Income and wealth inequality in the U.S. is substantial. How does the Federal Reserve affect such inequality when it pursues full employment and price stability through policy? In his paper, “Monetary Policy with Racial Inequality,” Makoto Nakajima of the Philadelphia Fed reports on differences in employment, income, and wealth by race and develops a model to examine how monetary policy affects various racial groups. Using this framework, he examines how monetary policy interacts with fiscal policy and the implications of using an alternative monetary policy target.

Nakajima examined monthly employment data from 1973 to 2021 by major racial group — Black, Hispanic, White, and Asian.1 The average unemployment rate during this period was 11.8 percent for Black workers, 8.8 percent for Hispanic workers, 5.5 percent for White workers, and 4.9 percent for Asian workers. The data further reveal that the unemployment rate for all groups of workers moved more or less in parallel, which means that cyclical changes in the economy affected all four racial groups similarly even though unemployment levels varied significantly by racial group.

The weekly earnings of Black and Hispanic workers were well below the median (19 and 26 percent below, respectively), while the earnings of White workers were 3 percent above the median, and the earnings of Asian workers were almost 19 percent above the median. Nakajima argues that education may explain Asian workers’ relatively high weekly earnings: Over half of Asian workers have a college degree, more than any other racial group, suggesting that, for Asian workers, median earnings include a large premium for attending college.

Nakajima also computed how many households in each racial group can be classified as “hand-to-mouth” — that is, consumers who exhaust all their readily available financial recourses in every pay period.2 He finds that almost half of Black and Hispanic households (47 and 49 percent, respectively) are hand-to-mouth. This compares to 25 and 28 percent of White and Asian households, respectively.

Among all households, mean total wealth (that is, the combination of liquid wealth, such as checking accounts, and illiquid wealth, like real estate) was $366,000.3 Comparing individual racial groups, the mean total wealth of Asian, Hispanic, and Black households was respectively about 82 percent, 23 percent, and 21 percent of the mean total wealth of White households.

Nakajima asks what is behind the higher unemployment rate among Black and Hispanic workers. To answer that question, he looks at the transition rates of potential workers into and out of three categories: employed, unemployed, and out of the labor force. He finds that both Black and Hispanic workers faced a higher rate of separating from a job and becoming unemployed than White workers. Furthermore, Black workers had a lower probability of finding a job. Nakajima also reports that Black and Hispanic workers had higher transition rates into and out of the labor force than White workers.4 This means that, among Hispanic and Black workers, employment was less stable and income was more volatile.

Next, he incorporated racial inequality in income and wealth into an existing model framework widely used at the Federal Reserve.5 By using this updated model, he could see how policy affects different racial groups. In addition, he used the model to study the interaction between accommodative monetary policy and fiscal policy, both of which provide buffers against economic shocks.

Nakajima finds that, although all racial groups benefit from economic stimulus induced by an accommodative monetary policy, such a policy lowers the unemployment rates of Hispanic and Black workers more than it lowers those of White or Asian workers. This is because the unemployment rates of Black and Hispanic workers are more sensitive to economic shocks, which also makes their unemployment rates more responsive to an accommodative monetary policy. Using this model, he was also able to replicate the results of a study by Bartscher et al. (2021) showing that the unemployment rate for Black workers declined 0.34 percentage point more than the unemployment rate for White workers following a 25-basis-point reduction in the policy interest rate.6

According to his model, an accommodative monetary policy benefits all racial groups, but through different channels. An accommodative monetary policy improved the welfare of Black and Hispanic workers more than it improved the welfare of workers from other racial groups, he finds, because workers from these two racial groups are more likely to be hand-to-mouth, and thus, changes in income affect their expenditures more directly. More specifically, a drop in the unemployment rate and an increase in wages associated with an accommodative policy led to a disproportionate rise in consumption for Black and Hispanic workers.7 On the other hand, Nakajima found that White workers gained more from the appreciation in asset prices associated with an accommodative monetary policy. In the model, these positive effects outweigh the costs of temporarily higher inflation induced by accommodative monetary policy, although his model does not currently capture the differential changes in inflation rates across racial groups.

Nakajima further shows that Black and Hispanic workers benefited more from an accommodative monetary policy when income-dependent fiscal transfers were reduced or absent.8 He explains why: When help from fiscal transfers is limited, a higher income, thanks to monetary accommodation, benefits Black and Hispanic workers more because their consumption is more sensitive to changes in income.

Finally, he finds that a monetary policy that targets the Black unemployment rate (which is more volatile) rather than the overall unemployment rate (which includes all races) is essentially equivalent to a monetary policy that more aggressively targets the overall unemployment rate.

  1. The views expressed here are solely those of the author and do not necessarily reflect the views of the Federal Reserve Bank of Philadelphia or the Federal Reserve System.
  2. The author used monthly data from the U.S. Census Bureau’s Current Population Survey. He follows the definitions of races used by the Census Bureau in terms of the three racial groups used in the empirical analysis — White, Black, and Asian. He excludes American Indians, Alaskan Natives, Native Hawaiians and Other Pacific Islanders, and Two or More Races from the analysis. These groups made up 2.1 percent of the labor force, on average, between 2003 and 2018. Hispanic is an identity, and a Hispanic person can be of any race. When the author computes numbers for Hispanic workers, he includes all individuals who identify as Hispanic, regardless of the race of the individuals. When he computes numbers for White, Black, and Asian workers, he excludes those who identify as Hispanic.
  3. For more details, see Greg Kaplan, Giovanni L. Violante, and Justin Wiedner, “The Wealthy Hand-to-Mouth,” Brookings Papers on Economic Activity, 45:1 (2014), pp. 77–153.
  4. Median total wealth was $90,000. Both figures are in 2010 dollars.
  5. No similar data were available for Asian workers.
  6. Specifically, Nakajima’s model is an extension of the New Keynesian dynamic stochastic general equilibrium (NK-DSGE) model.
  7. Alina K. Bartscher, Moritz Kuhn, Moritz Schularick, and Paul Wachtel, “Monetary Policy and Racial Inequality,” Federal Reserve Bank of New York Staff Report No. 959 (2021).
  8. Nakajima confirms the findings of Ganong et al. (2020): Consumption among Black workers was 50 percent more sensitive to economic shocks and monetary policy accommodations than consumption among White workers. See Peter Ganong, Damon Jones, Pascal Noel, et al., “Wealth, Race, and Consumption Smoothing of Typical Income Shocks,” Becker Friedman Institute Working Paper No. 49 (2020).
  9. Specifically, Nakajima considered the following income-dependent fiscal transfers: the Earned Income Tax Credit, various transfers to households with children, and income-dependent welfare programs.