Monetary Policy and Inequality

At the Institute for New Economic Thinking’s highly interesting conference today on “Finance and Society,” Esther George, the President of the Federal Reserve Bank of Kansas City, and Claudia Buch, Deputy President of the German Bundesbank, both rejected the notion that monetary policy should aim to reduce inequality. Both forthrightly stated the mandates of their respective policy-making institutions: in George’s case, the Fed’s dual mandate of price stability and economic growth, and in Buch’s, the notorious European single mandate of “inflation close to but not above 2 percent.”

In flatly rejecting a role for monetary policy in dealing with economic inequality, the two seemed to be hinting at disagreement with Janet Yellen, who gave a widely-touted speech on inequality in the autumn in which she said it “greatly concerned” her. Earlier at the INET conference, Yellen spoke broadly about the role of financial regulation in both the economy and society. To be sure, Yellen has never advocated expressly that reducing inequality be adopted into the mandate of the Federal Reserve. But nonetheless, George and Buch indicated that any wider consideration of the impact of their policy-making beyond the mandates they’re each bound to follow amounts to stepping out of line and into the realm of politics or public debate.

There are two problems with George and Buch’s point of view: first of all, even if they don’t want to admit it, monetary policy mandates already do intimately affect the distribution of wealth and income. And second, there’s some evidence that rising inequality is itself an important contributing factor to credit cycles, which have been the cause of macroeconomic volatility, most notably in the 2008 Financial Crisis. So macroeconomic policy makers will have to come to grips with the empirical reality of the policy environment they confront, which intimately involves inequality

The central bank mandates of price stability and economic growth already step into the realm of the income and wealth distribution. Quite simply, inflation harms creditors, who tend to be rich, and benefits debtors, who tend to be poor, while unemployment disproportionately harms the poor. (Whether it benefits the rich is a far more contentious question for another day.) It may be convenient to proclaim that inequality is unrelated to monetary policy, but the fact is that the two are closely linked.And proceeding as though monetary policy can be made without reference to inequality only serves to exacerbate inequality and possibly undermine explicit central banking mandates.

I’ve created a few charts illustrating the relationship between different measures of the unemployment rate in the US and growth in labor income across the distribution of labor income. The data on the latter were created by my colleague Carter Price, who combined the Current Population Survey (which is top-coded, and hence is only accurate up to the 95^th-98^th percentile, depending on how income is calculated) with the World Top Incomes Database (assembled from tax data, which is very informative about the rich but pools the entire bottom 90% of the income distribution together).

Figure 1: The CDFs of US Labor Income illustrate both the rise in tail inequality and the increasing share of the working-age population that earns zero income from labor. Between 1990 and 2000, the distribution improved for everyone. Between 2000 and 2012, it got worse for everyone except the very top. At least by this measure, everyone in the bottom 70% was worse off in 2012 than in 1990.

Figures 2-4 plot the “predicted” percentage deviation in labor income at a given earnings percentile from its mean over 1990-2012 as a function of either the civilian unemployment rate or the U6 rate, which is a broader measure of unemployment and under-employment that takes workers forced into part-time work and marginally attached to the labor market into account.

The key finding is that the bottom half of the labor income distribution depends a great deal on low unemployment to secure income gains. Hence, prolonged, slack labor markets are bound to worsen inequality. And insofar as monetary policy-makers have it in their power to bring about full employment, or prevent it from occurring by prioritizing other policy objectives like price stability and seeking any excuse available to curtail monetary stimulus, they are choosing to increase inequality.

Figure 2: The predicted percentage deviation of labor income from its 1990-2012 mean, by percentile, as a function of the civilian unemployment rate. The prediction is the result of a regression of labor income by percentile on a polynomial in the unemployment rate.

Figure 3: This figure is the same concept as Figure 2, but with a different set of income percentiles reported.

 

Figure 4: The same pattern holds when U6 is used instead of the core unemployment rate.