On July 28, 2023, the U.S. banking agencies issued additional guidance on liquidity risks and contingency planning. The agencies “encourage[d] depository institutions to incorporate the discount window as part of their contingency funding arrangements.” They also noted that if the discount window was part of the institution’s contingency funding plan, the DI should be operationally ready to borrow. Even though being unable to borrow from the discount window contributed significantly to the disorderly failures of SVB and Signature Bank (see “Improving The Government’s Lender of Last Resort Function: Lessons From SVB and Signature Bank”), the banking agency announcement simply repeats what was already included in guidance and examination manuals. The agencies should have taken modest but still important steps to encourage banks to be willing and ready to borrow from the discount window but chose, instead, to do essentially nothing almost five months after the bank failures.
A contingency funding plan (CFP) sets out the bank’s strategy for addressing liquidity shortfalls in an emergency situation. All DIs are required to have a CFP. In early 2003, the Fed revised the discount window to make primary credit, the main lending program, a “no questions asked” facility.[1] Later that year, the banking agencies issued an “Interagency Advisory on the Use of the Federal Reserve’s Primary Credit Program in Effective Liquidity Management” or SR 03-15, available here. The letter encouraged depository institutions to consider adding primary credit to their contingency plans and listed five attributes of the program that made it a potentially useful addition. The letter stated that if a bank incorporates the discount window into its CFP, it should be operationally ready to borrow. In 2010, the agencies issued guidance on CFPs that again reminded banks that if they included the discount window into their CFP, they should be operationally ready to borrow. The Fed’s examination manual includes all the same language on the discount window and CFP, and it includes the discount window as a funding source in a CFP template.[2] Indeed, we know that SVB had incorporated the discount window into its CFP, although also that it was not operationally ready to borrow.[3]
So what’s in the recent announcement is unremarkable; what is remarkable is what it did not include. For example, as stated in the announcement, credit unions with more than $250 million in assets are required to demonstrate access to the discount window or the Central Liquidity Facility (the credit union’s discount window). There is no such requirement for commercial banks of any size. There is no statement that institutions should make use of the discount window and examiners should view such occasional use as unremarkable, or any attempt to reduce the stigma associated with using the discount window (see “A Major Limit on the Fed’s Crisis Toolkit: Shame”). By contrast, the Bank of England recently reiterated such guidance and even clarified that an institution can turn to the discount window first when implementing its contingency plan (see “Bank Examiner Preferences are Obstructing Monetary Policy”).
And there is no change to the internal liquidity stress tests that banks are required to perform regularly and report the results to their examiners. In particular, banks are not allowed to plan on using the discount window, or even the standing repo facility (SRF), as the means by which they would convert assets into cash in their ILSTs. Despite the similarity between the CFP and the monetization requirement in ILSTs, banks can plan on using the discount window in the former and not in the latter. Reportedly, SVB had been interested in signing up for the SRF but decided not to because it could not plan on using the facility to monetize its agency MBS in its ILST, even though that is the purpose of the SRF (see “Two Important Fed Programs that Should Be Mutually Reinforcing Are in Conflict. Why?”).[4] On August 1, 2023, the Fed released the results of a Senior Financial Officer Survey that included questions about banks’ preparedness to borrow from the discount window. The second most significant factor discouraging banks’ interest in the discount window was “supervisory or regulatory treatment.” Similarly, there is no change to how the discount window is incorporated into resolution liquidity requirements (see “Central Bank Contingency Funding in Resolution Plans”) or in the liquidity coverage ratio requirement (see “CLF Notes – What is a Committed Liquidity Facility?”).
So why did it take the agencies five months to issue a few paragraphs repeating existing guidance? Perhaps they disagreed. Perhaps some agencies wanted to go further, and others balked. As noted by the Treasurers of BPI banks, examiners really seem to hate the discount window and the SRF (see “Bank Treasurers’ Views on Liquidity Requirements and the Discount Window”). The backstory behind SR 03-15 illustrates how hard it can be to overcome that prejudice:
I was on the team at the Board that revised the discount window in early 2003, with the goal of reducing stigma. After the Board implemented the changes, we monitored fed funds transactions to identify banks that paid more than the discount rate to borrow in the fed funds market. If we found one, I would call up its funding officer to talk about the discount window. After talking to a large West Coast bank that had been paying up in the market, the funding officer got back to me and indicated that he had spoken with their examiner and the examiner had said he was not comfortable with the bank borrowing from the discount window. I then worked with the banking agencies to publish SR 03-15. The funding officer brought the SR letter to the attention of his examiner. The examiner still preferred that the bank not borrow from the discount window.
It is extremely hard to overcome examiner distrust of the discount window. And stigma does not just come from examiners, it comes from bank management as well, stemming from decades-long concerns about potential consequences from borrowing. One of the most significant factors discouraging banks’ interest in the discount window was public disclosure of borrower-level information about each loan followed by “supervisory or regulatory treatment.” A treasurer told me that when he was hired he was told that if he borrowed from the discount window there would be two phone calls, one from the president of the New York Fed to the bank’s CEO to ask why they borrowed, and one from Personnel to him to tell him to clear out his desk. Moreover, management and examiner attitudes toward the discount window can be mutually reinforcing. For example, the 1955 revision to the Fed’s Regulation A (the regulation covering the discount window) was designed to increase stigma by taking advantage of bank management’s longstanding “tradition against borrowing” (see “Discount Window Stigma: We Have Met the Enemy, and He Is Us”).
Nevertheless, as illustrated by the bank failures in March, it is critical that the Fed reduce stigma so that banks are prepared and willing to borrow when necessary. If SVB and Signature Bank had been able and willing to borrow, their failures likely could not have been avoided, but they would have been orderly with much reduced systemic consequences. To reduce stigma, a good first step would be a clear and forceful statement from the banking agencies that borrowing was a business decision and examiners should not view it negatively. A crucial further step would be for Fed leadership to convey the same message to the public and to Congress.
[1] Madigan, Brian and William Nelson, “Proposed Revision to the Federal Reserve’s Discount Window Lending Programs,” Federal Reserve Bulletin, July 2002. https://www.federalreserve.gov/pubs/bulletin/2002/0702lead.pdf.
[2] See pdf page 78 of the manual: https://www.federalreserve.gov/publications/files/cbem-3000-202205.pdf
[3] See the bottom of page 5 https://www.federalreserve.gov/supervisionreg/files/svbfg-liquidity-planning-target-supervisory-letter-20231102.pdf Regarding the CFP of SVB: “Additionally, several listed funding sources such as brokered CDs and discount window access have not been tested.”
[4] The NYU Stern whitepaper on the bank failures stated “[W]hile the discount window was available to Silicon Valey Bank, the Fed did not appear ready to lend to it and the bank did not appear ready to borrow.” SVB and Beyond: The Banking Stress of 2012,” NYU Stern, July 3, 2023. p. 215-216. https://www.stern.nyu.edu/experience-stern/about/departments-centers-initiatives/centers-of-research/volatility-and-risk-institute/research/svb-and-beyond-banking-stress-2023
