- Annalisa Fedelino
- Published Date:
- October 2010
In Bolivia, the initial drive behind the decentralization process was political. In the late 1980s—after decades of dictatorship and neglect of the poorest regions—a need emerged to strengthen democratic structures, starting at the lower levels of government. Decentralization’s initial stage included the introduction of municipal elections and the development of a legal framework for fiscal decentralization in the mid-1990s. A second stage brought a renewed focus on poverty, linked to the Heavily Indebted Poor Countries (HIPC) Initiative, and an attempt to address growing fiscal imbalances, including the rising subnational debt. Finally, significant changes at the regional level (prefecturas) took place in 2005, when the first popular elections of regional governors were held, and when a new hydrocarbon law changed the revenue-sharing regime by assigning a significant share of natural resource revenue to the regions (discussed later in this chapter). These changes in financing arrangements for subnational governments, however, were not accompanied by a redefinition of spending responsibilities across government levels, severely eroding the central government’s flexibility in budget management. As a result, subnational governments receive about one-third of general government revenue, but are responsible for only about one-fourth of total spending. A very large share of subnational resources is in the form of intergovernmental transfers (Table 4.2).
|Type of government||Unitary|
|Population (million, 2007)||9.4|
|Area (million km2)||1.1|
|Levels of government||3|
|Average municipality size (population)||84,172|
|Minimum municipality size (population)||1,287|
|IMF technical assistance missions on||1997, 1998,|
|fiscal decentralization||2001, 2004|
|Total subnational revenue||33.8||10.7|
|Transfers from the central|
|Total subnational expenditure||25.1||8.5|
|Overall balance (before transfers)||n.a.||-6.9|
Transfers include shared revenue.
Transfers include shared revenue.
As of late 2008, in the context of significant regional tensions and constitutional reform, Bolivia is in the midst of a national discussion on a new system of intergovernmental fiscal relations. After a protracted process that began in 2006 and triggered serious confrontations between the government and opposition regions, a new constitution was approved in January 2009. This constitution introduced broad elements of a new system of intergovernmental fiscal relations conducive to more decentralization and included a transitory clause mandating that a provisional regional-autonomy regime be developed until the presidential elections in December 2009. After these elections, a new law on fiscal decentralization was expected to be enacted by the newly elected congress.
The 1990s: Increasing Municipal Autonomy Leads to Fiscal Imbalances
The decentralization process in Bolivia started with municipalities, the lowest tier of government. The first popular elections for mayors were held in 1987; however, fiscal decentralization only truly started in the mid-1990s. The laws of Popular Participation (1994) and Decentralization (1995) set the legal framework for new intergovernmental relations. Municipalities were given new spending and revenue responsibilities, including local taxes, and revenue-sharing agreements were established (municipalities were assigned 20 percent of most national taxes). The status of regions (prefecturas), however, was not as well defined. While they also received new spending and revenue responsibilities (a share of revenues from natural resources and excises), they remained largely under central control, with regional governors appointed by the president.
The decentralization system, although providing poorer local governments with greater resources, suffered from significant vertical imbalances, and only partially addressed horizontal inequalities. Before the reform began in 1994–95, a few cities (regional capitals) received almost all devolved revenues; after 1995, smaller municipalities saw their resources increase substantially, as much as 300 percent in some cases. Nevertheless, local governments remained highly dependent on transfers from the center (including revenue-sharing agreements and earmarked transfers), with own-revenues representing less than one-third of local expenditures during the 1990s. Transfers were mostly based on municipalities’ population size, with no effective mechanism to address inequalities in income and revenue capacity.
Although the weaknesses in the system contributed to increasing deficits at the subnational level during the 1990s, the costs of the decentralization were not initially perceived as a threat to macrofiscal stability. The heavy reliance on central transfers and weak controls on subnational borrowing led to soft budget constraints at the subnational level and a gradual deterioration of the municipalities’ accounts. Because Bolivia was undergoing significant reforms in the 1990s and economic growth was accelerating, the cost of the decentralization was viewed as relatively limited, and easily compensated for by surpluses at the central level. However, subnational debt, particularly short-term debt, started to rise rapidly, especially in the largest municipalities.
As a result of growing concerns with the rising levels of subnational debt, IMF staff were called to provide technical assistance on subnational public finances in the late 1990s. A first technical assistance mission in 1997 focused mainly on the need to develop effective administrative controls on local debt, though it noted that the incentives for overborrowing were due to the lack of clear spending responsibilities (leading to overspending) and the dependence on the central government (and potential bailouts).1 A follow-up mission in 1998 provided a broader analysis of intergovernmental relations and recommended a series of actions to reduce the underlying weaknesses and improve overall fiscal management. The recommended actions included reducing reliance on transfers by increasing local revenues; better defining spending responsibilities, taking into account institutional capacities of smaller municipalities; and improving the transfer system to reduce vertical and horizontal inequalities.
The legal framework was improved, partly reflecting IMF advice and attempts to control subnational debt, but the main weaknesses of the system were not tackled. In 1999, a new law for budget administration reinforced the powers of the Ministry of Finance to control subnational debt and reduced the debt limits. In addition, the authorities made changes in the transfer system to address some of the imbalances. In particular, poverty indicators were to be included when distributing cofinancing resources and HIPC-related revenues (see below); however, the main transfers (revenue-sharing agreements) remained unchanged. The efforts to coordinate overall fiscal policy and to control debt levels continued to be undermined by the absence of comprehensive, reliable, and timely information, and by the lack of political will at the center to enforce restrictions on subnational borrowing.
As the largest municipalities’ financial positions deteriorated, the central government started a debt-restructuring program for subnational governments in 1998. Given the lack of effective budget constraints and the difficulty of monitoring subnational debt, several of the largest municipalities (representing more than three-quarters of local spending) continued to accumulate deficits, with some breaching the debt limits by the late 1990s. Debt-restructuring programs were agreed to between the central and local governments (and one prefectura). In exchange for central government financial support, the municipalities committed to take measures to increase local revenues and control spending. The debt-restructuring program also focused on changing the debt composition, from mostly short-term debt (with private contractors and public agencies) to longer-maturity debt (from the Corporación Andina de Fomento and the World Bank).
Although the debt-restructuring program helped improve municipalities’ financial positions, a 2001 IMF technical assistance mission stressed the need to strengthen mechanisms to prevent future problems. The mission also emphasized that the inclusion of a central bailout component could lead to moral hazard issues. The mission advised establishment of credible sanctions against local governments breaching the rules and significant improvement of the reporting of fiscal and financial data to the Ministry of Finance (including acceleration of the implementation of the new information system, SIGMA, which was being supported by the World Bank and the Inter-American Development Bank).
The Early 2000s: Fiscal Unraveling
Bolivia’s fiscal position deteriorated in the early 2000s as a result of its economic and political crisis, and the ensuing need for a large fiscal adjustment highlighted further tensions in the decentralization system. At the end of the 1990s, Bolivia entered a period of economic and political crisis—with negative per capita growth in 1999–2002—and the fiscal deficit ballooned, reaching 8–9 percent of GDP in 2002–03.
When fiscal adjustment finally began in 2004, it fell disproportionally on the central government, mainly reflecting difficulties in coordinating fiscal policy across government levels.2 Almost any revenue measure was partially offset by higher subnational spending (because revenue-sharing agreements mandated transferring part of increased collections to subnational governments). In addition, the central government’s ability to reduce spending was limited by the lack of clarity in spending responsibilities and of coordination with subnational governments. For example, subnational governments, which controlled teachers’ hiring and working hours, had no incentives to find savings (e.g., by reducing fraud and waste) because the central government had to pay the large wage bill.
Meanwhile, the Enhanced HIPC Initiative led to a new push in the decentralization process, further compounding challenges for fiscal management. Under the 2000 National Dialogue law, savings from the Enhanced HIPC Initiative were to be redirected to municipalities based on poverty indicators. While these resources represented a welcome windfall for the poorer municipalities, the associated earmarking rules introduced rigidity in their use. Furthermore, the HIPC initiative also created unintended fiscal pressures at the central level because the increase in debt (resulting from the fiscal crisis of 2000–02) effectively eliminated any savings from the HIPC debt relief—but not the obligation to transfer funds to municipalities.
The Late 2000s: Stronger Prefecturas Demand Control of Hydrocarbon Resources
Although fiscal decentralization initially focused on municipalities, over time regions have demanded greater autonomy and revenue devolution, especially in the context of booming hydrocarbon prices during 2004–08. Starting in 2005, two significant changes took place: First, regional governors (previously appointed by the president) were popularly elected, marking greater political autonomy for regions. Second, a new Hydrocarbon Law assigned more resources to regions, but without a corresponding transfer of expenditure responsibilities. The hydrocarbon-producing regions have been the main beneficiaries of this new system because a large share of the royalties accrues to them.3
Regions rely primarily on the sharing of hydrocarbon revenues with the center. Royalties on oil and natural gas represent almost two-thirds of total revenue, creating significant horizontal imbalances; for example, Tarija, with 4 percent of the country’s population, receives almost 50 percent of total royalties distributed to regional governments.4 The heavy dependence on energy-based transfers poses important challenges to regions, which may be ill-equipped to manage their volatility; in addition, regions have little control over the hydrocarbon revenue take, which depends on decisions made by the central government (e.g., on the fiscal regime and contract specifications).
The existing revenue-sharing structure has resulted in low incentives to exploit tax bases at the subnational level. Both municipalities and regions have limited autonomy regarding the tax base and the tax rates, which are mostly set by the central government, thus encouraging dependence on transfers and central government decisions. In addition, heavy reliance on central transfers undermines subnational incentives to raise revenue through better tax administration and through increasing revenue. The changes in revenue-sharing introduced by the 2005 Hydrocarbon Law have only exacerbated these problems.
Excessive earmarking of transfers provides incentives for nontransparent accounting and inefficient spending. Most programs earmark transfers in fixed proportions to capital and current expenditures or to a specific sector. This creates incentives to distort the definition of current and capital expenditures to meet defined percentages. The absence of a functional classification of expenditures and weaknesses in the subnational public expenditure management systems also constrain transparency in the fiscal accounts.
Significant overlaps in spending responsibilities create inefficiencies and reduce accountability. Health and education are characterized by extensive concurrency among the central government, regions, and municipalities. The central government sets the norms and the curricula, and pays for salaries of teachers and medical personnel. Municipalities are responsible for construction and maintenance of schools and health premises and for educational and medical equipment and supplies. Regional governments are responsible for the implementation of norms and standards set by the center. Selection and hiring of teachers and medical personnel are performed at the regional level, but the certification of hours worked occurs at the municipal level. Lack of coordination has led to spending inefficiencies; one of the consequences of this setup is that, while new construction generates new requests for personnel to run the premises and to provide the services, separation of responsibilities results in frequent discrepancies between the building of new premises and their staffing. More generally, investment spending decisions are not coordinated across levels of government.
In late 2007, the subnational revenue-sharing system was modified again to change the distribution of the hydrocarbon tax (IDH). IDH transfers to regional governments were cut by about 60 percent, based on the argument that the center needed funds to finance a new transfer program; this was also an opportunity for the central government to reassert some authority over the regional governments, especially those with autonomy aspirations. Half the amount clawed back was reallocated to the municipalities as increased transfers.
The recommendations from the most recent technical assistance mission in 2004 remain largely valid. They include the following:
- Clearly define spending responsibilities (avoiding duplication) and improving accountability. In general, one level of government should control both the size and quality of a spending item and its financing. For example, the dispersion of responsibilities for public education and health contributes to the relatively high education wage bill and the lack of quality control in these sectors.
- Simplify the transfer system (reduce overlapping programs) with well-defined policy objectives. The transfer system could be further improved by introducing a new equalization system that would take into account the revenue capacity and expenditure needs of subnational governments. Hydrocarbon taxation should accrue to the central government level—a goal toward which the constitutional assembly has not succeeded to date in making progress (regions that are resource-rich and dynamic would like to retain a large share of the natural resource revenues, while the central government would like to claw back the revenues to strengthen the finances of the treasury and for redistribution).
- Assign subnational governments significant ownsources of revenues (i.e., enabling them to set rates for local taxes and impose, at the margin, surcharges on national taxes) in line with redefined spending responsibilities. Appropriate sequencing is needed in giving subnational governments access to new own-revenue sources and transferring additional responsibilities. Only subnational governments that accept new responsibilities and perform them adequately should be given continued access to new tax resources.
- Strengthen the rules limiting subnational borrowing and muster a political commitment to a no-bailout policy. Steps that can be taken to improve fiscal rules to limit borrowing include (1) setting the debt stock limit as measured by present value, given the concessional nature of most of the debt, and including in it any guarantees given by subnational governments or related public enterprises; (2) making debt limits comprehensive by referring to all entities belonging to a department or municipality; and (3) calculating the debt-service-to-revenue ratio for the entire term of a new operation or the next 10 years (whichever is shorter). However, the political commitment to a no-bailout policy is the key element. The no-bailout commitment should include strict implementation of the legal framework for subnational debt control and the application of sanctions to violators. In particular, municipalities that are unable to generate timely and comprehensive reports on their debt and finances should be barred from borrowing.
According to the 1997 budget law, debt service was not to exceed 25 percent of revenues and debt stock was to be lower than 250 percent of revenues.
For example, focusing spending cuts at the central agencies level undermined the functioning of critical agencies and public services. In addition, the authorities adopted a financial transactions tax (excluded from revenue-sharing) after attempts to pass other measures met stiff resistance, also by regions and municipalities.
A new Direct Tax on Hydrocarbons (IDH) was introduced in 2005; although it has resulted in an increase in the overall government take from the hydrocarbon sector, close to 60 percent of IDH collections is transferred to subnational governments. In addition, the share of royalties going to subnational governments was increased from 45 percent to 55 percent. Regional governments have been the largest beneficiaries of these increases (universities have also benefited from additional earmarked revenues from the IDH).
In 2005, shared revenue represented about 90 percent of total net resources transferred from the central government to subnational governments, while grants represented the remaining 10 percent. The Regional Compensation Fund is funded by 10 percent of the hydrocarbons products excise tax, and its objective is to reduce inequities in revenues across regional governments. However, large disparities remain after the transfer of these resources, contributing to disparities in spending and poverty rates across regions.