Bank runs happen slowly at first, then fast. The recent failures of Silicon Valley Bank and Signature Bank seem no different. Their rapid asset growth over the past three years, mostly in long-term bonds, was fueled by unsecured deposits tied to their undiversified base of loan clients. While the unprecedented scale of Federal Reserve and government stimulus following the pandemic clearly contributed to the explosion of bank deposits, the promise of low-for-long rates led to a search for yield and bank investment in long-term bonds. Effectively, banks seem to have manufactured tail risk again, this time not by underwriting risky mortgages but by taking on interest rate risk.
Regional banks tend to always come under strain when rates rise and local economies are hit. But the business models of many large global banks have also been found wanting as the era of easy money is coming to an end. Regulators have managed to arrest depositor runs for the time being by implicitly extending guarantees to uninsured deposits that no longer have sufficient private bank capital backing them.
Nevertheless, market uncertainty remains high. Investor expectations of the economy have switched within a couple of months from soft landing to no landing to a possible hard landing. Credit conditions are tightening in response, both at banks and in capital markets.
One hopes that things will calm down, but it is better to be prepared for further stress. What can the Fed (and other regulators) do if confidence in the banking system continues to erode?
Experience from rescue measures adopted in the fall of 2008 following the collapse of Lehman Brothers suggests that simply guaranteeing deposits and backstopping bank creditors does not suffice. These steps, even if necessary, will not restore confidence in the health of the banking system to a point where it can provide adequate intermediation to the real economy and serve as a reliable counterparty in financial transactions.
Indeed, bank credit default swap spreads and option implied volatilities remained extraordinarily high until March 2009. The market barometers of financial stability declined to normal-time levels only after the Fed stress-tested the banking system to further adverse losses, provided transparent credible estimates of capital shortfall calculations and incentivized banks with capital shortfall to raise equity, with a back-up from the Treasury in case they were unsuccessful.
After this “supervisory capital assessment program” was implemented from February to May 2009, around $75 billion of private capital was required to be raised by the U.S. banks. Much more was raised soon after. The risk of further financial fragility subsided and macroeconomic calm was restored. Treasury funds set aside for recapitalizing banks were not deployed.
This regulatory playbook can serve as a model today, but a key difference is that the Fed should stress test the entire banking system to the risk of a stagflation.
In currently employed regulatory stress scenarios, economic recessions are associated with low interest rates that boost the value of banks’ securities investments. This is, however, counterfactual at the present juncture. Reflecting reality, the stress scenarios instead need to feature an economic slowdown with a high level of rates and possibly even further hikes, which may be essential to arrest above-target inflation. Not just loan losses but also mark-to-market losses on securities should be transparently recognized.
Some concession in marking to market could be considered formulaically based on whether the bank has a stable, insured retail deposit base. Scenarios also need to factor in the ongoing secular shift away from commercial real estate.
Most large banks with high asset quality and diversified lines of business will likely fare okay, but there might be some surprises, as in the summer of 2009. Some that have invested more heavily in long-term bonds may be capital-deficient and should be asked to raise public equity without further ado.
The most exposed banks might even look entirely decapitalized and may have to be sold to healthier banks that might be willing to pay to “purchase and assume” their deposit and loan franchises. Such sales may require some backstop from the authorities.
Smaller, regional capital-deficient banks may not be able to access public markets and may have to be handled by the Federal Deposit Insurance Corporation’s prompt corrective action and orderly resolution frameworks. If done right, the capital-raising and asset-and-deposit reallocation measures would stabilize the system as well as the economy. As in 2009, government guarantees might not be availed in the end, reducing the burden to the taxpayer.
Bank capital is a form of private deposit insurance. If we are not to fully socialize the economy-wide risks from banks engaging in carry trades, then bank capital will have to play a substantial role in restoring confidence. Marking capital honestly, stressing it plausibly and raising it adequately, appears to be a feasible regulatory plan of action so that depositors and creditors do not remain concerned about banks’ solvency risk. It would be better to design and have such a plan in position to roll out before we see a large bank failure.
Viral V. Acharya is the C. V. Starr Professor of Economics in the Department of Finance at New York University’s Stern School of Business and a former deputy governor of the Reserve Bank of India.