Dissenting Statement of Commissioner Daniel M. Gallagher Concerning Re-Proposal of Rules Implementing the Credit Risk Retention Provisions of the Dodd-Frank Act

Commissioner Daniel M.Gallagher

U.S. Securities and Exchange Commission

Aug. 28, 2013

Dissenting Statement of Commissioner Daniel M. Gallagher Concerning Re-Proposal of Rules Implementing the Credit Risk Retention Provisions of the Dodd-Frank Act

Today, a majority of the Commission will approve a re-proposal of the rules implementing Section 941 of the Dodd-Frank Act, the so-called "risk retention" provision.[1]  I cannot in good conscience support today´s re-proposal, and I respectfully but vigorously dissent. 

The re-proposed risk retention rules, if adopted, will ensure that the vast majority of mortgages in the United States are insured or owned by the government, will introduce another flawed government imprimatur of creditworthiness into the markets, and will disincentivize proper risk management and due diligence in the mortgage markets.  It is unfortunate that, even with the financial crisis still so fresh in the collective memory of policymakers and taxpayers, regulators seem determined to repeat the mistakes of the past.

My dissent today is based on the substance of the re-proposal.  I note at the outset, however, that I am pleased to see that the agencies involved in this rulemaking are re-proposing these rules rather than adopting flawed rules.  I believe that the government must re-propose and seek comment on all proposed rules when, as is the case here, it is confronted with feedback from commenters that reveals a need to fundamentally reexamine the proposal.[2]  This is not merely good government, but also what is required by relevant law, including the Administrative Procedure Act.[3]  And, I will note, many of the agencies approving today's joint rulemaking re-proposal, including the Commission, are also involved in the so-called Volcker rule rulemaking.  The Volcker rule proposal, which has also received an extraordinary level of feedback evidencing the need for a fundamental reexamination, has become the poster child for re-proposal in lieu of adoption.  I strongly urge, as I have done in the past, that the implementing regulations for the Volcker rule be re-proposed as well. 

As required by Section 941 of the Dodd-Frank Act, both the original March 2011 risk retention proposal and today´s re-proposal were crafted jointly with the federal banking regulators, the Federal Housing Finance Agency (FHFA), and the Department of Housing and Urban Development (HUD).  The original proposal included a strict definition of the key term "qualified residential mortgage" (QRM).  In addition to excluding loans with certain product features such as negative amortization, interest-only payments, or significant interest rate increases from the definition of QRM, the original proposal also required QRMs to have high quality underwriting standards, including a minimum down payment, a maximum loan-to-value ratio, maximum front-end and back-end debt-to-income ratios, and good credit history on the part of the borrower.  The re-proposal completely abandons the definition of QRM set forth by the agencies in the original proposal and instead adopts the deeply flawed definition of "qualified mortgage" (QM) set forth by the Consumer Financial Protection Bureau (CFPB) earlier this year.  As explained further below, I am unable to support the re-proposal.

BACKGROUND

As a baseline matter, I do not believe that the federal government should be dictating prescriptive risk management standards, as the re-proposed rules would do for mortgage securitizers.[4]  However, given the dominant role of the federal government in the U.S. housing markets, it is perhaps too much to expect that the federal bureaucracy will refrain from extending itself into areas it is not well suited to regulate, such as the prescription of centralized risk management standards.[5]  And although I disagree with the idea that the government should be setting standards for so-called "skin in the game," at least this debate – unlike others facing the Commission today, such as corporate political spending disclosures and CEO pay ratio – concerns an actual cause of the financial crisis rather than a political wish-list issue.  Indeed, failed federal housing policy was the primary driver of the financial crisis, and many other factors, such as regulatory failures and private market excesses, fed off of this and other failed government policies. 

This rulemaking could have gone in two other very different directions.  The agencies could have proposed a very stringent definition of QRM, such as the one proposed in 2011.  This would have created a narrow class of QRM loans, thus allowing for a sizable, vibrant non-QRM mortgage market.  Or, the agencies could have proposed a very loose QRM definition, effectively making the 5% retention charge moot.  In either scenario, one could expect an active private market. 

Unfortunately, the re-proposal before us today, presented as the product of the Commission in consultation with the banking regulators, the FHFA, and HUD, abdicates the task of defining a Securities Exchange Act term to the Consumer Financial Protection Bureau and defers to its hugely problematic definition of "qualified mortgage."  The re-proposing release candidly admits, on the one hand, of the loans originated from 2005 to 2008 that would have qualified as QMs, a staggering 23 percent experienced a 90-day or greater delinquency or a foreclosure by the end of 2012.  On the other hand, the release clarifies that "the agencies are not implying that they consider a 23 percent default rate to be an acceptable level of risk."[6]  I have no cause to disagree with that statement, for I believe that by deferring to the CFPB´s definition of qualified mortgage, the agencies are demonstrating that they are not considering default rates or risk levels at all.  Instead, this definition is motivated by a desire to return to the failed federal housing policies that proved to be a direct and decisive cause of the financial crisis. 

Many of those policies, I should add, were largely the product of a bipartisan consensus – but bad bipartisan policy is still bad policy.  The Housing and Community Development Act of 1992, passed with overwhelming bipartisan support,[7] required that 30 percent or more of the loans purchased by Fannie Mae and Freddie Mac (the GSEs) be made to borrowers at or below the median income in their communities.  In 1995, HUD exercised its authority under that legislation to increase the requirement to 40 percent of all loans purchased by the GSEs the next year and 42 percent in subsequent years, which was followed by an increase to 50 percent in 2000.[8]  By 2008, HUD had increased the requirement to 56 percent.[9]

In order to meet these requirements, the GSEs had to continually lower their underwriting standards, eventually creating an incentive for lenders to make zero-down-payment loans to borrowers with ever-decreasing credit scores.  By 2008, half of all mortgages in the U.S. were subprime or similarly weak.[10]  Through the alchemy of loan securitization and the utter incompetence of the rating agencies, these subpar loans were repackaged and resold in triple-A rated tranches of residential mortgage backed securities (RMBS).  The rest, as they say, is history.

A SUBSTANTIVELY FLAWED RE-PROPOSAL

History, of course, repeats itself.  It was Karl Marx, of all people, who added that history does so first as tragedy, then as farce.[11]  But while the premise of today´s re-proposal is indeed farcical, its real-world results will be anything but.  The theory behind the risk-retention provision is that by ensuring that securitizers have "skin in the game," they would share in losses suffered by poorly performing securitizations and would therefore be dissuaded from securitizing poor quality loans in the first place.  In order for a risk retention rule to have any meaning, however, any exemptions from the application of such a rule would have to be made on a limited basis to only the highest-quality loans – that is, those that simply don´t need the protections the rule is designed to offer.  

As former FDIC Chairman Sheila Bair, an early and vocal advocate for risk retention requirements, writes in reference to the March 2011 proposal, "Of course the QRMs, those requiring 20 percent down, were meant to be a small part of the market – the exception, not the rule.  Mortgages that banks kept in their own portfolios or securitized while retaining 5 percent of the risk could have lower down payments."[12]  The logic behind that statement is simple.  Meaningful risk retention requires that only loans highly unlikely to result in default should be exempt from the rule.  For all other loans, the securitizer must be confident enough in the quality of the securitized loans to be willing to put its own skin in the game.  This, in turn, requires lenders who anticipate selling their loans to securitizers to exercise reason and good judgment in deciding to whom they will lend.  It therefore encourages a return to preeminence of what was, until a couple decades ago, the cornerstone of loan underwriting: the ability and willingness of the borrower to repay the loan.

Writing last year, former Chairman Bair expressed her concern, prophetically, that regulators "will succumb to political pressure and lower the QRM standard."[13]  Proposing to define QRM as co-extensive with the CFPB´s QM definition does exactly that by eviscerating the ability-to-repay criterion and replacing underwriter judgment with a new, de facto government standard for mortgage lending.  A meaningful definition of QRM would result in two general categories of mortgages:  those so high in quality that the risk retention requirement would be unnecessary, and those for which a lender and securitizer have exercised judgment to deem them worth making and on which the securitizer is willing to risk taking losses.  The QM definition, however, removes all incentive to perform meaningful borrower-level reviews to determine, on a case-by-case basis, whether a loan is likely to be repaid and, therefore, should be made.  Instead, it focuses almost exclusively on the product features of the loan, essentially permitting lenders and securitizers to create loans with the QM imprimatur at will, regardless of the borrower´s ability and willingness to repay, merely by avoiding "gimmicks" such as negative amortization and balloon payments. 

A NEW GOVERNMENT SEAL OF APPROVAL

And make no mistake – the QM, and now the QRM, label will carry with it exactly the type of government imprimatur that, in the context of credit ratings issued by nationally recognized statistical rating organizations (NRSRO), created the massive moral hazards that were a critical cause of the financial crisis.  In the case of ratings, the implicit seal of government approval conveyed by the designation of a rating agency as an NRSRO led to the securitizers of loans and the buyers of those securitized products both eschewing the application of their independent judgment in favor of rote reliance on what we now know to have been tragically flawed ratings.  Likewise, by adopting a single QM/QRM standard based on an unrealistic and dangerously broad use of the term "qualified" and exempting securitizers of QM/QRM loans from the risk retention requirement, we are delivering exactly the same type of implicit endorsement that led to the massive expansion of subprime RMBS.  The QM designation, so to speak, is the new NRSRO triple-A rating.

In addition, the sheer breadth of the definition will impede the creation of a pool of non-QRM loans large enough to support securitizations, and it could have a profoundly negative effect on the mortgage lending industry.  A tightly defined QRM standard would ensure that lenders, based on their individual analysis of potential borrowers and their determinations of which of the much larger pool of non-QRM borrowers are capable of repaying their loans, would continue to provide mortgages to borrowers who do not meet the high standards of QRMs.  The QM definition, however, essentially establishes a new, government-mandated standard low enough to encompass the vast majority of loans.  It is difficult to see the incentive for a lender to provide a mortgage to an individual who couldn´t even meet that low standard, especially given that such a mortgage would not be subject to the protections offered to lenders for loans made under the QM standard.  The result will be a single pool of loans, none of which are subject to the risk retention requirement but many of which are likely to end in default.  What is the point of promulgating a risk retention standard and then exempting everything from it?

It is also very likely that all the loans in this single pool will ultimately be owned or guaranteed by American taxpayers through the GSEs.  In May, the FHFA announced that it was directing Fannie Mae and Freddie Mac to limit their future mortgage acquisitions to QM loans,[14] a circular requirement given that any loan that is acquired – or even simply eligible to be purchased or guaranteed – by a GSE is automatically deemed to be a QM under the CFPB´s definition.[15]  Given the likelihood, as I discussed above, that defining QRM as co-extensive with QM would result in a single pool of loans, coupled with the exemptions for and special treatment of loans owned or guaranteed by the GSEs, most notably their automatic designation as QM loans with the protections from liability that attach to that designation, it seems all but inevitable that virtually every loan will be either owned or guaranteed by a GSE.  It is also worth noting that, somewhat redundantly, today´s re-proposal would exempt GSE loans from the risk retention requirements of Section 941 of the Dodd-Frank Act altogether. 

If defining QRMs and QMs coextensively had been the goal of Congress, it could easily have said so in the Dodd-Frank Act.  Instead, the Act requires the agencies to jointly define the term "qualified residential mortgage" in a manner "no broader than the definition of ‘qualified mortgage.´"[16]  While on its face this language does not preclude the agencies from defining QRM in a manner identical to QM, the use of a separate term, the detailed language of the criteria to be considered in defining that term, and the explicit acknowledgement of the separate standard that would be set by the definition of QM certainly, at a minimum, imply that Congress envisioned two distinct definitions.

I am well aware of the fact that, as stated in the re-proposal, "An overwhelming majority of commenters criticized the agencies´ proposed QRM standard."  But rulemakings are not referenda, and while independent regulatory agencies like the Commission must always be mindful of the points raised by commenters, ultimately, they must apply their experience and expertise regardless of the volume of negative comments.  The vast majority of the comments came from particularly interested – that is, not unbiased – parties such as low-income housing advocates and the real estate industrial complex. I respect the points they have raised in support of defining QRM as loosely as the CFPB´s QM designation.  However, while increasing the rate of homeownership is a laudable goal, a return to the failed housing policies that served as the primary driver of the financial crisis is too high a cost to pursue that goal. 

Regulators must have the courage to make the hard decisions that the drafters of Section 941 of the Dodd-Frank Act, as well as policymakers during the pre-crisis years, did not; in this case, that would mean voting against – and, indeed, publicly decrying – any return to the failed housing policies of the past.

A DEEPLY FLAWED PROCESS

As I stated earlier, my dissent is based solely on the substance of the proposal.  However, it is worth noting that the process through which this re-proposal was presented to the Commission represents yet another incursion on the Commission´s independence.  The Dodd-Frank Act, despite placing the risk retention provision in a new section of the Securities Exchange Act, vests responsibility for this joint rulemaking process with the Chairperson of the Financial Stability Oversight Council.[17]  The result was that the SEC´s Commissioners were presented with a fully-baked re-proposing document, given less than a month to review it, and told that they could not make any changes –a take-it-or-leave-it proposition that flies in the face of the Commission´s status as an independent regulatory agency and runs contrary to the deliberative, interactive rulemaking process we are legally obligated to undertake. 

To compound the problem, the re-proposed rule would define the term QRM to mean QM as defined through the CFPB´s implementing regulations as may be amended from time to time­ – meaning that adopting the rule as re-proposed would leave the QRM definition subject to the whims of the CFPB.  I find it unconscionable that the Commission would abdicate to another agency, in perpetuity and under any circumstances, its responsibility and authority to define such a key Exchange Act term, much less to allow that other agency to permanently enshrine the housing policy mistakes that were a central cause of the financial crisis.  And, given that any future amendments to the QM definition made by the CFPB will not be subject to the regular, statutorily-controlled rulemaking process of the Commission, I question, as a matter of administrative law, the validity of any such amendments. 

CONCLUSION

Albert Einstein is said to have been the source of the quote, "Doing the same thing over and over again and expecting different results is the very definition of insanity." I am surprised and disappointed that in voting to issue today´s re-proposal, the Commission as well as the five other federal agencies joining in this rulemaking are prepared to permit the mistakes of the past to be repeated.