Earlier this year, the viewpoint of political conservatives that the credit quality of the mortgages being purchased by the two government-sponsored enterprises (GSEs) of Freddie Mac and Fannie Mae is poor became very prominent. In a recent hearing of the Senate Banking Committee about the Trump administration’s proposed housing reform plan, one conservative Senator even went so far as to call it a “dumpster fire.” With a Republican administration in office and a conservative leading the Federal Housing Finance Administration (FHFA), such a viewpoint seems to have found a ready audience. In fact, the Washington Post on October 2nd included an article entitled “Federal Government Has Dramatically Expanded Exposure to Risky Mortgages” which fully advocates this conservative viewpoint.
In A Case Study of GSE Politicization: The Flawed Narrative of Loose Credit I examine the credit quality of mortgages purchased by the GSEs and the argument being made by conservatives that their quality is poor. Utilizing a 25-year history of GSE credit, recently produced by the FHFA as a research paper, as well as other publicly-available sources, I show how the loose credit narrative is fundamentally flawed and inaccurate.
Behind this viewpoint is the extreme politicization of housing finance. After all, with $11 trillion outstanding residential debt impacting tens of millions of families spread throughout every congressional district in every state, and with the US government backing around two-thirds of all mortgages, such politicization is to be expected. One aspect of this politicization has long been dueling narratives—even more than a decade into conservatorship—about the credit quality of mortgages bought by the two GSEs which, between them, own about $5 trillion of first mortgages, and the likelihood of the US taxpayer needing to bail them out again.
During my seven years leading Freddie Mac (May 2012 to June 2019), I learned that conservative housing finance advocates viewed GSE mortgage credit quality as poor, to one degree or another, at all times, and at the extreme believed that in a significant economic downturn the GSEs would suffer massive losses and therefore pose a risk to the stability of the entire financial system. I similarly learned that liberal housing finance advocates continually believed the opposite, that the two GSEs are too restrictive in extending credit, and that they are passing up or overcharging for many mortgages with decent credit quality that could help many people, especially those of color, obtain the long-term benefits of homeownership.
Obviously, both can’t be right.
The data show conclusively that the GSEs pose less risk to the financial system than they have in years—with stress testing losses done under the Dodd-Frank Act (which are administered by the FHFA) declining by 80 percent since such tests began in 2013. The data also show that increases of the credit risk profile of single-family mortgages purchased by the GSEs since 2011-2012 reflect a return to normalcy after an extremely tight mortgage credit environment in the years immediately following the 2008 financial crisis. Also, the advent of the Qualified Mortgage (QM)—a regime created by Dodd-Frank and implemented through the Consumer Financial Protection Bureau—has dramatically improved credit quality for mortgages in America, because the product restrictions (prohibiting interest-only periods, balloon payments, and certain other features) have eliminated many of the credit-loss-causing weaknesses in the financial crisis. While not well known, the two GSEs in fact have adopted the product restrictions of QM.
In the mortgage industry, the credit quality of the GSE single-family mortgage book is, in fact, commonly referred to as “pristine”—a term that perhaps slightly overstates its quality.
My new paper shows how the conservative narrative of loose credit has two fundamental flaws. First, it is based upon a trend beginning in 2012—the tightest mortgage credit period in recent decades—rather than a longer view, which biases the results. Second, it is based on a single factor that may impact future credit performance: DTI, or the debt-to-income ratio, which is the percentage of household income that goes to the monthly mortgage payment, regarded in the industry as a second-tier predictor of future credit performance. In fact, the narrative usually ignores the other factors that contribute to future credit quality, especially the historic “big two”: credit rating (FICO score) and the LTV (loan-to-value ratio). In particular, FICO scores on GSE loan purchases are much higher after the financial crisis than before it.
Finally, the concept of credit risk transfer (CRT) is rarely, if ever, mentioned in the conservative narrative—as indeed it was not during the recent Senate hearings. This is the program, pioneered by Freddie Mac in 2013, by which the clear majority of the credit risk of new mortgage purchases by the GSEs are sold off to institutional investors and not retained by the GSEs. This not only means that the credit risk does not stay with the taxpayers, who heavily support the GSEs while they are in conservatorship, it also means that GSE mortgages are subject to what is called “market discipline,” where the investors in CRT vote with their money how they view mortgage credit, in terms of whether they will invest and, if so, at what price. The CRT market continues to purchase GSE credit risk unabated, in direct contrast to the conservative narrative of poor GSE mortgage credit quality.
The paper concludes that policymakers should focus on other important housing issues—like the continuing and deep shortage of the production of new housing stock, which is the major underlying cause of housing affordability issues today—rather than the red herring narrative of loose GSE credit.