Mortgage Analytics and Performance Dashboard
The Mortgage Analytics and Performance Dashboard (MAPD) gives policymakers at the national, state, and local levels the ability to see where owner-occupant homeowners in their jurisdictions have fallen behind on mortgage payments or used mortgage forbearance as a means of economic relief during the COVID-19 pandemic.
In the previous recession, tracking mortgage delinquency measures was one of the primary methods for understanding the amount of distress in the mortgage market. Unlike the previous downturn, where there was no uniform and widespread response, mortgage forbearance has been one of many tools used by the federal government to provide economic relief to U.S. homeowners. Mortgage forbearance is a policy that provides borrowers the flexibility to miss mortgage payments without immediate penalty during a specified period of time, normally six months. These missed payments usually are rolled into a repayment plan of some kind after the forbearance period has ended. Forbearance is typically available to borrowers experiencing temporary hardship, and historically it has been used only in areas where homeowners suffer from natural disasters such as hurricanes.
On March 10, 2020, the Federal Housing Finance Agency, which regulates Fannie Mae and Freddie Mac, released a public statement encouraging mortgage borrowers nationwide to take advantage of forbearance programs in light of the growing economic impact of COVID-19. Borrowers whose loans are not backed by Fannie Mae or Freddie Mac were also offered forbearance plans similar to those offered by the government-sponsored enterprises. By the end of April, nearly 4.5 million mortgages in the United States were in forbearance.
In addition to the extended payment plan for missed payments, mortgage servicers generally do not report missed payments to credit bureaus when a mortgage is in forbearance. (A mortgage servicer is the company responsible for administering the payment plan for mortgages; it is sometimes the lender and sometimes a third party.) The Mortgage Analytics and Performance Dashboard can estimate forborne mortgages using a combined servicer-credit bureau data set called CRISM; a forborne mortgage will be delinquent according to the servicer but current according to the credit bureau. CRISM is a product of Equifax Credit Risks Insight Servicing and Black Knight McDash data sets. The Atlanta Fed acknowledges Lara Loewenstein at the Federal Reserve Bank of Cleveland for the idea of using mixed signals of delinquency to estimate forbearance.
The tool will be updated monthly with a lag of approximately two months from the time the Atlanta Fed gets the latest CRISM data set.
Explore the Data
How the Dashboard Works
The dashboard presents two maps and two bar charts. The map and chart on the top, in shades of blue, illustrate the share of outstanding first mortgages estimated to be in forbearance. The map and chart on the bottom, in shades of orange, illustrate the share of outstanding first mortgages that are 30 or more days past due. Users have the ability to select the geography (state or core-based statistical area) and time period of interest. The map displays the share of mortgages that are delinquent by month at the ZIP code level.
The user can select the level of aggregation. For example, if the user selects Georgia and Alabama in the control panel, the time-series bar chart will reflect aggregated forbearance and delinquency rates as defined above for the whole of Georgia and Alabama.
Data Definitions and Sources
The loans in this data set are taken from CRISM (data from Equifax Credit Risk Insight for Servicing and Black Knight McDash), which is a combination of mortgage servicer data and credit bureau data. Lara Loewenstein at the Federal Reserve Bank of Cleveland observed that mortgage forbearance can be identified in the data when the mortgage servicer reports a loan as delinquent and the credit bureau does not, because forbearance is not meant to trigger adverse impacts to the borrower's credit score. Therefore, we will use the following formula in each ZIP code in each month to estimate the rate of forbearance:
We define an active mortgage as one that is current or in any state of delinquency (according to the servicer). We also limit the analysis to first mortgages (no second liens) with an owner-occupant, and we exclude any ZIP code months with fewer than 50 loans meeting these criteria.
The delinquency rate is defined in each ZIP code month as the fraction of loans in which the servicer and credit bureau agree that there is a delinquency. In this case, the borrower has not paid his or her mortgage, and we do not identify the borrower as being in forbearance. (One caveat is that an already-delinquent borrower may have signed up for forbearance, and we would show this borrower as delinquent, not forborne.) We also cannot identify borrowers who have signed up for forbearance but who continued paying their mortgages in full, which research by Black Knight, Inc. has shown is a substantial portion of those mortgages in forbearance.
In this tool, the following definitions apply:
Active loans are defined as owner-occupant first mortgages that are either current or delinquent.
The rate of forbearance is equal to the fraction of active loans that are delinquent, according to the servicer but not delinquent, according to the credit bureau.
The rate of delinquency is equal to the fraction of active loans that are delinquent, according to both the servicer and the credit bureau.
These rates are reported for the relevant geography selected by the user. Each ZIP code in the map reports the rates of forbearance and delinquency for that ZIP code; the time-series bar charts show the aggregate rates of forbearance and delinquency for all ZIP codes in the geography selected in the control panel.