Skip to main content

2021, No. 19
Posted 2021-07-16

Mortgage Forbearance and Economic Recovery from the Pandemic

by Miguel Faria e Castro and Olivia Wilkinson

The March 2020 CARES Act was a large fiscal package intended to provide financial relief to U.S. households during the COVID-19 pandemic. Importantly, this law has permitted qualifying households with federally backed mortgages to request mortgage forbearance, which reduces or pauses monthly mortgage payments.1 

This essay documents the impact of the CARES Act forbearance measures and investigates whether this relief assisted the recovery from the COVID-19 recession.2 We show that, unlike in past recessions, aggregate mortgage delinquency rates on credit reports did not rise during the 2020 recession. However, once forbearance is taken into account, actual delinquency rates are 2.6 times greater. We also find that the extent of forbearance is positively related to the economic recovery across U.S. states, even when controlling for other factors.

Forbearance programs do not forgive loans: Households still owe missed payments. Rather, these programs allow households to reduce or pause mortgage payments while not affecting the households' credit scores.3 

We start by asking how widely consumers used these mortgage forbearance programs. To do this, we compare 90+ day mortgage delinquency rates between two datasets4: Black Knight's McDash and the FRBNY/Equifax Consumer Credit Panel (Equifax).5 McDash collects data directly from mortgage servicers and thus records whether payments were made regardless of forbearance status; Equifax, on the other hand, collects data from consumer credit reports. In the absence of a forbearance program, missed mortgage payments are reported in the Equifax dataset. If there is a forbearance program in place, reduced payments will be reflected only in the McDash dataset. We can therefore use the difference between the delinquency rates in the two datasets as an estimate of the size of these forbearance programs.

The figure plots aggregate 90+ day delinquency rates in the two datasets, starting in 2005:Q1. For most of the sample, the McDash rate exceeds the Equifax rate, which suggests that there was already some forbearance before the pandemic. Importantly, however, the two lines diverge substantially starting in 2020:Q1: While the aggregate Equifax delinquency rate remains roughly constant (and actually falls), the aggregate McDash delinquency rate rises substantially. At its peak, the gap between the two is 4.08 percentage points. This implies that nearly 9 out of 10 borrowers who did not make their full monthly payments were benefitting from some type of forbearance program.6

We then ask whether forbearance may have affected the recovery from the COVID-19 recession. By giving households more flexibility in managing their financial situation, forbearance programs may have allowed them to reduce spending on goods and services by less than they would have otherwise. Less of a reduction in spending in turn may have moderated the destruction of local jobs and helped support the employment recovery. To investigate this, we develop a statistical model of the recovery that uses variation in forbearance usage rates, unemployment rates, household income, and population across U.S. counties to determine how policies affected the recovery. We find that forbearance has a positive and statistically significant impact on employment recovery: With all else constant, a relative increase of 1 percentage point in our measure of forbearance is associated with a relative decline in the local unemployment rate of 3.5 percentage points between April 2020 and April 2021. 

To conclude, the CARES Act mortgage forbearance programs seem to have been widely used, as evidenced by the gap between the McDash- and Equifax-reported delinquency rates; and there is some evidence that providing households with some financial flexibility at a time of great uncertainty and income loss may have played a role in the recovery from the COVID-19 recession.


1 See this document from the Consumer Financial Protection Bureau for a non-technical description of the program:

2 Although the National Bureau of Economic Research has not yet assigned an end date to the COVID-19 recession, the U.S. unemployment rate peaked in 2020:Q2 and has decreased since then. We therefore call the period since 2020:Q2 the "recovery" from the recession. 

3 While this essay focuses on the federal mortgage forbearance programs that were implemented under the CARES Act, lenders and loan servicers voluntarily initiated forbearance programs during the COVID-19 recession for private mortgages and other types of credit, such as credit cards, auto loans, and student debt. 

4 The 90+ day mortgage delinquency rate is the percent of mortgages with payments at least 90 days late. 

5 FRBNY is Federal Reserve Bank of New York. 

6 At the peak, 90+ day delinquencies made up 4.81 percent of loans. The gap is 4.08, so 4.08/4.81 = 0.85, which we round to 9 out of 10 for simplicity.

© 2021, Federal Reserve Bank of St. Louis. The views expressed are those of the author(s) and do not necessarily reflect official positions of the Federal Reserve Bank of St. Louis or the Federal Reserve System.