- About us
- Research Review
- Latest Research
- Info Center
- Teaching Resources
This article is part of the Online Forum: The Redistributive Impact of Fiscal Policies
There has been much recent interest in falling income inequality in Latin America over the past decade. Scholars have been trying to understand such decline in a region historically characterised by high, persistent inequality (López Calva and Lustig, 2010). However, little has been said about the very top of the distribution. To the extent that the overwhelming majority of the literature uses households survey data, which underestimate income concentration, a reassessment is in order. In a recent paper (Alvaredo and Londoño Vélez, 2013), we study the shares of top incomes in Colombia between 1993 and 2010 using tax data.
The case of Colombia is worth studying on several grounds. The country has traditionally been identified as having one of the highest Gini coefficients in Latin America. It is the first country in Latin America to provide micro-data from the personal income tax for a relatively long period of time. Finally, Colombia has undergone key changes in the political arena since the 1990s; the 1991 Constitution established progressiveness as the foundation of the tax system, and as a result, all the subsequent (and numerous) tax reforms have been discursively presented as serving such principle.
The study obtains four main empirical results. First, income in Colombia is highly concentrated. The top 1% accounted for 20.5% of household income in 1993, and 20.4% in 2010. To put it bluntly, despite years of strong economic growth over 2003-2010, income is as concentrated in 2010 as back in the early 1990s.
Top income shares are at the highest level in any recent year in the World Top Incomes Database sample, except for the US, which has overtaken Colombia for several years in the late 1990s and the 2000s. Figure 1 contrasts the income share of the top 1% in Colombia with those of Argentina, Japan, Spain, Sweden, and the United States.
Figure 1. Top 1% income shares in Colombia, Argentina, Japan, Spain, Sweden and US, 1993-2011
Notes: Estimates for Japan, Spain, Sweden and US include capital gains.
Sources: The World Top Incomes Database and Alvaredo and Londoño Vélez (2013).
Second, high-income individuals in Colombia are, in essence, rentiers and capital owners. This feature differs from the pattern found in several developed countries in recent decades, where the large increase in top shares has been mainly due to spectacular increases in executive compensation and high salaries, and to a lesser extent to a partial restoration of capital incomes. While the working rich have joined capital owners at the top of the income hierarchy in the United States and other English-speaking countries, Colombia remains a more traditional society where the top income recipients are the owners of the capital stock.
Third, while households’ surveys show that inequality went down when 2007 and 2010 are compared, tax-based results offer a different picture, in which concentration at the top has remained stable over the same period. When survey based Gini coefficients are adjusted to take into account top incomes reported in tax files, inequality levels are higher than previously measured, and the recent reduction in inequality is less pronounced. The net-of-tax top 1% share is 20.1% in 2010, which can be compared with the figure from the household survey: 13.5%.
Fourth, personal taxation does little to reduce inequality. The income tax burden is very low at the upper end of the distribution, even if, in theory, the tax is structured around a progressive tax schedule comparable to OECD averages. In practice, the erosion of the tax base is due to a multiplicity of legal tax reliefs, even without considering the effects of evasion. To illustrate this point, Figure 2 compares taxable and non-taxable incomes for different sub-groups within the top 1% in 2010 (the situation is similar in the remaining years of our sample). Around 60% of reported incomes are treated as non-taxable; the fraction increases with rank, the ultra-rich having only one-tenth of their income considered taxable.
Figure 2. Taxable and non taxable income across top groups in Colombia, 2010
Notes: The figure assumes that 33% of income reported as "ingresos no constitutivos de renta" come from taxed dividends.
Source: Alvaredo and Londoño Vélez (2013).
Given these large tax reliefs, how much does the top actually pay? Figure 3 presents the average rates of income tax and social security contributions for different fractiles within the top 1% in 2010. The income tax paid is shown separately for regular and irregular income, and social security contributions are shown separately for employees and self-employees. The concavity illustrates the lack of progressivity in the Colombian tax system among those who pay the tax. The bottom half of the top percentile pays roughly 12%, while this percentage falls to 8 for the top 0.01%.
Figure 3. Income tax and social security contributions at the top in Colombia, 2010
Notes: SSC stands for social security contributions. It is assumed that 33% of 'ingresos no constitutivos de renta' come from taxed dividends. If dividends were 75% instead of 33%, the average tax rate for the top 0.01% would be 14% instead of 8%, resulting in an almost flat average tax rate for all individuals in the top 1% group. Irregular incomes are donations, long-term capital gains, and inheritances.
Source: Alvaredo and Londoño Vélez (2013).
These results are not a novelty from the qualitative point of view, in the light of the well-known high inequality levels and distortive tax systems in Latin America. However, they challenge the general scepticism regarding the use of tax data from developing countries to study inequality. Our estimates should be regarded as a lower bound, to take into account the effects of evasion and under reporting. Nevertheless, they show that incomes reported to tax authorities can be a valuable source of information, under certain conditions that require a case-by-case analysis. In Colombia, the average income tax rate effectively paid by the top 1% is so modest by OECD standards that the incentives to hide income could be much more limited than previously thought. The supportive evidence is given by the estimated levels of top shares. Our results also indicate that when high incomes are properly taken into account, optimism about declining inequality in Latin America should be somewhat dampened.
The most recent tax reform was passed in December 2012, supposedly with the aim of increasing progressivity, and thus respecting the principles expressed in the constitution. However, changes have been extremely moderate. Complexity and administrative costs have increased even more, benefiting those who can afford financial and accountancy services. Given the observed differences between statutory and effective tax rates, it will be necessary to conduct an evaluation of the reform as soon as data are made available for income year 2012.
Alvaredo, F., Atkinson, A. B., Piketty, T. and Saez, E. The World Top Incomes Database. http://topincomes.parisschoolofeconomics.eu.
Alvaredo, F. and Londoño Vélez, J. (2013). High Incomes and Personal Taxation in a Developing Economy: Colombia 1993-2010. CEQ Working Paper 12.
López Calva, L. F. and Lustig, N (2010). Declining inequality in Latin America: A decade of progress?. Brookings/UNDP.