Thank you, Mr. Chairman. Financial conditions have worsened considerably in recent days. Credit spreads have widened, equity prices have declined, and market functioning has deteriorated sharply. Although there are many factors that can be cited to explain what we are seeing—including the acute weakness in the U.S. housing sector, a deteriorating macroeconomic outlook, and the loss of faith in credit ratings and structured-finance products—we may have entered a new, dangerous phase of the crisis. Major financial intermediaries are pulling back more sharply and along more margins than previously—shrinking their collateral lending books and raising the haircuts they assess against repo collateral. For a time, this adjustment was occurring in a relatively orderly way, but we appear to have passed that point about ten days ago. The failure of Peloton—a major hedge fund—and the well-publicized problems of Thornburg Mortgage and Carlyle Capital Corporation in meeting margin calls have triggered a dangerous dynamic.
That dynamic goes something like this: Asset price declines—say, triggered by deterioration in the outlook—lead to margin calls. Some highly leveraged firms are unable to meet these calls. Dealers respond by liquidating collateral. This puts downward pressure on asset prices and increases price volatility. Dealers raise haircuts further to compensate for the heightened volatility and the reduced liquidity in the market. This, in turn, puts more pressure on other leveraged investors. A vicious circle ensues of higher haircuts, fire sales, lower prices, higher volatility, and still lower prices, and financial intermediaries start to break as a liquidity crisis potentially leads to insolvency when assets are sold at fire sale prices.
This dynamic poses significant risks. First, it impairs the monetary policy transmission mechanism. We have seen that in recent weeks in the sharp widening between mortgage rates on an option-adjusted basis and Treasury bond rates. Second, as hinted at above, there is a systemic issue. If the vicious circle were to continue unabated, the liquidity issues could become solvency issues, and major financial intermediaries could conceivably fail. I don’t want to be alarmist, but even today we saw double-digit stock price declines for Fannie Mae and Freddie Mac. There were rumors today that Bear Stearns was having funding difficulties: At one point today, its stock was down 14 percent before recovering a bit. Third, the problems in one financial market disturb others. We have seen the problems move from subprime to alt-A mortgages to jumbo prime mortgages and now even agency mortgage-backed securities. Commercial-mortgage-backed security spreads and corporate credit spreads have also widened, and we have seen considerable distortions in the municipal market.
The deterioration in market function can be seen in a number of ways. First, term funding spreads have widened back out. For example, the one-month LIBOR–OIS spread today is 56 basis points, up from its low point in 2008 of 16 basis points, which was reached in January. Second, haircuts for residential MBS have increased sharply, and if anything, the rate of deterioration in terms of haircuts has accelerated markedly in the last week. Third, bid-asked spreads for transactions on many types of financial instruments have widened, indicating a growing liquidity problem in the market.
To address these issues, the Federal Reserve has responded by increasing the size of the TAF program and by implementing a large, term, single-tranche RP program. Together, these two programs will likely cumulate to total outstandings of about $200 billion. In addition, as the Chairman mentioned, the ECB and the SNB today have submitted requests to increase their foreign exchange swap draws and restart their term funding auctions. But there are limits to what these programs can do. The TAF provides liquidity only to depository institutions—this liquidity is not necessarily passed on readily to primary dealers and to other financial institutions. Although term RPs do provide some assistance to primary dealers, these operations are limited to the highest quality collateral—Treasuries, agencies, and agency mortgage-backed securities. Moreover, as both programs are scaled up, there is a large impact on reserves that must be offset by Treasury redemptions, sales, or reverse repurchase operations. Frankly, there are limits to our ability to adjust our portfolio quickly without our actions becoming a source of disruption to financial markets.
For this reason, the staff has proposed a new facility, the term securities lending facility, or TSLF. A memo from the New York Fed staff and a term sheet were circulated to the FOMC earlier today, and these documents discuss in some detail this proposal. Let me give a summary of what I see as the most important points. In brief, this facility would expand the Federal Reserve’s securities lending program for primary dealers by lending securities secured for a term of 28 days, rather than overnight, by a pledge of other securities—Treasuries, agencies, agency mortgage- backed securities, or AAA-rated private-label mortgage-backed securities. The last category is not currently eligible for open market operations (OMO). Currently, our securities lending program is overnight and exchanges only Treasuries for Treasuries. The purpose of this facility is to help alleviate the rapidly escalating pressures evident in term collateral funding markets.
So how would this facility help to accomplish this? By providing the ability to swap illiquid mortgage-backed collateral for Treasury securities, the program would reduce the uncertainty among dealers about their ability to finance such collateral. The expanded supply of Treasuries obtained in the collateral swaps would improve the ability of primary dealers to finance the positions on their balance sheets. This should, in turn, increase the willingness of dealers to make markets across a range of securities. Better market-making, in turn, should lead to greater liquidity for these securities. This, then, should reduce price volatility and obviate the need for dealers to assess higher haircuts against such securities. The liquidity option provided by the TSLF should reduce liquidity risk more generally. The program should help slow, or even reverse, the dynamic process of reduced liquidity, greater price volatility, higher haircuts, margin calls, and forced liquidation.
Why does the staff recommend that the scope of collateral be broader than OMO- eligible collateral? The staff believes that a program based only on OMO collateral could help improve liquidity in those markets. An improvement in liquidity in these core markets could help other related markets. Despite this, the staff recommends that the TSLF go one step further and also accept AAA-rated private-label residential- mortgage-backed securities in this program. The staff believes that it is important to take this additional step because the level of dysfunction in the non-agency mortgage- backed securities market is pronounced, this market is large, and steps to improve market function in this asset class are likely to have positive consequences for the availability and the cost of mortgage finance. In other words, improvement in this area would make monetary policy more effective and would likely generate significant macroeconomic benefits.
To limit the credit risk exposure of the Federal Reserve, the facility for non- OMO-eligible collateral would be limited to AAA-rated residential-mortgage-backed securities assets not on review for downgrade. In addition, the securities would be repriced daily, and appropriate haircuts would be applied against such securities. Why not go further? Although the SOMA lending facility could be extended to include other asset classes such as commercial-mortgage-backed securities, corporates, and municipals, the staff recommends against such a broader extension for two reasons. First, these markets are not under the same degree of duress as the residential-mortgage-backed securities market. Second, adding additional asset classes would increase the operational complexity and risk of the program—for example, by requiring additional auction cycles.
What are the risks of such a program? We think there are several risks that are particularly noteworthy. First, we cannot be sure that the program will have its intended impact. Experience with the TAF suggests that it will, but there is no guarantee of this. Second, the TSLF could increase moral hazard. If the program is successful in preventing losses that would have arisen from an inability to obtain funding, the TSLF would be a form of insurance that could conceivably induce broker–dealers to run smaller liquidity cushions during normal times. Third, Federal Reserve credit risk would increase as the SOMA portfolio accepted lower-quality collateral from primary dealers.
On the first point—Will it work?—the TSLF would be quite large, perhaps cumulating to $200 billion, the same size as the TAF and the term RP program combined. We can make it even bigger if we desire. So we think we have the muscle here to have an impact on term funding markets. On the second point—the moral hazard issue—the staff believes that the TSLF will increase moral hazard somewhat. Ideally, this type of program would also be accompanied by prudential regulation to ensure that primary dealers hold adequate liquidity buffers across the typical business cycle. On the third point—the issue of credit risk to the Federal Reserve—we conclude that there will inevitably be some increase in credit risk. But this risk should be controllable. We know our counterparties, we are not accepting securities that are on watch for downgrades, and the non-agency MBS securities will be AAA- rated. Moreover, we plan to limit the exposure to the more robust AAA-rated MBS by not accepting private-label MBS with CDO-type structures and characteristics. And if securities are put on watch, we can demand substitution.
Fully cognizant of these risks and others—which are outlined in more detail in the memorandum circulated to the FOMC earlier today—we conclude that the benefits are likely to significantly exceed the costs of the program. The staff believes it is important to take those steps necessary to restore the monetary policy transmission mechanism to working order and to short-circuit the vicious dynamic now evident in financial markets. Debby Perelmuter will now describe the TSLF program in more detail, focusing on how it will operate in practice, now that I’ve outlined the theory. Debby.