Skip to main content

Statement at Open Meeting: Credit Risk Retention

Commissioner Kara M. Stein

Oct. 22, 2014

I want to thank the staff in the Divisions of Corporation Finance and Economic Risk Analysis for their work on this rule, and in particular, Kathy Hsu and Arthur Sandel for your diligence and commitment to this interagency effort, and for the time you spent working with my staff. I sincerely appreciate your efforts, as well as the diligence and thoughtfulness of the numerous other staff members at the other agencies.

The rule we adopt today is the culmination of a joint effort and a lot of hard work by our own staff, as well as staff at the Office of the Comptroller of the Currency, the Federal Reserve System, the Federal Deposit Insurance Corporation, the Federal Housing Finance Agency, and the Department of Housing and Urban Development. It is based on a very fundamental and common sense principle — something we all instinctively know and understand — something we call “skin in the game.” The origin of that phrase is unknown, and speculation on its roots ranges from William Shakespeare to Warren Buffett.[1] But its logic is simple and inescapable: those with a strong enough stake in a particular outcome will be motivated to accomplish that outcome.

In many ways, that same logic actually contributed to the havoc that securitization products helped wreak on our financial markets during the financial crisis. The seeds of this havoc were sown by what is known as the “originate to distribute” model of securitization. Originators, who made the decisions about credit risk, had little or no stake in the outcome of those decisions. That dubious privilege fell to the investors. Investors, however, were several steps removed from the underwriting process, and thus had difficulty detecting issues and problems. Meanwhile, originators and securitizers were compensated based on volume, not quality. They had a stake in volume, and, not unpredictably, this led to the churning out of trillions of dollars of asset-backed securities with far more regard for quantity than quality. Misaligned incentives and information asymmetries combined together to form dangerous risks to investors. When these risks materialized, they both precipitated and exacerbated the financial crisis.[2]

This rule attempts to realign the incentives of securitization market participants back toward quality, back into alignment with investors, and I support it. Essentially, we want those who make the credit risk decisions to have “skin in the game.” That is the simple logic at the core of what we are trying to accomplish with this rule. But that logic is where the simplicity ends.

Securitization is beneficial in numerous ways, including expanding credit availability, lowering the cost of credit, and increasing liquidity. We all want to see this market revive and thrive. To do so, we seek to strike the right balance between overly restrictive risk retention requirements, which could constrain the formation of credit, and overly permissive standards, which would do little to address the problems that the financial crisis laid bare. This is no simple task. That is why I strongly support the periodic review of, among other things, the definition of a Qualified Residential Mortgage (or QRM), and I appreciate the Chair’s leadership in including this review in the rule.

The rule requires this review to occur no later than four years after the rule’s effective date, and every five years thereafter. Importantly, the rule also permits a review at any time upon the request of one of the agencies. We should all remain vigilant as to the need for such a review. In undertaking these reviews, we should take a comprehensive approach, analyzing not just the definition of QRM, but how that definition is working within the larger context of the securitization markets. We should consider, for example, the structures of securitization transactions, the relationship between, and roles undertaken by, the various transaction parties, and any regulatory arbitrage that exists between the private label market and the Government-Sponsored Enterprises — all with a view toward a healthy securitization market, sound investor protection, and financial stability.

I am pleased to see a number of other provisions in this rule. For example, I strongly support the use of the Legal Entity Identifier (or LEI) that is included in this rule. This important global initiative has advanced significantly over the past 18 months. LEIs are now available to market participants through a number of operating utilities worldwide that have been endorsed by the Global LEI Regulatory Oversight Committee.[3] By providing unique identification of parties to financial transactions around the world, LEIs foster greater transparency which can improve risk management at firms, facilitate orderly resolution, promote market integrity, help contain abuse and financial fraud, and support higher quality and more accurate financial data. I would like to commend the Chair here as well for her leadership in supporting the adoption of LEI, and also the other agencies for their support. The partnership between regulators and industry to develop a consistent and integrated approach has accelerated the development of LEI. Legal Entity Identifiers now present a meaningful means for monitoring risks in a fragmented system. The future has arrived, and we should include LEI as fully as possible in every rule and initiative we consider going forward.

While I support the rule, as I stated, these are complicated issues, and I do have some concerns in certain areas. For example, the rule provides a safe harbor for foreign-related transactions that might be exploited if large swaths of U.S. assets are securitized by foreign subsidiaries of U.S. entities and sold to foreign subsidiaries of U.S. entities. Such a practice could represent a purposeful, form-over-substance evasion of risk retention for what is essentially a U.S.-based transaction.

Just as I expressed when we adopted the Cross-Border Security-Based Swap Rules and Guidance in June of this year, we cannot ignore the sometimes inextricable links between U.S. parents and their foreign subsidiaries, and we must guard against permitting hidden risks from these relationships to boomerang risk right back into the United States. In this regard, I appreciate the language in the preamble to this rule which signals that the agencies will be monitoring this safe harbor, and, where appropriate, considering the applicability of the anti-evasion provisions.

Significantly, I note that this rule represents one of several important provisions in The Dodd-Frank Wall Street Reform and Consumer Protection Act (“Dodd-Frank”) designed to address the risks and shortcomings in the securitization market brought into focus during the financial crisis. As we all know, credit rating agencies failed to adequately assess the risks of these complex financial products, and for a number of reasons, including a limited investment decision-making window, investors placed too much reliance on these ratings. We recently adopted significant governance and other changes to our rules for credit rating agencies,[4] and I hope that these reforms, as a package, will ultimately lead to better performance in the rating of securitization products.[5]

We also recently adopted revisions to Regulation AB which require disclosure of more granular, loan-level data in some asset-backed securitization transactions,[6] but more on that front is needed. We need to extend these loan-level requirements to additional asset classes. And, as I have previously stated, I strongly support providing investors with a software engine to aid in the calculation of waterfall models, or an “electronic waterfall.” Waterfall models can help investors assess the cash flows from the loan level data in a more accurate, and potentially less costly, manner. They are also particularly important given that investors only have a three-day window between the preliminary prospectus and the first sale in which to analyze the available data and make an investment decision.

We must also move forward to complete our rule-making mandate under Dodd-Frank Section 621,[7] which would prohibit certain conflicts of interest in securitization transactions, essentially preventing securitizers, at least for a certain period of time, from betting against the products they create at the expense of investors. While Section 621 is directed at different behavior, it supports the same goal as the rule we adopt today — that of promoting sounder underwriting and encouraging better aligned incentives.

Finally, we also must move forward with our fellow regulators on the incentive pay reforms of Dodd-Frank Section 956. Many of the problems in the securitization market emerged from compensation structures that incentivized firms and their employees to engage in poor securitization practices. [8] While “skin in the game” helps focus the attention of management on the quality of the product being securitized, to be truly effective, risk retention must be paired with meaningful incentive pay reforms. Otherwise, employees may place retained risk onto a firm’s balance sheet with little concern for the ultimate performance of those assets. I hope the agencies can move forward on those reforms as soon as possible.

As for today’s rule, it is one step in a larger effort to repair and revive the securitization market, and our task is not complete. I look forward to hearing from the public and working with my fellow Commissioners to do all that we can to support a strong, vibrant and flexible securitization market in which incentives are better aligned, and risks are fully disclosed and accurately assessed. Thank you.

[1] See, e.g., Politi, Mark, Skin in the Game, May 31, 2014,, retrieved October 22, 2014.

[2] See, e.g., Geithner, Timothy F., Chairman Financial Stability Oversight Council, Macroeconomic Effects of Risk Retention Requirements, January, 2011, available at ; Purnanandam, Amiyatosh K., Originate-to-Distribute Model and the Subprime Mortgage Crisis (April 2010) AFA 2010 Atlanta Meetings Paper, available at SSRN: or .

[3] See, e.g., Press release of the Global LEI Regulatory Oversight Committee, available at, dated June 30, 2014; press release of the Global Legal Entity Identifier Foundation, available at, dated June 30, 2014.

[4] See Nationally Recognized Statistical Rating Organizations, Release No. 34-72396, 79 FR 55077, August 27, 2014 (the “NRSRO Governance Rule”).

[5] As I stated in my remarks at the adoption of the NRSRO Governance Rule, however, we are by no means finished with our work in this area. In fact, there is still significant work to be done to address the fundamental conflicts of interest present in both the issuer-pay and subscriber-pay business models of NRSROs.

[6] See Asset-Backed Securities Disclosure and Registration, Release No. 33-9638, 79 FR 57184, September 4, 2014.

[7] See Proposed Rule on Prohibition Against Conflicts of Interest in Certain Securitizations, Release No. 34-65355, 76 FR 60320, September 29, 2011.

[8] See, e.g., Wilmarth, Arthur E., Jr., The Dark Side of Universal Banking: Financial Conglomerates and the Origins of the Subprime Financial Crisis, 41 Conn. L. Rev. 963 (2009); The Financial Crisis Inquiry Report: Final Report of the National Commission on the Causes of the Financial and Economic Crisis in the United States, February 25, 2011, pp. 61-64, available at