Inequality: a different perspective

We document novel empirical relationships: in a cross-section of advanced economies, the ratio of the current account balance to GDP and inequality are negatively correlated. The negative relationship between savings and inequality is the main driver of this correlation. We demonstrate that low private savings in countries with high inequality lie at the heart of this result.
We show that a model with endogenously incomplete domestic markets can rationalize these findings. In countries with high income inequality higher levels of debt can be sustained, since the value of autarky is lower. The opposite is true in countries with low income inequality. Consequently, the level of income inequality in a country is positively correlated with the supply of assets and the interest rate in that country. When international financial markets are open, to clear the world financial market at a common interest rate, capital flows from (previously) low-interest rate countries to high-interest rate ones. Equivalently, capital flows from equal to unequal countries.

This paper presents evidence of a new propagation mechanism for wealth inequality, based on differential responses, by education, to greater inequality at the start of economic life. It is motivated by a novel positive cross-country relationship between wealth inequality and perceptions of opportunity and fairness, which holds only for the more educated. Using unique administrative micro data and a quasi-field experiment of exogenous allocation of households, the paper finds that exposure to a greater top 10% wealth share at the start of economic life in the country leads only the more educated placed in locations with above-median wealth mobility to attain higher wealth levels and position in the cohort-specific wealth distribution later on. Underlying this effect is greater participation in risky financial and real assets and in self-employment, with no evidence for a labor income, unemployment risk, or human capital investment channel. This differential response is robust to controlling for initial exposure to fixed or other time-varying local features, including income inequality, and consistent with self-fulfilling responses of the more educated to perceived opportunities, without evidence of imitation or learning from those at the top.