By Gary Acosta, Jim Parrott and Mark Zandi
September 17, 2019 at 6:00 a.m. PDT
The housing market isn’t keeping up with the changing face of America. As the nation becomes more diverse and the way that its citizens earn a living and pay their bills more varied, the rules that determine who gets a mortgage and on what terms remain frozen in another time. If these rules and those who write them don’t adapt, our country’s moribund homeownership rate will fall further, particularly for the populations of color that will drive most of the demand for housing for generations to come.
To put this into broader economic perspective, housing typically contributes almost one-fifth of the nation’s gross domestic product, and families of color are projected to make up more than three-fourths of new demand over the next several decades. If we can’t figure out a way to better serve these communities, we are headed for significant economic consequences.
Yet the situation threatens to get worse, not better.
The Consumer Financial Protection Bureau, the federal agency set up to ensure consumers have access to appropriate financial products and services, is poised to decide what kinds of mortgage loans are safe enough that lenders cannot later be sued by borrowers claiming they should never have received a loan in the first place. As most lenders are not willing to take on such legal risk, the CFPB is in effect determining who will have meaningful access to the mortgage market.
This is no easy task. We learned in the housing crisis of a decade ago that it is in no one’s interest for the agency to open the door to unsustainable lending. Yet restricting the ability of creditworthy borrowers to get loans doesn’t make sense either, since it constrains homeownership and all of the economic and social benefits that come with homeownership.
To thread that needle, the CFPB is considering limiting this legal safe harbor to loans made to borrowers whose monthly debt payment falls below a certain percentage of their reported monthly income. However, this makes little economic sense, because the risk that a borrower may be unable to pay their loan is driven by many factors other than the amount of debt they carry, including their credit score and the amount of savings they have in the bank.
Indeed, among credit score, the size of a borrower’s loan relative to the value of their home, and their debt relative to income, it is the latter, the so-called debt-to-income ratio, that has proved far and away the least important in predicting whether a borrower will default. Policymakers and lenders alike have struggled to determine what a borrower should even count as income when applying for a mortgage, making a borrower’s debt-to-income ratio at best an uneven measure of credit risk.
Even more disconcerting is the impact this approach would have on access to mortgage loans, particularly among families of color. More than one in five African Americans who have gotten a mortgage recently would have a much harder time under this cap, along with one in four Hispanic Americans. It would thus further constrain access to homeownership for the very groups for which we desperately need to find ways to expand it.
If the housing market is to keep up with the nation’s changing demographics and contribute to our economy’s success, policymakers must find ways to make it easier — not harder — for lenders to identify families who can become sustainable homeowners.
The CFPB should not set a debt-to-income ratio cap, but allow lenders to weigh all the relevant factors to determine a borrower’s risk. Doing so doesn’t just make more sense in assessing credit risk as we understand it today; it allows lenders to expand their reach to more creditworthy borrowers as we improve our understanding of credit risk in the decades to come.
Gary Acosta is co-founder and chief executive officer of the National Association of Hispanic Real Estate Professionals. Jim Parrott is a nonresident fellow at the Urban Institute and owner of Falling Creek Advisors. Mark Zandi is chief economist at Moody’s Analytics.