

	May 19, 2006 JS-4274 Statement of Under Secretary for Domestic Finance Randal K. Quarles to Bond Market Association Annual Meeting New York, NY-Thank you very much. Let me say first that we at the Treasury greatly value our interaction with market participants and the Bond Market Association and it is a great pleasure for me to be here today to address this distinguished audience. I plan to focus most of my remarks today on Treasury market issues, but I want to begin by again highlighting the Administration's views on the risks posed by government sponsored enterprises (GSEs) and the urgent need for reform. Government Sponsored Enterprises It is now widely recognized that the GSEs have relied upon their funding advantage to expand the size of their retained portfolios far beyond levels necessary to achieve their mission of supporting activity in the secondary mortgage market. To be sure, the mortgage securitization activity conducted by Fannie Mae and Freddie Mac does in fact provide a public benefit by increasing the amount of capital available to support mortgage credit, and broadly enhancing liquidity in the mortgage market. However, the retention of large investment portfolios does not further this purpose beyond what can be achieved through securitization. Indeed, these retained portfolios concentrate the prepayment and interest rate risks associated with mortgages and mortgage-backed instruments in entities that-as a result of the lower levels of capital they are required to hold-are substantially more leveraged than other financial institutions. The concentration of risk inherent in these portfolios along with the GSEs' thin capital structure is an important policy concern and a high priority for the Treasury. The ongoing financial reporting problems at the GSEs only heighten our concerns. The Administration's reform proposals are intended to ensure greater regulatory oversight, enhanced market discipline, and appropriate capital requirements for the GSEs. In order to protect against the systemic risk posed by the housing GSEs' mortgage investment business, the Administration strongly believes that limitations should be placed on the size of the housing GSEs' retained mortgage investment portfolios. The Treasury continues to urge Congress to take action soon to address these issues. We strongly support Senator Shelby's version of GSE reform and remain hopeful that by the end of the year we'll have a stronger regulator for the housing GSEs with the authority and direction to limit the size of their retained portfolios. All Americans have a stake in the success of these efforts, and certainly everyone in this audience. Fixed-income markets would no doubt be at the center of any crisis involving the GSEs, and it is imperative that Congress take action now to reduce this threat to our financial system and economy. Environmental Factors Affecting Treasury Markets Turning now to Treasury market issues, as we all know, the Treasury market is a critical national asset. The market operates remarkably well virtually all of the time, but there have been a few instances over the last twenty years that have been quite disruptive including episodes of attempted manipulative or questionable trading behavior, spikes in delivery fails during periods of very low short-term interest rates, and periods of severe market distress during major crises such as the fall of 1998 and the aftermath of the 9/11 attacks. Unfortunately, there are reasons to believe that the risk of each of these types of problems will not decline over time and may even edge higher. On the potential for manipulative behavior, for example, expansion and consolidation in the financial industry has increased the number of institutions that are capable of taking on very large positions in Treasury markets, raising the risk that a single firm may gain effective control over an issue. Similarly, we might expect that the incidence of very low nominal interest rates and the associated potential for systemic fails episodes could be somewhat higher in the years ahead. Average inflation rates are now quite low in most industrialized countries. As a result, central banks that wish to establish any given level of the real interbank rate will need to need to target a lower nominal interbank rate than in the past when average inflation rates were much higher. In short, we probably should not view the very low nominal short-term interest rates that prevailed in 2003 as a historical aberration-similar economic circumstances in the future may well lead again to very low nominal short-term interest rates. Finally, the risk of terrorist attacks and other disasters is regrettably likely to remain a significant concern for financial markets. A natural question then is how the basic Treasury market architecture might evolve over time in response to changing environmental factors. Current Treasury Market Architecture and Recent Initiatives First let me take a moment to review some of the key elements of the current Treasury market architecture. As you know, the Treasury follows a "regular and predictable" policy in issuing new securities with the goal of minimizing its funding costs over time and enhancing market liquidity. This policy implies that the quantity of securities provided at auction is set with a long-run view and does not respond greatly to short-run movements in market conditions. In addition, the Treasury does not opportunistically reopen securities that are highly sought after in the secondary market. Dealers and others typically finance their positions in recently-auctioned securities in the special collateral repo market. Securities that are in high demand typically trade at specials rates well below the rate on general collateral, but current contractual conventions in the repo market essentially establish a floor on repo rates at zero percent. One implication of this market structure is that the maximum degree of specialness in the repo market-that is, the spread of the general collateral rate over the specials repo rate-is a function of the level of the general collateral rate, which in turn is largely determined by the stance of monetary policy. When the central bank wishes to establish very low short-term rates, the maximum degree of specialness will be quite small. During those periods, we might expect to see greater incidence of fails episodes because the cost of failing is low. Conversely, when central banks wish to establish a high short-term rate, the maximum degree of specialness in the repo market can be quite substantial. During these periods, one might expect to see somewhat greater incidence of questionable behavior in the repo market given that the returns to gaining control over a security are higher. Some recent initiatives by the Bond Market Association (BMA) to promote negative repo rate trading are an important step toward allowing market forces to determine the degree of specialness, even during periods when the general collateral rate is very low. The BMA's recently-issued negative rate repo trading guidelines clarify the treatment of fails on the opening leg of a negative rate repo agreement and are an important milestone but much remains to be done. In particular, widespread negative repo rate trading seems unlikely to develop in the absence of additional changes in the treatment of repo fails along with changes in the treatment of cash market fails. If such changes are implemented, the potential for repo rates to dip into negative territory would be an important factor that could reduce the risk of systemic fails episodes. Other potential market developments such as the BMA's work on a margining system for fails could further enhance the functioning of the repo market and limit credit exposures during systemic fails scenarios. Negative Repo Rates and the Repo Market Architecture The potential for negative repo rates also raises broader questions about the possible evolution of the overall architecture for the repo market. To gain some insight into these issues, it seems useful to step back for a moment to consider the market architecture that most central banks employ in the interbank market for reserves. Central banks often operate by targeting a short-term interest rate. Unexpected demands for reserves, however, can push the level of the interbank rate above the target rate on any particular day. Many central banks maintain a standing lending facility to meet such unexpected demands. The lending rate is usually set at a fixed spread over the central bank target rate. The maximum degree of upside "specialness" in the reserve market is generally capped by the spread of the lending facility rate over the target interbank rate and thus is not a function of the level of short-term interest rates. When reserves are in short supply, this structure relies on market forces to efficiently distribute reserves up to the point at which the market interbank rate reaches or slightly exceeds the lending facility rate. At that point, banks often choose to borrow from the lending facility, thus creating additional supply of reserves to satisfy demand. As you know, the Treasury recently released a strawman proposal for a securities lending facility intended to stimulate public comment and discussion. As outlined in that proposal, a securities lending facility might play a role in the repo market quite similar to that of central bank lending facilities in the reserve market. Consistent with current market conventions, the strawman proposal assumed a lending rate of zero percent at the securities lending facility. In this structure, the specials market would operate much as it does today; the rate for highly sought after securities would fall below the general collateral rate. And when the specials rate dropped all the way to zero percent, market participants might choose to borrow securities from the securities lending facility. Some early market commentary on the strawman proposal has viewed the BMA's work on negative repo rates as a "market alternative" to the proposed Treasury securities lending facility. As emphasized in our public comment document, the Treasury has not taken any position at this point on the desirability of establishing a securities lending facility. But I would note that the BMA's negative repo rate initiatives and the proposed Treasury securities lending facility need not be viewed as competing proposals and, indeed, could well be complementary. For example, in the strawman proposal, market participants would not have a strong financial incentive to borrow from the securities lending facility even when the specials rate reaches zero percent because current market conventions allow market participants to fail to deliver the securities at an effective rate of zero percent. However, if market practices evolve so that market repo rates could be negative, then-just as in the case of central bank lending facilities-the potential for the market rate to move beyond the securities lending facility rate would provide an important financial incentive for market participants to borrow at the securities lending facility. If market participants took the steps necessary to foster active negative rate repo trading, there might even be a rationale for reconsidering the structure of a securities lending facility. For example, again following the example of central bank lending facilities, it might be desirable to establish a securities lending facility rate that was set at a fixed spread below the general collateral rate. In this framework, the maximum degree of specialness in the repo market would then be independent of the level of short-term interest rates. Of course, we are a long way yet from active negative repo rate trading and a Treasury securities lending facility has not even passed the proof-of-concept stage. My point is simply that these potential structural changes and the associated market implications are closely related and need to be considered in tandem. Securities Lending Facility Proposal: Some Potential Risks So far, I've described the possible directions for the evolution of the repo market architecture with something of a 30,000-feet-up perspective. But such grand views often obscure difficult terrain below and that is certainly the case with Treasury securities lending facility. As I have noted, a securities lending facility could potentially play a useful role in the Treasury market, but there are some important caveats to note as well. I will not review all the details of the Treasury's discussion paper here, but I would like to touch on some of the important issues that have surfaced in the early discussions about the merits of such a facility. Moral Hazard and Market Risks One contentious issue concerns the potential risk that a securities lending facility could exacerbate moral hazard. Some have argued, for example, that the establishment of a securities lender of last resort would represent a form of insurance that, in effect, "bails out" investors with short positions and thereby encourages risky trading behavior. Others have argued that the current contractual arrangements in the Treasury market already limit the potential costs of short-sellers. Moreover, the potential for moral hazard already exists, in part because during a systemic fails episode, market participants have been able to obtain regulatory forbearance on the rules applicable to aged fails. Still others have argued that moral hazard concerns may be misplaced because a well-designed securities lending facility would not really be an "insurance policy" that simply redistributes market risks, but rather a mechanism for reducing overall market risks. At this stage, the Treasury hasn't come to any firm conclusions on this point and we hope that public comments on the strawman proposal will be helpful in sorting through some of these issues. Gaming The potential for inappropriate use of a securities lending facility is also a concern. The white paper discusses a number of parameters of a securities lending facility such as forward delivery, prompt disclosure of the aggregate amounts borrowed from the facility, and potential reporting requirements during the period of borrowing that should help to mitigate the risk of gaming. These technical parameters may be effective in limiting the scope for some types of inappropriate usage, but even these safeguards are probably not air tight. The experience of other countries with securities lending facilities in place suggests that gaming has not been a large problem in practice-but we are cautious about drawing strong inferences for U.S. markets from foreign experience. Reopenings Another key question regarding the securities lending facility concerns whether or not the problems that it is intended to address could be resolved nearly as well by other, less intrusive means. For example, some have noted that the Treasury could simply reopen securities on those very rare occasions when systemic risks appear to be particularly threatening. Of course, reopenings are a blunt instrument in that they raise the supply of an outstanding security for a lengthy period and therefore represent a significant departure from the Treasury's regular and predictable issuance policy. Moreover, discretionary reopenings would no doubt prompt a great deal of speculation about the circumstances that would trigger a reopening. On the other hand, some have also noted that a reopening to address only very rare systemic crises could well be effective and would probably not entail the same ongoing concerns about moral hazard and gaming associated with a securities lending facility. Market Surveillance Finally, let me say that even in a world with negative repo rates and a securities lending facility, there would likely still be considerable scope for manipulative behavior in the market. As a result, the need for the type of market surveillance program that is currently in place would remain. As you may know, following violations of Treasury auction bidding rules in the early 1990s, policy makers formed an Interagency Working Group to perform Treasury market surveillance. This group continues to monitor trading in wholesale cash, derivative, and repo markets for Treasury securities. The Interagency Working Group holds routine biweekly surveillance calls to coordinate effort and to share observations and other materials among participants. The Interagency Working Group includes representatives from the Treasury, the SEC, the CFTC, the Federal Reserve Board, and the Federal Reserve Bank of New York. Coordinated surveillance among official entities, each with existing interests in Treasury market functioning and regulation was seen as the most appropriate response to issues raised by Treasury auction rule violations. By detecting anomalous price patterns, reviewing reported positions and trading volumes, and considering commentary from market participants, the members of the Interagency Working Group foster market transparency and efficiency without imposing undue regulatory burdens on the market. Since the end of 2004, questionable behavior in the financing and futures markets has increased including cases involving concentrated positioning and increased control of individual Treasury issues among a small subset of Treasury market participants. Through this period, the Interagency Working Group has shared information across markets and coordinated policy responses. When the Interagency Working Group has identified cases that merit further review, it has forwarded the information to the appropriate enforcement agency for investigation. The increased frequency of situations raising surveillance concerns makes this an opportune time to remind all Treasury market participants that questionable trading behavior has consequences and can result in increased regulatory scrutiny and referrals to enforcement authorities. So for the record, I would urge any traders that establish especially large positions in any Treasury issues to tread very carefully. We understand that very large positions in and of themselves may be benign, but large positions coupled with very tight conditions in cash and financing markets can lead to questions. So consult closely with your managers and compliance officers. And I would also urge senior managers to make sure that you keep a watchful eye out for any questionable activities across any of your trading operations. The Treasury along with our colleagues at the other regulatory agencies certainly will be. Conclusion To sum up, the basic architecture of the Treasury market is evolving in response to changes in environmental factors. Indeed, the Bond Market Association's efforts to facilitate negative rate repo trading is a very significant step in this process. And it may be that a Treasury securities lending facility could also be an important structural change that would enhance the resilience of the Treasury market. As I've mentioned, however, the jury is still hearing evidence on the merits of such a facility. We expect that market surveillance by the Interagency Working Group will continue to be an important function as the market evolves. All Treasury market participants have a responsibility to conduct business in a manner that supports market efficiency, liquidity and transparency; the Treasury along with our colleagues in the Interagency Working Group remain committed to fostering those key objectives. Thank you very much.