Thank you, Mr. Chairman. I am going to be referring to the handout in front of you. It seems to be getting thicker at every meeting. Since the June FOMC meeting, financial markets have been characterized by two distinct phases. Until the middle of July, share prices weakened substantially, and credit spreads widened. The financial sector’s difficulties were at the forefront as housing- price declines continued to pressure this sector. The IndyMac failure led to uninsured depositors taking losses, and this roiled the regional banking sector. The equity prices of Fannie Mae and Freddie Mac plummeted, and their ongoing viability was called into question. The passage of housing legislation that provided support to the GSEs then led to an improvement in investor sentiment and a modest recovery in share prices. As shown in exhibit 1, financial sector shares led the recovery. However, the overall improvement in the broad market indexes was very modest, both in the United States and abroad (exhibit 2). Moreover, no meaningful improvement was evident in the corporate debt or CDS markets. CDS spreads have not changed much, and spreads of asset-backed securities have begun to widen again (exhibits 3 and 4).
Despite intermeeting news that I would characterize on balance as more bad than good, this news did not trigger the type of risk-reduction spasms by investors that have sporadically plagued financial markets over the past year. Exhibits 5 and 6 compare the correlation of daily asset price changes across a broad array of asset classes in July to the corresponding period in March. The blue boxes denote correlations with absolute values of more than 0.5. As can be seen, asset price movements have become much less correlated.
Although the mood is slightly improved today compared with a few weeks ago, the underlying news, especially from the financial sector, remains quite bleak in most respects. In particular, there is little conviction that financial shares have reached a bottom. This can be seen in the unusually high volatility of financial share prices (exhibit 7) and the positive skew in options prices for financial firms, in which the price of a put has been much higher than an equivalent out-of-the money call (exhibit 8). Financial market participants are paying more to protect the downside than to participate on the upside.
Merrill Lynch’s recent experience is reflective of the challenging environment faced by financial firms. Merrill Lynch raised new equity capital and announced that it had sharply reduced its net ABS CDO exposure. Investors were initially pleased that the company had bitten the bullet, and the share price rallied in response. But further consideration tempered the initial enthusiasm—and there are more articles on this in the Wall Street Journal today. A closer look revealed some troubling aspects of the transactions. First, Merrill Lynch took an additional $4.4 billion of CDO write- downs from the June quarter-end valuation date. Merrill Lynch sold CDO exposures with a par value of $30.6 billion to Lone Star for $6.7 billion. At quarter-end, these positions had been carried on the books for $11.1 billion. Second, this transaction resulted in a drop in net CDO exposure of only $1.7 billion because Merrill Lynch provided 75 percent nonrecourse financing to Lone Star. Merrill Lynch gave away all the upside on these assets in exchange for a payment equal to 6 percent of their par value. Third, Merrill Lynch’s equity issuance reportedly resulted in 38 percent dilution to existing shareholders. The dilution was exacerbated by the terms of an earlier share issuance agreement with Temasek, a Singapore sovereign wealth fund. This agreement granted Temasek a “make whole” provision if, within 12 months, common stock was issued at a price below Temasek’s $48 per share investment. This resulted in Merrill Lynch’s paying $2.5 billion to Temasek, which Temasek then rolled into a new $3.4 billion share investment.
Over the past month, the troubles of Fannie Mae and Freddie Mac have taken center stage. Rising loan delinquencies for prime single-family mortgages caused share prices to plunge. This eroded confidence that the firms would be able to raise new equity capital and raised concerns about the viability of these firms. This, in turn, intensified the downward pressure on share prices. As a result, investors began to lose their enthusiasm for the firms’ debt. Investors in short-term discount notes were uninterested in taking on any potential credit risk. As a result, issuance volumes fell, and discount note rates climbed (exhibits 9 and 10). Some investors in the firms’ longer-term debt obligations—including some major foreign central banks—became unwilling to add to their long-term agency debt and agency MBS positions, and one or two of the central banks actually cut their positions somewhat. However, longer- term debt spreads did not change much because the loss of central bank demand was offset by buying from U.S. fixed-income asset managers, who believed that the implicit Treasury support of GSE debt was likely to be hardened (exhibit 11). Fannie and Freddie responded by issuing less debt. To husband their liquidity, the two firms have backed away from purchasing agency mortgage-backed securities for their own portfolios. The removal of this bid was one factor that caused the mortgage basis— the spreads between the option-adjusted yield on agency MBS and other benchmark yields, such as Treasuries and interest rate swaps—to widen significantly (exhibit 12).
The Congress responded by enacting housing legislation that included provisions that hardened the implicit government guarantee and, thus, reduced debt rollover risk. The Treasury now has authority to lend the GSEs an unlimited amount of funds, the magnitude being constrained only by the debt-limit ceiling. In response, discount note issuance costs have fallen in the most recent auctions. However, enactment of the legislation has not generated any comparable narrowing in the mortgage basis or resolved the longer-term outlook for the GSEs. The mortgage basis remains wide in part because Fannie Mae and Freddie Mac have few incentives to expand their balance sheets. Although their regulatory capital is still well above minimum levels, these capital standards are under review. Moreover, because further losses are likely in coming quarters, it is unclear how long this excess capital will be available to support portfolio growth. Of course, the two firms could respond by issuing new equity. However, the low level of these companies’ share prices makes this option unattractive. To raise sufficient funds to ensure long-term viability would cause massive dilution for existing shareholders. To put this in perspective, the current market capitalization of Freddie Mac is only about $5 billion. This compares with a book of business in terms of its portfolio and guaranteed book of $2.2 trillion. Because the GSEs will likely remain reluctant to expand their balance sheets in the near term, the mortgage basis will probably remain elevated, keeping mortgage rates high. This will intensify the downward pressure on housing activity and prices, which in turn will lead to greater loan delinquencies and losses. The consequence will likely put further pressure on Fannie’s and Freddie’s capital positions. So what’s the bottom line? In my view, the legislation has helped to avert—at least for now—a meltdown in the agency debt and agency MBS markets. But the passage is no panacea for ensuring the viability of Fannie Mae and Freddie Mac or in enabling the two firms to provide significant support to the U.S. housing market.
One consequence of the GSE-related turbulence was a temporary pickup in demand in the most recent schedule 1 TSLF auction, on July 25 (exhibit 13). As term mortgage agency repo spreads widened relative to term Treasury repo rates, the cost of borrowing via the TSLF became more attractive. This illustrates how the TSLF program can act as a shock absorber and reduce volatility in term repo rates. With the exception of the July 25 auction, the TSLF auctions continue to be undersubscribed with relatively stable bid-to-cover ratios.
In contrast to the turbulence evident in the financial sector, the bank term funding markets have been relatively stable since the June FOMC meeting. As shown in exhibits 14 and 15, the spreads of one-month LIBOR and three-month LIBOR to OIS remain elevated in the United States, Europe, and the United Kingdom. However, this masks the fact that forward funding rates appear to have risen significantly. As shown in exhibit 16, the forward three-month LIBOR–OIS spread has risen about 20 basis points over the past three months. This spread is now anticipated to remain elevated at around 50 basis points on a one-to-two-year time horizon, indicating that market participants expect term funding pressures to persist for the foreseeable future. Before the crisis, the spread was about 10 basis points.
The U.S. TAF auctions also show a stable trend. As shown in exhibit 17, the bid- to-cover ratio remains around 1.2 to 1, and the stop-out rate has been quite steady over the past five auctions. In contrast, as shown in exhibit 18, the bid-to-cover ratio for the ECB dollar auction continues to climb. As I noted in an earlier briefing, part of this rise reflects the fact that the ECB auction is noncompetitive. The bids are prorated, and the banks pay the U.S stop-out rate. Larger bids by European banks in the ECB auction do not affect the interest rate they pay for such funding, and that encourages more-aggressive bidding. Conversations with the ECB staff indicate that they are concerned that the outcome could be a bidding spiral. Individual banks could keep raising the size of their bid submissions to ensure a stable amount of dollar funding. It is possible that these pressures could eventually encourage the ECB to switch to a Swiss National Bank type of multiple-price auction. This would eliminate the incentives to bid more and more aggressively on the part of the European banks. However, such a change probably would result in a higher stop-out rate in the ECB auctions compared with the United States or Switzerland. ECB officials might not be fully comfortable with such an outcome.
Early reactions by primary dealers and depository institutions to the two refinements to our liquidity facilities—the $50 billion program of options on the TSLF and the introduction of the longer, 84-day, maturity TAF auctions—have been favorable. We recently—last Friday and this Monday—completed an extensive set of interviews with the primary dealer community about the TSLF options program and will be proposing final terms, within the set of parameters approved by the FOMC on July 24, by the end of this week. Of course, we will keep you fully apprised as we go forward on this.
As you know, our liquidity facilities have placed significant demands on the Federal Reserve’s balance sheet, as the Chairman mentioned. As the liquidity facilities have been expanded, we have reduced the size of our Treasury portfolio. We do this to drain the reserves added by our liquidity programs. The use of our balance sheet to sterilize these reserve additions has raised questions about whether sufficient capacity is still available to meet prospective demand—especially a large unanticipated rise in PDCF or PCF borrowing. Exhibit 19 illustrates the transformation of the Federal Reserve System’s balance sheet over the past year, out of Treasuries into non-Treasury lending. As shown in exhibit 20, the non-Treasury portion consists mainly of $150 billion of TAF loans, the $62 billion (in steady state) of foreign exchange swaps executed with the ECB and the SNB, and our $80 billion 28-day maturity, single-tranche repo program. Although the amount of Treasuries held in the SOMA portfolio still totals $479 billion, a majority of these securities are encumbered in one way or another. As shown in exhibit 21, from that $479 billion we need to allocate $45 billion of Treasuries to collateralize the foreign central bank repo pool, retain $35 billion of on-the-run Treasury securities to keep available for our traditional Treasury securities lending program, and set aside $175 billion for the TSLF program and now an additional $50 billion for the TSLF options program (TOP). When the options program is included, we have about $174 billion of unencumbered Treasuries available to offset additional PCF and PDCF borrowing or to fund further expansion of our liquidity programs. Obviously, further expansion of our TAF or TSLF auctions or single-tranche repo operations would be at our discretion and, thus, does not pose a meaningful problem in terms of reserve management. We wouldn’t expand these programs if we didn’t have the ability to conduct offsetting reserve draining operations. However, what would we do if faced with a huge rise in PCF or PDCF borrowing? An inability to drain the reserves added by such lending would cause the federal funds rate to collapse below the target.
Fortunately, we have a number of alternatives that would enable us to offset very large demands for PCF or PDCF borrowing. First, we could sell our remaining unencumbered Treasury holdings or use them to engage in reverse repo operations with the primary dealer community. This could be augmented by the $35 billion of on-the-run Treasury securities currently set aside for securities lending. Together, these two sources could be used to drain more than $200 billion of reserves. Second, we have made arrangements with the Treasury so that, if the need arises, the Treasury would issue special Treasury bills into the market on our behalf and take the proceeds and deposit them at the Federal Reserve. Putting the proceeds of such T-bill sales on the Fed’s balance sheet would drain reserves from the banking system. The potential scope here is large. The housing legislation raised the debt limit substantially. There is now about $1.2 trillion of headroom under the debt limit compared with only about $400 billion previously. Third, we continue to press for legislation that would accelerate the timing of the Federal Reserve’s authority to pay interest on reserves. Being able to pay interest on reserves would put a floor under the federal funds rate. In this case, an inability to drain additional reserves from the banking system would not result in the federal funds rate collapsing toward zero. Finally, we continue to explore the legal and operational feasibility of expanding our balance sheet in other ways. For example, could we engage in reverse repurchase transactions using the collateral obtained from our single-tranche repo and from our TSLF operations?
I wouldn’t say I am confident that we can handle any eventuality—after all, the triparty repo system provides trillions of dollars of funding to the primary dealers. In the unlikely event that it all came to us, we wouldn’t have the capacity to fully offset it at present. But we could accommodate hundreds of billions of dollars of demand if that proved to be necessary. That said, I want to go on record that any very large unanticipated demand for funding from the Federal Reserve by dealers or depository institutions might take a few days or more to offset by reserve-draining operations. Thus, in such a circumstance, the federal funds rate could temporarily trade below its target.
Turning now to interest rate expectations, monetary policy expectations have reverted back to the very slow path toward tightening that was evident before the April FOMC meeting. As shown in exhibits 22 and 23, the federal funds rate and Eurodollar futures curves have shifted down sharply since the June FOMC meeting. Our current survey of the primary dealers shows an even slower anticipated pace of tightening. As shown in exhibit 24, the average of the dealer forecasts in our most recent survey anticipates no tightening until the second quarter of 2009. But the change in the dealers’ forecasts since the June FOMC meeting is more modest than the shift in market expectations (compare exhibits 24 and 25). Looking at the probabilities of different rate outcomes implied by options on federal funds rate futures in exhibits 26 and 27, one sees that there has been a steady trend upward in the probability that the FOMC will keep its policy target rate unchanged at both the August and the September FOMC meetings. Note also that the probability assigned by market participants to further easing is lower than the probability assigned to tightening.
The past month has been marked by a significant decline in commodity prices. As shown in exhibit 28, although the energy complex has led the way down, agriculture and industrial metals prices have also declined significantly. These declines have spurred a large decline in breakeven rates of inflation measured by the spread of nominal Treasury and TIPS yields (exhibit 29). As shown in exhibit 30, longer-term market-based indicators of inflation expectations have increased a bit. Both the Barclays’ and the Board staff’s measures of five-year, five-year-forward inflation implied by nominal Treasury versus TIPS yields have drifted upward since the June FOMC meeting. However, both measures remain well below the peaks reached in early March.
There were no foreign operations during this period. I request a vote to ratify the operations conducted by the System Open Market Account since the June FOMC meeting. As always, I am very happy to take any questions.