Monetary policy and racial inequality
Alina Kristin Bartscher, Moritz Kuhn, Moritz Schularick, Paul Wachtel 10 February 2021
The gap between the income and wealth of black and white households in the US is large and persistent. According to the 2019 Survey of Consumer Finances (SCF), the median wealth of a white household is almost nine times higher than that of a black household. The income gap is smaller (1.7 times) but still large. Moreover, these gaps are as large as they were 50 years ago (Kuhn et al. 2020). Concern about racial inequality has increased recently with evidence that the COVID-19 pandemic is having a disproportionate effect on the black community (Bertocchi and Dimico 2020). These stark facts have attracted the attention of economists (e.g. Mayhew and Wills 2020, Chetty et al. 2018) and policymakers. For instance, Raphael Bostic, president of the Federal Reserve Bank of Atlanta, recently stated that the Federal Reserve “can play an important role in helping to reduce racial inequities and bring about a more inclusive economy”.1
A prominent line of thinking is that an accommodative monetary policy lowers unemployment rates and increases labour income for marginal workers, who are oftentimes low-income and minority households. This is what Coibion et al. (2014) call the earnings channel of monetary policy. More specifically, Carpenter and Rodgers (2004) show that a monetary policy accommodation reduces the gap between the unemployment rates of black and white households.
Yet at the same time, monetary policy has portfolio effects through its impact on asset prices. Asset price changes should affect the wealth distribution if portfolios differ systematically between black and white households, as is the case (Figure 1). Only a third of black households hold equity and less than half own a home. As a result, monetary policy that increases asset prices potentially has different effects on the portfolios of black and white households.
Figure 1 Portfolio composition (per cent of total for white and black households)
Much of the existing literature on the distributional consequences of monetary policy focuses on the income distribution, while the wealth distribution has largely been ignored because it was previously assumed that policy effects on asset prices were temporary. However, recent work suggests that these effects are persistent (Bernanke and Kuttner 2005, Paul 2020, Bernanke 2020, Cieslak and Vissing-Jorgensen 2020). The relative magnitudes of earnings and portfolio effects on the racial wealth and income gaps have not been explored in the literature.
Our new research (Bartscher et al. 2021) addresses this issue. We systematically examine the impact of a monetary policy expansion on the wealth of black and white households through the portfolio channel and on the gap between black and white income through the earnings channel. First, we estimate the impact of a policy shock on asset prices and the unemployment gap. Next, we use these estimates to examine the changes in wealth and income of the typical black and white household using data from the 2019 SCF to show how monetary policy affects the gaps. We find that although an expansionary monetary policy reduces the gap between black and white unemployment rates, it has only a small effect on the gap between average black and white earnings. At the same time, the policy expansion increases asset prices, resulting in capital gains which are orders of magnitude larger for white than for black households.
Monetary policy, asset prices and the unemployment gap
To estimate the effects of a policy shock on asset prices, interest rates, and the unemployment gap, we use instrumental variables local projections following Stock and Watson (2018) and Jorda et al. (2020). Several exogenous policy shock measures are taken from the literature, including our benchmark series from Romer and Romer (2004) as updated by Coibon et al (2017).
Figure 2 shows the effects of the monetary policy shock (a 100 basis point reduction in the federal funds rate) over a five-year period. Similar results are found with other shock measures. There is a substantial asset price boosting effect of surprise monetary easing, in combination with a reduction in the black–white unemployment gap. If white households profit disproportionately from such asset price increases, then it is possible that the portfolio effects of monetary easing go in the opposite direction of the income effects.
Figure 2 Effect of a monetary policy shock
Note: Effect of a 100-bps expansionary shock to federal funds rate on (from left to right): Stock prices (S&P 500), House prices (Case-Shiller Index), 10 year Treasury Bond rate and the gap between Black and White unemployment rates.
Earnings and portfolio effects of monetary policy
Based on our estimates of the asset price effects and the observed portfolio allocation, we derive the estimated effects of the monetary policy shock on wealth, which are shown in Figure 3. An unanticipated monetary policy accommodation leads to asset price changes that benefit white households to a much larger extent than black households because average white wealth is much larger, and a larger fraction is held in equities. The largest effects are after three years, reaching about $25,000 for white households and about one-fifth as much for black households. The biggest effect comes from the large and persistent effect on equity prices, while the house price effect diminishes over time. Bond effects are small because bond holdings are only a small fraction of total wealth for both black and white households.
Figure 3 Effect of a 100-bps monetary policy shock on wealth per household
In addition to the portfolio effects, i.e. the direct effects of capital gains from the monetary shock, there are additional indirect effects of monetary policy shocks if an accommodative monetary shock reduces mortgage interest rates and the interest earned on deposit-type assets. Making assumptions on refinancing and passthrough to mortgage interest rates, we derive the mortgage interest and deposit rate effects from the monetary shock. These effects are shown in Figure 4. Interestingly, black households, with small deposit balances to begin with, lose little from lower interest rates; the average black household gains more from mortgage refinancing. White household deposit interest losses, almost $600, are higher than the average gains from refinancing.
Figure 4 Additional effects of a 100-bps monetary policy shock
Our estimates consider the effects of asset price changes on the wealth of the average black and white household. Since portfolio gains are highly concentrated among wealthy households, one may suspect that the racial wealth gap among more ‘typical’ households might be less affected by asset price changes. To examine this, we look at black and white households around the median of their respective wealth distributions (defined as households between the 40th and 60th percentiles). The portfolio effects of a monetary policy surprise on black and white households around the median are shown in Figure 5. Comparing the effects around the median to the average effects, we find that gains are smaller in levels but that the relative differences between black and white households persist.
Figure 5 Effect of a 100 basis point monetary policy shock on households around the median
Quantifying the earnings effect
The average gap between black and white unemployment rates has historically been about six percentage points. Our estimates show that the gap is reduced by a 100-bps monetary policy shock, with a peak effect after 3 years of -0.34 percentage points. We use earnings data from the 2019 SCF and some conservative assumptions to estimate the impact of the reduction in the unemployment gap on the earnings gap. Specifically, we assume that each black household head who finds employment receives the average earnings of employed black households. The additional income gains of the average black relative to white household is computed by multiplying the estimated impact of the monetary policy shock on the unemployment gap with the average earnings gain. The result is less than $100 per household, or just 0.19% of annual total income for all black households.
Finally, we compare the relative earnings effect to the relative portfolio effect defined as the difference in capital gains accruing to black and white households. The earnings effect of 0.19% of annual income after three years can be contrasted with the corresponding differential portfolio effect after three years of around 17% of annual income. Hence, the differential in the capital gains effect is orders of magnitude larger than the earnings effect.
Monetary policy shocks that change asset prices have differential effects on the wealth of black and white households. White households gain more because they have more financial wealth and hold portfolios that are more concentrated in interest-rate-sensitive assets such as equities. At the same time, monetary policy shocks reduce the gap between black and white unemployment rates and bring larger earnings gains for black households. Bringing the two together, however, leads to one stark finding: the reduction in the earnings gap pales in comparison to the effects on the wealth gap. Our analysis therefore does not bode well for the suggestion that a more accommodative monetary policy helps alleviate racial inequalities. Clearly, this does not mean that achieving racial equity should not be an important objective of economic policy, but the tools available to central banks might not be the right ones.
Bartscher, A K, M Kuhn, M Schularick and P Wachtel (2021), “Monetary Policy and Racial Inequality”, CEPR Discussion Paper 15734.
Bernanke, B S and K N Kuttner (2005), “What Explains the Stock Market’s Reaction to Federal Reserve Policy?”, The Journal of Finance 60(3): 1221-1257.
Bernanke, B S (2020), “The New Tools of Monetary Policy”, American Economic Review 110(4): 943-83.
Bertocchi, G and A Dimico (2020), “Race and the COVID-19 pandemic”, VoxEU.org, 29 July.
Carpenter, S B and W M Rodgers (2004), “The Disparate Labor Market Impacts of Monetary Policy”, Journal of Policy Analysis and Management 23(4): 813-830.
Chetty, R, Nathaniel Hendren, Maggie R. Jones, and Sonya R. Porter (2018), “Race and economic opportunity in the United States”, VoxEU.org, 27 June.
Cieslak, A and A Vissing-Jorgensen (2020), “The economics of the Fed put”, NBER Working Paper.
Coibion, O, Y Gorodnichenko, L Kueng and J Silvia (2014), “Innocent bystanders? Monetary policy and inequality in the US”, VoxEU.org, 25 October.
Jorda, O, M Schularick and A M Taylor (2020), “The Effects of Quasi-Random Monetary Experiments”, Journal of Monetary Economics 112: 22-40.
Kuhn, M, M Schularick and U I Steins (2020), “Income and Wealth Inequality in America, 1949–2016”, Journal of Political Economy 128(9): 3469–3519.
Mayhew, K and S Wills (2020), “Inequality: What has happened, why care, and what can be done about it”, VoxEU.org, 18 June.
Paul, P (2020), “The Time-varying Effect of Monetary Policy on Asset Prices”, Review of Economics and Statistics 102(4): 690-704.
Romer, C D and D H Romer (2004), “A new measure of monetary shocks: Derivation and implications”, American Economic Review 94(4): 1055-1084.
Stock, J H and M W Watson (2018), “Identification and estimation of dynamic causal effects in macroeconomics using external instruments”, The Economic Journal 128(610): 917–948.