(Bloomberg Opinion) -- The U.S. Federal Reserve faces the daunting task of maintaining confidence in the banking system during one of the worst economic crises on record. Its latest round of bank stress tests has fallen woefully short. It must do better.
The coronavirus pandemic promises to be a challenge for banks, as people and businesses increasingly fall behind on their obligations. Stress tests are supposed to help in such situations, by showing how bad the losses can get, and by ensuring that banks have enough loss-absorbing capital to avoid distress and keep lending.
Back in 2009, during the subprime mortgage crisis, the original U.S. stress tests did well. They succeeded in part because officials chose to report bank-specific loss estimates -- a move that many regulators saw as radical and potentially destabilizing. But as Timothy Geithner, then Treasury Secretary, put it, “even bad news would be better than no news.” He was right: The level of transparency made the U.S. tests much more credible than their European counterparts, and banks were able to raise significant capital in private markets.
In recent years, though, the annual tests have become a different sort of exercise. The largest banks have been celebrated for passing and making large payouts to shareholders. Regulators have boldly concluded that the “capital building phase of the post-crisis era is now complete.” Yet some — including me — have doubted the results, in part because the tests rely on regulatory capital measures that are lagging and imprecise, and often at odds with what market-based measures suggest about banks’ health.
This week’s stress tests laid bare the limitations of the current regulatory regime. The tests’ original worst-case scenarios — devised in February, before the coronavirus crisis took hold — included shocks such as a 10% unemployment rate that seem laughably optimistic today. In an effort to provide a more real-time measure of financial resilience, regulators performed a separate “sensitivity analysis” that featured harsher scenarios. But rather than learning from the success of the 2009 tests, they chose not to release bank-specific results, citing the “limitations” and “considerable uncertainty” created by the pandemic.
Their rationale is indefensible. Surely, whatever uncertainties exist today, a decade into regular stress testing, pale in comparison to what officials were facing when they designed the original tests “on the fly” and nonetheless managed to disclose bank-specific loss estimates. The lack of transparency will not “bolster public confidence,” as Vice Chairman for Supervision Randal Quarles expects. On the contrary, it is already undermining trust, leading observers to speculate that potential losses must be severe if they cannot be made public. The result may be a damaging bifurcation, as the strongest banks disclose their results and the rest end up stigmatized.
Presumably also in the name of bolstering confidence, the Fed failed to use the stress tests to halt dividend payments. Instead, it will allow billions of dollars in equity capital to leave large financial institutions even though many firms could well find themselves undercapitalized in the coming months. And in a further abrogation of their duty to ensure financial stability, regulators approved measures weakening rules designed to curb speculative trading and protect government-insured depositary subsidiaries from losses on derivatives.
The Fed needs to change its approach, and fast. First, it should let the public know how individual banks performed in its Covid-19 sensitivity analysis. Second, it should prevent large banks from disbursing capital for the foreseeable future. Third, it should require banks that failed the COVID portion of the exam — which appears to be at least a quarter of the participants — to submit plans to address their capital deficiencies.
More broadly, the regular annual stress tests are clearly not sufficiently stressful. The exercise should be an ordeal, with realistic worst-case scenarios, rather than a celebration of banks’ dividend and share-buyback plans. The coronavirus has taught us that the scale and scope of crises are hard to know. So banks must have ample capital to weather the unexpected, and regulation should respond more automatically — for example, by cutting off capital distributions as downturns begin. This will remove the burden of action from officials wary of rocking markets in difficult times.
In a speech last week, Quarles recognized that regulators “simply would not have been doing our jobs” if they had performed stress tests as usual. On this point he is correct: The pandemic requires them to make stabilizing the financial sector their foremost priority. They should get to it.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
Natasha Sarin is an assistant professor at the University of Pennsylvania Law School and an assistant professor of finance at the Wharton School.
©2020 Bloomberg L.P.