The distance between America’s rich and the poor has grown in recent years—Berkeley economist Emmanuel Saez, for instance, finds that average real incomes for the top 1 percent of US earners grew by 80 percent from 1993 to 2014, compared to less than 11 percent growth for everyone else. US presidential hopefuls, eager to find solutions to a problem that frustrates many, hawk policies designed to shrink the gap—among them, raising taxes on the wealthy.
But under certain circumstances, hiking taxes might actually exacerbate inequality, according to research by two Chicago Booth economists. Lubos Pastor and Pietro Veronesi built a theoretical model suggesting that as tax rates rise from zero, income inequality and the level of stock prices initially rise with them before eventually falling. Their model also predicts that higher tax rates make the economy smaller but more productive.
Pastor and Veronesi describe a dynamic in which higher tax rates weed out would-be entrepreneurs with low skills and low tolerance for risk. The remaining highly competent, risk-comfortable entrepreneurs invest more heavily and skillfully, particularly in riskier, higher-return investments, according to the model. Their investments raise both the value of those assets and their own wealth, increasing income inequality until taxes are sufficiently high to lower it.
The researchers find that when taxes are high, only those with the highest skill and lowest risk aversion find it advantageous to become entrepreneurs. As a class, entrepreneurs become more skilled and less risk averse, and as a result, even richer than they would be if they faced lower tax rates, the study finds.
In practice, however, very few countries maintain tax rates low enough that raising them would mean an increase in inequality. To complement their model, Pastor and Veronesi analyzed tax rates between 1980 and 2013 in the 34 countries that make up the Organisation for Economic Co-operation and Development, measuring each country’s tax burden (ratio of total taxes to GDP), inequality level, and productivity, among other metrics. They find the actual relationships between tax rates, productivity, and inequality to be largely as their model predicts, but because they don’t observe tax rates below about 19 percent, in their data rising taxes always correspond with a drop in inequality.