Sense of Urgency

On April 28, 2023, Michael Barr, the Federal Reserve Vice-Chair for Supervision, issued his long-awaited post-mortem of the Silicon Valley Bank failure.  The FDIC issued its post-mortem for Signature Bank the same day.  There’s a lot to unpack here, with the Barr Report running 118 pages, plus numerous supporting documents.  I’ll cover the report in pieces.  Today’s post will focus on the long lags between identifying concerns in the field and taking supervisory action.

Barr’s report concentrates more on institutional weaknesses than on individual errors.  While this “mistakes were made” approach diminishes personal accountability, tossing a couple of unlucky regulators under the bus tends to do little good in the long run.  The report focused on institutional roadblocks to effective supervision..

As I noted in an earlier post, the pace of corrective actions, especially for larger banks, can be frustratingly slow.  At OCC, we issued Supervisory Letters within 45 days of completing field work on an exam.  I thought that was often too slow, but it qualifies as a rapid response compared to some of the Fed’s actions at SVB.

Consider the 2021 Liquidity Target Exam.  The FRB of San Francisco issued four Matters Requiring Attention and three Matters Requiring Immediate Attention.  The Fed’s definition of immediate is open to debate. The FRB completed fieldwork on August 27^th, but didn’t send out the Supervisory Letter until November 2^nd, 67 days later.  The report’s supporting supervisory materials only include those from the FRB to SVB and not SVB’s responses.  We don’t know how quickly SVB responded or the quality of their response.

It gets worse.  Going into 2021 exam, Liquidity was rated “1”, the highest rating.  A “1” rating is hard to get, with most healthy banks rated “2” on a 5-point scale.  The FRB waited until the next supervisory rating cycle, itself delayed, to revisit the liquidity rating.  Liquidity remained rated “1” until August 17, 2022, nearly a year after examiners identified deficient liquidity management practices.  The new rating was “2” – still satisfactory.  That rating is hard to square with the liquidity management weaknesses noted in the 2021 exam and the bank’s liquidity risk profile.

The Barr Report indicates that deterioration in SVB’s liquidity during 2022 and adverse findings from a liquidity horizontal (cross-bank) review would likely have led to a further downgrade of the Liquidity rating to “3.”  However, the bank failed before the Fed got around to making the change.

The supervisory framework was likewise several hundred days late and billions of dollars short when it came to capital stress testing.  Although SVB Financial Group crossed the $100 billion threshold in June 2021, it was not scheduled to receive its first supervisory stress test results until three years later, in June 2024.  Legislation in 2018 removed a requirement to conduct company run stress tests.

The San Francisco Fed’s role as both bankers’ bank and regulator creates the potential for a conflict of interest.  (SVB CEO Greg Becker also served on its Board of Directors.)  The Barr Report makes only passing reference to this potential conflict.  However, attempts to avoid a conflict may be partly behind what went wrong at SVB.  The San Francisco Fed had limited autonomy in making supervisory decisions on its own, especially for larger banks.  Multiple vetting sessions including both San Francisco and DC staff meant that supervision moved at a glacial pace.  The need to coordinate with regulators from the State of California likely slowed down things even further.

A light touch regulatory environment under Barr’s predecessor, Randy Quarles, may also account for the long lags between findings and supervisory actions.  As noted on page 9 of the report, “supervisory policy placed a greater emphasis on reducing burden on firms, increasing the burden of proof on supervisors, and ensuring that supervisory actions provided firms with appropriate due process. Although the stated intention of these policy changes was to improve the effectiveness of supervision, in some cases, the changes also led to slower action by supervisory staff and a reluctance to escalate issues.”

Examiners need to support their findings.  However, few judgments can be made with metaphysical certitude.  That’s especially true when evaluating risk.  It’s one thing to observe that the bank is losing money or that many loans are becoming past due.  Calling out excessive risk taking is another matter.  You can point out that risk taking is extreme based on prior experience or a comparison with other banks.  But you can’t say for certain that the bank’s luck will run out – until it does.  By then it’s too late.

The FDIC’s report on Signature’s failure provides less detail and is less self-critical than the Fed’s SVB post-mortem.  Slow supervisory responses to emerging problems emerged as a common theme.  A liquidity target started in October 2022, was still in process when the bank failed in March.  The FDIC attributes the slow pace of supervision to employee turnover and ongoing staffing shortages.  The report notes that from 2020, an average of 40% of Large Financial Examiner (LFI) positions in the New York metro were vacant or filled by temporary staff.  Other regulators have also found positions in the New York area hard to fill.  The FDIC is also the primary federal regulator for First Republic Bank.  First Republic is based in San Francisco, another high-cost area where staffing shortages are endemic.  The staffing issue may come up again.

The Fed was too slow to address liquidity weaknesses at SVB, but at least were able to diagnose gaps going back to 2021.  The same cannot be said for interest rate risk, where glaring deficiencies were left unaddressed until it was too late.  I’ll discuss interest rate risk at SVB in an upcoming post.