roughly 80% of the whole amount is spent by municipalities and cities of county status, which confirms their prominent role in service provision. The budget of counties is just a small fraction of municipal expenditures, and the regional governments spend less than 10% of all decentralized resources. The marginal role of regions in service provision can be contrasted with their prominent role in capital spending. Overall, sub-national capital budget in Poland (including EU funds) is by far larger than the investment spending of the central government. Municipalities remain the largest sub-national investors. However, as explained earlier in the report, an important share of municipal investments is to some extent dependant on the regions, who develop and implement regional development strategies and allocate significant grants to municipalities through the ROPs. 4 Regions themselves, with almost 6 billion PLN (ca. 1.5 billion euro, or over 11% of all sub-national capital spending) investments in 2018, are also important actors responsible for large public investment projects. 4 The figure reflects all capital spending, i.e. Including projects co-financed by EU structural funds. If we take into account only investments with no financial support of EU funds the share of regions in total sub-national budgets capital spending drops from over 11% to 7.5%. 13 Figure 4. Expenditure structure by subnational tier Source: Authors calculations based on the data by the Central Statistical Office The functions provided by regional governments may be also illustrated by the structure of their budget spending using the functional classification of expenditures. Figure 5. Functional structure of regional governments’ expenditure 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% current capital Education Social protection Culture Health Care Transport Administration Agriculture Other Source: Authors based on data from the Central Statistical Office Transport (in this case spending is mostly subsidies to regional railway systems) accounts for more than one third of total operating spending of the regions. This is followed by spending on regional administration, culture institutions (theatres, museums etc.) and agriculture. In the case of capital spending, the domination of transport (in this case mostly renovations and constructions of regional roads) is even more visible, since this sector absorbs well over half of all investment spending. It is followed by much lower investments in health care and agriculture5. 5 The structure of regional investments slightly changes if we take into account only projects with no EU structural funds support. In such a case transport projects are still the most numerous, but their share in total investments drops to 47%. The next largest sector is health care which absorbs 27% of total “own revenue” capital spendi ng of regions. In case of operating spending the role of EU structural funds is minimal, so the sector structure of spending is roughly the same regardless including or excluding EU funding. 14 Most of functions/ services are delivered directly by the regional administration or by companies owned by the regional government. However, there are also examples of involving private or non-governmental sectors in service provision. Most of investments (such as regional roads modernization and construction) are contracted-out to private companies selected in the tender procedure. In at least one region the regional railway services have been outsourced to the company with majority of private ownership. Some of activities in the area of culture of social services are contracted to non-governmental organizations. However, the scope of possible PPP-type of activities is larger in case of municipal governments, who are responsible for a wider set of service functions. Revenue sources of regional governments The financing of regional governments is regulated by the Law on Revenues of Local Governments and Law on Public Finance. As regards the revenues, the present system has been stable over the last 15 years (since 2003). More recent changes – as it is discussed later in this section – have concerned regulations on sub-national borrowing. Own revenues and tax sharing In contrast to the municipal tier – which enjoys limited power of taxation (maximal level of the local taxes is defined by the law, but local government may decide upon tax rates, granting exemptions etc.) – regions do not have their own tax revenues. Strictly speaking, own revenues are limited to very small fees and revenues from property, and constitute just over 5 per cent of the total budget6. The list of regional own revenues is listed in the art. 6 of the Law on Revenues of Sub-National Governments and include revenues from properties owned by regional governments, small local fees (e.g. fee for licences required to sell alcohol, fees for administrative procedures), part of the fees for exploitation of natural resources, interest on bank deposits. In individual regions the share of own in total revenues can vary from 1.99% in Opole region to 10.2% in Western Pomerania. The main source of revenues is provided by the tax sharing arrangement, which consists of 1.6% of revenues from (the central government) Personal Income Tax (PIT) and 14.75% of revenues from Corporate Income Tax (CIT). It is worth mentioning that tax sharing are also important sources of funding for the municipal and county tiers, but in that case, it is the PIT (not CIT) which is by far the most important source. The Polish system of tax sharing meets following criteria: • Sub-national governments share the risk of cyclical changes of revenues from the tax • Sub-national have full discretion to decide upon spending allocation of obtained revenues • The formula of sharing is stable (the same formula throughout several years, determined by the law for undefined period) 6 In Poland’s experience, (in contrast to more generalized lessons from other countries), the lack of regional taxes (which wou ld be autonomously regulated by regional governments) is not necessarily a big problem. Most of theoretical arguments of fiscal federalism for sub-national tax autonomy refers to accountability, closeness between tax-payers and authorities levying the public duties and to possibility of better fit between fiscal policies and local needs and environment. Those arguments better apply to local (municipal) than regional tier in Poland, which is more distant for average citizen and attachment to which is in Poland much weaker than to local municipalities given the nature of expenditure responsibilities of the regional tier. 15 • Shares are based on origin (as opposed to the formula allocation assuming certain amount of redistribution) - the yield received by the region depends on the amount collected within the same region. The specific formulas of tax sharing for all subnational tiers are provided below: 38.08 ( ) , −2 , = 10.25 ( ) × , × , −2 . ( ) 6.71 ( ) ∗ , −2 , = 1.4 ( ) × , × , −2 . ( ) where: t – current budgetary year, i – individual LG; Blochliger and Petzold (2009) define such a system as “strict tax-sharing”. Official budget classification in Poland treats tax-sharing as part of the revenues from own sources, however due to no local influence on tax rates, this classification seems to be very controversial. As shown in Figure 6, tax sharing provides almost half of all regional budget revenues, and together with small “strict own revenues” it is just over 50% of the total budget funding. The current system of CIT and PIT sharing among tiers of sub-national governments is not questioned in general. However, there are at least two issues which are often raised by local governments, and which from time to time are issues of debates. • The first one concerns vulnerability of sub-national budgets (the so-called vertical fiscal externalities) to consequences of tax policy decisions made on a national level. It is indicated that for example reduction of tax rates or voting on new tax exemptions related to CIT or PIT have an impact on sub-national budgets, since roughly have of revenues from those sources is allocated to regions, counties and municipalities. Local government associations suggest that central government initiates from time to time decisions which would be popular among voters, but the costs of the policy are to be borne by sub-national governments, without any compensation. They demand either a compensation of any central decisions leading to diminishing local revenues, or a clear separation of sub-national income taxes, which would be regulated at the regional and local levels. However, none of the central past administrations has so far positively reacted to these demands. • The second issue concerns the demand of a sub-national share in the Value Added Tax (VAT). The argument for introducing such a new tax sharing mechanism is that it is the tax which demonstrates the highest positive dynamics of revenues. The change has been considered by some of the central administrations; however, no decisions have been eventually made so far. Sub-national governments seem to overlook two potential dangers related to this form of revenue sharing. First, is related to the fact that also VAT revenues are vulnerable for cyclical fluctuations. It means, that the revenues may grow fast in times of positive cyclical growth, but they may drop fast in times of slow-down or economic crisis (although such cyclical volatility is lower than of corporate income tax). Second, VAT sharing makes no sense in the form of strict tax-sharing – as defined above. The 16 revenues collected from VAT within certain territory are poorly connected with the production activity in the given region. Obviously, it would be possible to introduce VAT sharing based on population or other allocation formula (as it is e.g. in Germany or Spain) but such option has not been discussed in Poland thus far. Moreover, introducing such an allocation formula would make the sharing system closer to the transfer (general purpose grant) with a defined amount to be allocated among sub-national jurisdictions than own revenue, as currently classified under the Polish nomenclature. Figure 6. Structure of regional government revenues (2018) own revenues 7% 9% PIT share 34% CIT share General purpose grant - vertical 37% General purpose grant - horizontal 2% 11% Specific purpose grants Source: Authors based on data from the Central Statistical Office Apart from excessive volatility, the there is one additional issue related to the present system of CIT-sharing, which makes the system questionable (even though it has not been widely discussed thus far). It relates to how CIT is distributed - according to the Law, the CIT is shared with subnational governments by origin. However, there are many companies (and especially the largest tax-payers such as banks or mobile telephone networks) which operate in more than one sub-national unit. Who should receive a share in the paid CIT in such a case? Formally the issue is regulated by the paragraph 10 of the Law on Sub-National Revenues. It stipulates that the amount of CIT is distributed among all sub-national governments in which the company operates, proportionally to the number of employed in each of regions (or municipalities in case of municipal shares). But in practice this regulation is not really implemented. Nobody bothers to allocate e.g. shares in CIT paid by a large bank among all 16 regions, over 300 counties and more than 2000 municipalities. In practice the share goes to the sub-national government in which the company is registered, exacerbating horizontal inequalities in revenues. This explains why city of Warsaw (where most of the largest, nation-wide companies have their headquarters) and the region of Mazovia have by far the highest revenues from this source. Per capita CIT revenues of Mazovia are over twice as high as of the next region in the ranking and over 5.5 times higher than of the region with the lowest CIT revenues. In case of revenues from PIT sharing the disproportions are much lower. Transfers to regional budgets The rest of regional revenues originates from the transfer system, which may be divided into three parts: (1) Vertical general-purpose grant (2) Horizontal equalization mechanism in the form of “Robin Hood tax” (3) Various specific purpose transfers. 17 Vertical general-purpose grant – i.e. the transfer which is funded by the national budget and which may be spend freely (with no control over its allocation) by the regional government. This transfer consists of two elements: • Equalization transfer, which partially compensates the regions with revenues per capita from tax sharing which are below the national average. The basic part of the equalization transfer compensates 70% of the difference between actual revenues from tax sharing per capita and the national average. However, the transfer is received only by regions with less than 3 million residents. It is difficult to find a justification for this limitation, which has been never spoken-out by the authors of the law, but its practical consequences are very limiting. There are only three regions with population over this threshold, one of them is Mazovia which is the richest Polish region, which would not be eligible for equalization transfer anyway; the two other are also relatively affluent. There is also “supplementary part” of the equalization based on per capita formula, but constructed in the way which favours the small regions (i.e. those with number of residents below 2 million). • Education grant which is part of the general-purpose grant (i.e. with no limitation for spending allocation) but with an intention to support provision of education services. In case of regions these responsibilities are very limited – they include mostly institutions responsible for on-job continuous training of teachers, adult education (after completing secondary schools – szkoły policealne) and some of special schools for mentally handicapped. But the education grant is very important for the revenues of municipalities and counties which are responsible for operation of primary and secondary schools. The detail formula of education grant is regulated by the Minister of Education on the annual basis (however with very small variation from one year to another). Basically, the formula is “per pupil” with several adjusting coefficients, the most important of them applied to rural schools, which are expected to have higher per pupil costs. The education grant being part of the general-purpose transfer is in fact an example of transfers which cannot be easily classified within the general purpose – specific purpose dichotomy. Boerbook and Huigsloot (2010) call this type of hybrid “earmarked general grant”. As we see on Figure 6, this part of transfers (equalization transfers plus the education grants) provides over 10% of total regional revenues. Horizontal equalization through the “Robin Hood tax” formula. As it is clear from Figure 6,the role of regional horizontal equalization is nowadays very limited (just over 2% revenues of all regions), but its political salience has been very high in public debates. While the equalizing part of the general-purpose grant represents vertical equalization (it is financed by the state budget), the balancing part introduces horizontal equalization between local governments. In the case of municipalities, it is financed by payments made by some of these local governments, whose per capita local and shared taxes are higher than 150% of national average; the formula to determine their contribution is relatively simple and it raises only very limited controversy (the details of calculating the “Robin Hood” payments are presented in the Table 4). But in the case of counties and regions, the marginal rate of “Robin Hood Tax” is very high, and therefore it has generated considerable political controversies. One of the criticisms is related to the fact that the formula for determining payments takes into account the revenue side only, totally ignoring the possible variation of spending needs. As explained below, the spending needs are somehow addressed in the formula of allocation of collected “Robin Hood tax”, but also in this case it is done in a very inaccurate (if not opaque) way. But the issue of the variation of spending needs is more important in the case of municipalities and counties (where there are more disparities in terms of population 18 density or intensity of social problems, such as unemployment; these issues are less prominent at the regional level. The calculation of the payments by the richest local governments takes into account the tax base of the most important local fees and taxes (see Table 5). In case of the regions, the scale of payments is not as progressive as in the case of the counties, but much more progressive than in the case of municipalities. Regions in which per capita revenues from shared taxes are higher than 110% of national average have to pay 70% of the “surplus”, while in the case revenues are over 170% of the national average, the rate of the “Robin Hood tax” rises to 95%. Table 4. Revenue thresholds used for calculations of payments of the “rich local governments” for the balancing grant. Municipalities Counties Regions Threshold of Rate of the “tax Threshold of Rate of the “tax Threshold of Rate of the “tax per capita own on surplus per capita own on surplus per capita own on surplus revenues income” revenues income” revenues income” compared with compared with compared with national national national average average average 150% 20% 110% 80% 110% 70% 200% 25% 120% 95% 120% 80% 300% 30% 125% 98% 170% 95% Table 5. Revenues taken into account in determining “Robin Hood” tax payments Municipalities Counties Regions Shares in CIT and PIT Shares in CIT and PIT Shares in CIT and PIT Property tax Tax on agriculture Tax on forests Tax on motor vehicles Tax on civil transactions Tax on exploitation of natural resources Stamp duty The Law does not provide a clear picture of the allocation of the grant funded by “Robin Hood scheme”, but delegates precise regulations to the ordinance to be issued by the Ministry of Finance. It says only, that the formula should take into account needs related to social welfare services and in particular to housing subsidies paid to the poorest tenants (see also summary in Table 6). Also, in case of counties, the precise allocation of the balancing grant is decided by an ordinance of the Ministry of Finance, and the formula should take into account needs related to social welfare services and roads’ maintenance. Table 6. Criteria for allocation of revenues from balancing (regional) grant Municipalities Counties Regions Spending on social welfare, Spending on social welfare - 20% for regions with especially on housing benefits Spending on county roads unemployment rate above 110% of national average 19 - 10% for regions with GDP per capita below 75% of national average - 40% for regions with high density of regional roads - 30% proportionally to spending on regional railways In case of regions, the amount collected from the Robin Hood contributions of the relatively richer regions is allocated in the following way: • 20% is allocated among regions with unemployment rate higher than 110% of national average; • 10% is allocated among regions with GDP per capita level lower than 75% of national average; • 40% is allocated among regions with higher than average density of regional roads; • 30% will be allocated in a way following Ministry of Finance ordinance which will take into account needs related to regional rail passenger transport. The practice shows that the criteria decided by Ministry of Finance reflects historical patterns of spending and location of the existing infrastructure to be financed—that is, measures of existing capacity, rather than objective indicators reflecting variation in spending needs. According to 2017 data the ‘Robin Hood” type payments were made by: • 97 municipalities • 22 counties plus 36 cities of county status • 4 regions. Beneficiaries of this part of general-purpose grants were: • 1421 municipalities • 379 counties and cities of county status (out of the total number of 380) • 14 regions. The list provided above clearly indicates several issues which have raised doubts and have been highly debated in Poland. The controversial issues may be summarized in following points: 1. Unnecessary (and costly) complication of the system arising from the fact that there are several local governments which first contribute with their “Robin Hood payment” and next they receive some part of the balancing grant. Sending money to the central government in order to get part of it back in the form of other grants would seem to please the banks charging their fees for the transfers, but it is not a transparent system of inter-governmental fiscal relations. 2. Very high marginal rate of the “Robin Hood tax” (up to 95% for regions and up to 98% for counties) is considered unfair by many politicians and part of the public opinion. The issue of dis-incentives for development policies is raised in that context as well. It seems that establishing high shares in income taxes was at the root of the problem. Politically it might be easier to decide for a lower local government share in CIT and PIT, mitigate the horizontal transfer systems and leave more money in the central government budget to strengthen the vertical grants payments. But from today’s perspective, the reduction of local government shares in CIT and PIT does not look like a politically feasible solution either. 20 3. The “time gap” issue has been perhaps the most debated (and criticized) during last few years, i.e. after the economic slowdown associated with the Great Recession has begun. It is probably the most salient issue, which has led to recent vibrant conflicts. The base for calculation of the “Robin Hood” payment of the affluent local governments is their revenue from relevant taxes collected in the base year, which is (with small simplification) two years before the budget year in which the payment is to be made. For several years it has not been seen as very urgent problem. But if three following conditions are combined: a. stagnating economy, which negatively influences tax revenues b. marginal rate of the payment is very high (95% or 98% respectively in case of regions and counties respectively) c. the tax in question is strongly vulnerable to cyclical fluctuations (as it is in case of Corporate Income Tax, which is the main tax revenue of Polish regions) the accumulated impact may be devastating. It explains why Mazovia regions (the main and until recently the only donor region) has protested the most loudly. And in 2013 Mazovia region has refused to pay part of its contribution which led to serious tensions with the Ministry of Finance. Several years back some of local governments contested the equalization regulation before the Constitutional Court, which in March 2014 rules that some of the regulations on horizontal equalization were unconstitutional. At the same time, the Courts gave the central authorities 18 months (i.e. till Autumn 2015) to prepare a new regulation. The Ministry of Finance suggested a temporary solution which would reduce the burden of the most affected local governments,7 and announced the preparation of the new, permanent solution. However, as of today, the temporary regulation is still in effect and discussions on the new, permanent solution are still far from being finished. Those temporary regulations change both amounts the collected by “Robin Hood” and the allocation of the regional grant. The marginal “tax rate” has been reduced from 95% to 59% and the system has imposed additional cap on the Robin Hood tax, which nowadays must not exceed 35% of total tax revenues collected by the regional government. The number of criteria for allocation of regional grant has been reduced from 4 to 2. Details of the change are presented in the Table 7. Table 7. The horizontal equalization of regional governments – temporary solution Calculation of payments by “rich” regional governments Criteria for allocation of regional grant Permanent (until 2015) Temporary (for 2015-2019) Until 2015 For 2015-2019 period Threshold of Rate of the “tax Threshold of Rate of the “tax - 20% for - 52% for per capita tax on surplus per capita tax on surplus regions with regions with revenues income revenues income unemployment unemployment compared with compared with rate above rate above national national 110% of 110% of average average national national average average 110% 70% 125% 49% - 10% for - 48% for regions with regions with tax GDP per capita revenues per below 75% of capita below national 125% of average 7 In case of regions, the horizontal equalization provides currently about 2.2% of total budget revenues (see also fig. 4). Before the temporary reduction of contribution of the most affluent regions the share was above 4% of total revenues of regions. 21 120% 80% 150% 59% - 40% for national regions with average high density of regional roads 170% 95% In addition, total payment cannot - 30% exceed 35% of tax revenues proportionally to spending on regional railways The third component of the transfer system consist of several, sectoral specific purpose grants. As it is clear from the Figure 6, these conditional transfers provide more than one third of all regional budget revenues. This funding source has been criticized for being very fragmented, for being heavily controlled by sectoral ministries, and for rarely providing fully transparent rules for the allocation of resources among subnational units. In addition, these transfers are not predictable. Obviously, the specific purpose grants should not be totally eliminated. They are justified – as for a limited period of time - in case of transfer of new functions. But more fundamentally to support well-defined sectoral policies and programs of the central governments that can be better delivered by subnational governments and to address externalities and other well-known justifications for conditional central government transfers. Nevertheless, the current situation the system is definitely too complex, not sufficiently transparent and constitutes an unnecessarily large part of total revenues of the Polish regions’ budgets. In 2018 the total value of specific grants for regional governments was almost 6 billion PLN. The vast majority (4.5 billion PLN) was related to EU supported project – either as direct grant from operating programs or as central budget support to finance matching funds necessary in order to apply for the EU funding. About 55% of all specific grants were allocated as capital grants and 45% to support current expenditures. The most significant specific grants programs include: 1. 1.75 billion PLN for transport. This amount allows to cover: a. 715 million for investments and 220 mln for maintenance of regional roads b. 600 million to support regional bus transportation (mostly to refund costs of reduced fares for students, pensioners etc.) c. Over 125 million to support operating costs of regional railways 2. Almost 150 million of various grants for social care 3. Over 100 million of various grants for agriculture (mostly amelioration of lands). Most of specific grants is operated by Ministry of Development, but some programs are also handled by Ministry of Finance, Ministry of Labor and Social Protection, Ministry of Culture and other line ministries, and discretion of ministries in allocation of some of the grants among regions is considerable. The amount of grant is typically provided by the Ministry in the stage of budget preparation (i.e. before November of a year before the budget year), but it happens that the sum is being changed during the budget year (typically the increase is financed form the budget reserve line). Technically speaking, most of the central grants are transferred to regional governments through the regional governor, but his (her) role is almost purely technical, with no discretion to change the amount or formulate additional conditions. 22 Subnational borrowing Sub-national governments in Poland have a right to take out loans or issue bonds in the financial market without asking for a priori permission from the central government. However, sub-national loans are subject to regulations and ex post control. The first fiscal rule concerning borrowing is written in the Polish Constitution of 1997. Poland is perhaps the only European country which adopted the Maastricht recommendation of total public debt not exceeding 60% of GDP in its Constitution. In addition, the Law on Public Finance specifies several borrowing limits and fiscal rules, which automatically apply when total (consolidated) public debt exceeds 50% of GDP. More severe rules apply when total debt exceeds 55% of GDP. The rules and limits apply to both central and sub-national government debt; however, these limits are not in any way divided between central and sub-national budgets. The practice of the last 20 years has demonstrated that the bulk of the limit has been eaten up by central government, often utilizing debt to finance its social programmes. The share of sub- national governments in total public debt has never exceeded 10% of the total. The Law on Public Finance specifies also more detailed rules for sub-national borrowing. These regulations have undergone two clear phases: (i) before 2013, and (ii) after 2013. Until 2013 the law specified the limit of the size of accumulated debt by the end of the fiscal year, which could not exceed 60% of the total budget revenues of local government unit. In addition, the debt service in a given year could not exceed 15% of total budget revenues. For some time, the law did not distinguish between the different purposes of borrowing; there was no separation of capital and operating budgets, so the loan could be taken both for investment purposes or in order to finance current expenditures. That situation changed in 2013 when the new Law on Public Finance adopted in 2009 was fully implemented. The first change was to require a balanced operating budget of local governments, which indirectly implied that borrowing would be available for capital investment purposes only. The second, and perhaps the most important change was to rescind the limits for the total size of borrowing, and to adopt in its place the model in which borrowing capacity is dependent on the operating surplus of the budget. The new rule – with some simplification – may be summarized in the following way: (i) The mean of operating surplus for last three years is calculated for each sub-national unit (ii) The new loan is allowed if the annual service of debt in the future will not exceed the mean calculated in step (i). All subnational borrowing is subject to the control of RIO (Regional Chambers of Audit), which are described in the more detail way in the next section. These rules apply to the budget debt, but the companies owned by sub-national governments are exempted from these regulations. It happens that the willingness to avoid debt limits has been among the motives for the corporatisation (establishment of companies) for local public services’ delivery; however, this phenomenon (if ever) applies much more commonly at the municipal level as opposed to the regional tier. Another controversial regulation concerns the procedures used in cases in which sub-national entities have problems with servicing their debt. In such a situation, the local government has to approve a recovery programme, which needs to be agreed with the Regional Chamber of Audit. It often includes painful cuts in spending. But at the same time, the local government can receive a low-interest loan from the central budget which allows them to repay the outstanding debt. Note that there are no regulations concerning potential bankruptcy of sub-national jurisdictions. It means that the risk for the creditor (usually bank) is next to non- 23 existing. Even if a local government cannot serve its debt, it is ensured that it will receive a support from central government and the risk for the bank to not receive its money back is almost zero, which introduces a serious moral hazard problem. In practice, so far there has been no cases of insolvency of regional governments, and only very few cases of municipalities or counties which got into troubles related to their debt. The most spectacular was a case of quasi-bankruptcy of a small, rural municipality which eventually was liquidated (its territory was divided between two neighbouring municipalities). But in general, local governments borrowing policies have been usually prudent. Between 2004 and 2017 the total amount of low-interest loans from the central budget to subnational governments (in order to support their debt service) has been about 700 mln PLN (over 150 mln euro) and they were incurred by less than 30 local governments (ca. 1% of all sub-national jurisdictions), usually at the municipal level. Figure 7 illustrates the trend in the level of accumulated debt of sub-national governments during last 25 years. It shows that in the decade of 1990s Polish subnational governments were very reluctant to use borrowing instruments, and borrowing policies were limited to a small group of innovative local governments (usually major cities). In the first decade of 21st century the level of debt grew dynamically, but it did stabilize after 2010. Figure 7. Level of sub-national debt in Poland (million PLN) 80000 60000 40000 20000 0 2009 1995 1998 2000 2001 2002 2003 2004 2005 2006 2007 2008 2010 2011 2012 2013 2014 2015 2016 2017 2018 Municipalities Cities of county status Counties Regions Source: Authors based on data from the Central Statistical Office Figure 8 confirms that the level of sub-national debts still constitutes only small fraction of total public debt and it is a minor issue from macro-economic point of view. It has been never higher than 5% of GDP and during the last few years it has been close to 3.5%. It also shows that the regional debt is a small fraction of sub-national borrowing (most of the sub-national debt is related to loans taken by municipalities, especially major cities). Figure 9 confirms the marginal role of sub-national in total public debt. 24 Figure 8. Debt of sub-national jurisdictions as proportion of GDP in Poland 5 4 3 2 1 0 2013 2014 2015 2016 2017 2018 Regions Other sub-national jur. Source: Authors based on data from the Central Statistical Office Figure 9. Public debt as proportion of GDP 60 50 40 30 20 10 0 2013 2014 2015 2016 2017 2018 sub-national central govt. Source: Authors based on data from the Central Statistical Office The main form of borrowing in Poland are loans from commercial banks. Bonds are less popular, and even if they are issued, they are usually bought by one bank and they are often some related forms of bank loans, which are also called bonds for formal reasons, because some regulations make bond issuing cheaper for both banks and city governments than the typical commercial loan. Nevertheless, several Polish sub-national governments apply for rating scores from international rating agencies (Fitch being perhaps the most popular in the country to a large extent because of promotion reasons). As already indicated above, for example, in 2010 five Polish regions had an A- rating from Fitch. Bonds constitute no more than 15-20% of the total debt of the regions. Budget process – the role of the MEF The rules of budget preparation, implementation and budget reporting are provided in the Law on Public Finance. In case of regions, the Regional Executive Board is obliged to prepare the draft of the budget for the next year before 15 November of the preceding year. This draft is sent to the regional council and to the Regional Chamber of Audit to ensure it complies with all legal requirements. The draft budget is discussed 25 by the council, which may change the allocation of expenditures and even the projection of revenues; however, it has no right to propose changes which would result in an increase of the budget deficit (to be financed by borrowing). As a matter of principle, the budget for the next year should be adopted by the regional council before the end of December. But limited delays in the budget resolution are acceptable; in such cases, in the first months of the new year the draft budget is the base used to authorize any necessary spending. However, if the budget is not agreed and voted by the end of February, the Regional Chamber of Audit (RIO) may impose the budget prepared by the RIO. The communication between Ministry of Finance and regional government is important for the budgeting process. Large part of revenue projections, especially related to shares in PIT and CIT revenues, as well as general purpose grants, is based on the forecasts provided by the Ministry. Since regions – opposite to municipalities – do not have their own regional taxes, the role of their own revenue projections is much smaller. The budget may be amended during the year, but the only institution allowed to propose amendments is the Regional Executive Board. After the end of the year, the Regional Executive Board prepares report on budget execution which has to be submitted to the regional council (and also to the Regional Chamber of Audit) before the end of March of the subsequent year. The detail report follows the official budget classification and is publicly available on the web-site of the Ministry of Finance. (The reports are sent to both Ministry and Finance and RIO not only after full budget year, but also after every quarter). The 2009 Law on Public Finance also requires the preparation of a multi-year prognosis of budget revenues and spending. The minimal time horizon for such a prognosis is 3-years, and it is prepared int the “rolling system” (the document for next three years adopted on an annual basis). Most of sub-national governments prepares these prognoses in parallel with their multi-year investment programmes. Supervision, transparency and accountability Financial supervision Ensuring the fiscal probity of subnational governments is always a challenge. One way to meet this challenge is to make local governments finance themselves, placing the regulatory burden on taxpayers and creditors. Another is to make a government agency or a specialised court responsible for financial oversight. Poland’s reformers never seriously considered the first, because they knew local governments would be heavily dependent on transfers. But they also rejected the second because they were afraid that a national government agency might use financial oversight to meddle in local politics. Instead, a new financial oversight institution was developed – Regional Chambers of Audit (RIOs). In 1992 that seventeen RIOs were actually created. They were made independent of all line ministries and their boundaries extended across regions, making their jurisdictions different from all other government institutions. Each RIO operated independently, though their directors belonged to a Collegium designed to co-ordinate their activities. Most importantly, while the Prime Minister nominated their Directors, but it has always been non-political position and the leadership of the RIOs remains unchanged in the case of changes in government (there are still 2 or 3 Presidents of RIO who have hold their posts since very beginning of RIO’s operation in 1992). In addition, the nominees had to be approved by the Sejmiki, who also named half of their managing boards. The result was a decentralized, co-governed institution whose “hybrid” structure and operational independence were explicitly seen as ways to prevent state overreach while fostering workable financial standards. Also, the staff of RIO was recruited from the recent college graduates or accountants and lawyers drawn from industry to ensure independence from the central government. 26 The primary aim of the RIO is to oversee the “financial economy of local governments”, which is divided into four main functions: Control, Oversight, Opinion Giving, and Information and Training. Under “Control” they review the legality – not substance – of all local budget resolutions (e.g. budget plan, report on budget execution, resolutions on rates of municipal taxes, compliance with legal rules concerning local borrowing). They can declare resolutions illegal and even take temporary control of a municipality’s finances if its Council fails to comply with applicable law. In the cases of conflicts over legality, the final decisions belong to the court system. All checks on legality are performed by the so called “collegium of RIO”, which consists of President, Vice Presidents, and experts (usually lawyers and economists from local universities) invited by the President. Under “Oversight” they must audit the finances of all local government once every four years and they must issue non-binding opinions – “Opinion Giving” – on a local government’s ability to service prospective debts. Finally, they are responsible for developing annual training programs for their jurisdictions. Transparency and participatory mechanisms Perhaps the most important regulation concerning the transparency of subnational governments is related to public access to information. This is implemented in practice by several means, including: - Compulsory publication of several information on web-sites of regional governments; - Right of citizens to participate in the meetings of regional council. Agendas and minutes of meetings are also published in the internet, and several regions allow to watch the council meetings live in the internet - Very detail reports on budget execution of each of sub-national government are publicly available on the web-site of Ministry of Finance - Citizens have a right to demand access to documents and information on the basis of the “right of access to public information”. As regards the direct impact of residents on decision making, the most far going regulation concerns the possibility of local and regional referenda. However, so far there has been only one case of a regional referendum. However, several local referenda – usually at the municipal level – are organized every year. This only case of a regional referendum concerned the building of a regional airport in Podlasie region (north-eastern Poland), held in 2017. The law on referenda provides slightly different regulations concerning two types of referenda. The first one is related to early termination of the regional council and regional executive board. Several conditions need to be met in order to decide on early termination of the council: - Petition to organize referendum signed by no less than 5% of eligible voters; - Majority of votes for early termination - Turn-out in referendum no less than 3/5 of the turn-out in the most recent regular elections. In addition, such a referendum must not be organized in less than one year after the regular election, less than half a year before the next regular elections, nor less than a year after the previous referendum. All other referenda (on any other issue than early termination of regional council) are organized according to very similar rules, with one important difference concerning the threshold of turn-out required in order referendum to be valid. In this case, the threshold is set on the level of 30% of eligible voters. The tool which has gained considerable popularity in Poland in recent years has been participatory budgeting. It has been voluntarily introduced by most of Polish large-- and also by several smaller-- towns, 27 and in 2017 it has become compulsory for cities of county status. Overall, this form of community involvement as well as several other forms of participatory democracy seem to be much more suitable for a local (municipal) level than for regions. Nevertheless, there is at least one Polish region which has used participatory budgeting mechanism. Usually in Poland the tool has involved very limited amount of funding (typically no more than 1% of the budget). Another line of criticism concerns the fact that contrary to original assumptions, participatory budgeting has little to do with public deliberation. Instead, it is rather a form of plebiscite or “beauty contest” in which citizens vote for projects submitted by various NGOs or local activists. There is also gradually increasing role of various consultative bodies functioning on a regional (and even more on a local) level which are a form of involvement of civic society in the process of governance. One example of such civic institutions are the labour market councils (LMCs), which have been created on the basis of a national law, the Act on Employment Promotion and Labour Market Institutions, adopted in 2004. These councils have been created at the state, the regional and the county levels. The councils work with the regional and local labour offices, a sectoral administration subordinated to the regional and county governments providing services to the employers and unemployed. The labour market councils were established in response to the unemployment crisis observed in early 2000 in Poland following the massive collapse of the state-owned companies in the first years of the economic transition after 1989. Another example of consultative institutions is the “Public Benefit Councils” which are a for a to discuss cooperation of regional or local governments with non-governmental organizations. Their functioning has been regulated by the Law on Public Benefit and Volunteerism from 2003. And since 2015 their creation has been obligatory on request of local NGOs. An important part of the role of civic society in preparation and implementation of development policies is related to the EU-co-funded Regional Operating Programmes. Each of the Programmes has its Regional Monitoring Committee which includes representatives of local businesses, societal organizations and experts from local universities. The Committee is the forum for discussion of development priorities and details of the programmes’ implementation. Other country-specific observations Role of regions in regional development policies The role of regions in sub-national investments is not very important (less than ¼ of total sub-national investment spending). The major regional investments concern regional roads, hospitals and amelioration of arable lands. The evolution of investment spending of subnational governments is shown on Figure 10. 28 Figure 10. Capital spending of sub-national governments (mln of PLN, constant prices of 2018) 50000 40000 30000 20000 10000 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 municipalities cities of county status counties regions Source: Authors based on data from the Central Statistical Office But, as previously discussed, the major role of regions is not related to their own investment projects but to strategic planning, which has an impact on projects implemented by other – public and non-public – actors. The concept which was discussed already during 1998 reform was the concept of regional contracts, to a large extent borrowed from the French experience. The first contracts were prepared already in 1999 and were based on central government funding made available to regional governments for agreed development priorities. The legal base for regional contracts were defined in the May 2000 Act on Rules of Support for the Regional Development (Ustawa o zasadach wspierania rozwoju regionalnego). The Act obliged central government to prepare the National Strategy for Regional Government (such a strategy was approved in November 2000). At the same time, regional governments were obliged to prepare their own regional development strategies. Those were the basis for negotiation of the regional contracts. One of the main lines of criticism of regional governments’ role in the formulation and implementation of the regional policy in that period was the observation of a very weak financial position of the regional governments themselves. As a result, they were clients (or beneficiaries) rather than partners of the central government in the contract negotiations. Michał Kulesza (2002) spoke about institutionalization of centralism based on resources rather than (as before the 1998 reform) on administrative powers – the center gives up managing people, functions, and services but keeps the power to manage through financial resources. This criticism seems to be valid but a bit exaggerated. Regional governments were gradually building their political power, also on the basis of their role in regional contracts. But there is no doubt that the amounts of funds involved in that period were very low, and since regional governments were financially weak, the shape of the contracts was to a large extent imposed by the central government, who was the stronger partner in the negotiations.8 The situation started to change gradually from 2000 with the process of accession to European Union. First, from 2000 Poland had access to pre-accession programmes (effectively spent on investment projects from 2002). Since 2004 Poland has gained access to EU Cohesion Fund and Structural Funds, a considerable part of which has been allocated to regional development. Regions have played an important role in 8 For more about contracts in the first years of regions’ operation see Swianiewicz 2009. 29 implementing those development policies. Formally, the funding still goes through the system of regional contracts, but now the negotiations over the contracts are trilateral – with participation of regions, central government and the European Commission – and the bulk of financial resources comes from the EU budget which is planned for 7-years perspective (in the case of Poland 2004-2006, 2007-2013 and 2014-2020). As shown in Figure 11, the amount of EU funds for local governments was very small in the pre-accession period (2000-2004), but has grown significantly in the following years. The direct share of regions in EU structural funds’ absorption is rather small, but this picture would be misleading from the point of view of the role of regions in the whole process. The majority of EU funds allocated to municipalities and major cities is dependent on the strategic decisions adopted by the regional government, and the funds are distributed through the Regional Operating Programme, which is prepared and managed by the region. Figure 11. EU funds in sub-national government spending (million of PLN) 18000 16000 regions 14000 counties cities of county status 12000 municipalities 10000 8000 6000 4000 2000 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Source: Author based on information from the Ministry of Finance and Ministry of Regional Development Figure 12 illustrates the role of EU funding (mostly of Regional Operating Programmes) in financing investments of subnational governments in Poland. EU grants have provided usually about one fifth of the total funds for investments, but in some years (2011-2015) it has been closer to 30%. The role of EU funding though not dominant is definitely very important. Figure 12. EU funds as % of total capital spending of sub-national governments 35 30 30.6 29.0 29.2 28.4 27.9 25 24.9 20 19.1 18.2 15 14.5 13.9 12.0 10 9.7 5 6.3 4.6 0 2004 2006 2008 2010 2012 2014 2016 Source: Author based on information from the Ministry of Finance and Ministry of Regional Development 30 The first EU co-funded comprehensive programme focused on regional development, in which Poland could participate, was the Integrated Regional Operational Programme (IROP) for 2004-2006 period. The IROP was managed nationally by the Ministry of Regional Development. But regional governments were offered some amount of discretion to decide on the amounts of money devoted for certain objectives. Regarding implementation, most tasks were vested with regional government institutions. The Marshal Office (regional self-government) was responsible for projects identification and selection, whereas the regional state administration (governor) performed its role in audit, monitoring, payment verification and certification. According to procedural assumptions, the Regional Steering Committees are the main instruments for horizontal cooperation within the regions, advising regional governments on particular investments and initiatives within the regional contracts. The Committees were appointed by executive boards of the regional governments, chaired by the head of regional self-government administration (Marshal), and consisted of representatives of regional NGOs, business, regional state administration, ministries and respective self- governments: counties and municipalities. Similar structure holds for the IROP Monitoring Committees. It consists not only of the representatives of Ministries, Region Offices and regional self-governments (marshals) but also of the representatives of self- governments at the county and municipal levels. Additionally, there exist 16 regional monitoring subcommittees that are responsible for the control of the effectiveness of the appraised projects and as such are not involved in the appraisal of the projects. The management structure of IROP is summarized on the Figure 13. Projects from the whole region were submitted to the regional self-government administration which forwarded them to Panel of Experts for technical assessment. The Panel of Experts was an independent body assisting the IROP Managing Authority in the evaluation process. Each region had its own set of panels – one for each support area. Panels consisted of three persons that were chosen by the regional marshal in accordance with the following key: one expert appointed from the list elaborated by the marshal (head of regional self-government administration), one from the list of a regional governor, and one from a list prepared by other (relevant) regional institutions. It should be mentioned that those experts were expected to work in two stages: in the first, they should look at the proposals judging if they contribute to the development of the whole region, and then assess them from a technical, specialist point of view. 31 Figure 13. IROP (Integrated Regional Operating Programme) 2004-2006 management structure European Commission Ministry of Finances Ministry for Regional National (Financial Institution) Development (Managing Monitoring Authority) Committee Regional Executive Regional Monitoring Committee Governor/ Central administration Board (Intermediate Body) Regional Steering Committee (appointed by Regional Executive Board, chaired by a Marshal Panel of Experts appointed by a Marshall Regional Marshal Office Final Beneficiaries Applications’ flow Financial flow signing agreements Source: Author After having been assessed and ranked, the projects were presented to the Regional Steering Committee, which could change the ranking especially taking into consideration their cohesion with and significance for the regional development strategy. The approved ranking list was forwarded to the regional executive board which also could have changed the ranking again and it made a final recommendation of the regional level. At this stage the final list was presented to the Ministry. Finally, the regional executive board was informed on the selected projects and forwarded them to the regional governor’s office, which was signed the Agreements Granting Structural Funding with Final Beneficiaries. That procedure adopted for the IROP 2004-2006 was considered unnecessarily complicated, and the division of responsibilities between involved institutions was not always clear. This resulted in difficulties during the implementation phase. The peculiar position of the regional governor and regional self- government in the implementation of the IROP resulted in the necessity of introducing institutions which had not been prescribed by EU regulations. 32 In the 2007-13 period and the following the 2014-2020 budget the responsibility for preparing and implementing the regional development programs has been to a large extent delegated to the regional self- governments (Figure 14). Figure 14. ROP (Regional Operating Programs) 2007-2013 and 2014-2020 management structure European Commission Ministry of Finance Ministry of Regional National Monitoring (financial institution) Development Committee Governor Regional Panel of Experts Executive Board appointed by a Marshall Cooperation Conference (headed by a state governor (Marshal is a vice-chairman) Regional Monitoring Committees Final Beneficiaries Source: Author The IROP was divided into 16 Regional Operational Programmes (ROPs) elaborated by the regions themselves (although in negotiations with European Commission and central government). The Marshal office is directly responsible for preparing a draft of the ROP. The only restrains that come from upper tiers are the rules of European Commission and some basic guidelines and recommendations of central government (as for the percentage of fund to be dedicated to social aims or to realisation of Lisbon Strategy goals). The Marshal office is also obliged to carry out consultations with potential beneficiaries. Additionally, a kind of advisory role is played by the Governor’s office. The new law abolished the Regional Steering Committees leaving decisions in the hands of the Regional Executive Board (within Marshal Office). Initially, those new regulations on regional development policies raised a lot of controversy related to the burning issue of the division of competences between the regional self-government and the governor. Hot debate followed regarding the parliament proposal that the governor would be allowed to veto (even not giving a reason) the selection made by the regional government. This suggestion was strongly criticized by several experts, as well as local and regional politicians (including the Association of Polish Regions), 33 some of them suggesting that it might not comply with the Polish Constitution and/or EU regulations. Interestingly enough, the “veto clause” was proposed by a group of MPs from the governing coalition, while the Ministry for Regional Development was against such a regulation. Eventually, after long debates the veto right for the governor has not been implemented, and regional elected self-governments have gained considerable discretion in selecting the individual projects to be implemented. Considering a huge role of EU funding for the regions, it is important to ask the question what the evolution of regional governments after the EU funding expires would be. It is clear that the level of funding available in 2014-2020 budget will decrease substantially in 2021-2027 budget, and after 2027 it is even less certain. In one of the recent reports on decentralization (Hausner 2013) it has been stated that “regions have to re - invent themselves” thinking about the “post EU structural funds era”. But the reflection on this issue is relatively scarce. One may speculate that the political role or regions will diminish to the similar which was played in 1999-2002 period, before the EU funding became significant. Regional governments will deliver services they are responsible for, but the role in regional development programs – supported just by relatively minor funding offered by regional contracts signed with central government - will be relatively marginal. Lessons learnt and insights for other countries 1. It is generally believed that the decentralization to regional governments devolving more discretion and powers to elected self-governments has had a positive impact on effective and the efficient implementation of policies. Poland is the only of EU “New Member States” (countries who joined EU in 21st century) which decided to give considerable autonomy to regional governments in the preparation and implementation of their own Regional Operating Programmes. Comparison of disbursement rates in Poland with another countries suggests that it has helped to achieve higher absorption rate of available funds. This claim is commonly accepted and repeated in several EU level documents on regional policies. Also, the evaluation reports of ROPs usually confirm the good management of the programmes. Finally, in my own research from 2011-2013 most of the researched municipal beneficiaries of the EU supported projects have indicated that the programme managed at the regional level have been more “user-friendly” and better tailored to local environment than programmes managed directly from the centre (Swianiewicz et al. 2013). It is difficult to find hard, quantitative evidence illustrating gains of regional decentralization. Searching for such evidences is methodologically very complex. It is easy to illustrate progress in service delivery or in development projects, but it is hard to prove causal relationship between decentralization and those positive trends. The indirect evidence might be searched in comparison with other countries in the same part of Europe which adopted alternative (more centralized) models of managing regional development policies. This comparison confirms gains of decentralization, but there are so many other factors which should be controlled in the comparison that methodologically rigid proof of the causal relationship is extremely difficult. 2. Financial decentralization to the regional level has been to a large limited to expenditure decentralization Revenue autonomy of regions (in contrast to the municipal level) is minimal – more than 1/3 of regional budgets is funded by earmarked grants and the rest originated either from the general-purpose block grant or tax sharing. The high role of conditional grants is still a problematic issue. Of course, there are reasons for some role of this type of transfers for example for well-defined social programs; however, the size of those programmes still seems to be too large. Moreover, the rules of funds allocation are neither sufficiently transparent nor stable. The whole system is fragmented among small programmes run by various ministries. 34 3. Horizontal equalization through the Robin Hood mechanism has always been politically controversial. This is experience of other European countries using this kind of mechanisms (Denmark, Sweden, Norway, Germany, Switzerland, Spain) and it is also confirmed in the Polish case. This political sensitivity remains even if the amount of funding redistributed in that way is relatively limited (as it is in Poland). A different design of the regional tax sharing might allow to avoid these political difficulties, without negative consequences for the disparities in the level of regional governments’ income. 4. The lack of hierarchical subordination between sub-national tiers seems to be an important feature of the Polish system. The positive consequence of such a solution is a clearer local accountability of elected subnational governments. In Poland before 1990 there has been a long period of a strict vertical subordination of lower tiers, and breaking those hierarchical links seemed to be natural after 1990 political turn-over. It has been almost identified with democratization. There are no obvious negative side-effects of such a design. Regional programmes and mechanisms of investment grants allocation through ROPs and other programmes secure considerable amount of regional coordination allowing coherent strategic development planning. 5. Polish regions are not protected by the Constitution. The Constitution adopted in 1997 mentions only the municipal level, leaving the decision on existence of counties and regions to the will of the Parliament. The reason of this fact is that the Constitution was adopted before the debates on number and territorial shape to sub-national tiers had been completed. Anyway, the current legal protection of regions is relatively weak. This issue may become more important nowadays, since the central government in power since 2015 has not been supportive of decentralization. On the contrary, most of the key political actors of the current governing party do not trust subnational governments and have a tendency towards re-establishing stricter control by the centre. The weak protection of regions makes them even more vulnerable towards re-centralization tendencies. In the current situation one may indicate at least a few examples of slow, crawling re-centralization limiting regional government powers and competencies. One is the taking control over the centre control over Regional Funds and Environment Protection and Water Management (Wojewódzkie Fundusze Ochrony Środowiska i Gospodarki Wodnej – WFOŚiGW), which were previously subordinated to regional governments. Another similar example is related to the Agricultural Advisory Centres ( Ośrodki Doradztwa Rolniczego – ODR). Last but not least, the role of regional governments in managing regional development programmes in the 2021-2027 European budget has not been clarified yet. The danger of reducing the regional discretion in this respect and stricter control by the centre is still a possible option. 35 References: Blöchliger, H., Petzold, O. (2009) „Finding the Dividing Line Between Tax Sharing and Grants: A Statistical Investigation“, Paris: OECD Network on Fiscal Relations Across Levels of Government. Boerboom, H., Huigsloot, P. (2010) „Earmeked General Grants and General Earmarked Grants in the Netherlands” ” [w:] J. Kim, J. Lotz, N.J. Mau (eds.) General Grants versus Earmarked Grants: Theory and Practice, Albertslund-Seul: The Korea Institute of Public Finance and The Danish Ministry of Interior and Health, p. 394-414. Emilewicz, J. & Wołek, A. (2002) Reformers and Politicians: The Power Play for the 1998 Reform of Public Administration in Poland, as Seen by its Man Players, Elipsa: Warszawa. Hughes, J., Sasse, G. & Gordon, C. (2003) “EU Enlargement, Europeanisation and the Dynamics of Regionalisation in the CEECs” [in:] M. Keating, J. Hughes (eds.) The Regional Challenge in Central and Eastern Europe, PIE-Peter Lang: Bruxelles-Bern-Berlin-Frankfurt-New York-Oxford- Wien. Kowalczyk, A. (1993) “Dotychczasowe doświadczenia i obecny stan prac badawczych nad refomrą podziału Polski na regiony” [in:] G. Gorzelak, B. Jałowiecki (eds.) Czy Polska będzie państwem regionalnym?, Warsaw University. Hughes J. Sasse G., Gordon C. (2003) “EU Enlargement, Europeanisation and the Dynami cs of Regionalisation in the CEECs” [in:] M. Keating, J. Hughes (eds.) The Regional Challenge in Central and Eastern Europe, PIE-Peter Lang: Bruxelles-Bern-Berlin-Frankfurt-New York- Oxford-Wien Kulesza M. (2002) “Methods and Techniques of Managing Decentralisation Reforms in CEE Countries: The Polish Experience” [in:] G. Peteri (ed.) Mastering Decentralisation and Public Administration Reforms in Central and Eastern Europe, Open Society Institute: Budapest Mażewski, L. (1992) „Przyszłość województwa” [in:] Samorząd terytorialny w Polsce. Materiały na seminarium 29-30.09.1992, Warszawa: Senat R.P. Regulski, J. (2003) Local government reform in Poland: an insider story, Budapest: LGI – Open Society Institute. Swianiewicz, P. (1999) The long history of radical change: local government and regional policies in Poland [In:] M. Tatur (ed.) The challenge of regional policies: actors and institutions of the local and regional level in post-socialist Europe, Deutsche Gesellschaft fur Technische Zuzammenarbeit, Eshborn. Swianiewicz, P., Krukowska, J., Lackowska, M. (2013) Profesjonalizacja absorpcji funduszy: beneficjenci samorządowi centralnie zarządzanych programów operacyjnych w okresie 2007-2013, Warszawa: Ministerstwo Rozwoju Regionalnego. 36 Further readings in English: Emilewicz, J. & Wołek, A. (2002) Reformers and Politicians: The Power Play for the 1998 Reform of Public Administration in Poland, as Seen by its Man Players, Elipsa: Warszawa. Kulesza, M. (2002) “Methods and Techniques of Managing Decentralisation Reforms in CEE Countries: The Polish Experience” [in:] G. Peteri (ed.) Mastering Decentralisation and Public Administration Reforms in Central and Eastern Europe, Open Society Institute: Budapest. Levitas, A., 2018, Local government reform as state building: What the Polish case says about “decentralisation”, Public Governance, No. 3(45) Regulski, J. (2003) Local government reform in Poland: an insider story, Budapest: LGI – Open Society Institute. Swianiewicz, P. (2013) “Poland: Creeping Decentralization of the Unitary State” [in:] J. Loughlin, J. Kincaid, W. Swenden (eds.) Routledge Handbook of Regionalism and Federalism, Milton Park – New York: Routledge, p. 331-340. Swianiewicz, P. (2011) Poland [in:] H. Heinelt, X. Bertrana (eds.) The Second Tier of Local Goverment in Europe: Provinces, Counties, Departments and Landkreise in Comparison, London: Routledge. Swianiewicz, P. (2010) Poland: Europeanization of Subnational Governments, [in:] J. Loughlin, F. Hendriks, A. Lidstrom (eds.) The Oxford Handbook of Local and Regional Democracy in Europe, Oxford University Press, p. 480-504 Swianiewicz, P. (2006) Local Government Organization and Finance: Poland, [in:] A. Shah (ed.) Local Governance in Developing Countries, Washington: The World Bank, p. 302-346. 37 Annex 1. Basic information about Polish regions Name Population in GDP PPS per Capital city and its population (in ,000) million (2017) capita in euro (2017) Lower Silesia 2.86 22,100 Wrocław 639 Kujawsko- 2.06 16,300 Bydgoszcz 352, Toruń 203 (*) Pomorskie Lublin 2,12 13,700 Lublin 340 Lubusz 1,01 16,700 Gorzów Wlkp 124, Zielona Góra 140(*) Łódź 2.48 18,600 Łódź 690 Lesser Poland 3.33 18,100 Kraków 767 Mazovia 5.32 31,700 Warszawa 1,765 Opole 0.95 15,900 Opole 128 Podkarpackie 2.08 14,000 Rzeszów 190 Podlasie 1,16 14,100 Białystok 297 Pomerania 2,28 19,300 Gdańsk 464 Silesia 4,52 20,700 Katowice 296 Świętokrzyskie 1.24 14,300 Kielce 197 Warmia-Mazury 1.41 14,200 Olsztyn 173 Greater Poland 3.45 21,700 Poznań 539 Western Pomerania 1.68 16,700 Szczecin 4040 (*) in two regions (Kujawsko-Pomorskie and Lubusz) there are two capital cities – one is a seat of the regional governor and the second of the regional elected council and its executive board. 38 Fiscal decentralization of regional governments – the case of Spain prepared under the World Bank TA RAS for Chile1 1 The note was written by Santiago Lago-Peñas, GEN – Universidad de Vigo (España), http://slagop.webs.uvigo.es/.. 1 Contents 1. Introducción y Contexto ....................................................................................................................... 3 2. Estado Actual del Modelo de Descentralización .................................................................................. 3 3. Cambios Relevantes en el Modelo de Descentralización: el Iter del Proceso ...................................... 5 4. Asignación de Responsabilidades de Gasto .......................................................................................... 7 5. Las Fuentes de Ingreso.......................................................................................................................... 8 6. Endeudamiento Regional y Estabilidad Fiscal.................................................................................... 15 7. El Proceso Presupuestario ................................................................................................................... 19 8. Auditoría, Control y Rendición de Cuentas ........................................................................................ 19 9. El Impacto de la Descentralización: Evidencia Empírica ................................................................... 21 10. Lecciones del Caso Español.............................................................................................................. 22 Anexo I: La Economía de las Regiones .................................................................................................. 24 Anexo II: Evolución de las Cuentas Regionales ..................................................................................... 29 Anexo III: Índice General de Transparencia ........................................................................................... 32 Anexo IV: Información Administrativa .................................................................................................. 33 REFERENCIAS ...................................................................................................................................... 34 1. Introducción y Contexto 1. Tras un largo período de casi cuatro décadas de dictadura, la Constitución aprobada en el año 1978 define un nuevo marco democrático, en el que la descentralización política y financiera y la creación de un nuevo escalón regional de gobierno en España ocupa un lugar relevante. La redacción concreta del texto constitucional posibilitaba evoluciones distintas, tanto en lo que atañe al número de regiones, como a sus fronteras y competencias. Esta indeterminación ayudó, sin duda, a un respaldo muy mayoritario de los ciudadanos a la nueva Constitución, porque pocos agentes de la época podían siquiera imaginar la intensidad y profundidad del proceso; empezando por los defensores de la tradición centralista. 2. En la actualidad, existen tres niveles de gobierno en España, al margen del escalón supranacional europeo. La Administración central, que incluye la Seguridad Social; 17 regiones, llamadas oficialmente Comunidades Autónomas (CCAA)2, y 2 pequeñas Ciudades Autónomas situadas en el Norte de África;3 y las Corporaciones Locales, que incluyen 8131 municipios y 39 Diputaciones Provinciales. Las Diputaciones proceden de la anterior estructura administrativa (similar al modelo francés, importado en la primera mitad el Siglo XIX) y han desaparecido en las Comunidades Autónomas uniprovinciales, que han absorbido sus competencias y financiación. Su función principal es el apoyo a los municipios más pequeños. Existe democracia directa en el caso de las CCAA y los municipios. En contraste, los gobiernos de las Diputaciones se escogen indirectamente a partir de los resultados de las elecciones municipales. Antes de 1978 existían regiones, pero sin relevancia administrativa ni política. Ese mapa regional sirvió de base para la demarcación territorial de algunas Comunidades Autónomas, pero no todas, que nacieron realmente con el proceso descentralizador. 2. Estado Actual del Modelo de Descentralización 3. A pesar de no ser formalmente un país federal y de que hace 40 años era un país fuertemente centralizado, el grado de descentralización a escala regional en España es hoy de los más elevados del Mundo, de acuerdo con los indicadores cuantitativos más habituales.4 Comenzamos con el RAI (“Regional Authority Index” o “Índice de autoridad regional” en español), un índice relativamente reciente, pero de uso muy extendido por dos motivos fundamentales. Primero, por la amplitud temporal e internacional de la base. Segundo, porque abarca los aspectos económicos y fiscales, pero también institucionales y políticos, de forma que ofrece una pintura general de la descentralización en cada país. Los resultados para España son extraordinarios. De aparecer entre los Estados más centralizados y a enorme distancia de las naciones federales, se pasa al segundo puesto mundial en el arranque de este siglo. Solo Alemania ha estado por delante en el conjunto del período. Federaciones como Bélgica, Estados Unidos, Canadá o Australia estarían hoy por detrás de España según el índice (Gráfico 1). 2 Andalucía, Aragón, Asturias, Cantabria, Castilla La-Mancha, Castilla y León, Cataluña, Comunidad Valenciana, Canarias, Extremadura, Galicia, Islas Baleares, La Rioja, Comunidad de Madrid, Murcia, Navarra y País Vasco. 3 Ceuta y Melilla, ambas con una población por debajo de los 100.000 habitantes. 4 En la panorámica elaborada por Martínez-Vázquez et al (2017) se aborda la cuestión de la medición del grado de descentralización. Gráfico 1: Evolución del Índice de Autoridad Regional (RAI) 40 35 30 25 20 15 10 5 0 1971 1950 1953 1956 1959 1962 1965 1968 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 Alemania Australia Canada EEUU España Suiza Fuente: Elaboración propia a partir de Hooghe et al (2016) 4. De forma complementaria, el Gráfico 2 muestra el porcentaje del gasto público en manos de los gobiernos regionales en los países de la OCDE descentralizados a escala regional desde 1995 hasta 2016. En este caso, España se mantiene lejos de Estados Unidos y Canadá, pero en el TOP-5 mundial. Gráfico 2: Porcentaje del gasto público en manos de los gobiernos regionales 70% 65% 60% 55% 50% 45% 40% 35% 30% 1999 2008 1995 1996 1997 1998 2000 2001 2002 2003 2004 2005 2006 2007 2009 2010 2011 2012 2013 2014 2015 2016 Alemania Australia Canadá EEUU España Suiza Fuente: Elaboración propia a partir de OECD (2019) 5. Es importante resaltar que en España existen dos regímenes de descentralización regional con lógicas y estructuras muy diferentes: el régimen foral, que se aplica a las regiones de País Vasco y Navarra (que son residencia de alrededor del 6% de la población española) y el régimen común, para el resto. El foral es de naturaleza confederal y hunde sus raíces en acuerdos fiscales anteriores a la dictadura que fueron restaurados y actualizados. Los Gobiernos subcentrales de País Vasco y Navarra son los que recaudan la práctica totalidad de los impuestos en su territorio (incluidos el IRPF, el IVA, el Impuesto sobre Sociedades 4 y los impuestos sobre consumos específicos) y transfieren al Gobierno central una cuantía anual por los servicios no descentralizados. Es un caso atípico en el Mundo. El régimen común sí es de tipo federal. En este caso, las Comunidades Autónomas recaudan una parte de sus ingresos y reciben transferencias de diferente tipo para complementar sus ingresos. 6. Esta dualidad de sistemas ha sido una fuente de distorsión permanente. Fundamentalmente, porque la cuantía anual a transferir desde las Comunidades forales es muy reducida5 y su PIB per cápita se sitúa entre los más altos de España. Como resultado de lo anterior, con unos tipos impositivos algo inferiores a la media, los gobiernos de las regiones forales cuentan con muchos más recursos que los demás, al no contribuir a la financiación de la nivelación interterritorial como le correspondería dada su elevada capacidad fiscal. Y eso acaba generando sentimientos de agravio comparativo en las regiones con un PIB per cápita elevado, pero sujetas al régimen común (en particular, Cataluña). En este documento la atención se centra en el caso de las regiones sujetas al régimen común, por ser el de mayor interés potencial para una estrategia descentralizadora en Chile. 7. La descentralización del gasto en España ha ido acompañada, desde el primer momento, por una especial atención y cuidado de que la descentralización no generase diferencias entre las regiones (de régimen común) en su capacidad financiera para prestar servicios públicos. Además, tradicionalmente, se ha incidido en la existencia de diferencias significativas en el coste per cápita de la prestación de los servicios autonómicos. 8. Como consecuencia de lo anterior, la nivelación interregional es muy intensa y persigue la nivelación plena de las necesidades de gasto (“expenditure needs”) y no solo de las capacidades fiscales. En ambos sentidos (nivelación plena y preocupación por las necesidades de gasto), España se parece a Australia. Pero a diferencia de Australia, no existe un organismo independiente que estime las disparidades en los costes de prestación de servicios. La negociación política sobre las variables a tener en cuenta y sus ponderaciones sigue pesando mucho; si bien es cierto que se ha ido mejorando con el tiempo sobre el fundamento técnico de las cifras. 9. La descentralización tributaria ha ido siempre por detrás de la del gasto. Aunque se han producido avances notables, como examinaremos posteriormente, existe un margen sustancial de mejora. 10. La política regional es muy intensa, por la confluencia de las medidas e instrumentos del Gobierno central y los de la Unión Europea (UE). España ingresa en la Comunidad Económica Europea en 1986 y es a finales de los años ochenta cuando se refuerza este vector de la política comunitaria. Durante muchos años, hasta la ampliación a países del Este europeo, España era el país que recibía más fondos de la Unión Europea para la implementación de una política regional ambiciosa. 3. Cambios Relevantes en el Modelo de Descentralización: el Iter del Proceso 11. En noviembre de 1975 fallece el dictador Francisco Franco y arranca la transición a la democracia. En 1978 se aprueba la nueva Constitución española y se abre el proceso de creación de las Comunidades Autónomas, que había arrancado con el establecimiento de algunas pre-autonomías ya en 1977. 12. En 1980 se promulga la Ley Orgánica de Financiación de las Comunidades Autónomas (LOFCA), que dibuja de forma muy general su marco financiero. Con varias reformas parciales, sigue vigente. En 5 Los cálculos de Zabalza y López-Laborda (2017) sitúan alrededor del 30% esa sobrefinanciación. No existen fundamentos económicos para ello, tan solo políticos. 5 paralelo y hasta 1983, se van aprobando los Estatutos de autonomía regionales, una suerte de Constituciones o Cartas regionales definidas y votadas por su poder legislativo y que, posteriormente, deben ser refrendadas por el poder legislativo nacional. El nuevo mapa autonómico aparece formado por 17 Comunidades Autónomas, cuyas fronteras solo en parte coinciden con las anteriores regiones geográficas definidas en el siglo XIX, y dos ciudades autónomas en el Norte de África (Ceuta y Melilla). El proceso de traspaso de competencias de gasto gana fuerza en todas las Comunidades Autónomas. 13. En 1984 se regula el funcionamiento del llamado Fondo de Compensación interterritorial (FCI), un programa de transferencias condicionadas para financiar tanto el desarrollo de las regiones más pobres como inversiones en las competencias descentralizadas en las más ricas. 14. Desde 1978 hasta 1986, la prioridad era dotar de recursos suficientes (transferencias incondicionadas y recaudación de algunos tributos sin otorgar capacidad normativa) a las Comunidades Autónomas para la financiación del coste efectivo de los servicios que se iban descentralizando a ritmos distintos en cada Comunidad Autónoma, negociando bilateralmente con la administración central. La autonomía fiscal de las CCAA es en esos años prácticamente nula. A finales de 1986 se aprueba el primer sistema de financiación quinquenal (1987-1991) que sigue sin proporcionar autonomía fiscal significativa. Se empieza a hablar de las necesidades de gasto para la prestación de servicios como criterio de reparto de los recursos. En paralelo, se traspasan nuevas competencias a algunas regiones, incluyendo sanidad y educación. 15. En 1991 se actualiza el modelo anterior para el quinquenio 1992-1996. Lo más sustancial de este acuerdo tiene que ver con el endeudamiento autonómico, a fin de cumplir los compromisos con la Unión Europea en materia de estabilidad fiscal en el camino hacia la creación del Euro. Se reforma el FCI para convertirlo en un verdadero instrumento de política regional, excluyendo a las CCAA más desarrolladas. 16. En 1993 se modifica parcialmente el sistema 1992-1996 para ceder una parte (15%) del Impuesto sobre la Renta de las Personas Físicas (IRPF), sin que las regiones cuenten con capacidad normativa para alterar tipos impositivos o deducciones. 17. En 1997 entra en vigor el sistema de financiación 1997-2001. Marca un claro punto de inflexión en la descentralización tributaria. Se cede capacidad normativa sobre el tramo de IRPF cuya recaudación fue cedida en 1993 y en los Impuestos sobre Patrimonio, Sucesiones y Donaciones, Impuesto sobre Transmisiones Patrimoniales y Actos jurídicos Documentados, y los impuestos sobre el juego, sobre los que previamente no existía la posibilidad de cambiar tipos impositivos o deducciones fiscales. 18. En enero de 2002 comienza a aplicarse la reforma de la financiación autonómica aprobada en diciembre de 2001 y finaliza el proceso de traspaso de competencias educativas y sanitarias a todas las Comunidades Autónomas. Se elimina la referencia a la revisión quinquenal del sistema de financiación y se pretende que el sistema de financiación sea permanente. Además, se profundiza en las posibilidades legales para el ejercicio de la autonomía tributaria las CCAA, se incrementa el porcentaje del IRPF hasta un tercio y se cede como coparticipación de ingresos (revenue sharing) una parte de la recaudación del IVA y los impuestos sobre consumos específicos. 19. En 2009 se produce una reforma sustancial del sistema de financiación autonómica, para dar pasos adicionales en la descentralización tributaria de 2002. En concreto, se cede la mitad del IRPF a las CCAA y se amplía el porcentaje de coparticipación de ingresos en el IVA y los impuestos sobre consumos específicos. 20. En 2017 se crea una Comisión de expertos pactada por el Gobierno central y las Comunidades Autónomas de régimen común para acordar una nueva reforma del sistema de financiación autonómica. La Comisión emiten su informe final en el verano de 2017 (Comisión de expertos, 2018). Desde entonces, los 6 Directores generales regionales competentes en la materia y los técnicos del Ministerio de Hacienda han estado trabajando sobre el Informe de la Comisión de Expertos, pero se han producido aplazamientos sucesivos en la fase política final de la negociación debido a la fragmentación electoral en el Congreso de los diputados español y la inestabilidad electoral del Gobierno central. 4. Asignación de Responsabilidades de Gasto 21. La regulación del reparto de responsabilidades de gasto está recogida en la Constitución Española de 1978. Así, en su artículo 149 se enumeran las competencias exclusivas del Estado o Gobierno central; en el 148 aparece el listado de competencias que las Comunidades Autónomas pueden asumir si así aparece recogido en su Estatuto de autonomía; y en el 150 se abre la posibilidad de la delegación de competencias estatales incluidas en el artículo 150, a través de una ley orgánica que debe aprobarse en las Cortes españolas. En la práctica, lo anterior se traduce en la existencia de una triple casuística: • Competencias legislativas y ejecutivas exclusivas. Es el caso de servicios sociales, agricultura y ganadería, pesca interior, industria, comercio, desarrollo económico, turismo, juventud o deportes. • Competencias para el desarrollo de la legislación básica del Estado, así como para la ejecución de esta legislación. Es el caso de medio ambiente, política económica, protección de los consumidores, educación, asistencia sanitaria o salud pública. • Competencias para la ejecución de la legislación aprobada en exclusiva por el Estado. Este es el caso de las políticas de empleo. 22. Para el despliegue de todas estas competencias, existen diferentes organismos e instrumentos para la cooperación horizontal (entre Comunidades Autónomas) y vertical (con la Administración central): • La Conferencia de Presidentes, de la que forman parte el Presidente del Gobierno y los Presidentes de las CCAA. • El Consejo de Política Fiscal y Financiera de las Comunidades Autónomas (CPFF), como órgano de coordinación del Estado y las CCAA en materia fiscal y financiera. • Las Conferencias Sectoriales sobre materias concretas, en las que participan el Estado y todas las CCAA. • Las Comisiones Bilaterales de Cooperación, entre el Estado y una determinada Comunidad Autónoma • Las Comisiones Sectoriales, Grupos de Trabajo y Ponencias de composición técnica y adscritos en gran medida a las conferencias Sectoriales. No obstante, existe un amplio consenso en que, en la práctica, este marco institucional es insuficiente y no funciona con la eficacia necesaria. A pesar de sus deficiencias de recursos y competenciales (Comisión de expertos, 2018), el CPFF es el instrumento más dinámico, eficaz y relevante para la gobernanza multinivel. 23. Consecuentemente, las regiones son hoy el principal prestador de servicios públicos para los ciudadanos. En cambio, el grueso de las transferencias monetarias (pensiones, seguro por desempleo) siguen en manos estatales. El Gráfico 3 muestra el gasto público como porcentaje de PIB en España y su distribución por escalones de gobierno. 24. Ese protagonismo en la prestación directa de servicios hace que las regiones dominen claramente la distribución del empleo público (Gráfico 4). El 58% depende de los Gobiernos regionales y solo uno de cada cinco empleados sigue dependiendo de la Administración central. La Administración central se ha especializado en la gestión de los principales programas de transferencias monetarias (pensiones, seguro por desempleo), los bienes públicos clásicos (defensa nacional, orden público, relaciones exteriores…) y la inversión en las grandes infraestructuras de transporte. 7 25. Finalmente, y como ya hemos apuntado, la política regional en España ha sido muy intensa en las últimas tres décadas. Dotada con recursos estatales y, sobre todo, de la Unión Europea desde 1986, es diseñada y ejecutada por los gobiernos regionales, que deciden sobre las actuaciones, dentro de las reglas generales de los instrumentos de política regional española y europea. En esas actuaciones predomina la inversión en capital físico, pero existe una relevancia creciente de apuestas por el capital humano y la I+D+i. No obstante, el ingreso de nuevos países en la UE con niveles de PIB per cápita claramente inferior a los españoles está reduciendo de forma continua los fondos con destino a España. Y todo apunta a que a partir de 2020 la reducción será drástica. (1) Gráfico 3: Distribución del gasto público por niveles de la administración Año 2018, expresado como porcentaje del PIB español. 45% 40% 41% 35% 30% 25% 20% 20% 15% 10% 15% 5% 6% 0% Total Administración Comunidades Corporaciones central (Incluyendo Autónomas Locales Seguridad Social) Fuente: Elaboración propia a partir de “Avance de la actuación económica y financiera de las administraciones públicas en 2018”. IGAE Gráfico 4: Empleados en las administraciones públicas españolas en enero de 2019. TOTAL PORCENTAJ E Administración Central 507,830 20% Comunidades Autónomas 1,495,153 58% Corporaciones Locales 575,405 22% (Municipios, Diputaciones y otros entes) Fuente: Elaboración propia a partir de datos del Ministerio de Política territorial y Función Pública. 5. Las Fuentes de Ingreso 8 26. Las fuentes de ingresos de las regiones incluyen: a. Tributos propios. Desde el principio del proceso descentralizador, las CCAA cuentan con la posibilidad de crear impuestos propios, sin necesidad de autorización del gobierno central.6 Sin embargo, no pueden gravar hechos imponibles previamente gravados por el Estado o los municipios, a fin de evitar un problema de doble imposición. Eso ha hecho que haya habido numerosos conflictos entre el Gobierno central y los autonómicos sobre figuras de nueva creación, generando inseguridad jurídica en los contribuyentes, y que esta vía de ingresos sea de relevancia marginal en términos recaudatorios (menos del 2% de todos los ingresos regionales). Los tributos propios se concentran en el campo de la imposición medioambiental; sobre residuos y contaminación ambiental de todo tipo. b. Tributos cedidos por la administración central. Esta cesión se produce en porcentajes diversos (existen impuestos compartidos) y con diferentes grados de autonomía normativa. c. Transferencias incondicionadas de nivelación horizontal y vertical de la administración central. d. Transferencias condicionadas de la administración central y la Unión Europea, destacando la financiación de la política regional. e. Deuda pública. 27. En lo que atañe a los tributos cedidos, en algunos casos existe transferencia de recaudación o coparticipación de ingresos sin capacidad normativa. El más importante es el Impuesto sobre el Valor Añadido (IVA) y los impuestos especiales sobre consumos (hidrocarburos, tabaco y alcohol). Las Comunidades reciben el 50% de la recaudación (el 58% en el caso de los especiales), pero no pueden modificar los parámetros del impuesto. La normativa europea prohíbe la existencia de tipos diferenciados entre las regiones de un mismo país. Sería, por tanto, una fórmula de “tax sharing”. También hay impuestos donde las CCAA tiene capacidad para alterar tipos impositivos o deducciones, se quedan con la recaudación, pero la gestión y recaudación es realizada por la Agencia Tributaria estatal (como el propio IRPF). En la Tabla 1 se sintetiza toda la información sobre descentralización de los tributos cedidos. 6 La Constitución Española establece este poder tributario regional y la Ley Orgánica de Financiación de las Comunidades Autónomas regula su ejercicio en su artículo 6, el cual establece que: “Uno. Las Comunidades Autónomas podrán establecer y exigir sus propios tributos de acuerdo con la Constitución y las Leyes. Dos. Los tributos que establezcan las Comunidades Autónomas no podrán recaer sobre hechos imponibles gravados por el Estado. Cuando el Estado, en el ejercicio de su potestad tributaria originaria establezca tributos sobre hechos imponibles gravados por las Comunidades Autónomas, que supongan a éstas una disminución de ingresos, instrumentará las medidas de compensación o coordinación adecuadas en favor de las mismas. Tres. Los tributos que establezcan las Comunidades Autónomas no podrán recaer sobre hechos imponibles gravados por los tributos locales. Las Comunidades Autónomas podrán establecer y gestionar tributos sobre las materias que la legislación de Régimen Local reserve a las Corporaciones locales […]. 9 Tabla 1: Estructura de la descentralización tributaria autonómica Porcentaje de Capacidad Gestión del Criterios de reparto entre cesión de los normativa impuesto CCAA o puntos de rendimientos conexión Impuesto sobre Sociedades 0% No No --- IRPF 50% Sí No Residencia del contribuyente IVA 50% No No Consumo en la Comunidad Impuestos sobre consumos 58% No No Consumo en la específicos Comunidad Impuestos sobre el consumo 100% No No Consumo en la de electricidad Comunidad IVMDH/IH tramo 100% Sí No Consumo en la autonómico(*) Comunidad Impuesto sobre 100% Sí No Residencia del matriculación de contribuyente vehículos(**) Impuesto sobre Patrimonio 100% Sí Sí Residencia del contribuyente Impuesto sobre Sucesiones y 100% Sí Sí Residencia del difunto o Donaciones del donatario, ubicación de los inmuebles ITPAJD (***) 100% Sí Sí Hecho imponible en la Comunidad Tributos sobre el juego 100% Sí Sí Juego en la Comunidad (*) Comenzando en 2013, el impuesto sobre las Ventas Minoristas de Determinados Hidrocarburos (IVMDH) desaparece como tal y pasa a integrarse en el Impuesto Especial sobre Hidrocarburos, para el que se establece un tipo autonómico que puede variar de una Comunidad a otra. (**) Aunque la ley lo permite, ninguna Comunidad autónoma ha asumido la gestión de este impuesto. (***) Impuesto sobre Transmisiones Patrimoniales y Actos Jurídicos Documentados. Fuente: Comisión de expertos (2018) 28. Para sintetizar la información anterior y comparar a España con los países federales de la OCDE, en el Gráfico 5 se pone en relación la descentralización tributaria (medida como la proporción de ingresos tributarios de cada país en manos de los gobiernos regionales) con el porcentaje de los ingresos regionales procedentes de impuestos sobre los que los gobiernos subcentrales cuentan con capacidad para modificar tipos impositivos y deducciones sin necesidad de consultar con el Gobierno central. En la UE, España es hoy el líder en descentralización tributaria efectiva. Las regiones en Italia y los Länder alemanes no disfrutan de autonomía real; y en Austria y Bélgica los tributos en manos regionales son escasos. España estaría en el Top-5 mundial, solo por detrás de Canadá, Suiza, Estados Unidos y Australia. 10 Gráfico 5: Grado de descentralización tributaria y capacidad normativa plena en 2014 Fuente: Elaboración propia a partir de datos de OECD (2019) 29. Existe un amplio consenso técnico (respaldado por la Comisión de expertos creada en 2017) de que es necesario avanzar en la autonomía tributaria. Pero el objetivo no debería ser tanto cuantitativo como cualitativo. En síntesis, hay que ajustar los procedimientos para incentivar el uso efectivo de la autonomía legal existente,7 resolver problemas de competencia fiscal nociva en la imposición patrimonial; reorganizar y potenciar la imposición medioambiental y reformar el tax sharing en el IVA y los impuestos sobre consumos específicos, para que las regiones puedan decidir colectivamente los tipos impositivos en su parte.8 30. Las transferencias incondicionadas de la Administración central a las regiones tienen por objetivo entremezclado la nivelación horizontal (entre regiones) y la vertical (entre niveles de administración). Su 7 Pensemos en el principal instrumento tributario, el IRPF. Si el Gobierno central decide subir tipos impositivos incluso a lo largo del ejercicio, esa subida se traslada en poco tiempo (incluso semanas) a las retenciones que soportan los contribuyentes. En cambio, si un Gobierno autonómico toma la decisión en otoño de 2019 de subir los tipos en el ejercicio 2020, las retenciones no se alteran y el incremento en la recaudación solo la notaría el contribuyente al presentar su declaración de la renta anual correspondiente a 2020, ya en los meses de mayo y junio de 2021. ¡Y al Gobierno regional se le transferirían los recursos adicionales a partir de julio de 2022! Otro desincentivo al uso al alza de la capacidad normativa (e incentivo para rebajar impuestos) es la expectativa de “restricción presupuestaria blanda”: las regiones saben que cada vez que se renegocia el sistema, habitualmente cada cinco años, los recursos transferidos desde la administración central a las regionales aumentan (Fernández Llera et al. 2013). Finalmente, el hecho de que la Administración central haya aceptado financiar la deuda autonómica en la última década a tipos de interés muy bajos es un incentivo a endeudarse más que a subir impuestos a la hora de financiar el gasto. 8 Ante la imposibilidad de ceder a las CCAA competencias sobre tipos impositivos en IVA y en impuestos sobre consumos específicos, existe un amplio consenso técnico, que se refleja en el Informe de la Comisión de Expertos (2018), de que sería un avance que las CCAA pudiesen decidir colegiadamente sobre la parte del IVA o impuestos sobre consumos específicos en sus manos. Frente a la situación actual, en la que los tipos impositivos son decididos por la Administración central, cabría la posibilidad de que existiesen dos tramos y los tipos impositivos de uno de ellos sería decidido por los gobiernos regionales en su conjunto, respetando los tipos mínimos exigidos por la legislación europea. De esta manera, las CCAA tendrían acceso directo a un objeto imponible (el consumo) que, de hecho, supera cuantitativamente al propio IRPF (Gráfico 6) 11 dinámica se vincula a la de los ingresos generales del Estado; es decir, se define un índice de evolución que toma como referencia la recaudación de los impuestos estatales. 31. El sistema de nivelación es técnicamente complejo, porque existen varios mecanismos que persiguen, al mismo tiempo, la corrección de desequilibrios horizontales y desequilibrios verticales. Además, en la medida en que la referencia son las necesidades de gasto estimadas para cada región y los porcentajes de cesión tributaria son idénticos, las regiones más ricas “devuelven” una parte de la cesión tributaria (el exceso sobre sus necesidades estimadas) que nutre el Fondo de Garantía de Servicios Públicos Fundamentales (FGSF). Por tanto es de hecho, un sistema de tipo “Robin Hood”. Dado que la gestión de los tributos implicados (IRPF, IVA, impuestos especiales) es responsabilidad de la Administración tributaria central, los recursos nunca llegan a entrar en las arcas de las regiones contribuyentes netas y esas transferencias horizontales no son muy visibles. En general, son solo conocidas por los expertos. 32. Para complicarlo aún más, en la práctica se producen desviaciones sustanciales de la distribución de recursos que resulta de esas necesidades de gasto estimadas. Las condiciones en las que en su momento se negoció el traspaso de cada competencia en cada región siguen pesando. Cada vez que se reforma el modelo de financiación, la financiación percibida con el sistema anterior se toma como consolidado (aplicándose la cláusula del “hold harmless”), y así reforzando el statu quo desde las primeras etapas del proceso descentralizador.9 Es decir, si los recursos que resultan de la nueva fórmula de reparto (con sus variables y ponderaciones) difieren significativamente del reparto actual entre regiones se acaban introduciendo en la negociación política entre la administración central y las regiones ajustes ad hoc y modulaciones que desactivan parcialmente los efectos de cada reforma. 33. Otro ejemplo de la influencia de la negociación política sobre los resultados financieros finales fue la introducción del criterio de la nivelación parcial en la reforma de 2009. El texto de la ley recogía la idea de que solo se nivelarían los llamados servicios públicos fundamentales (sanidad, educación y servicios sociales, que suponen alrededor del 70% del gasto regional) para así atender a las demandas de Cataluña en el frente de la nivelación, que pretendía reducir así su contribución al fondo común. Pero en la propia normativa se introdujeron modificaciones para que de facto se mantuviese la nivelación plena. Hoy la Comunidad Autónoma de menor PIB per cápita (Extremadura) sigue teniendo unos recursos presupuestarios por habitante claramente superiores a los de las regiones más ricas de Madrid o Cataluña. 34. También existen transferencias condicionadas procedentes del nivel central para la financiación de actuaciones de política regional, que son definidas por los gobiernos de las regiones de menor PIB per cápita, y para financiar servicios o materias sobre las que las Comunidades Autónomas tienen la competencia de gestión; entre las que destacan cuantitativamente las correspondientes a la formación laboral continua para desempleados y a la llamada “ley de dependencia”, aprobada en 2006 y que tiene como objetivo proteger y atender a las personas que, por edad, enfermedad o discapacidad, precisan de atención y ayuda para la vida diaria. 35. Por último, aparecen las transferencias condicionadas de la Unión Europea para la financiación de actuaciones de política regional y otras (como la política agraria).10 En la Tabla 2 se sintetizan los principales programas de transferencias a las regiones. 9 Las negociaciones bilaterales entre cada región y la administración central sobre el coste efectivo de cada competencia que se iba transfiriendo dieron como resultado financiación por habitante de las competencias (y por agregación, de la financiación total) muy diferentes, dependiendo de la habilidad de los negociadores en cada momento, de la coyuntura económica en cada momento (y, por tanto, el margen para la administración central) y del propio mapa de prestación de servicios en cada territorio: si un servicio centralizado estaba particularmente bien dotado en una región, esa ventaja se trasladaba al ser descentralizado. Las estimaciones sobre el efecto de las transferencias condicionadas apuntan a que en su mayor parte (alrededor del 90%) aumentan el gasto en las áreas que se pretende y en torno al 10% se traducen en un menor déficit (Lago-Peñas, 2006). 12 Tabla 2: Programas de transferencias a las CCAA Transferencias incondicionadas El Fondo de Garantía de Servicios Públicos Fundamentales tiene por objeto asegurar que las Comunidades Autónomas puedan ofrecer un nivel similar en la prestación para financiar los servicios públicos fundamentales (sanidad, educación y servicios sociales). Es financiado con recursos de las propias CCAA y de la Administración central El Fondo de Suficiencia Global es un instrumento de nivelación que opera como recurso de cierre del sistema, asegurando que las necesidades globales de financiación de cada Comunidad en el año base (esto es, sus necesidades de gasto estimadas) se cubren con su capacidad tributaria, la transferencia del Fondo de Garantía y el propio Fondo de Suficiencia Global. Financiado por la Administración central. Los Fondos de Convergencia, que se subdividen en el Fondo de Competitividad y el Fondo de Cooperación. El primero se crea con el fin de reforzar la equidad y la eficiencia en la financiación de las necesidades de los ciudadanos y reducir las diferencias en financiación homogénea per cápita entre Comunidades Autónomas, al mismo tiempo que se incentiva la autonomía y la capacidad fiscal y se desincentiva la competencia fiscal a la baja. Por su parte, el segundo beneficia a las CA de menor PIB per cápita y las que tienen una dinámica poblacional especialmente negativa. Transferencias condicionadas Transferencias Condicionadas de la Unión Europea para el despliegue de la política regional y otras actuaciones comunitarias. Fondo de Compensación Interterritorial Otras transferencias condicionadas Fuente: Elaboración propia a partir de información del Ministerio de Hacienda (www.hacienda.gob.es) 36. Un aspecto técnico a tener muy en cuenta tiene que ver con el hecho de que la mayor parte de los ingresos de las regiones son gestionados por la Agencia tributaria estatal, como se reflejaba en la Tabla 1. Eso obliga a que el sistema de ingresos de las regiones de régimen común descanse en un mecanismo de anticipos e ingresos a cuenta que permite adelantar a las regiones sus ingresos tributarios. Pero el mecanismo afecta también a las transferencias incondicionadas, que dependen de los ingresos del Estado. Solo cuando se conoce la recaudación final se procede a la liquidación definitiva de las diferencias. Por ejemplo, en julio de 2019 se ha liquidado el año 2017, con un saldo que puede ser negativo o positivo para cada región; y de un tamaño significativo en periodos de cambio en el ciclo económico. 37. Como muestra el Gráfico 6, la composición de los ingresos ordinarios (dejando al margen el déficit y endeudamiento) de las Comunidades Autónomas de régimen común muestra el peso dominante de los recursos tributarios; si bien es cierto que cuando se analizan individualmente, en las CCAA más pobres el peso de las transferencias es mayor por el efecto de los flujos fiscales de la nivelación interterritorial y la política regional. Gráfico 6: Estructura por instrumentos de la financiación ordinaria (excluida deuda pública) de las administraciones regionales 13 100% 4% Otras transferencias 7% 90% condicionadas 12% Transferencias condicionadas 80% UE y política regional española 70% 12% Transferencias incondicionadas 60% 10% Otros tributos 50% 25% 40% Impuestos sobre consumos 30% específicos IVA 20% 30% 10% IRPF 0% Fuente: Elaboración propia a partir de “Haciendas Autonómicas en cifras” (2016) 38. A fin de facilitar la comprensión del sistema de nivelación, en el Gráfico 7 se representa la financiación en euros per cápita de la región más rica (la de Madrid) y la que tiene un menor PIB per cápita (Extremadura) en 2016. Se diferencia entre cuatro categorías de recursos: ingresos tributarios, transferencias incondicionadas, recursos europeos y de la política regional española y otras transferencias condicionadas. La recaudación per cápita por tributos en casi el doble en la región de Madrid, pero esa diferencia es más que compensada por los recursos europeos y la política regional y la nivelación horizontal. Mientras que la Administración central transfiere recursos a Extremadura (1.247 euros per cápita), detrae recursos a Madrid por importe de 635 euros per cápita. Gráfico 7: Recursos financieros por habitante en 2016 4,500 3,804 4,000 3,500 2,959 3,000 2,447 2,500 2,000 1,666 1,500 1,247 1,000 755 500 14 137 110 0 -500 -1,000 -635 Tributos Transferencias Recursos Otras Total Recursos incondicionadas europeos y transferencias política regional condicionadas Extremadura Región de Madrid Fuente: Elaboración propia a partir de “Haciendas Autonómicas en cifras” (2016) 14 6. Endeudamiento Regional y Estabilidad Fiscal 39. En los últimos veinte años, los avances en la legislación y el marco institucional han sido sustanciales, como consecuencia del endurecimiento de las reglas fiscales en la Unión Europea y, por trasposición legal obligatoria, en España. Estas reglas incluyen límites al déficit total y el déficit estructural y reglas de gasto que limitan su crecimiento. 40. El “Índice estandarizado de regla fiscal” (FRI) computado por la Comisión Europea permite cuantificar de forma sintética el grado de exigencia de las reglas atendiendo a su base legal (norma constitucional, ley ordinaria…); facilidad para modificar la reglas o suspenderlas temporalmente; naturaleza del organismo encargado de la supervisión del cumplimiento de las reglas; automatismo de los mecanismos de corrección de las desviaciones, y margen de adaptación de la regla a shocks externos ajenos al gobierno. 41. Observando la evolución del FRI para España en el período 1990-2017, se aprecian saltos de nivel en España en 2002, derivados de la nueva normativa de estabilidad presupuestaria que se aprueba en diciembre de 2001 (véase recuadro adjunto), y el cuatrienio 2011-2014, de la mano de los cambios impulsados por la Comisión Europea en el bienio 2011-2012, como respuesta a los efectos de la Gran Recesión sobre las cuentas públicas de los países miembros de la UE (Gráfico 8). España se situó en el año 2017 en el octavo puesto en la UE en el valor del FRI (Gráfico 9). 15 Recuadro. Rasgos básicos de las Leyes de Estabilidad Presupuestaria aprobadas en diciembre de 2001 (Ley 18/2001 y Ley Orgánica 5/2001). Principios generales: · Estabilidad presupuestaria: equilibrio o superávit de las AAPP. · Plurianualidad: presupuestos elaborados en un marco trienal. · Transparencia: información suficiente para verificar la estabilidad. · Eficiencia: gestión presidida por la eficacia, eficiencia y calidad. Procedimiento para el conjunto del Sector Público: · En el 1er cuatrimestre de cada año: remisión por el Gobierno a las Cortes de los objetivos de estabilidad para los tres años siguientes. · Antes del 1 de septiembre de cada año: informe de la IGAE sobre el grado de cumplimiento del objetivo de estabilidad del año anterior. · Planes de corrección o saneamiento para los que tengan déficit y, en su caso, imputación de la parte que les toque en las responsabilidades por incumplir el PEC. Normas para el Sector Público Estatal: · Límite máximo anual de gasto del Estado acordado por el Consejo de Ministros. · Fondo de Contingencia a cargo del cual correrán las modificaciones de crédito (máximo del 2% del límite de gasto no financiero del Estado). · Los superávit deben destinarse a reducir deuda pública o al Fondo de Reserva de la SS. Normas para el Sector Público Territorial: · Fijación de objetivos de estabilidad previo informe del CPFF o de la CNAL, según los casos. · Planes de saneamiento si se incumplen los objetivos. · Imputación de su responsabilidad en los incumplimientos del PEC. · Central de información y condicionamiento de autorizaciones de crédito o deuda. Fuente: Caamaño y Lago-Peñas (2008) Gráfico 8: Evolución del FRI en España (1990-2017) 2.50 2.00 1.50 1.00 0.50 0.00 -0.50 -1.00 1995 2011 1990 1991 1992 1993 1994 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2012 2013 2014 2015 2016 2017 Fuente: Elaboración propia a partir de la Fiscal Rules Database elaborada por la Comisión Europea. 16 Gráfico 9: Valor del FRI en 2017 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 BG NL IT LT PT SK RO ES IE AT DE EE MT BE PL SE LV FI CY CZ DK FR LU HU EL SI UK HR Acrónimos usados: Austria(AT), Bélgica(BE), Bulgaria(BG), Chipre(CY), República Checa(CZ), Alemania(DE), Dinamarca(DK), Zona Euro (EA19), Estonia(EE), Grecia(EL), España(ES), Unión Europea 27(EU27), Finlandia(FI), Francia(FR), Croacia(HR), Hungría(HU), Irlanda(IE), Italia(IT), Lituania(LT), Luxemburgo(LU), Letonia(LV), Malta(MT), Países Bajos(NL), Polonia(PL), Portugal(PT), Rumanía(RO), Suecia(SE), Eslovenia(SI), Eslovaquia(SK), Reino Unido(UK) Fuente: Elaboración propia a partir de la Fiscal Rules Database elaborada por la Comisión Europea. 42. Complementariamente, la creación y puesta en funcionamiento de la Autoridad Independiente de Responsabilidad Fiscal (AIReF) en 2014 ha supuesto un reforzamiento notable en la tarea de control y supervisión de las cuentas públicas (Von Trapp et al, 2017).11 43. Los Gobiernos nacionales de los países de la Unión Europea tienen autonomía para concretar la aplicación de la normativa de estabilidad fiscal en sus países. En el caso de España se ha optado, en general, por una extensión de las exigencias aplicadas al Gobierno de España por la Comisión Europea. La normativa vigente otorga al Gobierno central una elevada capacidad de control sobre la actuación de los Gobiernos regionales, pudiendo incluso suspender la autonomía de una región y que el Gobierno central intervenga su gobierno. Sin embargo, el efecto de la Gran Recesión ha desbordado la normativa legal.12 La 11 La misión de la AIReF es garantizar el cumplimiento efectivo por parte de las Administraciones Públicas del principio de estabilidad presupuestaria previsto en el artículo 135 de la Constitución Española, mediante la evaluación continua del ciclo presupuestario y del endeudamiento público. La ley orgánica de creación de la AIReF otorga al Presidente un amplio ámbito de responsabilidades en la organización de la institución, su gobernanza y la toma de decisiones en todas sus competencias. En la actualidad, lidera un Comité de Dirección en el que se integran tres directores, uno por cada una de las tres Divisiones en que ha quedado organizada la AIReF: Análisis Presupuestario, Análisis Económico y Jurídico Institucional. Además, existe un Consejo Asesor integrado por un total de once miembros, nombrados por el presidente de la institución por un periodo de tres años entre personas de reconocido prestigio nacional e internacional con al menos diez años de experiencia en materia de análisis presupuestario, económico y financiero. La AIReF elabora y difunde evaluaciones independientes que atañen a todas las fases del proceso presupuestario y todos los niveles de gobierno. En particular, sus informes abarcan: las previsiones macroeconómicas de los gobiernos centrales y autonómicos, la metodología para calcular las previsiones tendenciales de ingresos y gastos, el proyecto de Programa de Estabilidad del Reino de España que se envía a la Comisión Europea, el análisis (mensual y trimestral) de la ejecución presupuestaria, deuda pública y de la regla de gasto, los planes económicos-financieros y planes de reequilibrio de la Administración Central y de las comunidades autónomas, los proyectos y líneas fundamentales de presupuestos de las Administraciones Públicas, o la aplicación de los mecanismos de corrección. 12 El efecto corrector impulsado por el severo endurecimiento en la legislación española sobre estabilidad presupuestaria que se produjo en 2011-2012 no fue suficiente per se en el escalón regional. Dado el coste político que hubiese supuesto para el Gobierno central aplicarla de forma estricta y permanente en el tiempo, no se agotaron, ni mucho menos, las posibilidades legales existentes. Y ello, a pesar de los incumplimientos recurrentes y graves de algunas CCAA entre 2012 y 2016. La experiencia española muestra la importancia de introducir un elevado automatismo en los mecanismos de control y penalización para evitar problemas de “economía política”. En segundo lugar, los mecanismos extraordinarios de liquidez que la administración central puso en práct ica desde 2012 para que las CCAA pudiesen financiar sus déficits en un escenario en el que los mercados financieros estaban cerrados 17 deuda pública ha aumentado de forma extraordinaria, la regional se sitúa entre las más altas del Mundo (Gráfico 10) y el déficit estructural en España sigue siendo muy elevado (Gráfico 11). Gráfico 10: Deuda pública regional sobre PIB en los países de la OCDE Fuente: Elaboración propia a partir de OECD (2019) Gráfico 11: Déficit (-) o superávit estructural (+) del conjunto del sector público en 2018. Valores en porcentaje del PIB. 4 2 0 -2 -4 Acrónimos: Austria(AT), Bélgica(BE), Bulgaria(BG), Chipre(CY), República Checa(CZ), Alemania(DE), Dinamarca(DK), Zona Euro (EA19), Estonia(EE), Grecia(EL), España(ES), Unión Europea 27(EU27), Finlandia(FI), Francia(FR), Croacia(HR), Hungría(HU), Irlanda(IE), Italia(IT), Lituania(LT), Luxemburgo(LU), Letonia(LV), Malta(MT), Países Bajos(NL), Polonia(PL), Portugal(PT), Rumanía(RO), Suecia(SE), Eslovenia(SI), Eslovaquia(SK), Reino Unido(UK) Fuente: Elaboración propia a partir de estimaciones de la Comisión Europea. para la mayoría o exigían condiciones desorbitadas para la colocación de deuda, contribuyó a agravar el problema de la restricción presupuestaria blanda y relajar los ajustes presupuestarios. 18 44. En síntesis, el problema del déficit ha afectado muy intensamente a todas las Administraciones Públicas españolas, salvo la administración local.13 Por tanto, las regiones en su conjunto han puesto en jaque la estabilidad fiscal en los últimos años. El enorme shock fiscal que supuso la Gran Recesión hizo insuficientes las reglas y los mecanismos de control. El Gobierno central tuvo que facilitar la financiación de sus déficits cuando los mercados financieros estaban cerrados para la colocación de deuda pública autonómica. 45. En estos momentos, el principal acreedor de las regiones es el Gobierno central, aunque no se ha producido una quita del principal de la deuda. Los créditos concedidos se mantienen como pasivos financieros de las administraciones regionales. 46. Los ajustes de las Comunidades Autónomas han sido en todo caso intensos y han pivotado sobre todo (en conjunto, más del 90%) en recortes de gasto más que en incrementos en la recaudación tributaria a nivel regional. 7. El Proceso Presupuestario 47. El proceso presupuestario en las regiones es completamente autónomo y separado del de la Administración central, pero esencialmente se sigue el mismo proceso que en el caso de los Presupuestos Generales del Estado. El ciclo comienza en los meses de junio y julio con la comunicación por parte del Ministerio de Hacienda de la previsión de recursos de las Comunidades Autónomas en concepto de transferencias e ingresos tributarios regionales gestionados por la Agencia tributaria estatal. En paralelo, se discute multilateralmente entre el Gobierno central y los autonómicos en el Consejo de Política Fiscal y Financiera y se vota el objetivo de déficit que le corresponde a cada Comunidad Autónoma; si bien el Gobierno central cuenta con la mayoría de votos y, por tanto, puede imponer su criterio. Finalmente, las regiones establecen individualmente su “techo de gasto”14 coherente con la llamada “regla de gasto” que se aplica desde el año 2013 y que es hecha pública por el Gobierno central durante el propio mes de julio. 15 En el mes de octubre se presenta el proyecto coherente con todo lo anterior en el Parlamento regional y comienza la discusión y votación, primero de las enmiendas a la totalidad y luego de enmiendas parciales. El proceso dura alrededor de 2 meses. Si es exitoso, los presupuestos se aprueban en diciembre y entran en vigor el 1 de enero de cada año, abarcando el año natural completo. La ley contempla la posibilidad de prorrogar los presupuestos del ejercicio anterior si el Parlamento regional no vota favorablemente el proyecto de presupuestos. 8. Auditoría, Control y Rendición de Cuentas 48. Las Comunidades Autónomas disponen de una Intervención general de cuentas propia (“ General regional comptroller”) que supervisa el cumplimiento de obligaciones contables y legales de todo tipo, con personal cualificado y seleccionado para esa función. No obstante, su independencia puede verse limitada, 13 El superávit de los municipios registrado en los últimos años se explica por tres motivos. En primer lugar, los municipios cuentan con una base fiscal más estable y menos sensible al ciclo. En segundo lugar, los municipios estaban gastando una parte sustancial de sus presupuestos (alrededor de un cuarto del mismo) en competencias no obligatorias por ley, pero demandadas por los ciudadanos. El Gobierno central impuso en 2012 que ese gasto debía dejar de hacerse. Finalmente, los municipios están también sujetos a la llamada “regla de gasto” que se explica en el apartado siguiente. 14 Esto es, el máximo nivel de gasto que pueden contemplar en su presupuesto. 15 La regla de gasto se aplica a todas las administraciones menos a la Seguridad Social y, en esencia, determina que el gasto público no puede crecer más que el PIB tendencial o de referencia de medio plazo para España, si no se adoptan medidas discrecionales que eleven los ingresos (en particular, subidas impositivas). Para un mayor detalle puede consultarse https://www.airef.es/es/noticias/quieres-conocer-la-regla-de-gasto/ 19 en ocasiones, por el hecho de que la Intervención es un departamento integrado en los Gobiernos regionales y su máximo responsable es nombrado por estos. 49. Adicionalmente, en 12 de las 17 regiones se han creado organismos dedicados al control externo de las cuentas y de la gestión económico-financiera de las instituciones y entidades del sector público autonómico y local (“Regional Audit Office). Estos organismos coordinan su actuación con el Tribunal de Cuentas de la nación. En este caso, la elección de sus miembros depende de los Parlamentos regionales. La principal crítica al funcionamiento de estos organismos de control es el retraso recurrente en la publicación de sus informes. 50. Tanto el control interno desarrollado por las intervenciones generales, como el de los organismos externos se centra en la legalidad de las actuaciones y muy escasamente en la eficiencia y la rentabilidad social. Esta es otra debilidad que, no obstante, es extensible a muchos otros países de la OCDE. 51. La Unión Europea ha sido un motor de cambio muy positivo para la rendición de cuentas en las Comunidades Autónomas. Especialmente en aquellas Comunidades Autónomas más beneficiadas por fondos europeos. La trasposición de directivas comunitarias sobre transparencia, contratación, buen gobierno, etcétera ha sido generalizada y también su puesta en práctica. En este sentido, las evaluaciones de Transparencia Internacional sobre las Comunidades Autónomas reflejan una puntuación en general alta o muy alta. 52. Los presupuestos participativos en España tienen una historia de más de 20 años, aunque su difusión se ha acelerado en el último quinquenio y todas las grandes ciudades reservan una parte de sus inversiones para ser repartidas tras un proceso de deliberación ciudadano. Sin embargo, en las Comunidades Autónomas los presupuestos participativos son más bien anecdóticos. 53. Las encuestas disponibles ponen en evidencia problemas de falta de información sobre la asignación de responsabilidades de gasto y tributarias. El acierto en la atribución de responsabilidades es sustancialmente mayor en el caso de las correspondientes al gobierno central que en el de los autonómicos. Alrededor de la mitad de ciudadanos no tienen claro quién es el responsable principal en materia sanitaria y educativa. Al contrario, la responsabilidad del Gobierno Central sobre pensiones y seguro por desempleo es, en general, bien asignada. Los resultados son similares en lo que atañe a los impuestos. El acierto es sustancialmente mayor en los impuestos exclusivos del Estado (Impuesto sobre Sociedades) que en los autonómicos o los compartidos (IVA e IRPF). No obstante, con el paso del tiempo, los ciudadanos tienden a asignar mejor entre los distintos niveles de gobierno las responsabilidades de gasto y los impuestos. Entre 2005 y 2015 hay una clara tendencia creciente en las cuatro competencias seleccionadas (Gráfico 12). 20 Gráfico 12: Porcentaje de asignación correcta de responsabilidades en prestaciones por desempleo, pensiones, educación y sanidad en 2005, 2010 y 2015 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Seguro desempleo Pensiones Educación Sanidad 2005 2010 2015 Fuente: Elaboración propia a partir de encuetas del Instituto de Estudios Fiscales. 9. El Impacto de la Descentralización: Evidencia Empírica 54. La descentralización regional es valorada muy positivamente por los españoles. La Gran Recesión supuso un retroceso significativo pero transitorio de esta evaluación general. Dentro del debate sobre el ajuste fiscal, se criticó el sobrecoste que supone la descentralización, pero en los últimos años ha remontado de nuevo la valoración positiva y son amplia mayoría los partidarios. 55. La descentralización ha contribuido a una mayor convergencia en la calidad de los servicios públicos (Costa-Font, 2010). La fuerte nivelación horizontal y la política regional han cohesionado socialmente y han hecho que los territorios con menor PIB per cápita cuenten con servicios educativos, sanitarios o sociales similares a los de las regiones más ricas. 56. La participación electoral en las elecciones regionales es solo ligeramente inferior a la de las elecciones generales. El espacio político autonómico es muy relevante para los españoles. 57. La descentralización regional en España ha permitido un mejor ajuste de la prestación de los servicios a las particularidades de cada región en cuanto a su demografía, su orografía y las preferencias sociales (Solé-Ollé, 2009). 58. La descentralización también ha permitido ensayar fórmulas diferentes para la organización de los servicios públicos, especialmente en educación y sanidad (Gallego y Subirats, 2011). 59. La descentralización ha ayudado a encajar la diversidad lingüística y cultural existente en España y que se traduce en la existencia de partidos políticos de carácter regionalista y nacionalista. No obstante, y como indicamos en el apartado siguiente, la descentralización no ha sido una panacea para resolver tensiones territoriales centrífugas. 21 10. Lecciones del Caso Español 60. Aunque el mapa territorial y las fronteras regionales vienen la mayor parte de las veces dados, la existencia de diferencias extremas en el tamaño poblacional (La Rioja con 315,000 habitantes versus Andalucía con 8.4 millones) plantea desafíos si se pretende desplegar una descentralización intensa y simétrica. Para algunos servicios públicos (sanidad o educación universitaria) las economías de escala exigen tamaños poblacionales que superen un umbral. 61. Si bien la descentralización política y financiera es una solución razonable para acomodar diferencias étnicas, culturales o lingüísticas, es un proceso que también contribuye a reforzarlas. La búsqueda y creación de identidades regionales propias se refuerza con la descentralización. Por otro lado, la apuesta por soluciones asimétricas en el sistema de financiación o el acervo competencial para satisfacer a las regiones con mayor demanda de descentralización, pueden generar sentimientos de agravio comparativo. El “federalismo asimétrico” que busca garantizar mayor estabilidad política puede acabar generando el efecto contrario en el medio y largo plazo: la tolerancia a las asimetrías es variable en el tiempo (Bird y Ebel, 2006). 62. Es fundamental adaptar el marco institucional a la nueva realidad descentralizada. En particular, resulta clave contar con foros políticos y técnicos bien organizados para coordinarse vertical y horizontalmente, en los que el gobierno central y los subcentrales actúen con lealtad mutua. Además, es muy importante contar con organismos técnicos independientes que permitan resolver algunas cuestiones difíciles porque conllevan efectos económicos. Por ejemplo, la estimación de las diferencias en los costes de los servicios públicos o la medición de la capacidad fiscal en cada región. 63. El mantenimiento de una legislación nacional básica muy exigente permite garantizar unidad de mercado, derechos de los ciudadanos y homogeneidad en la prestación de servicios. Pero también puede limitar las ventajas de la descentralización (el mejor ajuste a las preferencias dominantes en cada región o la experimentación) y dificultar la correcta asignación de responsabilidades de los gobiernos. Cuando varios niveles de gobierno tienen competencias muy relevantes sobre un servicio, la rendición de cuentas (“accountability”) se complica. Y aumenta la probabilidad de que se produzcan conflictos políticos sobre el alcance de las competencias de cada nivel de gobierno que acaban en el Tribunal constitucional. 64. La descentralización regional conlleva la necesidad de reforzar reglas fiscales y contar con supervisores fiscales y organismos de coordinación para evitar problemas de estabilidad fiscal. Cuando aumenta el número de agentes que deciden sobre déficit y deuda y teniendo en cuenta que la estabilidad fiscal es un bien común, aumentan los incentivos al desequilibrio fiscal. Especialmente, si existen expectativas fundadas de rescate (“bail-out”). Por ello, en paralelo a la descentralización de los instrumentos fiscales, hay que adaptar el marco institucional para garantizar la estabilidad fiscal (Lago- Peñas et al, 2019). 65. Es importante equilibrar la elasticidad de gasto e ingresos regionales al ciclo económico. En particular, si se descentralizan gastos rígidos a la baja y difíciles de ajustar, pero los ingresos son sensibles al ciclo, se genera un problema de coherencia fiscal, que se agrava si las regiones no cuentan con instrumentos fiscales suficientes. 66. Resulta fundamental dejarse guiar por la teoría del federalismo fiscal a la hora de decidir qué impuestos se descentralizan; cuáles son las buenas opciones y cuáles no (Martínez-Vázquez, 2013). Aunque en general, España no se ha desviado significativamente de lo que serían las buenas opciones, los principales problemas han surgido cuando lo ha hecho. En concreto, la cesión del Impuesto sobre sociedades a las Comunidades Autónomas forales ha generado externalidades negativas y distorsiones en la localización 22 empresarial que han acabo siendo sancionadas por los tribunales. Por su parte, la cesión sin restricciones de los impuestos sobre sucesiones y donaciones y sobre patrimonio neto han provocado un proceso de competencia fiscal nociva, poniendo en riesgo la existencia de ambos tributos. Como sostiene la literatura, descentralizar impuestos que gravan bases altamente móviles como la riqueza no inmobiliaria es una mala idea. 67. Aunque la evidencia empírica internacional muestra que la descentralización tributaria (casi) siempre va más despacio y por detrás de la del gasto, la falta de autonomía y responsabilidad fiscal es un problema para la rendición de cuentas (“accountability”), la responsabilidad y estabilidad fiscal y el funcionamiento global del sistema. Idealmente, los procesos deberían converger lo máximo posible en el tiempo. En España el proceso de descentralización tributaria arrancó realmente veinte años más tarde que la del gasto y todavía hoy no está del todo resuelto. 68. La nivelación horizontal (“fiscal equalization”) es un reto mayor. Hay que ponerse de acuerdo sobre el grado (plena o parcial) y sobre si se nivelan solo las diferencias en la capacidad fiscal o si además de las necesidades de gasto generadas por distintos perfiles demográficos, etcétera también se tienen en cuenta las diferencias interregionales en el coste de los servicios estándares. Compensar las diferencias interregionales en el coste de prestación de servicios es más difícil, pero no imposible. Pero para hacerlo bien hacen falta organismos técnicos independientes que fundamenten las ponderaciones y las cifras. 69. Una buena política regional exige buenos planes regionales de desarrollo y evaluación ex-ante de la rentabilidad social de los proyectos mediante análisis coste-beneficio. La abundancia de recursos para invertir sin una cultura de análisis de eficiencia bien asentada conduce a malas decisiones de inversión, que van desde la infrautilización de las infraestructuras construidas a una carga de mantenimiento (gasto corriente) a menudo no prevista y que condiciona los presupuestos futuros. La experiencia española muestra que no es suficiente invertir dinero. Hay que optimizar su empleo. Y eso pasa por adaptar el marco institucional del gobierno decisor a la cultura de la evaluación rigurosa de la rentabilidad social del gasto público. 23 Anexo I: La Economía de las Regiones 70. Los gráficos 13 y 14, muestran las grandes diferencias en la dimensión poblacional y la distribución del Producto Interior Bruto (PIB) entre las regiones españolas. En 2018 la población total española alcanzó los 46.7 millones de personas. El PIB fue de 1.21 billones de euros. Cuatro regiones (Andalucía, Cataluña, Madrid y Comunidad Valenciana) suponen el 30% de la superficie de España y concentran alrededor del 60% de la población y del PIB. En todo caso, este patrón es frecuente y supone una concentración espacial de población y PIB inferior a la que se registra en Chile, por ejemplo. 71. También existen diferencias interregionales significativas en el PIB per cápita, pero en línea con lo observado en los países de la OCDE. De nuevo, España no es excepción (Gráfico 15). 72. Como sugiere la pendiente negativa de la recta de regresión en el Gráfico 16, la descentralización en España ha venido acompañada por una leve convergencia en los PIB per cápita regionales, que resulta más intensa cuando excluimos la región de la capital, Madrid. No obstante, la mejora de las regiones más pobres en 1980 (Extremadura, Galicia, Castilla y León) viene explicado en mayor medida por pérdida de población relativa que por el incremento en la cuota del PIB (Gráfico 17).16 Los mecanismos de mercado (los flujos migratorios) habrían sido comparativamente más relevantes que la intervención pública. Gráfico 13: Distribución regional de la población española en 2018. 20% 16% 12% 8% 4% 0% Fuente: Elaboración propia a partir de datos de la Contabilidad Regional de España elaborada por el Instituto Nacional de Estadística. 16 El PIB per cápita de la región i-ésima respecto a la media n puede expresarse como el producto de la cuota del PIB regional sobre PIBi el total y la inversa de la cuota de la población: POBi PIBi 1 . Para una región por debajo de la media en PIB per cápita, =  PIBn PIBn POBi POBn POBn la convergencia se puede lograr aumentando la cuota de PIB (“convergencia activa”) o reduciendo la de población (“convergenci a pasiva”) vía migración interregional, migración extranjera o diferencias en las tasas de natalidad. En el gráfico 17, las CCAA que ven aumentar su PIB per cápita relativo son las que aparecen a la derecha de la recta de regresión. Al contrario, las que se sitúan a la izquierda perderían posiciones relativas. 24 Gráfico 14: Distribución regional del PIB español en 2018. 20% 16% 12% 8% 4% 0% Fuente: Elaboración propia a partir de datos de la Contabilidad Regional de España elaborada por el Instituto Nacional de Estadística. Gráfico 15: PIB per cápita en las regiones españolas 40,000 35,000 30,000 25,000 20,000 15,000 10,000 5,000 0 Fuente: Elaboración propia a partir de datos de la Contabilidad Regional de España elaborada por el Instituto Nacional de Estadística. 25 Gráfico 16: Relación entre la variación en el PIB per cápita en el período 1980-2017 y su valor en 1980. Valores del PIB per cápita (PIBpc) expresados en porcentaje sobre el total nacional 20 Madrid r = -0.25 15 r* = -0.43 Extremadura 10 Variación PIBpc Galicia 5 Castilla y León Aragón 0 Cataluña País Vasco Andalucía Castilla-La -5 Asturias Navarra Mancha -10 Baleares Murcia La Rioja -15 Comunidad Valenciana Canarias Cantabria -20 50 60 70 80 90 100 110 120 130 140 PIBpc 1980 Nota: ρ es el coeficiente de correlación; ρ*es el coeficiente excluyendo a Madrid Fuente: Elaboración propia a partir de la Contabilidad regional del INE Gráfico 17: Cambios en las cuotas regionales del PIB y de la población (1980-2017), Valores en porcentaje sobre el total nacional 2 Madrid 1 Comunidad Canarias ρ= 0.72 Valenciana Andalucía Murcia Baleares Navarra Cataluña POB 0 Castilla-La La Rioja Mancha Cantabria Aragón Asturias Extremadura -1 País Vasco Galicia Castilla y -2 León -2 0 2 4 PIB Nota: ρ es el coeficiente de correlación Fuente: Elaboración propia a partir de datos de la Contabilidad regional del INE. 73. Gracias a la intensa política regional europea y a la apuesta por un mapa de infraestructuras vertebrador, el stock de infraestructuras (calculado por el método del inventario permanente) ha tendido a crecer significativamente más en las regiones que tenían un PIB per cápita más bajo en 1980 (Gráfico 18). 74. No existen diferencias notables en la desigualdad de la renta entre regiones. Todas se sitúan próximas a la media de España. En 2015, España tenía un índice de GINI para la distribución personal de la renta de 0.28, con todas las regiones en el intervalo 0.25-0.30 (Herrero et al, 2018). En cambio, la tasa de paro española es elevada en perspectiva internacional y existen diferencias notables entre regiones sin que ello motive migraciones interregionales (Gráfico 19). La explicación se encuentra en la fortaleza de las 26 redes familiares, los programas de garantía de rentas de diversos tipos (incluyendo los subsidios por desempleo) y el hecho de que la tasa media es alta, de forma que los diferenciales son menos incentivadores. 75. Finalmente, la redistribución del sector público vía impuestos y gastos (tanto transferencias monetarias como prestaciones de servicios educativos, sanitarios y asistencia social) es intensa en España (Gráfico 20) Gráfico 18: Relación entre la variación en las participaciones en el stock de infraestructuras públicas entre 1980 y 2015 y el PIB per cápita en 1980. 2.5 Andalucía r = -0.48 2.0 Galicia 1.5 Madrid 1.0 Murcia Cantabria 0.5 Baleares ExtremaduraCastilla-La Stock Comunidad Cataluña 0.0 Mancha Asturias Valenciana -0.5 Canarias Castilla y León -1.0 Navarra La Rioja -1.5 País Vasco Aragón -2.0 -2.5 50 60 70 80 90 100 110 120 130 140 PIB pc Fuente: Elaboración propia a partir de la base de datos de stock de capital de IVIE-BBVA y la Contabilidad regional del INE. Gráfico 19: Tasa de paro. España y Comunidades Autónomas. Primer trimestre de 2019 30 25 20 15 10 5 0 Fuente: Elaboración propia a partir de INE - Encuesta de Población Activa 27 Gráfico 20: Renta Primaria per cápita y Renta disponible per cápita en las regiones en 2014 25,000 20,000 15,000 10,000 5,000 0 -5,000 Fuente: Elaboración propia a partir de Herrero et al. (2018) 28 Anexo II: Evolución de las Cuentas Regionales 76. El gráfico 21 recoge la evolución del gasto distribuido por funciones. Sanidad y educación absorben alrededor de dos tercios del gasto de forma consistente en el tiempo. A continuación, las actuaciones de protección social (atención a la dependencia) y las de carácter económico (que incluye buena parte de la política regional) suponen, de forma combinada, alrededor de un quinto del gasto autonómico. 77. La distribución económica del gasto aparece en el Gráfico 22. En correspondencia con la naturaleza intensiva en factor trabajo de los servicios sanitario y educativo y de la predominancia de la prestación directa sobre los conciertos con el sector privado, los gastos de personal de las regiones suponen el 5% del gasto autonómico, más de un tercio del total. El gasto en capital (inversión directa y transferencias al sector privado para el gasto en capital) son muy significativas en términos macroeconómicas (entre el 1% y el 2% del PIB español a lo largo del período), aunque menores comparativamente. Además, se refleja con claridad como los gastos de capital han sido una partida de ajuste del gasto. Tres cuartas partes de la caída en la ratio gasto regional sobre el PIB español entre 2010 y 2017 se produjo en los gastos de capital (inversión y transferencias). En concreto, el gasto total se redujo -1.6% y el gasto en capital -1.2%. Gráfico 21: Evolución del gasto regional expresado como porcentaje del PIB español: Clasificación funcional 16% Actuaciones de Carácter 14% General 1.8% 1.6% 1.5% Actuaciones de Carácter 12% 1.4% 1.3% 1.3% 1.1% 1.2% Económico 10% 3.7% 3.4% 3.6% 3.4% 3.3% 3.3% 3.3% Cultura 3.3% 8% Educación 6% 5.6% 5.4% 6.1% 5.6% 5.4% 5.4% 5.4% 5.4% 4% Sanidad 2% 1.8% 1.7% 1.6% 1.4% 1.4% 1.4% 1.4% 1.5% Actuaciones de 0% Protección y Promoción 2010 2011 2012 2013 2014 2015 2016 2017 Social Fuente: Elaboración propia a partir de datos del Ministerio de Hacienda y del INE. 29 Gráfico 22: Evolución del gasto regional expresado como porcentaje del PIB español: Clasificación económica 16.0% 14.0% 1.2% 0.9% 1.0% 7. Transferencias 0.6% 0.7% 0.6% 0.6% 0.9% 0.7% 0.5% 0.5% de Capital 12.0% 0.5% 0.5% 0.5% 0.4% 0.4% 6. Inversiones Reales 10.0% 4. Transferencias Corrientes 8.0% 3. Intereses de la 2.7% 2.7% 3.5% 2.9% 3.0% 2.8% 2.8% deuda 6.0% 2.8% 2. Gastos Corrientes en Bienes 4.0% y Servicios 1. Gastos de 5.2% 5.1% 4.8% 4.9% 5.1% 5.1% 5.1% 5.0% Personal 2.0% 0.0% 2010 2011 2012 2013 2014 2015 2016 2017 Fuente: Elaboración propia a partir de datos del Ministerio de Hacienda y del INE. 78. Los Gráficos 23 y 24 representan la evolución del déficit autonómico y la deuda desde el año 2010 hasta 2018. En ambos casos, las cifras se ofrecen en valores absolutos y porcentaje del PIB español y como porcentaje del PIB español. Como hemos anticipado, la Gran Recesión dispara el déficit regional, que supera los 50,000 millones y el 5% del PIB español en el año 2011, algo que en parte tiene que ver con el funcionamiento del sistema: unas previsiones de ingresos infladas, la confianza de muchos responsables políticos autonómico en que la crisis duraría menos y por tanto el ajuste del gasto no era tan urgente, combinados con el ya explicado retraso de dos años en la liquidación del sistema de financiación regional explican la diferencia de timing entre la recesión y el déficit. Desde una perspectiva comparada, el caso español ha sido excepcional, en negativo. A partir de 2012 la situación comienza a normalizarse. La dinámica de la deuda refleja fielmente el comportamiento del déficit en el período. En la actualidad solo Canadá entre los países de la OCDE muestra una mayor ratio de deuda pública regional sobre PIB. 79. Finalmente, el Gráfico 25 detalla la composición de la deuda de las regiones a finales de 2018. El 60% está en manos de la administración central, que ha funcionado de sostén financiero a unas regiones con grandes problemas (en algunos casos, imposibilidad de facto) de acudir a los mercados financieros. 30 Gráfico 23: Déficit regional en millones de euros y como porcentaje del PIB 0 0% -20,000 -2% -40,000 -4% -60,000 -6% 2010 2011 2012 2013 2014 2015 2016 2017 2018 Déficit Déficit/PIBpm Fuente: Elaboración propia a partir del boletín estadístico del Banco de España. Gráfico 24: Deuda regional en valor absoluto (millones de euros) y como porcentaje del PIB 350,000 30% 300,000 25% 250,000 20% 200,000 15% 150,000 10% 100,000 50,000 5% 0 0% 2010 2011 2012 2013 2014 2015 2016 2017 2018 Deuda Deuda/PIBpm Fuente: Elaboración propia a partir del Boletín Estadístico del Banco de España Gráfico 25: Composición de la deuda según forma jurídica y acreedor en 2018 Valores representativos de 3% deuda 15% Prestamos con Instituciones Financieras 14% Préstamos con el resto del Mundo 60% 8% Fondo de Financiación CCAA Asociaciones Público-Privadas Fuente: Elaboración propia a partir del Boletín Estadístico del Banco de España 31 Anexo III: Índice General de Transparencia 80. La transparencia en la gestión de los Gobiernos regionales es muy elevada en perspectiva comparada (Gráfico 26). En 2016, último año con datos conocidos, se alcanzó en media un valor de 95 sobre un máximo de 100. De hecho, seis regiones consiguen la puntuación máxima y solo cuatro se sitúan por debajo de 90 puntos. Gráfico 26: Índice general de Transparencia en 2016. El rango va de 0 a 100, que sería la máxima transparencia. Transparencia Página web, Transparencia Transparencia Transparencia Derecho de activa e relaciones económico- en las en materias de acceso a la información con los financiera contrataciones, ordenación del información sobre la ciudadanos convenios, territorio, comunidad y la subvenciones urbanismo y autónoma sociedad, y y costes de los obras públicas participación servicios ciudadana Andalucía 94.7 95.8 93.8 100 95 100 Aragón 84.2 91.7 87.5 84.6 100 90 Asturias 92.1 100 96.9 92.3 100 80 Islas 100 100 100 88.5 100 100 Baleares Canarias 100 100 96.9 80.8 90 100 Cantabria 100 83.3 100 84.6 100 100 Castilla y 94.7 100 90.6 88.5 95 100 León Castilla-La 89.5 83.3 90.6 84.6 100 100 Mancha Cataluña 100 100 96.9 92.3 100 100 Extremadura 89.5 91.7 87.5 84.6 85 90 Galicia 92.1 87.5 100 84.6 85 100 La Rioja 94.7 100 100 100 95 95 Madrid 94.7 100 100 100 95 100 Murcia 100 95.8 100 92.3 95 95 Navarra 86.8 95.8 75 80.8 80 70 País Vasco 100 100 100 100 100 100 Comunidad 100 100 84.4 84.6 100 100 Valenciana Media 94.9 95.6 94.1 89.6 95 95.3 Fuente: Transparencia Internacional. 32 Anexo IV: Información Administrativa 81. La organización administrativa de las administraciones regionales en España es muy similar a la de la Administración central, aunque cada región cuenta con plena autonomía para configurar su estructura de gobierno. 82. Existe un Presidente, habitualmente uno o dos Vicepresidentes (pero no es obligatorio; en algunas Comunidades Autónomas no existe). A continuación, se distribuyen las competencias en consejerías, al frente de cada una de ellas aparece un Consejero, que sería el ministro regional del ramo. El número típico de consejerías se sitúa alrededor de diez. 83. Las consejerías se organizan, a su vez, en direcciones generales, al frente de las cuales se halla un Director General. Además, existen otros organismos que dependen de las consejerías pero que son distintas por su naturaleza jurídica o función. En algunas CCAA se ha creado la figura del Viceconsejero, que sería un cargo intermedio entre el Consejero y los Directores generales. Cada Dirección general se organiza en subdirecciones que, a su vez, se estructuran en jefaturas de servicio. 84. Finalmente existen organismos autonómicos independientes del poder autonómico: el Parlamento, el Organismo externo de control de cuentas regional, organismos consultivos, etc. 85. A modo de ejemplo, en el Gráfico 27 se representa la estructura organizativa del Gobierno regional de Galicia y en el 28 se detalla la organización de una de sus Consejerías, la de Hacienda. Gráfico 27: Ejemplo de organización del Gobierno regional de Galicia Fuente: Elaboración propia. Gráfico 28: Ejemplo de organización de la Consejería de Hacienda del Gobierno regional de Galicia Fuente: Elaboración propia. 33 REFERENCIAS BIRD R, y EBEL R. (2006): “Fiscal federalism and national unity” en AHMAD, E. y BROSIO, G. (Eds.): Handbook of fiscal federalism, Chentelham: Edward Elgar. CAAMAÑO, J. y LAGO-PEÑAS, S. (2008): "Evolución y problemas del marco de estabilidad presupuestaria en España: especial referencia a las haciendas autonómicas", en Villar, M. (Dir.). Estudios jurídicos en memoria de Don César Albiñana García-Quintana., Instituto de Estudios Fiscales, Madrid. COMISIÓN DE EXPERTOS (2018): Informe de la comisión de expertos para la revisión del modelo de financiación autonómica", Madrid: Instituto de Estudios Fiscales. https://www.ief.es/docs/destacados/publicaciones/libros/lb/2018_ReformaFinanciacionTerritorial.pdf COSTA-FONT, J. (2010): “Does devolution lead to regional inequalities in welfare activity?”, Environment and Planning C: Government and Policy, 28: 435-449. FERNÁNDEZ-LLERA, R., LAGO-PEÑAS, S. y MARTINEZ-VAZQUEZ, J. (2013): "La autonomía tributaria de las comunidades autónomas y su (des)uso: Presencia de una restricción presupuestaria blanda", en LAGO- PEÑAS, S. y MARTÍNEZ-VÁZQUEZ, J. (Dir.): La consolidación fiscal en España: El papel de las Comunidades Autónomas y los municipios (experiencias, retos y perspectivas), Madrid: Instituto de Estudios Fiscales. GALLEGO, R. y SUBIRATS, J. (2011): “Regional welfare regimes and mutli-level governance”, en GUILLÉN, A.M. y LEÓN, M. (Eds.) The Spanish welfare state in European context. Oxfordshire: Routledge. HERRERO, C., VILLAR, A. y SOLER GUILLÉN, Á. (2018): Las facetas del bienestar: Una aproximación multidimensional a la calidad de vida en España y sus Comunidades Autónomas 2006-2015. Bilbao: Fundación BBVA-IVIE. HOOGHE, L., MARKS, G., SCHAKEL, A. H., CHAPMAN-OSTERKATZ, S., NIEDZWIECKI, S., y SHAIR- ROSENFIELD, S. (2016): Measuring regional authority. A postfunctionalist theory of governance. Oxford: Oxford University Press. LAGO-PEÑAS, S. (2006), "Capital grants and regional public investment in Spain: Fungibility of aid or crowding-in effect?", Applied Economics, 38(15): 1737-1747. LAGO-PEÑAS, S., FERNÁNDEZ-LEICEAGA, X., y VAQUERO-GARCÍA, A. (2017). "Spanish fiscal federalism: A successful (but still unfinished) process", Environment and Planning C: Politics and Space, 35(8): 1509- 1525. LAGO-PEÑAS, S., MARTÍNEZ-VÁZQUEZ, J. y SACCHI, A. (2019): “Fiscal stability during the Great Recession: Putting decentralization design to the test”, Regional Studies, forthcoming. MARTÍNEZ-VÁZQUEZ, J. (2013): “Tax Assignments at the regional and local levels”, en AHMAD, E. y BROSIO, G. (Dirs.): Handbook of fiscal federalism and multilevel finance, Chentelham: Edward Elgar. MARTÍNEZ-VÁZQUEZ, J., LAGO-PEÑAS, S. y SACCHI, A. (2017), "The impact of fiscal decentralization: a survey", Journal of Economic Surveys, 31(4): 1095-1129. OECD (2019): OECD Fiscal Decentralization Database, https://www.oecd.org/ctp/federalism/fiscal- decentralisation-database.htm SOLÉ-OLLÉ, A. (2009): “Evaluating the effect of decentralization on public service delivery: The Spanish experience”, en AHMAD, E. y BROSIO, G. (Eds.): Does decentralization enhance service delivery and poverty reduction? Chentelham: Edward Elgar. 34 VON TRAPP, L. ; NICOL, S. ; FONTAINE, P. ; LAGO-PEÑAS, S. y SUYKER, W. (2017): OECD review of the Independent Authority for Fiscal Responsibility (AIReF), París: OECD, http://www.oecd.org/gov/budgeting/airef-review-en.pdf ZABALZA, A. y LÓPEZ-LABORDA J. (2014): “The uneasy coexistence of the Spanish foral and common regional finance systems”, Investigaciones Regionales, 37:119-152. 35