Fat profits for banks are actually a flashing red light warning of heightened risk, according to research by the US Treasury Department’s Ben Meiselman, Chicago Booth’s Stefan Nagel, and University of Michigan’s Amiyatosh Purnanandam.*

That’s because to generate high return on equity, known as ROE, banks have to make riskier investments. The researchers constructed a model to test ROE as a risk indicator and analyzed it in the context of the savings and loan crisis of the late 1980s, the 2008–09 financial crisis, and the 2023 banking turmoil that led to the collapse of Silicon Valley Bank.

ROE provides a robust indicator of risk from one downturn to the next, the researchers conclude. “A bank that earns high ROE in good times must have a combination of risky assets and high leverage,” they find. “ROE succinctly captures the combined effect of systematic asset risk and its magnification by leverage.”

The researchers argue that ROE offers a more accurate gauge of banks’ stability than the commonly used risk-weighted models favored by regulators, who have struggled for decades to come up with accurate and timely ways of measuring bank risk.

Historically, financial regulators have approached the challenge of measuring stability by devising rules to assign risk grades to various banking assets and activities. The value-at-risk model frequently used for bank trading books is one such example. Regulators then aggregate risks to form an overall picture of how an institution is likely to fare during troubled times.

Recommended Reading

From one crisis to the next, regulators have created and repeatedly revised these risk-based models. The exercise has turned into a perpetual game of cat and mouse, as different crises have grown out of losses related to interest-rate movements, mortgages, and sovereign debt exposure. Following the 2008–09 financial crisis, for example, regulators imposed safeguards designed to address weaknesses in the securitization market. But the models, even after being improved, can still be blind to newly emerging dangers. And when banks get into trouble, the economic damage can be widespread.

Focusing on ROE starts with the premise that high profits in good times signal a risky portfolio that’s likely to suffer in bad times. In testing their model, the researchers compared banks’ ROE a year before each of the three crises began with how it looked once they were underway.

ROE is a significant predictor of systematic tail risk exposure during crises, they argue, noting that the link between profitability and systematic tail risk is consistently strong. Their simple, profitability-based approach had significantly more predictive power than model-based measures.

But the ROE measure does have a weakness, the researchers acknowledge—it, like the risk-based models, can be slow to detect emerging dangers. The timing of risky activity does not always coincide with the timing of profits, which could delay when risk is detected.

Still, ROE-based risk analysis is a relatively simple approach that’s less vulnerable to manipulation, they argue, making it a potentially useful complement to model-based approaches.

*The researchers include the following disclaimer in their paper: “The findings, interpretations, and conclusions expressed here are entirely those of the authors and do not necessarily reflect the views or the official positions of the US Department of the Treasury.”

More from Chicago Booth Review

More from Chicago Booth

Your Privacy
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.