By signaling their intent to use the fair-lending equivalent of the "nuclear option," the disparate impact doctrine, in pursuit of fair lending violations, the CFPB and the Departments of Justice and Housing and Urban Development have placed lenders in a no-win situation.
QM's signature requirement—that to be eligible for qualified mortgage status a loan must not have a debt-to-income ratio exceeding 43%—is likely to have a disproportionate impact on some protected classes of borrowers. So lenders can make QM loans and risk being sued under the disparate-impact doctrine—or make non-QM loans and risk being sued under the ability-to-repay doctrine.
A way out of this policy misalignment is for lenders to sell their loans to Fannie Mae and Freddie Mac, given the exemption to QM provided for the government-sponsored enterprises under the Dodd-Frank Act. The result, however, is to increase reliance on the GSEs at a time when most agree that reducing their footprint is in the best long-term interest of borrowers and taxpayers.
From its inception, the 43% debt-to-income ratio requirement in QM was a poorly designed antidote to the inability to repay that characterized a vast number of borrowers during the mortgage crisis. The major reasons borrowers would not afford to pay were that lenders gave them loans without properly verifying their income or assets, and that the loans often had nontraditional structures in which initially low monthly payments reset much higher a few years later. Eliminating these attributes in QM would go a long way toward addressing potential ability-to-repay concerns.
However, settling on a 43% debt-to-income ratio completely ignored the concept of compensating factors, a longstanding approach used in prudent underwriting practices. A loan with a 44% DTI, a 750 credit score, and a 70% loan-to-value ratio is a high quality mortgage, according to the default risk profile, and should be approved. However, under QM, this loan is likely to be done only through the GSEs.
In this example, the QM DTI standard puts lenders at great jeopardy of violating fair-lending laws, since adhering to the 43% DTI requirement makes it difficult to pass the federal agencies' disparate-impact test. A bright-line requirement offers no flexibility for lenders to find compensating factors, such as strong credit or substantial down payments, which would eliminate the potential for protected classes to be disproportionately rejected for mortgages.
The GSE exemption to QM at least brings some measure of reasonability to the mortgage underwriting decision, with less potential to bring a disparate impact suit by the government. Both GSEs continue to rely on statistically-based automated underwriting systems which have been developed and tested to meet the requirements of the Equal Credit Opportunity Act's Regulation B provisions for fair lending. (I worked on such testing of Freddie Mac's automated underwriting system.) Not only do the underwriting scorecards and policy overrides associated with these underwriting tools undergo exhaustive fair lending testing, but the cutoff scores used to determine loan eligibility are examined carefully. The underwriting scorecard produces a mortgage score that reflects the likelihood of default based on all of the borrower's risk attributes and other variables. This includes the borrower's credit profile, size of the down payment and debt-to-income, among other factors.
These scorecards thus allow tradeoffs to be made in the borrower's risk profile and recognize that a 44% DTI with compensating factors may in fact be a lower risk than a 43% DTI without such factors. The scorecard cutoff establishes a level of acceptable default risk for the GSE. To ensure compliance with fair lending laws, a battery of tests are performed to observe concentrations of borrowers segmented by race, income and other characteristics at and around the score cutoff. A cutoff can be adjusted to reduce disproportionately higher rejection rates for protected classes, and in so doing ensure compliance with fair lending requirements while managing credit risk.
Setting a QM DTI standard that ignores compensating factors, coupled with aggressive application of the disparate-impact doctrine, provides little flexibility to lenders and sends a chilling effect throughout the mortgage market. To avoid becoming ensnared in this regulatory dragnet, lenders would be better off selling to the GSEs, thus meeting the QM standards while reducing their exposure to disparate-impact legal actions. Unfortunately, this outcome raises the importance of the GSEs in the mortgage market rather than reducing their presence, as is required to begin the process of restructuring the mortgage secondary market.
Government engineering of mortgage underwriting standards was a bad idea from the start and is only made worse now by the unprecedented application of aggressive fair lending legal doctrine.
Clifford Rossi is the Professor-of-the-Practice at the Robert H. Smith School of Business at the University of Maryland.