��ɫ��Ӱ

The ghost of Silicon Valley Bank lingers. Two years after a panic by depositors caused the bank’s spectacular collapse, there are about the health of smaller lenders in an era of high interest rates. And, once again, the jitters revive a key question surrounding Silicon Valley Bank’s failure and the regional banking scare that quickly followed: When is a bank facing a short-term liquidity crisis and when is its business model so broken that it is essentially insolvent?

“There’s still a lot of confusion about this, including among bank analysts and regulators,” says Arvind Krishnamurthy, a finance professor at ��ɫ��Ӱ Graduate School of Business (GSB) and a senior fellow at the ��ɫ��Ӱ Institute for Economic Policy Research (SIEPR). In new research, Krishnamurthy — along with Peter DeMarzo, who is also a GSB finance professor and SIEPR senior fellow, and Stefan Nagel of the University of Chicago — analyze the 2023 bank turmoil and the impact that rapidly-rising interest rates had on their balance sheets. Here, Krishnamurthy and DeMarzo talk about lessons from Silicon Valley Bank’s implosion and how their research informs banking stability today.

The 2023 crisis resulted in three of the four largest bank collapses in U.S. history. Can you remind us of what caused this?

DeMarzo: In the years leading up to the crisis, Silicon Valley Bank and others lent money in the form of mortgages and other loans at very low interest rates. At the same time, they invested much of their cash into long-term securities like Treasury bonds, which are loans to the federal government. When interest rates rose, the value of these loans fell because their promised cash flows were below market rates. For many banks, this decline was large enough that the value of their assets (these loans) fell below what they owed their depositors. As a result, Silicon Valley Bank and many others like it became technically insolvent.

But, even if the value of the bank’s asset falls below its liabilities, this doesn’t necessarily mean it should be shut down. If the bank’s business is fundamentally profitable, maybe it just needs time to work itself out of the hole.

Can you give a simple example of what this tension between ‘technically insolvent’ and long-term viability looks like?

DeMarzo: Imagine a bank has $100 in deposits (liabilities) and uses them to make loans worth $105. If the value of those loans (assets) falls from $105 to $90 for some reason, and the bank shuts down immediately, that’s a problem — it won’t be able to fully repay its depositors.

But let’s say the bank continues to operate and generates future profits — for example, from new loans and customer services — that are worth $20. Then the true value of the bank after the value of its loans drops is still positive: $90 of loans, plus $20 in future profits, versus the $100 it owes depositors. That is, while the bank would be insolvent if it shuts down immediately, it is potentially profitable as an ongoing business — it will be able to fill its current hole with its future profits.

How does your research help regulators better determine if a bank is doomed or not?

DeMarzo: We look at measuring what is known as a bank’s franchise value, which is the value of the expected future profits the bank can earn from making loans and providing services to depositors. Essentially, it’s the difference between the value of the bank if it shuts down today and the value of the bank if it continues operating into the future. Assessing this franchise value is necessary to determine if a bank is worth saving.

We think there are misperceptions about franchise values and banks’ sensitivity to changes in interest rates, so we combed through bank sector data and developed a framework to shed light on what really happened.

What does your framework say about banks’ franchise values in 2023?

Krishnamurthy: A lot of people, including regulators, had the view that franchise values went up when interest rates rose — by $1.6 trillion in total, according to one estimate — and that this increase served as a hedge against the decline in the value of their loans. But our modeling shows that the opposite happened: The franchise value for the median bank declined, further exacerbating its losses on investments in long-term securities. Even so, we estimate that banks’ franchise values at the time were still substantial enough that most small and regional banks remained solvent in 2023.

A second point we make in our paper is that banks and bank regulators need to pay attention to how interest-rate fluctuations impact the market value of banks’ loans, securities, and franchise value. There are lots of little pieces of bank regulations that have led banks to overly focus on current cash flows and stabilizing them, which made their market values more vulnerable when interest rates changed. It’s a lesson the banking system learned from in the 1980s, but has since been forgotten.

DeMarzo: Right. Savings and loans were hit with big interest rate increases in the 1980s and it put many of them in a similar position where they were technically insolvent because the value of their assets had gone down. A difference in that case was that many of the S&Ls did not have profitable business models. When regulators gave them more time, they doubled down on making risky, speculative loans that were unlikely to pay off. In the end, it meant they were sitting on much bigger losses when interest rates rose again. A warning going forward is that it’s fine for the Fed to step in and prop up banks that are worth saving, but the last thing you want is a bunch of zombie banks taking crazy risks because they see only the upsides of doing so when the downside is somebody else’s problem. One way to control for this risk is to force the bank’s shareholders to increase their equity in the bank (i.e. increase the bank’s capital) so that they will have more “skin in the game” for the bank’s success.

What does your research say about the banking sector today?

DeMarzo: If the banking sector’s business model continues to work the way it has in the last 20 years — where profits mainly come from spreads on deposits and loans — then most banks today should be able to recover from their recent losses. But if depositors change their behavior by, for example, moving their money to online banks that pay higher interest rates or if new types of lenders emerge, thereby eroding bank profits, that could change things. And, of course, there’s always the risk of broader changes in the economy that are out of the banking sector’s control.

Krishnamurthy: That’s an important caveat. If the traditional business model in banking stops working, our paper stops working.

__

DeMarzo is the Philip H. Knight Professor, Interim Dean and John G. McDonald Professor of Finance at the ��ɫ��Ӱ Graduate School of Business. Krishnamurthy is the John S. Osterweis Professor of Finance at the ��ɫ��Ӱ Graduate School of Business.