Refinancing Inequality during the COVID 19 Pandemic

By | May 5, 2021

Mortgage refinancing is one of the main channels through which monetary policy affects the pocketbooks of everyday Americans. When interest rates are sufficiently low, homeowners can refinance their loans, reducing their monthly payment and realizing a substantial reduction in borrowing costs over the life of the loan. The logic behind expansionary monetary policy in this context is that when homeowners refinance their mortgages, they will redirect some of their savings into spending, thus stimulating the economy. As a result, the effectiveness of the refinancing channel to stimulate the economy depends on borrowers’ actual refinancing ability, and on their propensity to spend out of the savings they realize.

However, it is also well documented there are frictions in the refinancing market by which borrowers fail to refinance even when interest rates are sufficiently low to justify refinancing efforts. These frictions range from limited financial literacy and behavioral biases, to credit quality and limited competition. It is thus logical to ask if savings from refinancing are reaching those who have the largest marginal propensities to consume, or instead, if savings from refinancing are concentrated on individuals who are less likely to inject them back into the economy.

Savings from Refinancing

In a new working paper, we document that during the first half of 2020, savings from refinancing were heavily concentrated in high-income borrowers, who tend to show lower marginal propensities to consume out of increases in wealth. We argue that the distribution of savings from refinancing across income groups observed during the first half of 2020 could have reduced the potency of expansionary monetary policy to stimulate consumption. 

Specifically, we study savings from refinancing across the income distribution of homeowners. We note that in previous periods of low interest rates, individuals in the top quintile of the income distribution received larger savings from refinancing than their counterparts in the bottom quintile of the income distribution. However, this difference in savings was explained in its entirety by off-the-shelf observable characteristics such as loan to value (LTV), FICO score, original interest rate, and loan balance. That is, borrowers in the top and bottom quintile of the income distribution with the same LTV, FICO score, original interest rate, and original loan balance were saving about the same amount.  In contrast, during the first half of 2020 the difference in savings from refinancing between the top and bottom quintiles of the income distribution grew about ten times.

Two factors can explain the large increases in refinancing inequality observed in the first half of 2020: first, individuals in the top quintile of the income distribution increased their refinancing activity more than their counterparts in the bottom quintile and second, conditional on refinancing, they also captured the largest improvements in interest rate differentials. Before 2020, individuals in the top and bottom quintile of the income distribution had basically the same probability of refinancing – estimated at 1.14% — after controlling for observable characteristics. During 2020, the bottom quintile of the income distribution increased its refinancing activity by 1.25 percentage points (pp), whereas the top quintile of the income distribution increased its refinancing activity by more than 8 pp. In addition, higher-income individuals also captured the largest improvements in interest rate differentials. Before 2020, individuals in the bottom quintile of the income distribution received a 1.66 pp reduction in interest rates, conditional on refinancing. This reduction reached 1.83 pp in 2020 (i.e., a 0.17 pp improvement). In contrast, individuals in the top quintile of the income distribution who refinanced their mortgages received reductions of only 1.49 pp before the pandemic but reached an average 1.87 bps reduction in 2020 (i.e., a 0.38 pp improvement).

Two Additional Analyses of Refinancing Inequality

Overall, we estimate a $5 billion gap in savings from refinancing between the top quintile of the income distribution and the rest of the market. If individuals in lower segments of the income distribution were receiving the same savings from refinancing as individuals in the top quintile of the income distribution, they would capture an additional $5 billion in refinancing savings. To understand the phenomena of refinancing inequality, we perform two additional analyses. 

First, we investigate whether the increase in refinancing inequality is uniform across the entire market, or if it depends on the severity of shocks to local economies. We find evidence for the latter: refinancing inequality is positively correlated with COVID-19 case rates, and 74% of this correlation can be explained through changes in local economic conditions. [LR1] [MPPDC2] [MPPDC3] 

Second, we investigate if increases in refinancing inequality are driven by changes in lenders’ behavior or by changes in borrowers’ behavior. While there was a large surge in applications that could have affected lenders’ ability to process and approve applications, we do not find evidence of lenders prioritizing applications of high-income borrowers. Instead, we find that high-income borrowers are overrepresented in the pool of applicants, suggesting that low-income borrowers are not applying at the same rate as their high-income counterparts.

Contribution

Understanding the dynamics of the market for mortgage refinancing is important to design effective economic policy. Our results suggest that there is room for targeted policies to promote refinancing among homeowners in the bottom segment of the income distribution, or more generally there is room for targeted transfers to low-income borrowers. These policies have the potential to increase the efficiency of monetary policy and to reduce wealth inequality.

Sumit Agarwal is the Low Tuck Kwong Distinguished Professor of Finance at the Business School and a Professor of Economics and Real Estate at the National University of Singapore. 

Souphala Chomsisengphet is the Director of the Retail Credit Risk Analysis Division within the Economics Department at the Office of the Comptroller of the Currency. 

Hua Kiefer is the Acting Chief of the Quantitative Risk Analysis Section at the FDIC Center for Financial Research in the Division of Insurance and Research.

Leonard Kiefer is the Deputy Chief Economist of Freddie Mac.

Paolina C. Medina is an Assistant Professor of Finance at the Mays Business School of Texas A&M University.

This post is adapted from their paper, “Inequality During the COVID-19 Pandemic: The Case of Savings from Mortgage Refinancing,” available on SSRN.

Leave a Reply

Your email address will not be published. Required fields are marked *