Risk Retention--Back to the Drawing Board

Originally published in the July 2011 issue of Asset Securitization Report

The regulatory agencies charged with implementing the risk-retention provision of the Dodd-Frank Act (DFA) recently announced that the comment period for their initial proposal will be extended to Aug. 1. As reported in the press, the delay resulted from widespread opposition to the definition of qualified residential mortgages (QRMs) as outlined in the March proposal. More fundamentally, it reflects the difficult tradeoffs involved in attempting to implement the risk retention provisions outlined in the DFA without limiting the access of large numbers of borrowers to mortgage credit and further damaging the housing market.

As part of its risk-retention provisions, the DFA requires the regulators to define QRMs, i.e., the loans that are either fully or partially exempt from the 5% minimum requirement. The regulators subsequently proposed a fairly narrow definition that included front and back debt-to-income ratios of 28% and 36%, respectively, along with a maximum LTV (for purchase loans) of 80%. It does not allow mortgage insurance to be taken into consideration, while exempting loans securitized by the GSEs as long as Fannie Mae and Freddie Mac remain under the FHFA conservatorship.

There are numerous issues with the interagency proposal. As I discussed last month, the principal capture cash reserve account (PCCRA) provision would thoroughly distort the consumer mortgage market. In addition, the language exempting loans securitized by the GSEs might effectively serve as a poison pill that would make it more difficult to remove them from conservatorship, and may wind up as another step toward making them permanent, nationalized entities.

However, the most complex challenge for the regulators is that they are being asked to create a broad and permanent definition of mortgages that will be exempt from required risk retention. Implicit in the surrounding controversies is the understanding that mortgages ineligible for QRM status will be significantly more expensive for consumers than QRMs, if available at all. The QRM definition proposed by the regulators was generally viewed as being overly narrow, and has generated widespread opposition from the lending industry. This was highlighted by a recent report in ASR sister publication National Mortgage News entitled “Regulators Could Be Ready for a ‘Redo’ on Risk Retention.” In addition to comments supporting relaxed down payment requirements (one industry official was quoted as saying that “(L)ow down payments did not cause this crisis”), others argued for front DTI ratios as high as 37%-38%.

Despite the fact that these contentions are clearly contrary to the historical experience, the reactions to the proposal nonetheless highlight the difficulties inherent in outlining any set of universal and invariable lending standards. In addition to the subjective nature of such criteria, loan underwriting itself is a complex and multifaceted process; an attempt to forge a set of concrete and inflexible lending requirements is likely to inadvertently exclude significant numbers of credit-worthy borrowers.

Moreover, the DFA’s decision to leave the interpretation of the risk-retention proposals to the regulators forces them to interpret the concept of “skin in the game” itself. As outlined in the March release, the initial proposal looks to “establish underwriting standards designed to ensure that QRMs are of very high credit quality.” Given the difficulties in establishing a set of broad standards as well as the potential dangers to housing of reduced mortgage credit availability, this standard is clearly too restrictive. I support an alternative interpretation: risk retention should serve to inhibit the more egregious lending practices that led to the mortgage crisis. This means that QRMs should be defined broadly, particularly with respect to down payments and LTVs. Regulators do not need to usurp the role of credit managers in order to protect the financial system from systemic risks