The Ups and Downs of Interest Sensitivity

Some banks say interest income sensitivity disclosures are useless. Jessica Watkins' research shows otherwise.

 

In 2012, federal regulators proposed a new rule requiring banks to publicly disclose the impact of a sudden change in interest rates, known as an interest rate shock, on their revenue. The banks pushed back.

In a public comment letter, the West Virginia-based Pendleton Community Bank argued that "projecting the effect on income based on interest rate moves is at best a shot in the dark, at worst misleading to the reader." Utah-based Zions Bancorp responded more colorfully, calling so-called interest income sensitivity disclosures "as accurate as a weather forecast."

The rule never went into effect, and interest rate sensitivity disclosures remain voluntary. But a research paper recently published in The Accounting Review shows that such disclosures are more useful than the banks acknowledge. Written by Notre Dame assistant accountancy professor Jessica Watkins, in collaboration with Leslie D. Hodder (Indiana University) and Mei Cheng (University of Arizona), the paper finds that interest rate sensitivity disclosures accurately predict changes in bank revenue resulting from interest rate shocks.

That's important because banks make most of their money through net income interest — the interest they make from loans, minus the interest they pay on deposits — rather than commissions or fees. When interest rates change, whether because of market conditions, inflation or Federal Reserve action, that change shows up on a bank's balance sheet. "Interest rate shocks are large, sudden changes in interest rates," Watkins explained. "These can be driven by inflation, or changes in the real economy due to supply and demand shocks."

If the shock is big enough, a bank could go bust. Even if it stays solvent, revenue can swing wildly. That's why banks develop sophisticated internal models to predict how even the smallest changes in interest rates would affect their balance sheet, allowing them to cushion the effects of a shock. "They don't know how much their interest revenue and interest expense will change, because they don't know the drivers of that," Watkins said. "But they have tools at their disposal to hedge that risk." For instance, banks can recalibrate their ratio of loans to deposits, or change the interest rates they charge.

Financial regulators already require banks to disclose some information about their interest rate risk, including maturities and rates for their deposits and loans. But disclosing interest income sensitivity is a relatively new idea. It was introduced by the Securities and Exchange Commission (SEC) in 1997, as part of a rule requiring financial institutions to provide more quantitative and qualitative disclosures about their exposure to market risk. Interest income sensitivity disclosures were voluntary, and early research found no correlation between firms' predictions and the actual outcomes — bolstering the argument that the disclosures were unnecessary.

But because few banks initially issued the voluntary disclosures, that early research, conducted by Watkins' co-author Leslie Hodder, suffered from a low sample size. When Hodder and her colleagues revisited the data in recent years, they found that many more banks were issuing disclosures, and that their accuracy had greatly improved, possibly because the banks' modeling had become more sophisticated.

Watkins and her co-authors were able to empirically test the accuracy of the banks' disclosures by measuring their sensitivity predictions against the banks' real-world performance. The researchers found that the disclosures were indeed predictive of how bank revenue changed in reaction to interest rate swings. What's more, they found strong evidence that financial analysts and investors were using the disclosures to evaluate the financial strength of banks. Analysts used the disclosures to forecast future net income, while equity investors calibrated their market response to interest rate shocks according to interest income sensitivity disclosed by banks.

"We found that analyst forecasts were more accurate in projecting future net income when banks provided the disclosures, relative to banks that did not provide them," Watkins said, "with the caveat that we can only assess the accuracy of the disclosure for the banks that provide them." Because not all banks disclose their net income sensitivity, Watkins and her co-authors acknowledge the possibility of selection bias; perhaps the banks that choose to disclose are simply better at predicting the impact of interest rate shocks than the banks that don't disclose. Although they attempted to control for this selection bias, Watkins admits their findings would be stronger if all banks were required to issue the disclosure.

Why might some banks resist disclosing their interest income sensitivity? Watkins said that some don't want to incur the cost in time and money to convert their internal models into the standard set of measurements requested by regulators. Others may want to protect sensitive firm information from competitors, or insulate themselves from legal liability if their disclosures prove inaccurate. Currently, interest income sensitivity disclosures are protected by a "safe harbor" clause, allowing firms to update incorrect information without incurring a penalty. But if they became mandatory, regulators could fine companies that release misleading statements.

"They might be comfortable having those models internally and using them to manage their risk, but providing them externally would require significant costs," Watkins said. "It raises the bar for verifiability and accuracy." On the other hand, some banks choose to voluntarily disclose the information, often to increase market transparency and bolster investor confidence.

Because there are unknown costs associated with making such disclosures mandatory, Watkins and her co-authors take no position on what action regulators should take. Their research establishes that interest income sensitivity disclosures are both accurate and useful — contrary to what some banks have claimed — but it doesn't perform the kind of cost/benefit analysis regulators conduct before issuing a new rule. "We view this research as a necessary step if regulators are to require these disclosures," Watkins said, "but we can't say it's sufficient. This gives regulators some reasons for why they might want to do this."

In the decade since regulators from the SEC and the Financial Accounting Standards Board (FASB) first proposed the disclosure, inflation has averaged about 1.89 percent a year. But in 2021 it jumped to 4.8 percent, the highest rate since 1982. The rapid rise in prices has raised public alarm and highlighted the importance of financial risk management. It may even prompt regulators to revisit the idea of mandatory interest income sensitivity disclosures.

"If these conditions continue, you could see this back on the SEC's agenda," Watkins said. "Ultimately, the squeaky wheel gets the grease. As this becomes more of a problem, you might see this revisited."

 

JESSICA WATKINS is an assistant professor of Accountancy. Her research centers on capital market participants' use of accounting information and disclosure.

 

Published

"Usefulness of Interest Income Sensitivity Disclosures"

The Accounting Review (2021)

Mei Cheng (University of Arizona); Leslie D. Hodder (Indiana University); and Jessica Watkins (University of Notre Dame)