US banking crisis of 2023 could easily happen again

Regulators need to recognise that depositor runs will be much faster, and outflow rates much higher, in an era of social media and 24-hour banking.

Ever since the demise of Silicon Valley Bank in March 2023, regulators have been focused primarily on increasing loss-absorbing capital at the largest US financial institutions. Much less attention has been paid to the problem that precipitated last spring’s banking crisis: banks’ vulnerability to sudden depositor withdrawals.

The SVB debacle exposed three weaknesses. First, depositors pulled their money much faster than assumed by requirements such as the liquidity coverage ratio, intended to ensure that banks have enough cash and easy-to-sell assets to survive 30 days of withdrawals. Second, the Federal Reserve couldn’t provide sufficient emergency discount-window loans, because banks hadn’t pledged enough collateral to the Fed. Third, uninsured depositors had ample reason to run, because they couldn’t be sure the government would make them whole: Such bailouts can happen only after a bank fails and regulators judge that the situation is bad enough to invoke the “systemic risk exception.”

What to do? Certainly, regulators need to recognise that depositor runs will be much faster, and outflow rates much higher, in an era of social media and 24-hour banking. Yet requiring banks to hold a lot more high-quality assets in response would be counterproductive. Tougher liquidity requirements would force banks to divert funds away from lending.

There’s a better solution. The Fed can require banks to pre-pledge enough collateral (such as securities and consumer and commercial loans) to cover all their runnable liabilities (everything but equity, long-term debt and insured deposits) on a day-to-day basis. The central bank would be willing to lend against this collateral, ensuring that banks could always obtain the cash they needed to meet depositor withdrawals. If a portion of the pledged collateral counted toward satisfying the liquidity coverage ratio, banks wouldn’t have to bulk up on safe assets and reduce lending.

This would have several advantages. With an explicit Fed backstop, uninsured depositors would have little incentive to run in the first place. Most banks could easily comply: Because smaller banks fund themselves mostly with insured deposits, they wouldn’t have to come up with much collateral to pledge. Others that were more constrained could adjust by raising equity, issuing more long-term debt, increasing their insured deposits or holding more high-quality assets. The Fed’s lender-of-last-resort function would be much improved: No last-minute scramble would be required to identify and mobilise collateral to pledge to the discount window.

How, then, to proceed? First, officials should do a detailed study to ascertain which banks would struggle to satisfy the requirement. In some cases, the Fed might need to make adjustments to avoid unintended consequences. For example, it might be appropriate to grant some relief for the big clearing banks, which process large volumes of payment and securities transactions for their customers.  These activities often require significant operational deposits, well beyond the insured limit. Because these deposits are needed to conduct the banking business, they’re unlikely to run.

Second, the Fed should overhaul the discount window. Banks are often reluctant to use it when they should, for fear of being seen as troubled. This stigma could be reduced by restricting use of the discount window to healthy banks, and diverting genuinely troubled banks to a new emergency lending facility. Beyond that, the Fed needs to modernise its lender-of-last-resort function so that it can quickly value and enable the rapid substitution of collateral.  Also, it needs to harmonise discount-window operations at the 12 regional reserve banks into a standardised “no questions asked” approach.

These reforms will require considerable time and effort, but they’ll be well worth it. Done right, they’ll reduce the risk of banking panics and, at the same time, make banks stronger and better able to compete with less-regulated, non-bank rivals. It’s time to move forward.

Bill Dudley, a Bloomberg Opinion columnist, served as president of the Federal Reserve Bank of New York from 2009 to 2018. Views are personal and do not represent the stand of this publication.

Credit: Bloomberg