Why Did SVB Fail? We’ve Been Teaching About It for Years.

Since 2017, students in both the base and accelerated sections of the managerial finance course we have taught at Stanford Graduate School of Business have learned in the early weeks of the quarter about bank balance sheets’ vulnerability to rising interest rates. First-year MBA students who took our classes might remember that in week three, after presenting some basics on how fixed-income assets are priced, we provide them with a very basic spreadsheet that shows a balance sheet roughly representative of different sectors in the U.S. banking system.

Using this spreadsheet, we ask our students to apply their knowledge to infer the fragility of each sector, specifically by looking at what happens to bank equity if interest rates rise by one or two percentage points. Through this example, we illustrate the basic concept of interest rate risk: when interest rates rise, the bonds that banks hold on the asset side of their balance sheets fall in value.

The longer duration the asset, the greater the rate sensitivity. Because the banks’ liabilities have short duration, the liabilities drop by less. In undertaking this exercise, our MBA1s learn that even relatively small increases in interest rates can leave a bank with more liabilities than assets, erasing the value of its equity.

In light of this, we were somewhat astonished to learn that both bank regulators and the management of several major banks seem to have ignored this basic insight: when there is a duration mismatch between assets and liabilities, bank equity can fall sharply when rates rise. The Federal Reserve apparently did not include scenarios in its stress testing that reflected its own policies of raising interest rates to combat inflation. Evidently, government regulators do not hire Stanford MBAs.

Joking aside, on March 30 the Biden Administration proposed reforms to safeguard the banking system following the collapse of Silicon Valley Bank. These include ensuring that large banks have enough capital to withstand rising interest rates. It is a pity that Federal Reserve Chairman Jerome Powell did not implement the simple stress tests we did in Fin 201 and Fin 205 to predict what effect his rate increases might have on bank solvency.

Some press coverage has characterized what happened to SVB and Signature Bank as a “novel situation.” But every regulator should know that when interest rates rise substantially, bank assets fall in value much more than liabilities. This is particularly true for banks that hold a lot of very long-term assets on their balance sheets, such as mortgages, and somewhat less so for banks that might more conservatively hold shorter-term assets. If depositors are not confident that they will continue to have liquid access to their funds, they will react to this state of affairs by demanding their money.

There is nothing novel about this asset-liability mismatch. It underlies many historical episodes of banking crises, most notably the savings and loan crisis of the 1980s, which ultimately cost taxpayers $160 billion, or $436 billion in today’s dollars. Arguably the trickier type of banking instability to detect is when banks hold opaque assets of uncertain quality, such as the newer versions of mortgage-backed securities that populated many financial sector balance sheets during the financial crisis of 2008-2009. But so far, in the case of SVB and Signature, the main problem seems to have been good old interest rate risk.

Stanford GSB finance professors have also done a great deal of research on this topic in particular. In recent years, some research from the field has argued that banks bear relatively little interest rate risk because of their market power in the activity of deposit-taking. However, our colleague Professor Juliane Begenau has demonstrated in her work that, in fact, banks have heavy interest rate risk exposure and that the above factors do little to mitigate it — findings that are clearly corroborated by recent events. Our colleague Professor Amit Seru has recently assessed the extent of interest rate risk and its potential to spark bank runs across the entire banking system.

There is a debate as to the right policy for banking regulation. Some point out that 2018 legislation adjusting the regulation weakened it, although the Bank Policy Institute has concluded that even under the original rules SVB probably would have received a “passing score.” This is because the liquidity coverage ratio pursuant to the 2010 Dodd-Frank legislation was focused on asset quality as opposed to interest rate risk. Yet the basic red flags were there for all of SVB’s regulators — the state of California, the FDIC, and the Fed — to see in plain view.

It seems that much of the regulation that emerged in the 2010s was geared towards the problem of opaque and low-quality assets that plagued bank balance sheets in the run-up to the financial crisis. It is often said that “Generals always fight the last war.” This usually refers to the idea that they think of the past while ignoring the future. In this case, people in the banking system seem to have forgotten the many banking crises before the global financial crisis of 2008-2009. Many of these arose simply from the interest rate risk that goes along with issuing short-term demandable deposits and investing in long-duration fixed-income securities.

Perhaps the right adage in this case is, “Those who cannot learn from history are doomed to repeat it.”