As prepared for delivery
New York - Kevin, thank you so much for that kind introduction and for inviting me to speak at today’s conference. After all the conversations with you and your colleagues as part of the credit rating agency exercise we released in February, it is the least I could do. Fitch was a great pleasure to work with, and I urge those of you who are not familiar with the exercise to find it on the Treasury web site.
It has been almost a year now since Secretary Lew announced that Treasury would take an active role in helping to catalyze the development of a vibrant, responsible private label MBS channel. As part of these efforts, we have engaged with countless market participants including investors, issuers, servicers, due diligence firms, trustees, and the credit rating agencies to learn about the key obstacles facing the market and how to best address them.
And let me just say now that obstacles are many and consensus solutions hard to come by. But we have viewed Treasury’s role in helping re-start this market as a high priority, using our good offices and convening powers as important means of illuminating structural issues that need to be addressed and encouraging market participants to directly engage each other to resolve them.
One such obstacle we saw was the lack of clarity around how the credit rating agencies would evaluate more diverse pools of mortgage collateral that were being rated and brought to market and consequently, what the economics of originating and securitizing such collateral would be. The quality of loans that collateralize post-crisis non-agency securitizations has generally been exceptionally high, and as a result, subordination levels on these transactions have not been particularly informative in evaluating the viability of an expanded private label MBS channel.
To address this issue, we invited the credit rating agencies to evaluate the collateral underlying Freddie Mac’s STACR 2014-DN3 transaction, which referenced the most recently originated collateral at the time the exercise was initiated, as if it collateralized a non-agency securitization. The exercise resulted in some interesting insights as highlighted in an Urban Institute analysis.
The first is that greater diversification would potentially reduce the required subordination to within the range observed on securitizations of jumbo collateral. The characteristics of the loans underlying the referenced transaction are weaker than those of loans that have collateralized post-crisis non-agency securitizations. And loss severities are projected to be higher given the lower balances of the referenced loans.
However, the greater geographic diversification was found to potentially yield lower subordination requirements. Furthermore, the credit rating agencies found that super conforming higher down payment loans would potentially have rating cuts in line with post-crisis jumbo loan securitizations. Both of these outcomes should give market participants greater confidence in the ability of the non-agency market to support a broader spectrum of collateral.
But while this exercise provided valuable information about the potential subordination levels those bonds in a securitization would be required to have in order to achieve various ratings, it said little about the returns that institutional investors would demand in order to purchase these bonds. Subordination levels are an important part of the equation, but if the returns that investors demand to own bonds are too high – whether due to general risk aversion or particular to this asset class – the economics of private label securitization will not make sense.
Indeed, what we learned from our initial engagement with market participants is that demand for the senior-most bonds in new issue private label MBS has been severely diminished because investor confidence in transaction parties evaporated in the wake of the financial crisis. The risk premium that these investors assign to the sector today is prohibitively high, making securitization an unattractive financing channel and resulting in the retention of many newly originated loans in whole loan form in portfolio.
To address this crisis of confidence, Treasury has worked to create an open, neutral forum in which to discuss deficiencies in the legacy securitization model and develop creative structural solutions. As part of our benchmark transaction initiative, stakeholders including institutional investors, issuers, servicers, and due diligence firms have been participating in weekly conference calls devoted to issues including servicing oversight, contractual breach enforcement, investor communication, and transparency.
The participants in these work streams have dedicated substantial time and effort to preparing discussion materials, deliberating with colleagues internally, and sharing their expertise and viewpoints on these calls. To me, this is clear evidence that there is a strong interest among market participants to bring back a more vibrant private label MBS market.
The revitalization of this market is something the Administration supports, as well. A thriving non-government supported securitization channel has the potential to stimulate competition, expand consumer choice, and drive down borrowing costs. And last November, I first proposed the concept of a benchmark transaction as a means of catalyzing the resurgence of this market.
To refresh your memory, the rationale behind this idea is that a transaction of sufficient size that is the product of a collaborative effort between investors and transaction parties would, by definition, embrace the specific structural reforms necessary to bring back senior bond investors and thus become the de facto industry standard. Since then, Treasury has facilitated a frank dialogue with some of the largest institutional investors in the market to identify legacy failures and develop structural reforms that would give them sufficient confidence to return to the market.
And so today, I would like to share with you some of the progress that has been made in these discussions, focusing on the role and responsibilities of a transaction party we refer to as the Deal Agent. This is not an entirely new concept. In contrast to the legacy regime, most post-crisis RMBS transactions have entailed more robust mechanisms for identifying and enforcing contractual breaches.
For example, some transactions have explicitly relied on an independent evaluator to review loans for breaches of representations and warranties at the direction of a controlling certificate-holder or the trustee, while others have relied upon a Transaction Manager who automatically reviews loans using specific, predefined procedures and documents – the so-called “reps and tests model” – upon the occurrence of certain triggers such as serious delinquency or loan liquidation at a loss.
Relative to the legacy securitization regime in which potential breaches of representations and warranties were rarely investigated or enforced in a timely manner, the introduction of these new parties and procedures represents real progress. However, our engagement with some of the largest institutional investors who have fled this market led us to conclude that further improvement was needed before they would feel sufficiently comfortable and confident to return to the market.
The now-widely recognized structural deficiencies in legacy private label securitizations that came to light during the financial crisis shattered the trust of market participants, leaving a wake of painful losses and protracted litigation. Ineffective enforcement mechanisms, weak or no oversight of transaction parties, poor investor communication, and lack of transparency all contributed to a loss of confidence among the largest investors in the asset class.
The poor governance of legacy securitizations only compounded the issues inherent to securitization arising from the misalignment of interests and asymmetries of information rather than working to correct it.
It is no surprise then that the principal precondition for investors to return to the market is a greater sense of accountability to their interests and rights. Initially, this led to support for trustee fiduciary duty, but as I have said on more than one occasion, we have concluded that the trustee is the wrong party to assume such a duty going forward.
The core competency of trustees is in carrying out administrative functions, not in investigating or making determinations about contractual breaches. These activities require subjectivity and judgment, which are ultimately qualities that a fiduciary must embody. In order for a trustee to execute its administrative functions to the satisfaction of investors, it should be sufficient to draft strong contracts with clearly defined duties. Investors made it clear to us, however, that something more was required.
Institutional investors consistently held to the view that there needed to be a transaction party whose role would extend beyond prescribed contractual duties like investigating and enforcing breaches of representations and warranties. Given the fragmentation of investors and their historical difficulties in effectuating actions by transaction parties, they argued for a new entity to oversee trust activities that would be independent and represent their interests, somewhat akin to a corporate board of directors.
Investors in private label MBS have grown all too familiar with unpleasant surprises and need someone empowered to address these unanticipated issues and circumstances as they arise on their behalf. The directors of a corporation do not always know how business conditions will evolve or what unforeseen developments will arise; they are expected to use their expertise and judgment to determine the most prudent course of action in any future scenario. Similarly, investors seek a more open-ended mandate for this third party, whom we refer to as the Deal Agent.
Under corporate law, directors must discharge two primary fiduciary duties: Duty of Care and Duty of Loyalty. In the context of private label securitizations, these duties seemed sensible and logical to us. Consistent with the Duty of Care, investors expect the Deal Agent to exercise expert judgment in its decision-making with an objective of maximizing the value of the assets to the trust as a whole.
In addition, consistent with the Duty of Loyalty, investors expect the Deal Agent to act in their best interests and be free from their own conflicts of interest. In this highly litigious environment, however, firms that take on the role of Deal Agent should also have some assurance that they will not be held liable for every judgment call that turns out to be wrong after the fact from the standpoint of not maximizing value of the trust.
It is reasonable to expect that, like a director of a corporation, a Deal Agent should not suffer legal action arising simply from a poor decision. Generally, corporate board members are protected by the business judgment rule, which shields them from liability as long as they acted in good faith with the best interests of the corporation in mind and on an informed basis.
Similarly, serious consideration should be given to developing a sustainable framework within MBS transactions that gives investors a comparable level of confidence that their interests are being protected without leaving Deal Agents exposed to unlimited legal risks that could once again leave the sector mired in costly litigation.
Beyond exercising expert judgment in unforeseen situations, the Deal Agent would also have more prescribed duties. As in recent securitizations, the Deal Agent would be responsible for the investigation of breaches of representations and warranties. In addition to initiating a review in the event of predefined triggers, the Deal Agent would also be expected to perform periodic sampling and would be empowered to perform additional reviews based on its discretion.
A key evolution of the Deal Agent’s role that is emerging from our market engagement that differs from third party investor representatives in some current PLS deals is in the realm of servicer oversight. Just as the directors of a corporation oversee a firm’s business activities as carried out by management, the Deal Agent would also oversee the activities of a securitization trust including importantly, servicer performance.
Servicers would be evaluated according to a set of transparent and objective key performance indicators that would be clearly identified in the deal documents. As you all know too well, market conditions evolve, and there should be scope for transaction parties to revisit these key performance indicators over the life of a securitization in order to ensure that they continue to properly capture the prevailing environment.
The Deal Agent would hold the servicer accountable for underperforming based upon these metrics, performing more active oversight, developing remediation plans, and ultimately, having authority to recommend servicer termination in the event of chronic and uncured failures. In order to improve transparency, the Deal Agent would also serve as an intermediary between investors and the servicer, facilitating the transmission of questions and answers regarding servicing activities and performance.
Finally, investors have been frustrated by mismanagement of trust assets that has been manifested in questionable expenses and erroneous cash flow allocations. To address this issue, the Deal Agent would also be expected to play a proactive role in vendor management and monthly cash flow reconciliation.
Taken together, this expanded suite of duties and responsibilities of the Deal Agent with respect to investors as a whole should improve trust governance and counteract the inherent misalignment of interests among transaction parties. By acting with Duty of Care and Loyalty to investors, the Deal Agent can help overcome their natural fragmentation and better protect their interests.
Accordingly, credit rating agencies can consider whether these enhanced investor protections through the activities of a Deal Agent might help reduce losses sustained by investors and consequently, merit reduced subordination requirements.
If performed by an independent party with sufficient expertise and accountability, the role of the Deal Agent may yield economic benefits similar to qualitative considerations in use today. Similarly, with enhanced protection of their interests, investors should reconsider the risk premium they currently assign to the asset class, which is one key reason why the economics do not favor private label securitization today.
As important as the negotiation of transaction terms – of which I have provided a brief glimpse – the benchmark transaction process has reset relationships among transaction parties and is fostering productive dialogue among market actors that have been largely estranged since the financial crisis. It is through these renewed lines of communication that the slow task of rebuilding trust is progressing.
Treasury hopes to keep that momentum going in the days ahead by hosting those participants who have helped develop the terms and standards in the benchmark transaction process that would underlie a potential future issuance for further discussion. These investors, issuers, servicers, and due diligence firms who have been coming to the table in good faith over the last four months and with a genuine interest in helping this market overcome its historical failures, will discuss their emerging areas of agreement.
While a broad consensus around the role and duties of the Deal Agent is evidence that this effort is bearing fruit, much work remains to be done before a live transaction can be consummated.
Substantial issues remain outstanding such as the selection and compensation of the Deal Agent and the development of standardized representations and warranties, and we will use the Treasury meeting to develop a consensus on the next steps that need to be taken in the industry’s efforts to bring to market a benchmark transaction, thereby helping to revive a non-government guaranteed mortgage-backed securitization channel.
Again, I would like to thank Fitch for having me here today, and I look forward to taking a few of your questions.