How coronavirus could alter Fed’s thinking about stress tests
WASHINGTON — The Federal Reserve plans to publish regular stress test results next month based on banks' pre-coronavirus condition, but a supplement to the annual assessment — focused on how institutions are handling the economic shocks from the pandemic — could get most of the attention, observers say.
On top of releasing the standard results for the Dodd-Frank Act Stress Tests and Comprehensive Capital Analysis and Review, the Fed has said it will conduct “sensitivity analyses” to examine banks' responses to COVID-19. The addendum will consist of “alternative scenarios and certain adjustments to portfolios to credibly reflect current economic and banking conditions.”
A month out from the publication of the 2020 stress test results, it’s still a mystery how the sensitivity analyses will be used and whether banks' performance in the additional exercises will be made public. Some have posited that the Fed could use the sensitivity analyses more than the standard DFAST and CCAR results to make decisions on whether individual banks can continue planning capital distributions.
“It could be that the Fed could take the sensitivity analysis, and if they saw results that were especially concerning for them, use that as the basis of individual engagement with a financial institution around its capital plan, as opposed to formally incorporating it into CCAR,” said Jeremy Newell, a partner in the financial services group at Covington & Burling.
This year's DFAST and CCAR tests have prompted some questions about the relevance of the exercises, since the assessments will not reflect the current economic reality: a historically sharp downturn resulting from the pandemic. As is typically the case, the Fed said it is running this year's tests for the 34 eligible banks with over $100 billion of assets based on their financial data from last year, before the coronavirus.
“In many ways we're putting the stress test to its own test of credibility, because, according to the Fed, the purpose of a stress test is to ensure that large bank holding companies are able to lend to households and businesses during a severe recession,” said Gaurav Vasisht, former senior vice president of the Volcker Alliance.
Yet the Fed is adding the accompanying sensitivity analysis to better understand how banks are responding to financial pressures that may have been brought on by COVID-19.
Fed Vice Chairman for Supervision Randal Quarles told the House Financial Services Committee this month that the planned sensitivity analyses would factor in several hypothetical economic outcomes related to both the pandemic and a subsequent recovery, including potential losses in different kinds of assets such as commercial real estate.
The stress tests generally assess a bank's ability to withstand simulated economic conditions, from baseline to severely adverse.
Observers have speculated that the supplemental analysis is designed to compensate for an inherent flaw in this year's test: What may have seemed like an extremely severe downturn last year is the current reality for 2020.
“What has happened is probably beyond the worst-case scenario that was ever envisioned for stress testing,” said Joseph Lynyak, a partner at Dorsey & Whitney. The Fed "is probably going to be looking at that very, very closely and analyzing bank capital positions and liquidity positions from what used to be a worst-case scenario, and now is a reasonably based adverse scenario.”
But with the release of the 2020 test results approaching, observers still have a lot of questions about the supplemental test. Chief among them may be whether the Fed discloses results from the sensitivity analyses to the market, or whether the additional exercise will be conducted confidentially like much of the Fed’s other supervisory work.
“If the function of the CCAR exercise is to provide more transparency and confidence to the market, that might suggest results of the sensitivity analysis will be released publicly,” said Newell. “If the goal really is a supervisory assessment designed to better the Fed’s understanding of the financial situation of the firm, that would likely be like all other aspects of the supervisory process, confidential and supervisory in nature.”
If the sensitivity analyses were to uncover anything suggesting an individual bank might be unprepared for additional stress, it would be in the Fed’s best interest to keep the results of those exercises private so as to not “stoke the public’s concern,” said Lynyak.
“They're going to be very, very circumspect about creating a public perception that there's a bank capital issue that is floating around,” he said. “I think it's going to be waiting to see on our part to be looking past the looming dark walls of the Federal Reserve.”
Ultimately, more information on what exactly the sensitivity analyses will entail would be helpful, said Vasisht.
“My reading [of a sensitivity analysis] is that the Fed wants to update the assumptions and the numbers in some way that would more accurately reflect reality,” he said. “If there's a way to do that, then I think the Fed should disclose so people can think about whether it’s an effective way of considering the new reality.”
Some observers theorize that the Fed could emphasize the sensitivity analyses more than the standard stress test results in determining any changes to a bank's capital distributions.
Banks are required to submit their planned capital distributions for the year to the Fed as part of CCAR. The Fed then tests a firm’s capital adequacy under different economic scenarios, factoring in how much money banks plan to use to pay dividends to shareholders or buy back stocks.
The nation’s eight largest banks have said that they will suspend share buybacks through the end of the second quarter, and JPMorgan Chase CEO Jamie Dimon even said Tuesday that American banks won’t resume their share buyback programs until executives can see "the white of the eyes of the recovery.”
But most banks have continued to commit to paying dividends to investors, and the Fed has made no move to prohibit the firms it regulates from doing so, although Quarles has indicated that the sensitivity analyses could be used to determine whether or not it would be appropriate for a bank to make capital distributions.
“Our regulatory framework requires us to do that analysis before we would … determine … the need of banks to conserve capital,” Quarles said in an exchange with Sen. Brian Schatz, D-Hawaii, while testifying before the Senate Banking Committee in May.
The Fed may indeed be more suited to use the sensitivity analyses instead of the CCAR exercises this year to determine whether individual banks would be able to both make capital distributions and continue lending, according to some observers.
Dividend payments are “going to be a real focal point” in the Fed’s sensitivity analysis of a bank, said Lynyak.
“There's probably going to be a debate as to whether or not they want to be limiting that,” he said. “It probably all is going to be on an ad hoc basis as to what do the individual banks look like after they project out potential write-downs and loan reserves that are going to be necessary.”
But Vasisht argued that even additional stress testing wouldn’t be the best approach to determine whether a bank should be paying dividends.
“I think we're in such uncharted territory that regulators should prohibit shareholder payouts, at least until we figure out where the crisis ends and where recovery begins,” he said.
Regardless, the coronavirus pandemic will almost certainly change the way stress testing is conducted in the future, some say. Not only will it add to the data set of recessions on which the Fed bases its annual stress test scenarios. It could also force regulators to rethink how to assess a bank’s capability to deal with a sudden economic shock.
“No one ever did an analysis on what does a bank’s capital situation look like if you shut down the retail economy for four months,” said Lynyak. “Now that we know that that's not the worst-case scenario, that's real world, you're probably going to see a reinvigorated stress-testing set of protocols to be getting a firm handle on the effects on banks over the next 18 months to two years.”
Although the coronavirus won’t alter the core stress-testing processes and scenarios this year, it could be something to look out for in the future, Newell agreed.
“I’d certainly expect that the current economic stress will be used to inform and ideally improve CCAR going forward,” he said. “The goal of the CCAR stress scenarios has never been to predict what the next crisis looks like, given that that’s an impossible and futile task, and instead designed to be a hypothetical exercise, based on what we know from past severe recessions more broadly.”