One year ago, in March 2023, Silicon Valley Bank failed, followed soon after by similar failures at Signature Bank and First Republic Bank. Measured by the nominal size of bank assets, these were three of the four largest U.S. bank failures in history. (The 2008 Washington Mutual Bank failure remains the largest.) Is this just a one-time problem, a problem that has already been resolved, or are the root causes still present?
Tobias Adrian, Nassira Abbas, Silvia L. Ramírez, and Gonzalo Fernández Dionis address these issues.The US banking sector since the March 2023 turmoil: Surviving the aftermath (IMF Global Financial Stability Note, March 2024).
I have already discussed the failure of Silicon Valley banks several times on this blog from various angles. For example, seeThe Federal Reserve’s Anatomy of Silicon Valley Banks” (April 28, 2023), “Was Silicon Valley Bank's bailout of depositors the right decision?” (June 6, 2023), “Why are US bank runs so fast these days?” (June 20, 2023), “Spread accounts across multiple banks for deposit insurance” (November 29, 2023). For more time to think about everything in his one place, I recommend Andrew Metrick's essay. “Silicon Valley Bank Failure and 2023 Panic” (Winter 2024 issue) Economic Outlook Journal (38:1, 133-52).
(Full disclosure: I have been editor-in-chief of the Journal of Economic Perspectives since 1986, so the articles published there are necessarily of interest to me. However, all articles are published by the American Economic Association All available for free as a tribute to the journal's latest issue to its first issue.)
In some ways, the banks that failed had problems that were not widely shared. U.S. bank deposits are insured up to $250,000 by the Federal Deposit Insurance Corporation. Therefore, those who have deposits less than that amount do not have to worry that their bank will fail. But companies sometimes have more money in their bank accounts, and companies in Silicon Valley banks in particular had large amounts of money they received from venture capitalists. In fact, a whopping 94% of his deposits at Silicon Valley Bank exceeded his $250,000 limit and were uninsured. This is not the case at most banks.
But some of Silicon Valley Bank's problems were more widespread across the banking sector. In particular, if you hold a financial asset that pays a fixed interest rate, such as most U.S. government bonds, and interest rates rise, the value of those low-interest bonds will fall. Many banks own U.S. government bonds, but Silicon Valley Bank owns more than any other bank.
As the authors of the IMF point out,
After the failures of SVB (Silicon Valley Bank) and SBNY (Signature Bank) in March, depositors and investors were asked first about liquidity, and secondly about liquidity, which matched a specific profile with various attributes such as: I was concerned about the financial health of the bank. (1) Large amount of deposit outflow. (2) High concentration of uninsured deposits. (3) increased reliance on borrowings and utilization of liquidity facilities; (4) significant unrealized losses; (5) High exposure to CRE (commercial real estate 0). Although high levels of uninsured deposits and large deposit outflows were characteristic features of failed financial institutions (SVB, SBNY, FRB (First Republic Bank)), our analysis shows that this group This is a group of small and medium-sized regional banks that account for a large amount of uninsured deposits, have a large amount of unrealized losses, are highly concentrated in CRE, and have become increasingly dependent on borrowings since the stress of March 2023.
Commercial real estate issues weren't a problem for Silicon Valley Bank. However, many local banks make large loans to people building commercial real estate. With the shift to a work-from-home economy, the value of commercial real estate has declined, increasing the risk of these loans. The authors state:
In addition to unrealized losses due to rising interest rates, the credit risks of some financial institutions, particularly their exposure to CRE (commercial real estate), are at the center of investors' concerns today. Small and regional banks are exposed to substantially two-thirds of the $3 trillion in CRE exposure in the U.S. banking system (Figure 4, Panel 1). In January 2024, changing market expectations about the timing and pace of U.S. interest rate cuts, combined with large losses announced by large banks with significant CRE exposure, caused the regional bank index to decline by 10%.
Concentrations of CRE exposure represent a serious risk for small and medium-sized banks amid economic uncertainty and rising interest rates, potential declines in real estate values, and deteriorating asset quality. …One-third of U.S. banks (primarily small and regional banks) have CRE exposures greater than 300% of their capital plus allowances for credit losses, which is a 16% of the total (Figure 4, panel 2).
The IMF authors are not saying anything catastrophic here. They call this problem the bank's „weak tail.“ This means that all five of the factors they mention must come together to cause a bank to fail. However, reading between the lines, we see that banking regulators are forcing some of the „weak banks“ – small and regional banks – to merge, and are trying to do so before the banks' financial situation becomes dire. I wouldn't be surprised either.