Inequality and Growth: From Micro Theory to Macro Empirics

To establish the nature of the link between income distribution and economic growth by means of a standard growth regression, one needs to collapse an entire income distribution into a scalar measure of inequality. Due to data shortages macro-economic research has typically been forced to use the gini coefficient for this purpose. Using a simulation set up we check how well different measures of inequality or poverty succeed in detecting the correct relationship. We find that the gini coefficient might not be the worst of choices, but the comparison of the explanatory power of different inequality measures can help to identify the theoretical mechanism through which inequality affects growth.