Don't eliminate ability to repay — reform it
Over the last few weeks, we have seen a flurry of activity dealing with the expiration next year and potential replacement of the so-called GSE QM patch.
The Consumer Financial Protection Bureau signaled its intent to let the patch expire and replace it with a revised definition of Qualified Mortgage that eliminates or significantly diminishes debt-to-income as a factor and move toward a metric that measures a loan's APR spread over the average prime offer rate.
This follows the contours of an approach suggested by a consortium largely of big bank and big nonbank lenders in a letter to the CFPB last September. At its core, this approach means eliminating or significantly diminishing ability to repay as a consideration, ironically, in the "ability to repay" (QM) statute.
How soon we forget the lessons of the subprime housing crisis.
Fortunately, there is an alternative to this approach. At almost the same time CFPB made its announcement, a broad coalition that included the National Association of Realtors, small lender groups including the Community Home Lenders Association, consumer groups and private mortgage insurers unveiled a sensible approach to reform, not end, use of DTI in QM in a letter.
This coalition warned about moving from a debt-to-income standard to an APOR standard, and proposed instead retaining DTI for QM but making it more flexible. This approach would protect both borrowers and taxpayers.
There is fairly broad consensus that the main driver of the 2008 housing and economic crisis was subprime lending to borrowers with inadequate income to pay the debt service on their mortgage loans. In response, Congress enacted the Qualified Mortgage ability to repay law, and the CFPB implemented it, setting a basic requirement that the DTI on a mortgage loan not exceed 43%.
There is also fairly broad consensus that DTI is not a perfect mechanism for measuring a borrower's likelihood of making timely mortgage payments — a fact recognized in the creation of the GSE patch.
However, it is certainly highly relevant as an underwriting factor — and probably the best measurement of a borrower’s "ability" to repay the loan. In contrast, an APOR benchmark does not measure this. It only measures investors' willingness to fund loans at reasonable rates and terms. As such, it clearly does not meet a borrower’s ability to repay — which after all, was at the heart of the statute.
For example, a low LTV loan to a borrower that can't make payments may not be that risky to an investor, and therefore, may meet the APOR standard that is ultimately established. But that does not mean the borrower can repay the loan, it simply means an investor may not lose money.
The subprime crisis leading up to the 2008 housing crisis showed that just because investors are willing to buy loans does not mean that those prices accurately reflect risk, or that the borrowers can repay the loans. While we don't know what APOR threshold the CFPB might set, a review of my firm's underwriting files from the subprime era raises concerns that the level that is chosen might have greenlighted a substantial number of alt-A credit and other risky loans from that era.
It appears the CFPB may have concluded that an approach that allows DTIs above 43% (or say 45%) but only with compensating factors may involve too much complexity. Some argue that this is not enough of a bright line standard for the hundreds of different lending sources that QM would apply to. This is a complex problem and a real challenge, possibly with no perfect answer.
However, this concern does not warrant throwing out a DTI metric altogether. CHLA believes that our coalition proposal to tie higher DTI thresholds to compensating factors reflects sound underwriting practices and would work well in the real world. Further, our members believe that modernizing Appendix Q to reflect economic trends, such as the gig economy, would address a meaningful portion of the prudent loans that a strict 43% DTI had precluded. This could be done without weakening consumer or taxpayer protections.
Alternatively, if the conclusion is that use of compensating factors is too complicated, it would still be preferable to raise the DTI threshold to some number up to as high as 50% — than it would be to jettison this important metric altogether.
The 2008 housing crisis taught us important lessons. One of the most important ones is simple: Don't eliminate ability to repay, reform it.