The Federal Reserve on Thursday released the initial results of its annual bank stress tests, in which it puts bank holding companies through various adverse economic scenarios to see if their regulatory capital can withstand severe hypothetical downturns. The Fed tested 33 banks, each with more than $100 billion in assets.

Investors were carefully watching this year's results because the coronavirus pandemic has created actual economic scenarios that were unfathomable just a few months ago. Because of the virus, the Fed conducted three more "sensitivity analyses" to see how a bank's balance sheet would perform in a V-shaped, U-shaped, and W-shaped recovery.

The results showed that while most banks are reasonably well capitalized, some came uncomfortably close to their required minimum capital threshold in the most severe scenarios. And investors in the sector should understand that some bank dividends may be at risk in the third quarter.

Here is what we learned from the Fed's report.

Federal Reserve

Image source: Getty Images.

The Fed is restricting dividends

Perhaps the biggest result of the stress testing was the Fed's decision to restrict capital distributions, including dividends. Seeing that the capital levels at banks may take a significant hit as the coronavirus persists, the Fed capped dividends "to the amount paid in the second quarter," meaning increasing the dividend is not allowed in the third quarter. Furthermore, the Fed said it would limit dividend payouts to "an amount equal to the average of the firm's net income for the four preceding calendar quarters."

That means some banks may have to cut their dividends because the last two quarters have been characterized in general by low profits, or maybe even losses, potentially limiting the amount that can be paid out to shareholders. This does not bode well for a bank like Wells Fargo (NYSE:WFC), which reported only $0.01 in earnings per share in the first quarter and is not expected to do much better in the second quarter.

The Fed also suspended stock buybacks, although that wasn't exactly a huge surprise considering banks already suspended them earlier this year and were not expected to resume them in 2020.

A range of results

One of the things the Fed watches in its annual stress test is banks' common equity tier 1 (CET1) capital ratio. That's the base level of capital a bank needs to maintain so that even after accounting for lots of unexpected loan losses, it can still continue to extend credit to people and businesses during a downturn.

As mentioned above, the Fed tested banks under three additional severe scenarios this year. Those included a V-shaped scenario categorized by a very sudden and harsh downturn followed by a quick recovery; a U-shaped scenario that includes a more gradual and prolonged economic downturn and recovery; and a W-shaped scenario that includes a second round of COVID-19 disruption that begins later this year. These were the assumptions it used behind these scenarios.

Bank Stress Test Economic Assumptions

Image source: Federal Reserve.

As you can see above, the Fed assumed the worst unemployment rate at 19.5% during a V-shaped scenario, and the worst GDP contraction of 13.8% during a U-shaped scenario. The 10-year treasury fell the lowest to 0.5% during a W-shaped scenario. Here is what happened to the aggregate CET1 ratios of the banks tested over a nine-quarter outlook under these various scenarios.

Stress Test Results

Image source: Federal Reserve.

After being put through these hypothetical scenarios, banks still must be able to maintain a base minimum CET1 ratio of 4.5%. As you can see, banks are largely capitalized well enough to sustain this level in all of the various scenarios. But the lowest quartile of banks during a W-shaped recovery gets dangerously close to hitting that 4.5% level, which regulators never want to see happen. Still, the aggregate CET1 ratio was 7.7% in the W-shaped scenario. The projections by the Fed, however, did not incorporate capital distributions, although the Fed noted that common equity distributions during the first half of 2020 would have depleted aggregate capital ratios by about 0.5%.

Loan losses at banks could get much worse

In recent projections made June 10, the Fed projected actual unemployment to jump to 9.2% and actual GDP to contract 6.5% by the end of 2020. The interesting thing is that's not so different from the Fed's severely adverse scenario in its stress testing, where unemployment rises 10% while GDP contracts 8.5%. In this severely adverse scenario, the Fed projected that the weighted average loan loss rate for all 33 banks over a nine-quarter outlook is roughly $430 billion.

But in a W-shaped scenario, the weighted average nine-quarter loan loss rate rose to about $680 billion, or 9.9%, and even higher in a U-shaped scenario, with a weighted average loan loss rate of 10.2%, or more than $700 billion in loan losses. So, while things could get bad as they stand, they could clearly get a lot worse too, which is likely why the Fed is making restrictions on capital distributions.

Just the beginning

The release of these initial results is really just the beginning. On Monday, many banks are expected to disclose their capital distributions plans including whether they can maintain their current dividends. Additionally, because conditions are changing so quickly, the Fed plans to require banks to go through another round of stress testing with updated capital plans later this year.