The State and inequality in Brazil

Politics and Economy
Poverty - Inequality - Aid Effectiveness


We have recently finished a set of studies on the role the State has on inequality in Brazil. Using survey data we estimate that the State is directly related to (contributes, technically speaking) one third of all income inequality in the country, even after accounting for the equalizing effects of taxes and other direct tributes such as contributions to the public pensions system.

The study focuses on direct transfers to and from the state, as pensions, social assistance, salaries and some tributes. If indirect transfers from the State, as subsidies and interests paid to public bonds, were accounted for, the contribution to inequality would probably be higher. We are not making an apology for a minimal state but rather arguing that in order to reduce inequalities the State must be both strong and egalitarian in its actions. A State that is only strong – in the sense of being stable and commanding a large amount of resources – can actually increase inequality.

The main objective of the studies was to measure to contribution of the State to income inequality in Brazil, and the dynamics of this contribution. For this we applied methods of static and dynamic decomposition of the Gini coefficient to survey data. With these methods they isolate the contributions of different sources of incomes, such as pensions, social assistance and wages, and of taxes. We also used Juhn, Murphy and Pearce decompositions to calculate the effect of public-private wage differentials, which have been increasing in favor of the public sector workers during the last decade.


  • Despite the decline in inequality since the second half of the 1990s, Brazil is still an extremely unequal country. The State contributes to about one third of this inequality. If the equalizing effect of taxation is not accounted for, the contribution of the State comes closer to half of total inequality.
  • Pensions contribute with around one fifth of total post-tax inequality.
  • Public sector salaries contribute with one quarter of total post-tax inequality; these workers earn better salaries because they are more better educated than the rest of the labor force but also because there is a wage differential that favors them, even when compared to equivalent workers. This differential, alone, contributes with 6% of the Gini coefficient, several times more than the progressive effect of social assistance programs.
  • Direct taxes are highly progressive, but the total direct tax load is low: reduces only one tenth of total inequality.
  • Most of pensions are public, but there are two separate public systems, one for workers in the private sector and another for those of the public sector. Public pensions for public sector workers are related to 9% of the Gini but benefit only 4% of the population; 28% of the population lives in households benefited by public pensions for private sector workers, but these pensions contribute to only 12% of inequality. Much of the difference is caused by different rules in each system that impose a ceiling for pensions for workers in the private sector, but not for those in the public sector. There are reforms to address this issue but they will not substantially affect inequality in the next decade.
  • Social assistance programs are effective to reduce poverty, but have little relevance for inequality. Their contribution is of 1% of the Gini coefficient; roughly speaking, if these programs were shut down, inequality would not increase much more than 1% in Brazil.

The project resulted in four studies, which are being published both in English and Portuguese at

Share this