This paper (available for download below) is part of a series of publications from an event the Latin American Program hosted in December 11, 2012.

Executive Summary

Brazil differs from other Latin American and emerging-market countries in that its tax burden is much higher, 36 percent of GDP (2012), the result of a rapid increase in the post-WW2 period.  Instead of trying to increase the level of taxation, its main challenge is to reduce taxation or, more feasibly, to make the tax system more efficient.  Yet despite the fact that Brazil has one of the most unequal distributions of income in the world, equity has not been a major issue in discussions of taxation in Brazil, even among academics.  Instead, attention has centered on spending and how to make it more progressive, with taxes relegated to the role of collecting more revenue.  Only recently, as equity has arisen as an issue in tax reforms in the region, and as international observers have returned to the subject, do we find a few studies of tax incidence in Brazil.  This work intends to sketch a brief update on tax equity (and the lack of it) in Brazil, in the hopes of awakening interest in the subject.    

Since the promulgation of the 1988 Constitution, much has been discussed about tax reform but little has been done.  While many other relevant changes took place, including privatizations, market-oriented reforms, and, at the turn of the century, the Law of Fiscal Responsibility, equity in taxation did not become an important issue either for governments or for organized civil society.  The reigning idea has been that spending was most important to progressivity, and taxes just needed to be effective at obtaining revenue.   But this assumption has to confront the fact that Brazil has now reached OECD levels of tax burden and social spending--along with an income distribution almost as unequal as Bolivia’s.  It is well known that spending is not progressive enough.  It is less well known, and rarely debated, that taxation in Brazil is regressive as well. 

Like the rest of Latin America, Brazil differs from advanced industrial countries most of all in its low level of income tax revenues, especially those from the personal income tax.  Wealth taxes are also very low in Brazil.  Brazil’s high overall tax burden is mainly due to its very high consumption taxes, above all the state-level ICMS (the object of more failed reforms than any other tax), which brings in around 7.1 percent of GDP.  At the federal level, social contribution revenues are larger than tax revenues.  Debt service also consumes a great deal of tax revenue.  And unlike OECD countries and most of its neighbors, many Brazilian taxes are poorly designed, “cascading” in their effect on income or production costs.

Apart from the fact that studies of Brazilian tax incidence are rare and relatively recent, measurement of incidence is hampered by the great complications of the system, above all because of differences among state-level taxes on goods and services.  In addition, governments release data sporadically and late—despite having an impressive system for automated processing, in the case of the personal income tax.  But we do have enough details from a few incidence studies to say that most tax increases between 1993 and 2004 were in indirect taxation and that these hit poorer households harder than rich ones.  (One study estimates that in 2004 the poorest decile paid 48.9 percent of household income in taxes, while the richest paid 26.3 percent.)  We also have good reason to suppose that the changes in taxation since the mid-2000’s have made things more unequal, increasing the relative burden on the middle class the most, on the poor somewhat less, and on the rich least of all.  Meanwhile, other work shows that in the fiscal system as a whole, the progressive effect of transfer programs is outweighed by the regressivity of social security and taxation.

What can be done?  Wealth taxes are one possibility, although experience from other countries gives reason for pessimism about their revenue potential.  In Brazil, the vehicle and urban real estate taxes together collect less than 1.5 percent of GDP; it would be politically difficult to raise this figure by much.  Income taxes are another challenge.  The efficacy of replacing corporate tax payments with revenues from personal income taxes would be reduced by the fact that many corporations are in fact people, taking on corporate form in order to avoid personal income taxation (which in Brazil has higher rates).  Moreover, raising rates at the top of the personal income tax table would probably have little effect:  although the government has not released good data, it is likely that this would hit a narrow slice of highly paid employees of the state and large private firms, and not the big fortunes with income from capital.  Meanwhile, the rise in formal employment seen in the last decade had added social security contributors mostly from the poorer households, so that this system is becoming even less progressive.  And to put all this in perspective:  if we take only the interest payments on Federal debt paid to households, this amounted to about 5 percent of GDP in 2006, or six times the amount spent on all social assistance (including Bolsa Familia), which was 0.8 percent.  Clearly, the problem is larger than one tax or one spending program. 

To begin, Brazil needs a more informed debate.  Too many Brazilians believe that those too poor to pay income taxes pay no taxes at all, when the poorest really pay a higher proportion than the rich.  Not enough identify the state-level consumption taxes as a key problem for equity.  The government should produce better national household surveys, not only of income but of quality of life and usage of public services.  Leaders have to get past the comfortable assumption—held by the current government even when its support in Congress was broader and its approval ratings higher—that taxation is a matter of good management, to be solved by a few adjustments here, a few tax exemptions there.  In short, the challenge now is to improve not the quantity but the quality of Brazilian taxation, as part of a thoroughgoing reform of fiscal policy.          

The Latin American Program thanks the Tinker Foundation for their support in making this project possible.