This study recognized interest rate banking risks five years before the crisis

The most surprising thing about the demise of Silicon Valley Bank (and Signature and First Republic) is that depositors, investors and regulators were taken by surprise. As in nearly all financial crises, the warning signs were abundant and hiding in plain sight. And as in nearly all financial crises, there were folks sounding those warnings, but their calls went unheeded.

Back in 2018, when interest rates around the world were being held abnormally low, I helped coordinate a study by experts from 19 central banks, under the auspices of the Bank for International Settlements (BIS), titled “Financial stability implications of a prolonged period of low interest rates.” The study assessed the risks to banks, insurance companies and pension funds under three scenarios: a baseline in which inflation and interest rates gradually rose to more normal levels; a “low-for-long” scenario in which inflation and interest rates remained depressed; and a “snapback” scenario in which interest rates and inflation stayed abnormally low until 2023, when they rose precipitously.

The snapback scenario turned out to be quite prescient, although, obviously, no one in the study group could have anticipated the global pandemic that kept inflation and interest rates low during 2020 and 2021, nor the later surge in inflation that triggered a corresponding surge in interest rates. The study group’s identification of risks associated with the snapback scenario also turned out to be prescient.

The paper noted, first, that rising interest rates in the advanced economies might trigger capital flight and asset selloffs in emerging market economies (EMEs). Indeed, many frontier and developing economies have run into trouble recently, although the major EMEs have been doing surprisingly well.

Second, rising yields could trigger a liquidity squeeze for insurance companies and pension funds that were using derivatives to protect against downturns in interest rates, because declines in the value of their contracts would compel them to post additional collateral with their counterparties. This is exactly what happened last fall in the United Kingdom, when a sharp rise in yields led to a surge in margin calls for pension funds invested in so-called “liability driven investment” schemes, triggering self-reinforcing fire sales of bonds that ultimately required intervention by the Bank of England.

Finally, the BIS paper worried that in the event of a snapback in interest rates, “banks would likely experience valuation losses on long-duration assets and credit losses on loans. Adaptations to maintain profitability during low-interest rate periods, such as lengthening asset maturities and shifting loans to the interest-sensitive real-estate sector, would exacerbate the effects of a subsequent snapback.” Fast-forward to the present: SVB, Signature Bank, First Republic Bank…voilà!

The warnings of the BIS paper largely fell on deaf ears. It drew scant attention when it was released in July 2018, and there was little follow-up by regulators around the world on its three main policy recommendations: (1) enhanced monitoring of financial institutions’ exposure to low-for-long and snapback risks, especially through stress tests; (2) collection and analysis of appropriate firm-level data to monitor exposures and risks; and (3) adoption of appropriate resolution strategies. Even as recently as last November, the Federal Reserve’s Financial Stability Report paid at best limited attention to the prospective effects on bank capital of the on-going run-up in long-term yields. And the “severely adverse” scenario in the Fed’s bank stress test for 2023 centered around a global recession and lower, not higher, interest rates.

Would an earlier focus on the dangers to bank balance sheets posed by rising long-term yields have helped prevent the disruptive events of early March? On the one hand, some observers argue that the demise of Silicon Valley Bank represented a failure by its supervisor, the Federal Reserve, to intervene forcefully enough, rather than a failure to appreciate the effect of rising interest rates on unhedged long-duration bonds. Indeed, SVB’s direct supervisor, the Federal Reserve Bank of San Francisco, had identified deficiencies in the bank in late 2021.

On the other hand, had concerns about the impact of rising long-term yields on bank balance sheets been more prominent and more widespread throughout the regulatory community, the problems with SVB, Signature and First Republic might have been identified earlier and the Fed might have acted more forcefully to induce them to address their problems.

Financial stability analysis requires the repeated consideration of scenarios that might seem quite remote from prevailing conditions. I expect that, going forward, balance-sheet exposure to changes in interest rates will receive considerably heightened scrutiny. The challenge for regulators will be to scrutinize for the risks that seem far-fetched today before they take us by surprise tomorrow.

Steven Kamin is a resident scholar at the American Enterprise Institute (AEI), where he studies international macroeconomic and financial issues.