Redistribution, inequality, and growth revisited

An influential paper by Berg et al., ‘Redistribution, inequality, and growth: new evidence’, uses the SWIID data to examine the impact of inequality and redistribution on growth in both developing and developed countries. It finds that while inequality is harmful for growth, redistribution does not hamper growth. This comment demonstrates that the redistribution and inequality data the paper uses are not credible. They are largely based on imputations, not actual observations. That is why one cannot really examine the research question for most of the countries in their sample, especially developing countries. Replicating their analysis on data not based on imputations—from the WIID dataset—changes some of the conclusions. In particular, inequality does not seem to lead to lower growth.