Opinion – Pablo Acosta: How to give more balance and cohesion to federative relations?

There is certainly a good case for fundamentally rethinking fiscal federalism in Brazil. In international comparison, Brazil has a high degree of revenue decentralization. But simulations show that transfers are not effective in reducing fiscal inequality between federal entities, with middle-income states receiving more resources from intergovernmental transfers than the poorest ones.

A reform of the main transfers from the Union to subnational entities has the potential to simplify and make transfers more equitable and, at the same time, bring balance and cohesion to federal relations in Brazil.

For example, the two largest transfers from the Union to subnational entities are the State Participation Fund (FPE), which in 2021 totaled R$ 125 billion, and the Municipal Participation Fund (FPM), which totaled R$ 146 billion. Together, these funds represented around 4% of GDP in 2021. In the same year, FPE transfers represented an average of 20% of the states’ gross revenues, but in some cases, such as Roraima, Acre and Amapá, this percentage reached 50%, revealing the fundamental role of this fund in the fiscal viability of these entities.

Based on the best international practices, and considering the vast existing literature on intergovernmental transfers, the recent study by the World Bank brings several simulations to enable a reform of the FPE. In Brazil, as in other countries, many transfer instruments, including the FPE, combine and confuse objectives of decentralization and distribution or equity.

Thus, in order to rationalize and provide more transparency to the system, the study proposes a reform of the FPE that considers three separate funds for each objective to be pursued: fiscal equalization, revenue sharing and performance incentives.

Revenue Sharing Fund

Most OECD countries and many developing countries implement some form of revenue sharing, which is the main instrument for promoting decentralization. Countries that do this – by allowing states to keep some of the resources where they are generated – are intended to provide incentives to develop state economies and provide some sense of balance and national cohesion.

In the study, a revenue sharing fund was simulated with approximately 10% of the total funds available for PEF. The criteria for sharing the fund’s resources take into account the percentage of the state’s population and its GDP, with different weights allocated to each component. Naturally, the definition of which share of resources would go to each of the instruments (objectives) is, to a large extent, a political decision.

Tax Equalizer Fund

Resources from an equalizing fund aim to close or reduce the fiscal gap of subnational entities, which is calculated as the difference between spending needs and fiscal capacity.

The calculation of spending needs took into account the main drivers of spending, which are generally related to the number of potential “clients” that use public services in the state. The calculation of tax capacity took into account the potential per capita tax revenue of the entities. The use of potential revenues instead of actual revenues prevents entities from adopting a strategic behavior. In other words, the objective is to prevent entities from exerting less effort in tax collection than they otherwise would.

Well-designed equalizing transfers automatically recognize, in their formulas, revenue-sharing funds as part of the fiscal capacity of states, thereby reducing or eliminating the size of equalizing transfers they receive. Therefore, in the simulations carried out, the calculation of fiscal capacity took into account revenues from the revenue sharing fund.

Performance Incentive Fund

Brazil has good examples of performance-based transfers being implemented within the intergovernmental transfer system. Both in the environment (ICMS-ecological) and in education (ICMS-education), mechanisms were implemented to encourage local governments to improve performance. Studies show, however, that performance-based transfers are more likely to succeed when they are part of a larger package that also includes technical assistance and, in some cases, inputs.

The Possible Reform

Results of the various simulations discussed show that the new distribution of resources, considering the three objectives (revenue sharing, fiscal equalization and performance incentives) would generate a more equalizing pattern in the allocation of transfers, with the poorest states receiving more resources per capita than those richer. In addition, by incorporating incentives into its design —indirectly, in the case of the fiscal gap, or directly, in the case of the performance incentive fund—, the proposal brings more balance and cohesion to federative relations in Brazil.

Finally, the simulations developed are not intended to be a final recommendation. All actual calculations and estimates can be further refined with better data as well as contributions from national experts. The calibration of the parameters used in the model (such as the size of each fund and the weight of technical criteria) ultimately depend on an agreement between the Federal Government and the states. The study offers tools to make this agreement possible.

_{This column was written in collaboration with my World Bank colleagues Shireen Mahdi, World Bank Lead Economist for Brazil, Kjetil Hansen, Senior Public Sector Specialist, and consultants Claudia Tufani and Jorge Martinez-Vasquez}