Decreasing inequality under Latin America's "social democratic" and "populist" governments: is the difference real?

The article addresses the claim that the "left populist" governments of Argentina, Bolivia, Ecuador, and Venezuela failed to effectively reduce inequality in the 2000s. The author examines the econometric evidence presented by McLeod and Lustig (2011) that the "social democratic" governments of Brazil, Chile, and Uruguay were more successful and shows that McLeod and Lustig's results are highly sensitive to their use of data from the Socioeconomic Database for Latin America and the Caribbean (SEDLAC). Conducting the same analysis using inequality data from the Economic Commission for Latin America and the Caribbean (ECLAC) suggests the exactly opposite conclusion. The contrast between the results obtained using SEDLAC and ECLAC data suggests that the choice of inequality data source is not immaterial and that the difference is probably driven by how the two sources handle the underreporting of income in household surveys. The key difference between SEDLAC and ECLAC data is that the latter correct for the underreporting of income while the former do not. Absent reasonable criteria for choosing between the two datasets, the author suggests that any econometric results pertaining to Latin American income inequality should prove robust to both data sources.