Mortgage Metrics Report

Summary

In a speech to the American Securitization Forum in New York, Comptroller of the Currency John C. Dugan unveiled a new Mortgage Metrics Report compiled by the Office of the Comptroller of the Currency (OCC) and focused on delinquencies, loss mitigation actions, and foreclosures in mortgages serviced by national banks.

Recognizing the need for more granular data to assess the state of troubled mortgage markets, the OCC began in February to require the nine largest national bank mortgage servicers to submit comprehensive mortgage data on a monthly basis. The report analyzes data submitted on each of the more than 23 million loans held or serviced by these nine banks from October 2007 through March 2008. The $3.8 trillion portfolio represents 90 percent of mortgages held by national banks and about 40 percent of mortgages overall. The participating national banks are Bank of America, Citibank, First Horizon, HSBC, JPMorgan Chase, National City, USBank, Wachovia, and Wells Fargo.

In creating the new report," the OCC seized the opportunity to improve the way mortgage performance is measured, producing better information for supervision of our banks, and better information for policymakers, other regulators, market participants, and the public," the Comptroller said.

Findings highlighted by Comptroller Dugan included:

• The overall mortgage servicing portfolio of the nine banks reflects credit quality that is relatively satisfactory and relatively stable. The number of current and performing loans remained at about 94 percent over the entire six-month period.
• While subprime mortgages constituted less than 9 percent of the total portfolio, they sustained twice as many delinquencies as either prime or Alt-A mortgages.
• Among loss mitigation actions, payment plans predominated, outnumbering loan modifications in March 2008 by more than four to one, but loan modifications increased at a much faster rate during the period.
• Subprime mortgages accounted for 43 percent of all loss mitigation actions at the end of March, while making up less than 9 percent of the portfolio. Loss mitigation actions exceeded newly initiated foreclosures among subprime borrowers by nearly 2 to 1.
• As in other studies, foreclosures in process are clearly on the rise – climbing from 0.9 percent of the portfolio to 1.23 percent – but the number of new foreclosures varied considerably month to month and was down substantially in March from a high in January.
• Seriously delinquent subprime loans had fewer new foreclosure starts than similarly delinquent prime or Alt-A mortgages, perhaps reflecting the national emphasis on developing alternatives and assistance programs for this class of borrowers.

The new OCC report improves upon other reports on the mortgage industry in three ways. First, its metrics are comprehensive, covering servicers and holders of mortgages and all mortgages, not just subprime. Second, the report is based on loan-level data rather than surveys that report aggregate or summary information submitted quarterly or less frequently. The loan-level data provides greater detail and reliability over time. Third, the agency established standardized definitions and data elements to ensure that information is reported consistently from bank to bank and from loan to loan.

"We are hopeful that the standard definitions and methodology used in this report will be applied more broadly across the U.S. mortgage market," the Comptroller said. "The more we can use standardized metrics across the board, the better we can measure, monitor, and manage mortgage risk."

Although this initial data set was provided on a "best-efforts" basis and includes some "noise" in the data, the OCC is working with the banks to validate the accuracy and fill gaps in the data gathered. The Comptroller also noted that some of the conclusions in the report may seem different from conclusions reported elsewhere, but with good reason. "The data in this report is more precise since it is on an individual loan basis; the population of mortgages held and serviced by these banks has some characteristics different than the overall population of mortgages; and the standard data elements and definitions in the report will also lead to differences in reported results."