How to Take the Stress Out of Stress Testing

If the outcomes of stress tests are predictable, why go through the time and expense of executing such complex and rigorous exams? That’s the question du jour in the wake of the Federal Reserve’s February 15 kickoff of its annual stress testing program, which this year will include four macroeconomic scenarios.

The most stressful “severely adverse” scenario in this year’s Dodd-Frank Act Stress Test (DFAST) is similar to last year's iteration, and banks continue to be robustly capitalized. Moreover, they are subject to the stricter lending standards and capital conservation practices that were adopted after last year's bank failures.

 Cristian deRitis

The idea of the 2024 test, as always, is to ensure that large banks have enough capital and liquidity to survive a variety of recessionary scenarios. Toward that end, the Fed has added exploratory scenarios to this year’s stress test that require banks to consider everything from deposit runs to funding pressures to supply-chain disruptions to the impact of declining commercial real estate and volatile interest rates. However, despite all that, we can still confidently predict that all 32 participating institutions will be deemed to have sufficient capital when the results are published this summer.

This brings us right back to the point about the purpose of stress tests. If large banks are simply not going to fail the annual capital test, then why go forward with it?

Well, firstly, stress tests highlight potential weaknesses. Even if a bank survives a severe recession, its balance sheet may be weakened to the point that it prompts depositors to switch to other institutions.

Given that much of a bank’s value is derived from its deposit franchise, its long-run viability is questionable. The stress test can illuminate particularly vulnerable asset classes, investment positions and counterparties that bank managers and regulators may want to scrutinize further.

Secondly, just as firefighters and police officers regularly run drills, banks need to conduct stress tests to keep their crisis management processes and procedures up-to-date and ready to respond when needed. Practicing the response process is beneficial, even if the next real-world threat encountered by a bank differs significantly from what is envisioned in the stress test.

These benefits notwithstanding, the current stress testing system is frustrating for banks, because parts of it are extraneous. There are, however, steps that banks and regulators can take to address this problem.

Unnecessary Stress: Three Steps for Simplifying the Process

While bank risk managers would say that regular stress testing of their portfolios is valuable, many would also agree that the way testing is conducted today is unnecessarily burdensome — and may, in fact, divert attention from a stress test’s objectives.

I offer three simple ways to make stress testing less stressful:

1. Regulators, including the Federal Reserve, should take a more structured approach to coordinating the stress testing process. Announcing scenario release dates and the number of scenarios in advance are small steps that can improve the quality of the results and eliminate the inefficiency of having bank staff on “standby” for weeks on end. Standardizing scenario formats and providing sufficient background details will ensure consistency between the expectations of banks and regulators.
   
   
2. Banks should run their own scenario analysis quarterly — or even monthly — to supplement regulatory stress testing. This will ensure they are able to evaluate scenarios in short order. Moreover, it will also allow bank risk managers to anticipate official stress test scenarios, giving them an opportunity to adjust their portfolios and mitigate potential threats well ahead of the official test.
   
   
3. Best practices for conducting stress tests should be shared across institutions. While banks compete in many areas and need to protect their proprietary models and trade secrets, stress testing should not be a competition. Encouraging an open exchange of best practices and processes can make the entire banking system more efficient and robust, promoting standardization and comparability of stress testing results by regulators and the investment community. Greater transparency regarding regulators’ own stress testing models and analysis can facilitate this process.

Exploring Stress

Unique to this year’s DFAST was the inclusion of two exploratory scenarios designed to evaluate the resiliency of banks against acute stress on their funding.

The first of these scenarios considers the situation where supply disruptions cause inflation and interest rates to rise during a moderate global economic recession. Higher interest rates are assumed to cause depositors to seek higher-yielding investments such as money market funds.

Banks would then be required to raise interest rates paid on deposits or to substitute deposits with more expensive sources of wholesale funding. Adding to the financial burden, the higher interest rate environment would cause mortgage originations to fall as demand for refinancing recedes. Lastly, under this scenario, the U.S. dollar is assumed to appreciate against most other currencies, pressuring exports.

The second exploratory scenario combines persistently high inflation and rising interest rates with a severe global recession, leading to even higher credit losses than the first scenario.

In addition to estimating the effects of these two scenarios on losses, revenues and capital, the eight globally systematically important banks (GSIBs) need to consider the impact of two sets of market shocks on their portfolios.

The first exploratory market shock is characterized by a sudden dislocation to financial markets, given expectations of reduced global economic activity and tighter financial conditions. An increase in anticipated defaults leads to a widening in credit spreads and equity price declines, while volatility rises from heightened market uncertainty. This, in turn, leads to higher margin requirements, causing some hedge funds to unwind their positions at a loss, pressuring banks with large exposures to these counterparties.

The second exploratory market shock considers similar effects to credit and equity markets, while altering the path of Treasury rates as inflation expectations decline. Commodity prices are assumed to fall under this scenario, while the price of gold and other precious metals rises as investors diversify their portfolios. Like the first exploratory market shock, this shock seeks to quantify each bank’s risk to their top counterparty exposures.

The addition of the exploratory scenarios to DFAST is a reasonable reaction to last year’s high-profile bank failures, which exposed the vulnerability banks can have to funding stress. Depositor flight is a particularly acute risk, as the loss of confidence can spread quickly and feed on itself. An in-depth assessment of these potential vulnerabilities will help both risk managers and regulators identify weakness at both bank and systemic levels.

Parting Thoughts

No amount of stress testing can eliminate risk, but more stress testing can significantly reduce it.

Given the substantial increase in bank capital following the global financial crisis, all of the banks participating in this year’s Federal Reserve stress test are likely to pass. However, the value of the stress tests isn't just in the result, but also in the process. The exercise provides an opportunity for banks to enhance their operations and identify potential weaknesses in their strategies that could expose them to disproportionate and undercompensated risks.

The introduction of two exploratory scenarios designed to probe particularly vulnerable aspects of banks’ balance sheets is a beneficial and welcome addition to the stress testing process.

However, while few would dispute the value of more testing, the execution of the test itself offers opportunities for improvement. The DFAST stress test is plenty stressful on its own; there is no need to add unnecessary complexity.

By streamlining the process, we can ensure that the focus remains on identifying and mitigating potential risks, strengthening the resilience of the banking system.

Cris deRitis is the Deputy Chief Economist at Moody's Analytics. As the head of model research and development, he specializes in the analysis of current and future economic conditions, consumer credit markets and housing. Before joining Moody's Analytics, he worked for Fannie Mae. In addition to his published research, Cris is named on two U.S. patents for credit modeling techniques. Cris is also a co-host on the popular Inside Economics Podcast. He can be reached at cristian.deritis@moodys.com.