Transcripts of the monetary policymaking body of the Federal Reserve from 2002–2008.

Thank you. The report distributed to the Committee on October 8 follows up on the January 2001 SOMA study to explore several issues. These include the structure of auctions of term discount window advances and their coordination with Desk operations; the perspective of bank supervisors on the risks posed by an auction credit facility (ACF); the methods proposed to control credit risk to the Reserve Banks; and the effect of ACF advances on Federal Home Loan Bank advances. In addition, at the request of the Crisis Management Steering Committee, the report considers whether an ACF could be used in a contingency as a substitute for the Desk’s open market operations.

The structure of ACF auctions could be quite simple or more elaborate, and there is in fact a fair amount of detail in the report about that. A simple structure would be preferable when first introducing an ACF. The report concludes that auctions of term advances could be easily coordinated with Desk operations and could be effective in supporting the secular growth of currency on the Fed’s balance sheet and handling seasonal swings in the demand for currency and reserves.

In consulting the supervisors about the issue of risk, those supervisors on the working group expressed concern that, if banks were permitted to borrow very large amounts from the ACF without any constraints, that could pose liquidity risks to banks that might create a “too big to get downgraded” issue. That is, examiners might become reluctant to downgrade a large bank’s CAMELS rating if that risked making the bank ineligible to roll over its ACF funding. Supervisors also noted that the Fed would shift from being an occasional lender of last resort to having an ongoing lending relationship with banks under an ACF. This might pose a potential conflict between using an ACF in the implementation of monetary policy and rating banks independently of monetary policy objectives. Supervisors stressed that such concerns could be substantially reduced, though perhaps not entirely eliminated, by keeping an ACF relatively small and by establishing the same set of risk controls that would be applied to reduce the Fed’s exposure to credit risk. Bank risk-taking, risk to the FDIC, and credit risk to the Fed could be limited by establishing a variety of controls that are described in the report. Those controls include restricting eligibility to financially strong depository institutions and imposing limits on how much each institution could borrow.

Even with these risk control measures, however, the report notes that extending ACF advances to a number of depository institutions would increase the likelihood that the Fed would at some point have to deal with an institution whose condition deteriorates so rapidly that FDICIA guidelines apply. Therefore, the report suggests that the Fed would want to consider developing special procedures for dealing with such an event.

A contingency auction credit facility or what we call a CACF would share many features of an ACF, but differences in the purposes of the two facilities and in the circumstances surrounding their use would require some different features as well. Those, too, are discussed in detail in the report.

Compared with backup facilities that duplicate the Desk’s open market operations, a CACF—like other discount window facilities—would broaden the range of counterparties in an emergency as compared with simply the number of primary dealers. In addition, a CACF would not be limited to dealing only in narrow classes of collateral, which could also be important in an emergency.

The report considers two emergency scenarios. In the first, in which financing markets are functioning but the Desk and its contingency facilities are not, a CACF could be an effective backup to the Desk in providing a predetermined aggregate supply of reserves to the banking system. A CACF could be similarly effective in the later stages of a more widespread crisis if many participants in financial markets were able to return to normal operations but recovery of the Desk’s operations were delayed. In the second scenario, in which financing markets are severely dysfunctional at the same time the Desk’s ability to operate is impaired, a CACF would more likely be used to elastically supply funds to depository institutions at a fixed interest rate. In such a scenario, the emergency primary credit facility would likely be as effective as or easier to use than a CACF. The main appeal of the CACF over other types of discount window lending in that type of scenario would come from its ability to visibly provide Fed credit earlier in the day, which might have a more positive effect on bank and market behavior during an emergency.

The report identifies several open issues that would have to be resolved if these facilities were to be implemented, and I will not go through all of those. In concluding, I want to acknowledge the efforts of the people who helped prepare this report. The group included Jim Clouse, Bill Nelson, Michael Martinson, and David Wright of the Board; Spence Hilton, who drafted a large part of the report, Sandy Krieger, and Bill Walsh of the New York Fed; Lois Berthaume from Atlanta; Chris Moore from Cleveland; Bob Hankins from Dallas; and Steve Meyer, who also drafted a large section of the report, from Philadelphia. My thanks to all of them for their work. That concludes my remarks.

Keyboard shortcuts

j previous speech k next speech