BPI Statement for the Record for House Hearing on the Fed Discount Window and Emergency Lending

Chairman Barr, Ranking Member Foster and members of the Subcommittee, thank you for the opportunity to testify today. My name is Bill Nelson, and I am the Chief Economist of the Bank Policy Institute, which is a research and advocacy group supported by banks with more than $100 billion in U.S. assets. Prior to joining BPI, I was a deputy director of the Division of Monetary Affairs at the Federal Reserve Board, where I worked for 24 years. Beginning in 1998, I helped develop the Fed’s discount window policy and eventually oversaw the lending operations of the 12 Federal Reserve Banks, including regular discount window lending to banks as well as emergency lending to nonbanks. In that role, I participated in the major revamp of the discount window in 2003 and oversaw all the Fed’s post-Lehman credit facilities. I have written extensively on the discount window and lender of last resort. I was also one of the architects of the liquidity regulation regime put in place after the Global Financial Crisis.

Congress created the Federal Reserve System in 1913 in large part for two reasons: 1) To provide banks a safe place to keep the reserves backing deposits and currency, and 2) To extend collateralized loans to banks so that reserves can expand and contract as needed by the system as a whole, as well as to provide a bank immediate liquidity when necessary under stress, especially if experiencing a run. Congress recognized that when the Fed stands ready to lend to a bank experiencing a funding shortfall, it allows the bank to devote more of its resources to lending to businesses and households, supporting economic growth, rather than the bank having to hold an excessive amount of vault cash or unproductive reserves. Over time, in most circumstances, the Federal Reserve has shifted to adjusting the aggregate level of reserve balances for monetary policy purposes using cash or repo open market operations of government securities, but discount window loans are still used to provide reserves to the system as a whole if necessary late in the day, and also as the Fed’s frontline tool for addressing banking system strains and financial crises.

Discount window loans are no longer extended as discounts of bank loans, although the name remains. Instead, the Fed extends collateralized advances under the authority granted under Section 10B of the Federal Reserve Act. The loans can generally be extended for maturities of up to four months and must be collateralized to the satisfaction of the lending Reserve Bank. The Fed seeks to take all bankable assets as collateral, provided that it can assign a fair value to the asset and obtain a perfected security interest. The Fed applies conservative haircuts to the fair value of the pledged asset to determine the lendable value – the amount of discount window credit the Fed will provide against that asset. As of February 2020, the last time the Fed provided the information, banks had collateral prepositioned at the discount window with a lendable value of $1.6 trillion. Most discount window collateral is loans rather than securities, with consumer loans alone making up 40 percent of pledged collateral, and business loans another 20 percent. While the prepositioning of collateral has gotten a lot of discussion lately, discount window collateral has been prepositioned for many decades. 

The Fed publishes aggregate data on its balance sheet, including discount window loans, once a week. In addition, the Dodd-Frank Act requires the Fed to publish information on each discount window loan, including the identity of the borrower, after two years.

The FDIC Improvement Act of 1991 restricts the Fed’s ability to lend to undercapitalized and especially critically undercapitalized banks. Simplifying a bit, the Fed generally should not extend a discount window loan to a bank beyond five days after the bank becomes critically undercapitalized.

In addition to extending regular discount window loans to banks, the Fed also extends emergency loans, usually to nonbanks. The loans, which are made under Section 13(3) of the Federal Reserve Act, can only be extended in unusual and exigent circumstances, must be provided through a broad-based facility, and must be secured sufficiently to protect taxpayers from losses. The facility must not be intended to support a specific company and must be approved by the Secretary of the Treasury. The Fed must confidentially provide Congress with information on each Section 13(3) loan, including the identity of the borrower, within seven days. The Fed is required to publish loan-level information one year after the facility is closed. The Fed can also lend to anyone under any circumstances against Treasury and agency collateral under Section 13(13) of the Federal Reserve Act, but in practice this authority is rarely used.

Most central bank lending to banks is unremarkable. Looking outside the United States, loans to banks were initially the main asset of the ECB, following the practice of the Bundesbank before it. Those central banks judged buying government securities to be monetizing the debt and so be inflationary and lending to banks to be the least distortionary investment option available. In the case of the United States, prior to the Global Financial Crisis, discount window lending was an important tool for monetary policy implementation. If the Fed inadvertently left the system short of reserves on any given day, the fed funds rate would rise up until some bank borrowed, creating more reserves. The fact that some bank borrowed was not the result of that specific bank’s liquidity situation. Smaller banks borrowed to address short-term funding needs, such as might be caused by losing a large municipal deposit, for example.

To be sure, the discount window is also and has always been the Fed’s first line of defense against broader financial turmoil. Generally, the first action the Fed has taken when systemic financial strains materialize is to remind everyone that the discount window is open and operating and available for banks to use. Doing so reassures banks about their own, and their counterparties’, abilities to secure the funding necessary to meet their obligations. That greater confidence reduces the likelihood that banks will curtail lending under stress, which would amplify the stress and reduce the supply of credit for businesses and households. The Fed has also provided discount window loans to individual banks that are experiencing funding strains, usually just temporarily, but in some cases so that the bank can operate until the FDIC shuts it down.

In part because of the dual nature of discount window lending, both unremarkable monetary policy tool and source of contingency funding, there has been a stigma associated with borrowing from the discount window at least since the 1920s. Tapping your contingency funding inevitably suggests something has gone wrong. Former Fed Governor Betsy Duke, who had been president and CEO of a community bank before becoming a governor, described borrowing from the discount window as like borrowing from your parents – you’ll do it if you have to, but nobody likes doing it. Several bankers have told me that even if the Fed assures a bank that it is OK to borrow, the bank’s examiners will subsequently take the borrowing as indication that there is a problem with the bank’s liquidity risk management.

Stigma got much, much worse in the aftermath of the GFC when borrowing from the discount window was equated with having received a bailout, even though the loans were fully collateralized, extended at an above-market rate and all repaid on time with interest. Because of the pillorying banks that borrowed received, many banks will now not borrow under any circumstances except perhaps if there is an obvious glitch affecting the entire payment system. Stigma was also worsened by the requirement in the Dodd-Frank Act that borrowing be made public, even with a two-year lag. One bank treasurer told me that if he borrowed from the discount window there would be two phone calls, one from the president of the local Federal Reserve Bank to the bank’s CEO asking why the bank borrowed, and one from human resources to him telling him to clean out his desk.

Consequently, one of the Fed’s most important monetary policy and financial stability tools doesn’t work well. While it is tempting to conclude that stigma is irrelevant because a bank will borrow if it has to, that is incorrect for two reasons. First, in many cases it is not the borrowing bank that is experiencing the liquidity strains but rather a third party, and the Fed is seeking to get banks to intermediate to that third party. That indirect approach will not work if banks won’t borrow. Second, a critical purpose of the discount window is to make banks confident that they will have the funds they need to meet draws on credit lines, deposit withdrawals or other cash outflows so that they are not forced to pull back from lending or sell assets at fire-sale prices, two actions that can turn liquidity strains into a systemic liquidity crisis. If using the discount window would get you fired, the fact that you technically have access to the window isn’t going to provide much comfort.

Another source of stigma is that the post-GFC liquidity requirements do not recognize access to the discount window and prepositioned collateral as a source of liquidity. Banks are not allowed to assume that they borrow from the discount window in either the liquidity coverage ratio (LCR) or their internal liquidity stress tests (ILSTs). Remarkably, although this may be changing, they are not even allowed to point to the discount window or the Fed’s standing repo facility as the means by which they would monetize their liquid assets in their ILSTs. Not only do these restrictions depart from reality – a bank prepared to use the window is more liquid than one that isn’t – they miss an opportunity to create a strong incentive for banks to be so prepared. Moreover, efforts by the banking agencies to convince banks that they should be willing to borrow from the discount window ring hollow if banks are told that contemplating such borrowing is verboten when the bank tests its liquidity needs under stress.

In part because of these shortcomings, two weeks before Silicon Valley Bank failed, the Bank Policy Institute published a note calling for a holistic review of liquidity requirements. Before the GFC, liquidity assessments focused on ensuring that banks had well-diversified reliable sources of funding and contingency plans that included being prepared to borrow from the discount window. After the GFC, liquidity assessments focused on ensuring that banks had large stockpiles of high-quality liquid assets. In the event, SVB failed while awash in liquid assets but with unreliable, concentrated funding, and unprepared to borrow from the discount window. Signature Bank, which failed two days later, was also unprepared to borrow from the window.

In the wake of the bank failures in spring 2023, there has been an increased recognition of the importance for liquidity risk management of being prepared to borrow from the discount window. There appear to be several different approaches floating around for encouraging such readiness. A month ago, a report by the Group of Thirty recommended that banks be required to have discount window borrowing capacity that when combined with deposits at the Fed exceeds uninsured deposits and short-term funding. The following week, acting OCC Comptroller Mike Hsu recommended that banks be required to have discount window borrowing capacity plus reserves in an amount at least as large as potential five-day outflows under severe liquidity stress. Another potential requirement that frequently comes up is that discount window borrowing capacity and deposits at the Fed exceed 40 percent of uninsured deposits.

The common thread across these and other related proposals is a recognition that a bank that is prepared to borrow from the discount window is more liquid than one that isn’t, and that weaving that thread into the weft of regulatory and supervisory assessments of bank liquidity will make those assessments more accurate and increase incentives for banks to be prepared to borrow. Moreover, addressing greater needs for short-term contingency funds with discount window capacity rather than reserve balances allows banks to devote more of their balance sheets to lending to businesses and households – loans that can then be pledged to the Federal Reserve as collateral – rather than expanding even further their loans to the government in the form of even larger portfolios of Treasuries or deposits at the Fed. Enabling banks to shift away from deposits at the Fed as a liquid asset has the added benefit of allowing the Fed to get smaller and less involved in the financial system on a day-to-day basis.

These recommendations all bear a resemblance to the pawnbroker-for-all-seasons proposal by Mervyn King, the former Governor of the Bank of England. However, King’s proposal goes much further. In particular, he recommends that each bank maintain discount window collateral with a lendable value equal to the total amount of all the bank’s liabilities. Under this proposal, there would be no need for liquidity requirements, capital requirements or even deposit insurance. While the proposal is elegant and clever, given that discount window haircuts on loan collateral are often 50 percent or higher, implementation would require banks to fund themselves with similarly sized amounts of capital, radically changing the banking industry and reducing economic activity.

A similar but somewhat less radical proposal that I have urged for over a decade is that the Fed sell banks committed collateralized lines of credit, potential draws on which could be factored into banks’ liquidity requirements and internal liquidity stress tests. One advantage of such so-called committed liquidity facilities is that banks would pay a market-based fee for the line, providing the bank a stronger incentive to reduce its liquidity risk, and raising funds for taxpayers. The interest rate on any draws on the lines would be well above market rates, creating a strong incentive for banks to only use the lines as a backup source of funding in contingencies and an incentive to repay any draws as soon as possible.

A critical ingredient of any such new requirement is reducing discount window stigma. It would be counterproductive to recognize the liquidity value of a bank’s being prepared to borrow from the discount window if the bank is unwilling to borrow when the need arises. While there is no easy way to reduce discount window stigma, as a first step, the leadership of the Fed and other banking agencies need to educate the public, Congress, bank examiners and bank investors that borrowing from the discount window is a business decision of the borrowing bank and neither a bailout nor an indication that the bank is in trouble. A step that Congress could take is rescinding the Dodd-Frank requirement that the Fed identify borrowers. Even with a two-year lag, the disclosure requirement adds considerably to stigma.

As part of any effort to increase banks’ preparedness to use the discount window, it would be worthwhile to investigate potential roadblocks to pledging collateral. The Fed has always sought to strike the right balance between ease of pledging collateral and ensuring that the Fed is protected from losses. For example, the Fed accepts loan collateral using what are called borrower-in-custody arrangements in which the bank pledges a large pool of loans but maintains possession of the loan documentation. The bank provides the Fed regular reports on the composition of the pledged pool and the Fed uses that information to determine lendable value. Nevertheless, the Fed should look, and no doubt is looking, for ways to improve the collateral pledging process.

One step each Federal Reserve Bank must complete when accepting loan collateral is resolving competing claims on the assets from the Federal Home Loan Banks. Congress has given the FHLBs the legal authority to secure their loans with blanket liens, meaning that all the assets of any bank with an FHLB advance outstanding are potentially encumbered. To overcome this challenge to getting a perfected interest in collateral pledged to the discount window, each Federal Reserve Bank and overlapping Federal Home Loan Bank work out a collateral-sharing agreement for each individual FHLB-member commercial bank that wishes to pledge collateral to the Fed. In those agreements, the FHLBs subrogate their interest for certain specific types of loans, such as consumer loans, to the Fed.

Any requirement that banks maintain minimum amounts of discount window borrowing capacity would, of course, need to go through a notice-and-comment process. In this case, given the novelty, importance, and systemic implication of any such requirement, a better approach would be through an advance notice of proposed rulemaking, which would allow the agencies to receive input on the general idea from bankers, central bankers, academics and other stakeholders before drafting a specific rule.