Mexico’s constitution provides for a federal system characterized by a complex network of intergovernmental fiscal relations. The federation includes the Federal District, 31 states, and 2,392 municipal governments.
During the 1990s, intergovernmental fiscal relations in Mexico were reorganized, with a major shift toward decentralization. This reorganization occurred against the backdrop of multiparty democracy taking root at all levels of government, and recovery from the 1994–95 financial crisis. However, the process is incomplete and has evolved little since then. Key challenges are to improve the efficiency of spending at the local level, notably through greater accountability and a clarification of responsibilities. Thus, while the political power of the state and local governments has increased, so has dissatisfaction with the effectiveness of decentralized public service delivery. At the same time, social programs at the federal level, particularly those targeted to poor and marginalized populations, have been expanded. These programs, such as Progresa-Oportunidades, have been highly effective in their poverty-reduction impact and are seen as examples of international best practice (Coady and Parker, 2002).
In Mexico, as in other Latin American countries, decentralization of key education and health care functions has been only partial. Although substantial spending responsibilities have been transferred, the federal government retains extensive regulatory and financing powers. This reflects, in part, the influence of national unions, plus the federal government’s concern that these critical functions may not be carried out effectively by subnational administrations. The response to partial decentralization has been to provide targeted transfers for education and health care, resulting in little incentive for subnationals to be fully accountable for the provision of these services, given these overlapping responsibilities.
The issues of education and health outcomes in the context of Mexico’s decentralization have been extensively studied. For example, the World Bank (2006) and the Organization for Economic Cooperation and Development (OECD, 2005) found that the decentralization process in Mexico remains incomplete. Despite a major increase in social program coverage, performance indicators in health and education remain well below OECD averages, and even Latin American levels.1 Furthermore, wide disparities are identified across states. The main challenges are, therefore, to improve the quality of service performance and correct inequalities at the national level.
Mexico: Indicators of Fiscal Decentralization
Mexico: Indicators of Fiscal Decentralization
| Type of government | Federal | |
| Population (million, 2007) | 103.1 | |
| Area (million km2) | 1.9 | |
| Levels of government | 3 | |
| States | 32 | |
| Municipalities | 2,454 | |
| Average municipality size (population) | 42,009 | |
| Minimum municipality size (population) | 102 | |
| IMF technical assistance mission on | 2007 | |
| fiscal decentralization | ||
Mexico: Indicators of Fiscal Decentralization
| Type of government | Federal | |
| Population (million, 2007) | 103.1 | |
| Area (million km2) | 1.9 | |
| Levels of government | 3 | |
| States | 32 | |
| Municipalities | 2,454 | |
| Average municipality size (population) | 42,009 | |
| Minimum municipality size (population) | 102 | |
| IMF technical assistance mission on | 2007 | |
| fiscal decentralization | ||
From a macroeconomic stability standpoint, subnational governments’ operations are not seen as an immediate concern. Following the resolution of the subnational debt crisis of the 1990s, debt contracted by subnational governments in Mexico—states and municipalities, including the decentralized entities, enterprises, and trust funds of these levels of government—has been relatively stable at low levels (1.6 percent of GDP in 2007, as estimated by the government) and accounts for a small fraction of the gross total public sector debt (38.3 percent of GDP in 2007). Strengthened banking regulations and supervision played a significant role in helping the country to emerge from the crisis and maintain stability. Furthermore, subnational debt is not subject to currency risk given that, as established in the legislation, it is entirely denominated in Mexican pesos.2 Still, some areas of concern, and room for further reform and progress, remain.
The Subnational Debt Crises of the Mid-1990s
The risk to federal government finances posed by subnational governments was brought to the fore in the 1994–95 financial crises. As a result of increases in interest rates, subnational government debt expanded significantly, by 50 percent in real terms in 1994 and another 10 percent in 1995, reaching the equivalent of 2.2 percent of GDP.3 Although still comparing favorably with debt stock as a percentage of GDP in other Latin American countries experiencing debt crises, Mexican states were not able to service their debt. This inability was due to the states’ heavy reliance on federal government transfers (participaciones) and their limited capacity to raise additional revenue more than to the size of the outstanding debt stock or short maturities.4 In 1995, the ratio of subnational debt to participaciones stood at 80 percent, significantly higher than 44 percent and 64 percent in 1993 and 1994, respectively.
After the 1995 crisis, the federal government took over the debt of the states. The Fund for Strengthening State Finances, developed as a federal government budgetary item, provided for extraordinary cash transfers, with a cost of Mex$7 billion in 1995; it continued at that level in real terms until 1998. States were required to restructure their debt in Udis, a new unit of account indexed to inflation. In return for the debt takeovers, the states were required to agree on a fiscal adjustment program with the Ministry of Finance. States were required to commit themselves to balance their budgets, reduce debt ratios, present their financial accounts in a uniform way, and update and publish state debt laws to regulate and limit debt.
Following the federal bailout, subnational governments’ debt stock across all states declined, with the exception of the State of Mexico and the Federal District. The total stock of subnational debt was reduced to 1.8 percent of GDP in 1997, representing 3.4 percent of total public debt and 62 percent of shared revenues.
Since the 1995 crisis, considerable progress has been made to move toward a market-based system to ensure fiscal discipline. Previously, subnational governments could run deficits or issue debt. A considerable change in the attitude to borrowing followed the subnational debt crisis of the 1990s, and banking regulations and provisioning requirements were strengthened. Additional requirements for credit ratings have also been a positive development.
Current Intergovernmental Fiscal Arrangements
The expenditure responsibilities of each government tier are not clearly defined. As indicated above, health and education are the two main areas of overlapping responsibilities. Subnational spending is dominated by personnel expenses, while spending on goods and services (including investment) is limited. The system of wage negotiations, with feedback effects between federal- and state-level negotiations, contributes to this problem. Overall, because subnational governments account for a high share of public spending, but have insufficient resources (especially own-taxes at the margin), significant vertical fiscal imbalances persist.
Subnational governments have limited own-tax handles. States have been given hard-to-tax bases, including the small taxpayers who yield little revenue while generating considerable political and administrative costs. States also have the option of piggybacking on the income tax, but this measure has only been adopted by three states because of the concern that, by doing so, they will lose competitiveness. For municipalities, the critical tax is the property tax. It raises remarkably little revenue (0.3 percent of GDP over the period 2000–04, compared with the OECD average of 1.9 percent of GDP). Low property tax collections are not attributable to low tax rates, but rather to administration (e.g., outdated cadastre and valuations of property, and low collection efforts).5 States keep the cadastre and all cadastral functions but have little incentive to update them because collections accrue to the municipalities. State legislatures also approve changes to the property tax rates.
States and municipalities are heavily dependent on federal transfers. Thus, the burden of raising taxes and revenues falls exclusively on the federal government and reduces accountability and incentives to efficiency for subnational governments. About 85 percent of states’ total revenues comes from federal transfers. The corresponding share for municipalities is 65 percent. Overall, federal transfers represent 8.1 percent of GDP.
The present transfer system is complex and its formulas are obsolete. The system is based on revenue-sharing (participaciones) and a myriad of additional transfers and earmarked grants (aportaciones). The overall distribution of the transfers is not particularly equalizing because it is based on initial conditions that do not reflect recent trends in needs or demand for public services, or fiscal capacities.
The pool of participaciones is composed of the main federal taxes, such as personal and corporate income taxes, the tax on assets, the value-added tax, and excises. The pool also includes revenue from oil and mines. A share of 21.06 percent of this pool is transferred to the states according to a complex formula.6
Earmarked funds are transferred under an increasing number of programs, the biggest of which was instituted in 1998 under the name of aportaciones. Federal earmarked funds represent the largest source of subnational revenues (57.7 percent of the total in 2006). The most important transfer (covering more than 60 percent of total aportaciones) is earmarked for education. The health sector also benefits from a basket of earmarked funds. As with education, the amount of transfers for health is determined according to wage and other expenditures in the health sector and is adjusted yearly according to increases in such expenditures.
Lack of information undermines monitoring of subnational government operations. The absence of budget and accounting standards across government levels limits monitoring by the federal government and the elaboration of information on general government finances. States also differ in their government financial management information systems (GFMISs). Seven states have already purchased the most modern GFMISs, which ensure full integration between budget appropriations and accounting, but other states use outdated information systems. Proliferation of bank accounts at the subnational level also hinders effective cash management. Implementation of treasury single accounts (TSAs) at the subnational level, which could address some of these problems, is hampered by federal requirements mandating the maintenance of separate bank accounts for earmarked transfers.
IMF Advice on Fiscal Decentralization
A staff mission in 2007 provided technical assistance regarding intergovernmental fiscal relations.7 The mission stressed that states should be given more-significant tax handles to promote greater accountability at the subnational level and to encourage the states to play a more active role in revenue generation. The mission analyzed different alternatives. One alternative was for the states to piggyback on the single-rate corporate tax (called IETU)8—a new tax that was approved in September 2007 and implemented in 2008—because of the potential revenue generation, its likely management by the federal tax administration agency, and the fact that it would represent a major tax handle for the states. If allocated on the basis of income taxes (ISR, Impuesto sobre la Renta), it would provide more revenue to the richer states, but if distributed according to simple apportionment procedures (for example, based on production or employment) it could reduce the “disequalization” from the piggyback. Other alternatives would be to put a floor on the income tax piggyback to eliminate concerns over tax competition; to consolidate all taxation of motor vehicles at the state level; and to introduce additional state excises on items permitted under the constitution. In contrast, an additional retail sales tax might not be viable, given the states’ limited administrative capacity. Furthermore, introducing a piggyback on the value-added tax would replicate existing problems with unequal bases.
Recommendations to overhaul the property tax were also discussed. Measures include strengthening the cadastre, improving the valuation procedures with the support of the states (with federal assistance) for the municipalities, and establishing control over rates at the margin for the municipalities (without the need to refer back to state legislatures—which could enact a band within which rates might be set).
The system of transfers should follow more-transparent and equalizing criteria. In particular, the mission advised improving the distribution of the aportaciones for health care and education by adapting the formulas to reflect actual public service needs, and increasing quality incentives. The distribution of participaciones (excluding those going to municipalities) could follow equalization criteria, such as population and inverse of income per capita, or spending needs and fiscal capacities. To achieve a political consensus, the mission proposed the use of a “hold-harmless clause,” which would convert the existing amount of the transfer pool into a lump-sum transfer fixed in nominal terms, based on data from the year preceding the reforms, to ensure that no entity would be made worse off by the reform.
Clarification of the responsibilities and increasing subnational accountability were also recommended. Functional spending responsibilities for lower levels of government should be clarified, while allowing full control over the choice of economic inputs (wages and salaries, operations and maintenance, and so on).
A number of measures were suggested with regard to subnational debt: establishing prudential limits for borrowing, linked to fiscal rules and targets; evaluating the budgetary and debt treatment of new instruments, including securitization; carefully assessing the fiscal risks associated with public-private partnerships and developing a sound legal framework for such operations; and improving the oversight of credit rating assessments.
A package of measures for public financial management would help improve transparency and good governance based on the following:
Introducing a common budget classification and accounting framework consistent with international standards, at all levels of government. Subnational governments should follow specific reporting standards, and a common database for their fiscal operations should be established. Sanctions for subnational entities not complying with the reporting standards should also be introduced. Standardized information would enable the federal government to monitor and consolidate subnational governments’ operations to produce accounts for the general government.
Establishing sound GFMISs to manage effectively public finances at all levels of government. While GFMISs may differ across states, there should be an agreement on minimum common standards and the design of interfaces for an effective flow of information.
Implementing TSAs at all levels of government would allow borrowing requirements to be minimized and cash to be managed effectively. Legal provisions requiring separate accounts for earmarked transfers should be reviewed. The federal government could explore the possibility of providing technical services on the implementation of a TSA to subnational levels on an agency basis.
Recent Progress and Remaining Challenges
A reform that modified the system of intergovernmental fiscal relations was approved in September 2007, making the allocation of transfers more transparent and redistributive. It also froze, in nominal terms, the existing transfers at prereform levels and applied the new formulas to any increase in federal revenues. Earmarked transfers for education and health were adjusted to reflect demand and state cofinancing to create incentives to strengthen states’ basic systems. The rest of the transfers were modified according to an equalization criterion, economic activity, and tax collection efforts.
However, the reform failed to give any significant tax handle to the states, instead assigning them additional transfers. The reform gave 30 percent of the revenue gains that would result from the two federal taxes (the single-rate corporate tax and a levy on cash deposits) that were created jointly with the intergovernmental reform to subnational governments. In addition, the reform also allocated to these governments all the collections from other newly created federal taxes, in particular, from the surcharge on diesel and gasoline, and from excises on gambling and lotteries.
The September 2007 reform also included measures to address subnational government public financial management weaknesses. The reform called for a constitutional amendment that would empower congress to issue a law establishing a standard accounting framework across all levels of government, and enable the Superior Audit of the Federation to audit the aportaciones. The constitutional amendment was approved in May 2008, and could pave the way for establishing a standard accounting framework across all levels of government, and introducing legal provisions requiring subnational governments to provide the federal government with timely information on their fiscal operations.
Although the 2007 reform marks significant progress in improving the system of intergovernmental fiscal relations, additional measures would be desirable. Increasing the own-taxing powers of subnational governments remains crucial to ensuring accountability and providing incentives for subnational governments to use these powers. Debt management could be improved by ensuring that all debt—as well as commitments associated with public-private partnerships and guarantees—is registered with the federal government, and by limiting the use of federal government transfers as a guarantee for subnational government borrowing. Finally, adopting a common budget classification and reporting standards, as well as developing modern GFMISs to interface with a new federal system, would allow monitoring of subnational finances and assessing the general government fiscal position.
Decentralized Service Delivery for the Poor (World Bank, 2006) and “Getting the Most Out of Public Sector Decentralization in Mexico” (OECD, 2005) are two examples of comprehensive studies of subnational spending assignments and decentralization in Mexico.
In 2006, subnational debt grew in real terms (5 percent growth in real terms and 0.7 percent above the national GDP growth at constant prices) while public sector debt declined. In 2007, both federal and local government debt declined as a percentage of GDP.
The one-month CETES rate rose from 14 percent in November 1994 to 75 percent in April 1995. A few states (Campeche, Coahuila, Chiapas, and Nuevo León) experienced debt increases exceeding 150 percent during 1993–95. Out of the 32 states, five (Federal District, State of Mexico, Nuevo León, Jalisco, and Sonora) were responsible for nearly 75 percent of the stock of debt.
Average maturity of subnational debt stood at 6.6 years by end-1994.
On average, property tax rates range around 2 percent of the value of property, close to prevailing property tax rates around the world.
The formula for participaciones is based on three components: 45.7 percent of the pool is distributed according to population; 45.7 percent is allocated on a historical basis—the previous year’s allocation—corrected by a tax effort indicator; and the remaining 9.6 percent is allocated according to the inverse of the per capita allocations of the previous components.
The mission included participation by World Bank staff.
Impuesto Empresarial a Tasa Unica.