Perspective on Risk - May 3, 2023 (More banks; Signature)

Capital is needed; Changes are needed to FDIC resolution regime; NYDFS review of Signature Bank

I’m getting a bit tired of writing these, but figured this is one of my few areas of comparative advantage. Lord knows how Matt Levine and others do this every day.

Capital Prevents a Credit Crunch

As we have been noting, there are vast differences between current events and the GFC. The current events are becoming more worrisome because they may exacerbate a credit-crunch amidst the coming recession.

Early on, I had expected some of the stronger regional banks to go out and raise capital, principally as a signaling device of their strength and to stem withdrawals. Didn’t happen.

Now, faced with continuing deposit outflow, banks are severely retrenching their lending as they seek to improve their loan-to-deposit ratio while facing pressures to shrink.

One significant feature of the GFC was the mandated capital raises by the largest institutions (combined with stress tests to size the possible whole). From a policymakers point of view, the mandate to raise capital was precisely to offset the credit crunch dynamic, and prevent excessive credit tightening.

Perhaps such an approach is needed now.

There are two ways to de-lever a financial system; close banks and have the FDIC absorb losses as with SVB, or raise capital.

Changes are needed to FDIC’s resolution regime

The WSJ has an article Why Washington Let JPMorgan Buy First Republic that describes why and how JPM ended up the winning bidder for First Republic. This article 1is troubling because the FDIC’s resolution incentives do not align with broader systemic risk concerns.

In the last Perspectives, I wrote about how the Riegel-Neal 10% cap on national deposit share was not applicable in a failed bank acquisition. Clearly, the intent of Congress was to limit the deposit share of the largest banks.

Now we read that the FDIC’s resolution procedure will provide a competitive advantage to the largest banks that can afford to acquire the entirety of organizations.

JPMorgan was also the only bank with the appetite to buy substantially all of First Republic at a competitive price, the people said, including mortgages that other banks didn’t want. That was a priority for the Federal Deposit Insurance Corp. because it removed uncertainty over any assets left behind that it would have to sell.

The FDIC, which orchestrated the plan to seize First Republic and sell it, typically must accept the bid that imposes the smallest cost on its deposit insurance fund. JPMorgan’s bid cost the fund some $13 billion, better than the others’ best bids, but not by an enormous amount, some of the people familiar with the matter said.

A regulatory official defended selling the firm to JPMorgan, saying regulators were hamstrung by legal requirements that the winning bank provide the lowest-cost bid. Deviating would have cost the fund an additional, not insignificant, amount of money borne by all banks, including small community lenders, the official said.

The FDIC Improvement Act of 1991 (FDICIA) requires the FDIC to use a "least-cost" resolution method when dealing with failing institutions. There is no consideration for the Riegel-Neal cap or for other possible systemic considerations.

As reported by the WSJ, there were three other bidding firms whose current systemic significance is lower than JPM.

In the end, the FDIC received final bids Sunday from three other banks—PNC Financial Services Group, Citizens Financial Group and Fifth Third Bancorp—people familiar with the decision said, but JPMorgan’s was deemed lowest-cost.

This needs to be addressed legislatively.

NYDFS Review of the Supervision of Signature Bank

Yet another post-mortem: NEW YORK STATE DEPARTMENT OF FINANCIAL SERVICES INTERNAL REVIEW OF THE SUPERVISION AND CLOSURE OF SIGNATURE BANK

The vast majority of the report is devoted to detailing what was identified by the examiners and conveyed to management during the 2019-2021 examination cycles. It is only in Appendix B where “Recommended improvements to bank supervision” is addressed.

Diagnosis of why Signature Bank failed

The DFS diagnosed Signature’s failure for the following reasons:

The Bank’s growth outpaced the development of its risk control framework.

The immediate cause of the Bank’s failure was a propulsive run on deposits instigated by the consecutive announcements, first on March 8 that Silvergate Bank (“Silvergate”) was liquidating itself, and then on March 10 that the California Department of Financial Protection and Innovation was taking possession of Silicon Valley Bank (“SVB”) following an unprecedented run on its deposits. The resulting panic caused a run on Signature that was faster than any other bank run in history, save the run that had just taken place at SVB. Driven by advances in digital banking and the rapid spread of information and rumors through social media, Signature experienced a runoff of $18.6 billion in deposits in a matter of hours, reducing the Bank’s deposit base by 20 percent in one day. The run placed a significant strain on Signature’s liquidity position.

Signature’s failure to remediate the outstanding liquidity management issues undoubtedly contributed to its collapse.

While the role of digital assets in Signature’s failure has been the subject of speculation, the percentage of digital asset customer withdrawals on March 10 was relatively proportional to the percentage of digital asset customers in the deposit base overall. The bigger issue for Signature was that the Bank had a high concentration of uninsured deposits and was perceived as a crypto bank

On the supervision of Signature Bank:

Regulators identified liquidity management issues

Issues relating to the Bank’s liquidity risk management were identified in the Reports of Examination issued by the Regulators for the years 2018 and 2019.

For 2018, the Regulators identified an MRBA related to several breaches of the liquidity risk metrics established by Signature’s Board of Directors (the “Board”). The following year, the Regulators added a new MRBA related to liquidity risk management that identified material weaknesses in Signature’s contingency funding plan (“CFP”) and liquidity stress testing, including unsupported critical assumptions and insufficient liquidity stress testing.

Regulators lowered the rating appropriately

At the conclusion of the 2019 examination cycle, given the number and seriousness of the findings, the Regulators downgraded Signature’s liquidity rating from a ‘2,’ representing a “satisfactory” rating, to a ‘3,’ representing a “less than satisfactory rating.”

The NYDFS identified the same timeliness and staffing issues as the FDIC did with First Republic

While Regulators conducted timely examinations of Signature, finalizing and issuing regulatory findings in the form of Supervisory Letters or Reports of Examination did not happen in a sufficiently timely manner.

One reason for the prolonged turn-around time to issue Reports of Examination and Supervisory Letters to Signature was the cumbersome review process. For example, before a Supervisory Letter was issued to the Bank, it had to undergo several rounds of reviews, with no established internal deadlines for completion. Often, the letters were not prepared by the examiners who conducted the target review, meaning that the drafter would have to take time to review the findings to be able to draft the Supervisory Letter

DFS staffing constraints contributed to these delays

Internal staff constraints limited DFS’s ability to adequately staff examinations.

The DFS appears to feel that it did not escalate its issues when Signature was slow to remediate identified deficiencies.

DFS’s internal processes lack clear guidelines that examiners must follow to escalate regulatory concerns or instances in which a bank fails to remediate findings in a timely fashion.

The DFS highlights issues with even the more rigorous LCR standard.

The mismatch between recent lived experience and defined regulatory expectations is apparent when considering recent depositor behavior versus the mandated assumptions of the liquidity coverage ratio

Under the LCR rule … requires subject institutions to assume a five percent outflow rate for fully insured operational deposits not held in an escrow account and a 25 percent outflow rate for other operational deposits (meaning those that are held in escrow accounts or those that are not fully insured).

The observed behavior of Signature’s depositors beginning on March 10 did not align with these assumptions, in many instances exceeding them, despite the fact that the rule is intended to provide a uniform and conservative liquidity risk-management framework.

What to Do About Liquidity

This last point in the DFS review is a central issue that needs to be addressed for the future.

Overly conservative liquidity management assumptions will constrain the ability of banks to provide maturity transformation desired in the markets, and required for long-term project financing.

This can potentially be solved several ways, and probably others as well:

1. Increased securitization and the tranching of assets by maturity, along with the development of a pool of investors willing to hold the long-dated tranche (such as life insurers).

2. Liquidity options provided by the Fed (or FHLB system)

This will require thought and work.