Political and Fiscal Decentralization in South America: A Comparative Analysis of Bolivia, Chile, Colombia, Ecuador, and Peru

Abstract

Since the 1980s, government decentralization has been one of the most important areas of reform in Latin America. Over the years, decentralizing processes have been developed at varying rates of speed and with different areas of emphasis and motivations. All of this has led to a regional variety that provides an interesting opportunity for comparative analysis.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Since the 1980s, government decentralization has been one of the most important areas of reform in Latin America. Over the years, decentralizing processes have been developed at varying rates of speed and with different areas of emphasis and motivations. All of this has led to a regional variety that provides an interesting opportunity for comparative analysis.

Despite this highly heterogeneous reality, many of the major decentralizing reforms have not met the expectations that were generated. Though there have been substantive advances such as democratic elections through direct voting for a large number of sub-national officials, following three decades of decentralization, centralism continues to be a common denominator in Latin America.

The purpose of this study is to compare the current state of political and fiscal decentralization at the intermediate and local levels of government in five countries: Bolivia, Chile, Colombia, Ecuador and Peru. These cases have been chosen based on two main criteria: they are constitutionally unitary governments that are administered in a decentralized fashion in contrast to Argentina, Brazil and Venezuela, which have federal models, and they belong to the same continental sub-region that, while highly diverse, shares a recent history with authoritarian governments (with the exception of Colombia) followed by democratic restorations that came about in a relatively parallel manner.

Decentralization is understood as a public policy process that transfers responsibilities, resources or authority from the central government to sub-national ones, endowing the latter with autonomy (Falleti 2005). There are three types of decentralization: political, fiscal and administrative. This analysis is focused on the first two and provides a brief explanation of the contexts in which the decentralizing processes and political-administrative divisions began in each country.

2 General Description of Decentralization Processes

One of the common characteristics of decentralization processes in Latin America is that they began to develop in the 1980s as democratic systems were recovered following dictatorial periods.

The designs of new governments following years of authoritarianism had to be modern, democratic and pluralist given that they would necessarily have to have a decentralized face (Mascareño 2008). As a result, the decentralization process began with a strong political motivation and included expectations such as improved allocation of public goods and services, the promotion of accountability mechanisms and the encouragement of citizen participation (Daughters and Harper 2006) in addition to opening up a wider range of options for plurality and territorial representativeness (Restrepo 2006).

All of this encouraged the proliferation of a wide range of mechanisms that converted several countries into unique cases for interesting analysis. The countries’ varying political contexts, institutional characteristics and demographic, geographic, ethnic, productive and other features meant that each country would follow its own path. This explains the differences that can be observed today in their administrative divisions, the mechanisms that have been introduced for electing sub-national officials, the design of local and intermediate government and the models that have been used to endow them with resources. Despite this broad diversity, in each of the cases studied, a significant portion of those objectives have not been met, leaving “la sensación que la descentralización no ha logrado generar los procesos de inclusión política, social y económica que se esperaban”¹ (Montecinos 2005, p. 84).

On occasion, large decentralization reforms have resulted in formal changes that failed to substantially modify the design in practice or to satisfy the high expectations generated among the populace. Central governments continued to make important decisions, transferred limited powers to the sub-national level and allocated resources to local and intermediate governments whose expenditures are highly conditioned by the center.

In addition, the generalized crisis of the second half of the 1990s forced a change in focus in what Gómez Sabaini and Jiménez call the second wave of decentralizing policies. “No se priorizaron los regímenes de coparticipación de impuestos (habitualmente de libre disponibilidad), como en los años anteriores, sino que se dio preferencia a canalizar los recursos federales a los gobiernos subnacionales para educación y salud, fortaleciendo los mecanismos de transferencia con asignación específica”² (2011, p. 15). As we will see, this type of transfer tends to reduce receiving governments’ power to manage resources at their own discretion, which ends up strengthening the position of the central government. There is no doubt that three decades after these policies were first implemented, Latin America is more decentralized than before. However, public affairs are still centralist in nature, and it seems that this situation will not change in the short term.

2.1 Bolivia

Following the 1982 return to democracy, Bolivia was again ruled in accordance with the Political Constitution of 1967, which allowed for the democratic election of mayors in the provincial capitals and the constitutional recognition of the governments of each departamento (department; region level). In the mid-1990s, decentralization was expanded at the municipal level, which meant that mayors throughout the country could be elected. However, the intermediate level of government faced a significant setback when the constitutional recognition of regional governments was eliminated in order to increase efficiency.

In 2005, the country was in the midst of a strong political and social crisis. Under the administration of Carlos Mesa, the regional governments were strengthened once again and the democratic election of their officials was allowed. These entities also were granted a series of attributes that allowed them to work with greater levels of autonomy. Following the rise to power of Evo Morales in 2006, the country experienced increasing political tension dominated by ethnic and territorial variables in which the opposition argued strongly for the autonomy of the departments and for Sucre to be considered the capital. In May 2008, Santa Cruz declared its autonomy and manifested its intention to charge special taxes, which the government declared unconstitutional (La Nación May 18 2008).

In this context, a new constitution was passed in 2009 that defined Bolivia as a unitary, plurinational, decentralized state with autonomies. The text states that due to the “existencia precolonial de las naciones y pueblo indígena originario campesinos y su dominio ancestral sobre sus territorios, se garantiza su libre determinación en el marco de la unidad del Estado, que consiste en su derecho a la autonomía, al autogobierno”³ (Asamblea Constituyente de Bolivia 2009, p. 3). While the Constitution addresses issues of decentralization, the administrative model has not managed to resolve substantive matters such as poverty and inequality. “Las críticas se concentran propiamente sobre este punto, el de la fragilidad endémica del Estado boliviano que no ha logrado consolidar su institucionalidad en casi dos siglos de existencia”⁴ (Herman 2010, p. 39). In addition to this new imbalance between expectations and results, there are ethnic and territorial tensions in the country which undoubtedly will generate more than one internal conflict in the future.

2.2 Chile

Following the return to democracy in 1990, Chile quickly began to develop a decentralizing process based on the territorial structure inherited from the dictatorship. Early on in Aylwin’s administration, the statutory laws of Municipalities (1991) and Regional Government and Administration (1992) were passed. These pieces of legislation set the stage for the institutional structure that continues to be in place to this day at the territorial level. The former restored the democratic election of local officials and eliminated authoritarian enclaves within municipalities. It also restored autonomous administrative capacities and partially modified sources of income. The latter created regional governments for each of the regions as an intermediate-level agency, though one that was not completely autonomous from the central government.

While the creation of regional governments has led to unprecedented innovation, its effects on decentralization have been very limited. This is due to the fact that officials are not elected through direct democratic voting processes. In addition, the regional governments have limited budgets that come almost entirely from central transfers (they are not authorized to introduce their own taxes). There is also a lack of autonomy from the central level as well as technical and human limitations that keep these governments from successfully completing the tasks assigned to them.

2.3 Colombia

This country has been characterized by marked territorial differences that have created regional and local leaderships throughout its republican history.

The Constitution of 1991 expanded a decentralizing process begun in 1986 by allowing for the democratic election of governors, granting greater political and administrative autonomy to departamentos and municipalities, modifying the distribution of responsibilities among the three levels of government and elevating the transfer of resources to sub-national governments to a level of constitutional import.

Under the administration of Luis Uribe (2002–2010), Colombia stopped the progress that it was making on its road to decentralization. The period saw, among other things, cuts in municipal budgets and an increase in the resources of the central government that were invested in municipalities. Matters of decentralization returned to the fore beginning in 2010, when José Manuel Santos was elected. In 2011, the Territorial Organization Law which promotes planning and competitivity in the regions was passed.

Colombia is considered one of the most decentralized countries in the region and is regularly compared to the federal states of Argentina and Brazil (Subsecretaría de Desarrollo Regional y Administrativo 2009). However, it would seem that the desired results are far from being achieved. The current model generated the conditions necessary for the sub-national governments to satisfy an adequate provision of services, decrease territorial inequities and promote citizen participation. However, a significant number of them have not managed to meet the basic needs of their population and present deficient performance, “lo cual está asociado a prácticas de captura de los aparatos públicos, de corrupción y de clientelismo”⁵ (Maldonado 2011, p. 1).

2.4 Ecuador

When democracy was restored in 1979, Ecuador returned to the practice of holding democratic elections for local and regional officials. The country also passed a new Political Constitution (1978) which stated that “Ecuador es un Estado constitucional de derechos y justicia, social, democrático, soberano, independiente, unitario (…). Se organiza en forma de República y se gobierna de manera descentralizada”⁶ (Asamblea Constituyente 1978, p. 16). In 1992 a set of fiscal reforms was approved that created important sub-national transfer funds. One year later, the National Council on Modernization was established in order to generate a national decentralization and de-concentration plan. In 1998 a new transfer tool known as the ‘15 % Law’ was created and the Special Law on Government Decentralization and Social Participation was passed.

However, this entire process was characterized by its voluntary nature, as sub-national governments were allowed to request greater jurisdiction; the meager organization of the methodology for doing so; the ambiguity with which the government would transfer – or not – the resources requested; and the lack of clarity with which it planned interaction with territorial stakeholders (Verdesoto 2006).

Recent institutional changes have reenergized the decentralizing process. These include the constitutional reform of 2008 and the Statutory Code of Territorial Organization, Autonomy and Decentralization (COOTAD, acronym in Spanish) of 2010. These two pieces of legislation set out the general structure of the current and incipient decentralizing process. One year later, the National Jurisdiction Commission was created to transfer responsibilities from the central government to sub-national ones. Finally, the 2012–2015 National Decentralization Plan was created. This instrument was based on massive citizen participation and is designed to develop strategic medium-range plans that promote greater inter-territorial equity.

While it is still too early to evaluate this entire new institutional framework, the current model already presents some weaknesses. For example, there are Decentralized Autonomous Governments with uneven institutional capacities. As a result, some of them cannot meet some of their mandatory responsibilities while others have already moved voluntarily towards a larger transfer of responsibilities in the earlier process. The latter must now slow down their pace in order to reach uniformity among the entities that share a level. The definition of the system of responsibilities is not yet complete. There are still loose ends, mainly in the area of fiscal decentralization (Serrano and Acosta 2011a).

2.5 Peru

In 1979, Peru passed a new Political Constitution that eliminated the authoritarian government, recognized a series of powers of local governments and restored the democratic election of their officials. Ten years later – and strongly influenced by electoral calculations that forecast a loss at the polls in the presidential election of 1990-, the government of Alan García strengthened the intermediate level of government, creating 12 regions. However, none of the changes that he proposed were ever implemented.

Peru’s democracy was halted during the Fujimori administration. His authoritarian government even closed the regional governments in 1993 along with Congress and the courts through a self-coup. One year later, following strong internal and international pressure, a new Constitution was passed that restored recognition of sub-national governments and established that Peru is a Republic that is “democrática, social, independiente y soberana. El Estado es uno e indivisible. Su gobierno es unitario, representativo y descentralizado”⁷ (Tribunal Constitucional del Perú 1993).

However, most of the initiatives that changed the territorial institutional structure and strongly promoted sub-national governments came in the twenty-first century. The Law on Regionalization was passed in 2002. It allowed for the direct election of regional officials. The Statutory Law on Regional Governments, which set out resources and jurisdictions for those entities, was passed the same year. The Statutory Law on Municipalities was put in place in 2003, strengthening their powers and resources, and the following year brought the introduction of the natural resources canon, the most important budgetary transfer system implemented for sub-national governments by Lima.

The election of Ollanta Humala brought a new focus on the regional level and the promotion of greater levels of autonomy. However, the agents and stakeholders involved in the process have proved critical of this approach, claiming that the practice is far from what is established in the law and that it leads to “dependencia de decisiones del nivel central, duplicidad de competencias, falta de visión estratégica regional y de capacidades de gestión en los equipos técnicos y profesionales, pérdida de vitalidad del componente participativo, entre otros”⁸ (Serrano and Acosta 2011b, p. 7).

3 Political-Administrative Divisions

It is interesting to observe the variety that exists in the region with regard to the countries’ political-administrative divisions. Some nations have a large number of territorial units, and some even allow for the creation of indigenous territories with distinctive administration models, such as Bolivia, Colombia and Ecuador.

Bolivia is organized into departamentos, provinces, municipalities, autonomous regions and rural indigenous territories.⁹ Colombia has departamentos, municipalities, districts, corregimientos departamentales, corregimientos municipales, and indigenous territories. In Ecuador, the new institutional structure introduced through the constitutional reform of 2008 and the COOTAD added six types of Decentralized Autonomous Governments, one for each territorial level: regional governments,¹⁰ provincial governments, metropolitan governments, municipal governments, juntas parroquiales and governments for areas with a high presence of indigenous and Afro-Ecuadorian peoples.

However, in all of the cases studied, there are two territorial units with their own governments that stand out because of their responsibilities, resources and territorial authority: the local ones (with their respective local governments, LG) and the intermediate ones (intermediate governments, IG), which can be viewed in Table 20.1.

Table 20.1 Intermediate Governments (IG) and Local Governments (LG)

An analysis of territorial units and their respective population densities reveals a notorious atomization at the municipal level. As the data in Table 20.2 show, 70 % of the districts in Peru and nearly half of the municipalities in Bolivia and Colombia have fewer than 10,000 inhabitants. While Chile and Ecuador have a lower percentage of LG under that standard, the number of units with over 50,000 inhabitants is relatively low. As one can imagine, the average population of the IG is much higher. However, in all of the countries studied, significant numbers of IGs rule areas with fewer than 500,000 inhabitants.

Table 20.2 Sub-national governments and populations

The creation of sub-national administrations that govern small populations decreases their strategic positions regarding the central government, facilitating the possibility that the latter may domesticate them. This is even more likely when peripheral governments depend on a high percentage of transfers from the center. As we will see in Sect. 20.5, these are usually highly conditioned on financial ministries.

4 Political Decentralization

Political decentralization is directly related to the mechanisms that allow the people elect sub-national officials without being dependent on the central government. This allows, among other things, for elected representatives to answer to their territorial citizenry and not the edicts issued from the center. Since the 1980s, all of the countries studies meet this criterion with the exception of Chile, which recovered its democracy in 1990 but does not yet democratically elect its highest-ranking regional officials (see Table 20.3).

Table 20.3 Democratic election of sub-national officials

Both LG and IG are composed of an executive branch (which is responsible for the highest level of administration of the sub-national government) and another branch that tends to be called legislative, but that in practice is more nominative and oversight-related.

At the LG level, mayors are democratically elected for periods of 4 years in Chile, Colombia and Ecuador and for 5 years in Bolivia and Peru. With the exception of Colombia (and the limitation of immediacy in Bolivia), they may be reelected. The situation of municipal council members (which are called regidores in Peru) is similar. They vary in number within each country depending on the population that the LG governs. However, there are differences. Some council members are not democratically elected in Bolivia but are selected by indigenous communities through their own electoral processes. In Colombia, council members can be reelected immediately and in Ecuador the mayor presides over the municipal council with a right to cast a vote in order to break a tie.

The democratic election of IG officials has come later with the exception of Ecuador, which introduced this process for both levels of government simultaneously. As a result, the governors in Bolivia and Colombia, prefects in Ecuador and regional presidents in Peru are elected directly,¹¹ with greater restrictions on their reelection compared to the local sphere. Chile is the only exception, as the intendente is appointed by the President of the Republic as a de-concentrated agent of the central government. He or she is also the highest ranking official in the regional government, thus playing a double role. This open paradox has meant that in practice the intendente prioritizes his or her role of representative of the Executive over that of the regional government. These officials remain in office at the sole discretion of the President.

In regard to legislative bodies, the members of the assemblies of departamentos in Bolivia and Colombia, provincial councils in Ecuador and regional councils in Peru also are elected democratically and have similar characteristics to those presented by their counterparts in the LG: some of these officials in Bolivia are selected by their indigenous groups through their own processes and prefects preside over provincial councils in Ecuador with the right to a tie-breaking vote.

The exception continues to be Chile, where regional council members are still selected by the municipal council members. A change was made to the Constitution in 2009 that establishes democratic elections in this case, but the law which regulates the election is still under discussion (it entered Congress in September 2011).

Finally, in Bolivia, Colombia and Ecuador, the highest-ranking sub-national officials may be removed from their positions by citizen initiatives via referendums.

However, the democratization of sub-national governments has not led the central government to cede significant amounts of responsibility. The LGs, which are frequently stronger than the IGs, tend to have nearly exclusive responsibilities with high levels of discretion over basic services (sanitation, potable water, drainage, public lighting and waste management, among other) as well as planning. However, in regard to social services such as health, education and housing, they share responsibilities with other levels of government, mainly at the central level, from which they receive the transfer of responsibilities and highly conditioned transfers.

5 Fiscal Decentralization

This type of decentralization is understood as a group of public policies designed to increase the income or fiscal autonomy of sub-national governments. Decentralization can be studied based on income source and in terms of expenditures. While “la asignación de responsabilidades del gasto a los gobiernos subnacionales es importante, la forma en cómo los gobiernos subnacionales financien tales servicios es una preocupación clave”¹² (Gómez and Jiménez 2011, p. 11). The focus of this study is decentralization based on income, specifically in regard to three main areas: self-generated resources (creation of sub-national taxes, delegation of tax authority to officials at this level and non-tax income), transfers (usually from the central government) and debt capacity (Falleti 2005).

Table 20.4 presents an investigatory summary of the fiscal reality of the intermediate governments of the five countries. The first matter to be considered is the capacity that the IGs have to generate their own resources. Only Colombia and Ecuador give these bodies the power to collect taxes; they may not do so in Chile, Peru and Bolivia. Not however that in Bolivia, while autonomous governments in the departamentos do not directly collect taxes, the central government is required by law to transfer the monies earned through the exploitation of natural resources in their areas. In other words, though the departamentos cannot collect taxes, they are assigned to them. Chile is the only country whose IG do not collect money through the payment of fees and licenses, though as we will see in the pages that follow, those sources of income are not truly significant in the other cases. Also Chile is the only country whose IG do not have debt capacity. It is important to note, however, that while this measure restricts their budgetary autonomy, it is also a barrier to a lack of balance in public finances. South America has presented critical cases in which central governments have had to severely restrict borrowing, such as Argentina and Colombia in 2001 and 1997, respectively (Subsecretaría de Desarrollo Regional y Administrativo 2009).

Table 20.4 Fiscal responsibilities of IG in South America

Table 20.5 presents information on the distribution of income and public spending by level of government. The fiscal centrality in both income and expenditures is notorious, as is the budgetary strength of LGs compared to IGs (with the exception of Bolivia) and the apparent fiscal responsibility of sub-national governments, which have balanced income and expenditures. However, these data do not reflect an important vertical fiscal imbalance. In other words, the income collected directly by peripheral governments is considerably lower than the amount that they must spend based on their obligations. This is only balanced through transfers (mainly from the central government) and royalties (also central transfers, but based on the exploitation of non-renewable natural resources).

Table 20.5 Share of income and public expenditures by level of government

Figure 20.1 shows transfers and royalties as percentages of the income of IGs and LGs. The latter far outweigh the former particularly in Chile due to the greater strengths of municipalities when it comes to generating their own income, particularly taxes. Finally, the Chilean case is notable in that it has the third-lowest income at the municipal level but is less dependent on transfers and presents greater capacities to generate own income.

Fig. 20.1

Transfers and royalties as percentages of total income of sub-national governments (Source: Generated by the author based on national budgets)

All of the countries have vertical transfers (from the central government to peripheral ones) related to the exploitation of natural resources. These programs generally benefit the areas in which the natural resources are obtained more than others. The exception is Chile. It almost does not have allocations of this type and there are no vertical transfers at the municipal level. There is a mechanism for horizontal transfers at the local level in Chile.

The resources produced by transfers and royalties represent a very high percentage of the total sub-national income and are also very much conditioned on their expenditure either due to their nature (regular or capital expenditures) or the specific sectors to which they must be allocated. As a result, autonomy in public spending at the sub-national level is highly conditioned by the basis for the transfer.

In order to analyze the situation of sub-national taxes by country, the self-generated income (tax and other) and transfers (including royalties, where applicable) are considered due to the fact that these are the most important line items in their budgets. There are, however, other sources of income including balances from previous periods, social contributions and debt capacity. Finally, all of the numbers presented here reflect the executions of the income of sub-national governments and not their budgets. This is done in order to allow the reader to appreciate how many of the resources effectively enter at the sub-national level with more certainty.

5.1 Bolivia

5.1.1 Self-generated Income

At the level of departamento, while autonomous governments are to receive taxes and fees on natural resources, they do not have the authority to collect them.¹³ The income that is generated through this route is received through transfers related to the hydrocarbon industry. The greatest sources of self-generated income at this level are collected through the sale of assets. However, in 2011, this source was responsible for just 1.25 % of the income of the departamentos, while fees and licenses represented only 0.26 %.

The Autonomous Municipal Governments receive shared tax transfers and collect taxes. The most significant taxes are those related to real estate, sales, vehicles and assets, among others. In 2011, this item represented 12.43 % of all income at this level.

5.1.2 Transfers

Bolivia’s sub-national governments receive a wide range of transfers from the central government. The most important is the Direct Hydrocarbon Tax (IDH, acronym in Spanish), which applies to all departamentos and municipalities. It represents 32 % of total hydrocarbon production, and the resources must be spent on education, health, infrastructure and economic development. Ten percent goes to the departamentos and 31 % goes to the municipalities, with priority placed on areas that produce hydrocarbons. The remainder is distributed to universities, the National General Treasury, the program Renta Digna and the Indigenous Fund. In 2011, resources collected through the IDH represented nearly 68 % of all transfers received by autonomous regional governments.

At this level there is also the Special Tax on Hydrocarbons and Their Derivatives (IEHD, acronym in Spanish), which corresponds to 25 % of the monies collected on internal sale of hydrocarbons and the importation of their derivatives. Half of these resources are distributed in equal parts while the other half is disbursed in relation to the number of inhabitants of each departamento. In 2011, the IEHD covered nearly 30 % of transfers at the departamento level. Finally, there is the Departamento Compensatory Fund, which applies to all of the areas that are under the national average for royalties per inhabitant. Funds are transferred from the National General Treasury until that average is met as long as it does not exceed 10 % of the monies collected through the IEHD. The expenditures related to this transfer also operate based on conditions: 85 % must be used on capital spending and 15 % is to be used on regular spending.

Income from royalties, by contrast, is obtained through the Hydrocarbon Law of 2005, which establishes that 18 % of monies generated through hydrocarbon exploitation is to be distributed as follows: 11 % to the departamentos where hydrocarbons are produced, 1 % to the departamentos of Beni and Pando, and 6 % to the National General Treasury.

In 2011, transfers to the autonomous governments of the departamentos represented 53.86 % of their income. Royalties totaled 26.55 %. These two sources represented 80.41 % of the budget. The departamento that was least dependent on transfers and royalties that year was Beni (67.27 %) and the rest showed high levels of dependency (over 96 %).

At the municipal level, in addition to the IDH (which represented 42.86 % of the income received through transfers in 2011), there are three other instruments that make transfers from the central government. The main one and the principal source of all municipal income is the Popular Tax Co-Participation Fund. Twenty percent of the monies that are collected through a total of eight taxes (included the value added tax, the corporate tax and tax on transactions) go directly to the municipalities and are distributed proportionally based on the number of inhabitants. Eighty-five percent of this fund must be spent on public investments. In 2011, tax co-participation represented nearly 50 % of the income received through transfers.

There are also resources that come from the National Dialogue Law of 2000. These funds are generated through external debt remission. They are resources that are allocated to remedy deficits in municipalities with the highest poverty and risk levels in the areas of health and education. Twenty percent of the monies collected is allocated to education, 10 % goes to health and 70 % is placed in the National Economic and Social Investment Fund. In 2011, these resources represented a little over 7 % of all municipal transfers. Finally, the Autonomous Municipal Governments receive transfers from the Hydrocarbon Licenses, which are charged to the Bolivian Government Oil Facilities. The central government transfers 50 % of the monies collected to the municipalities that produce this material to be spent on public investment and environmental management projects.

Transfers to the Autonomous Municipal Governments in 2011 represented 62.33 % of their income. That same year, over 80 % of the total budget of 233 municipalities was dependent on transfers.

5.2 Chile

5.2.1 Self-generated Income

Regional governments do not generate income through taxes or fees or licenses.¹⁴ In 2011, only 0.05 % of their budgets came from self-generated income. A small exception was introduced in 2005 with the budgetary category of “Self-generated Income Allocated by Law.” However, these are not necessarily new resources because a significant portion of them were already in place through transfers from the central government. Though the name suggests otherwise, the term self-generated income only refers to a very small portion of this category (mining and geo-thermal licenses, resources generated through the laws on gaming casinos and national goods, and others) and these monies are not collected at the regional level. As a result, the central government transfers them.

Municipalities do have self-generated tax income. In 2011, this source represented 41.28 % of all municipal income. The most important category is municipal licenses (mainly related to commerce, exploitation and cleaning contracts), which represent 26.93 % of the self-generated tax income, followed by territorial taxes (25.61 %) and taxes collected through vehicle registration permits and drivers licenses (11.04 %).

5.2.2 Transfers

In 2011, transfers were the source of 99.91 % of the income of regional governments. The most important of these was the National Fund for Regional Development (FNDR, acronym in Spanish), whose resources correspond to a portion of total public investment established in the National Budget each year.

There are two major types of FNDR: traditional or distributed and provisional. The former is composed of resources from the national budget that are mainly allocated to social infrastructure projects. This type of fund is the oldest and allows for the greatest amount of decision-making power at the regional level. It is distributed as follows: 90 % of the fund is set out in the Budgets Law in function of two criteria. They are that (a) 50 % of this amount is distributed in accordance with regional poverty and indigence indicators and (b) the remaining amount is allocated in accordance to indicators that express the region’s territorial characteristics (likelihood that the population will access public services and cost differentials of paving and construction projects, for example). The remaining 10 % is distributed equally based on two criteria: standards of efficiency (measured in the health and education sectors and based on the investments made by the FNDR) and emergency spending. In 2011, this represented 37.57 % of all regional government income.

The FNDR for provisions is composed of funds that are to be invested regionally in sectors that are identified in advance. In this case, the Ministry of Finance makes the investment decisions while the regional governments play a secondary role. This fund represented 35.39 % of the total income of regional governments in 2011.

Other less important mechanisms include the aforementioned Self-Generated Income Allocated by Law and other transfers that are not part of the regional governments’ budget but do require their approval so that the investments related to them can be made.

At the local level, there is the Common Municipal Fund (FCM, acronym in Spanish). In contrast to the rest of the inter-governmental transfer programs described in this study, the FCM is a horizontal municipal transfer tool meant to redistribute resources at this level. It has historically been composed solely of municipal contributions, though beginning in 2005 it has included resources from the government. It is distributed as follows: 25 % is allocated in equal parts to all of the municipalities in the country; 10 % is allocated based on the number of individuals living in poverty in the municipality; 30 % is granted based on the number of tax exempt properties; and 35 % is allocated proportionally to the municipalities with the lowest permanent self-generated income the preceding year.

In 2011, the FCM represented 28.77 % of all municipal income. However, according to data from the Municipal Information System (Sistema de Información Municipal, www.sinim.cl), the importance of this transfer system is more notorious when one considers that it represents over 50 % of the income of half of the country’s municipalities.

In addition, the municipalities receive a series of transfers linked to their responsibilities in the areas of health and education by law, as well as other programs (housing, urban improvement and others). In 2011, these sources represented 9.37 % of their income.

5.3 Colombia

5.3.1 Self-generated Income

The governments of the departamentos are authorized to collected taxes.¹⁵ The most important ones are beer and liquor consumption taxes (which represent nearly 50 % of all taxes collected at this level) followed by registration and licensing (11.38 %), cigarettes and tobacco (7.86 %), motor vehicles (7.52 %) and the tax on gasoline (5.65 %) as well as other smaller taxes. In 2010, that income represented 26.1 % of the total for the departamentos. There are significant differences among these units. For example, nearly 45 % of the income of the valley or Atlantic departamentos comes from self-generated income, while less than 10 % of the income of Vichada, Guainia, Arauca, Vaupes, Putumayo and Amazonas comes from this source. Non-tax self-generated income represented 5.7 % of the total income.

The municipal tax system is composed of 19 taxes, including those applied to commerce and industry (41.52 % of the total municipal taxes in 2010), property (30.66 %) and the tax on gasoline (10.1 %). In 2010, 30.76 % of all municipal taxes corresponded to the collection of taxes at this level. That same year, non-tax self-generated income totaled 5.49 %.

As a result, the total self-generated income of the departamentos represented 33.42 % of all of their income while that of the municipalities was 39.31 %.

5.3.2 Transfers

For 10 years following the passage of the Constitution of 1991, Colombia had a system for transfers based on two mechanisms. The first was the Situado Fiscal, which was directed at the departamentos, and the second was the municipalities’ share. The increase in fiscal imbalance and diverse deficiencies in the allocation of these resources led the government to replace the two mechanisms with a single General Participation System (SGP, acronym in Spanish). The SGP is responsible for a significant portion of the budgets of sub-national entities. In 2010, it represented 46.88 % of the municipal budget and 47.18 % of the budget of the departamento.

Meanwhile, royalties are distributed in two groups. Some go to the territories that produce them and others go through the National Royalty Fund. In 2010, royalties represented 14.95 % of the total income of the departamentos and 5.22 % of that of the municipalities. Royalties and SGP transfers combined represented 66.58 % and 60.69 % of their budgets, respectively.

Budgetary dependence on these sources of income is considerable in sub-national governments. According to data from the National Planning Department, transfers and royalties represented nearly 90 % of the income of four departamentos while 90 % or more of the income of nearly 200 municipalities came from these income sources (Departamento Nacional de Planeación 2010).

5.4 Ecuador

5.4.1 Self-generated Income

Prior to the passage of COOTAD, provincial councils had very limited tax income generation capacities, receiving only 1 % of the alcabala sales taxes and 0.1 % of registrations.¹⁶ In 2010, self-generated income (taxes and other sources) in the provinces represented only 10.4 % the entities’ total income.

By contrast, the municipalities have a broad fixed tax base that is applied to local residents under the principle of fiscal co-responsibility. These include taxes on urban and rural property, vehicles, registrations and licenses, assets, public performances, gaming, sales tax and gains through the purchase and sale of real estate. COOTAD modified the fiscal decentralization model, assigning tax authority on these same taxes to provincial councils.

Self-generated income for municipalities other than taxes mainly consists of fees, investment income, rent and fines. Municipal self-generated income represented 24.8 % of total income for these units in 2010 (40.5 % of self-generated income came from taxes).

5.4.2 Transfers

Ecuador has 18 transfer funds that shift resources from the central government to sub-national ones, contributing most of the income of the latter. In 1998, the Political Constitution was modified and it was established that fiscal transfers would only be made with the corresponding transfer of responsibilities. This meant that transfers would be highly conditioned. The most important of these mechanisms are the Special Transfer Law of 15 % of the National Government to Local Governments (better known as the ‘15 % Law’), the Provincial Development Fund and the Local Development Fund.

The 15 % Law transfers that percentage of the net income to provincial councils (30 %) and municipalities (70 %) based on criteria such as surface area, population and need. It is mainly meant to finance investment plans for economic, social and cultural development.

Transfers represented 89.5 % of the provinces’ total income in 2010. The main mechanisms were the 15 % Law (48.8 %) and the Provincial Development Fund (16.3 %). Municipal transfers represented 75.1 % of all income for those entities and the 15 % Law represented over half of these resources.

5.5 Peru

5.5.1 Self-generated Income

While regional governments are given a share of import taxes by law, they cannot collect them and receive them only through transfers.¹⁷ (Note that in 2011, such income represented only 0.005 % of the total for the departamentos.) As a result, at the level of the departamento, all income is non-tax in origin (fees, fines, administrative interests and the sale of foods, mainly). These sources represented 9.27 % of all income for these entities in 2011.

By contrast, the municipalities can collect taxes. Self-generated income from local taxes represented 8.82 % of these entities’ total income in 2011. The main tax collected is the property tax, which represented 61.94 % of self-generated income at this level in 2011. The tax on production and consumption took a distant second place, representing 5.54 % of self-generated income. If non-tax self-generated income is included, the total self-generated income for municipalities in 2011 represented 12.83 % of total income for this level.

5.5.2 Transfers

At the departamento level, the main transfer mechanism is the Regional Compensation Mechanism, which constitutes nearly all of the budgets of the Regional Governments (Subsecretaría de Desarrollo Regional y Administrativo 2009). It is distributed based on factors of poverty, unmet needs, location on a border, population, tax contribution to the government and performance indicators related to the execution of investments. The resources from this fund go directly to regional investment projects mainly in the areas of agriculture, fishing, mining, energy, tourism, roadways, communications, health, education and the environment.

Thirty percent of the Camisea Socio-economic Development Fund, which is focused on areas through which the main hydrocarbon channels pass, is distributed among the departamentos affected by that infrastructure and Ucayali. With the exception of the last one,¹⁸ the resources are distributed based on indicators linked to population, social need and longitude of the channels and are to be spent mostly on regional investment projects and economic infrastructure maintenance.

Regional governments receive significant income in the form of royalties, which come in two forms: the canon and the mining royalties. The former is a share that sub-national governments receive as a result of compensations generated by the economic exploitation of non-renewable natural resources based on the income obtained by the central government as a result of that activity. There are several types of canons: mining, hydro-energy, gas, fishing, forestry and oil. In addition to mining royalties, each of these is highly conditioned on being spent on regional investment projects. As a whole, transfers and royalties represented 87.73 % of the total income of the regional governments in 2010.

At the municipal level, there is a Municipal Compensation Fund. It is distributed based on criteria such as population, mortality rate, unmet basic needs and rate of rural population and benefits the country’s most outlying and depressed municipalities. In addition, municipalities have significant resources made available to them through the canon concept. While 25 % go to regional governments, 75 % go directly to local ones according to the following format: 10 % goes to local governments in which the resource is exploited; 25 % goes to the local governments that belong to the same province as place in which the exploitation takes place; and 40 % goes to the local governments that belong to the same departamento where the exploitation takes place. In 2011, municipal transfers and grants represented 86.13 % of all income.

6 Conclusions

Three decades after the process began there is no doubt that Latin America now presents a more decentralized face. Proof of this is, for example, the democratic election of key sub-national officials; the political and administrative recognition of ethnic groups in some cases; and the greater – though incomplete – levels of autonomy and budgetary and jurisdictional responsibilities that have been assigned to peripheral governments, at least compared to the early 1980s. In the preceding pages, we have seen how some major decentralizing reforms are limited to formal aspects that do not manage to generate practice or fail to meet the high expectations created, which once again generates incentives for the center to take on greater leadership. Proof of this is the generalized tendency to promote sub-national financing through conditioned transfers from the capitals, the limited transfer of significant responsibilities to the periphery and the institutional weaknesses that persist in sub-national governments. While important efforts to make progress in this area have been observed in all of the cases studied, it is clear that there is a need for more than a set of reforms to untangle a sort of centralist culture that poses more obstacles from those inherent to a decentralizing process.

Regarding the specific analysis of the five countries selected, significant differences and common patterns are observed. One overarching characteristic is the presence of two sub-national levels of government: local and intermediate. The first has greater levels of political, administrative and financial authority while the second reflects somewhat weak strategic positions, constituting their political autonomies later and reflecting practically no opportunity to generate their own tax income (with the exception of Colombia and, recently, Ecuador).

It also was found that the countries studied present significant vertical fiscal imbalances that stem from a diversity of transfers and benefits that come from the center almost without exception. In addition these limit the fiscal and administrative autonomy of peripheral governments by highly conditioning the expenditure of the resources allocated to them and representing a significant portion of their budgets. It also was determined that there was a generalized horizontal fiscal imbalance in that there are substantial differences in the level of budgetary dependence shown by sub-national governments from the same level regarding transfers from the central government.

In regard to the differences observed, the diversity of the administrative units is noteworthy. Colombia and Bolivia present strong territorial traditions as well as high levels of decentralization and have integrated ethnic diversities into territorial policies. In Bolivia, this is very much in line with its policy towards its nations of origin, as indigenous peoples are given the opportunity to govern themselves using autonomous statutes in their territories. Colombia and Ecuador also recognize their ethnic and cultural diversity by providing an opportunity to establish special territories in which representatives of these groups have greater participation in local politics. Chile is on the opposite side of the spectrum, as is Peru, though to a lesser extent, as these are both traditionally centralized nations. The latter has presented a recent openness towards greater levels of political decentralization at the regional level, while the former is still on its way to achieving that and is the only country whose highest-ranking regional officials continue to be appointed by the central government.

From a fiscal perspective, the case of Colombia stands out in that it is the only country that has a wide range of taxes collected by its intermediate-level governments. (Ecuador is following in its footsteps.) This is true to a lesser degree in Bolivia and to a much lesser extent in Peru, where intermediate governments do not have the authority to collect takes but benefit from the tax monies transferred to them by the central government.