Board of Governors of the Federal Reserve System

The strong capital buffers associated with Dodd-Frank reforms have enabled banks to play a constructive role in responding to the COVID-19 pandemic. It is a mistake to weaken banks' strong capital buffers when they are clearly proving their value in the first serious test since the global financial crisis. This is a time for large banks to preserve capital, so they can be a source of strength in a robust recovery. I do not support giving the green light for large banks to deplete capital, which raises the risk they will need to tighten credit or rebuild capital during the recovery. This policy fails to learn a key lesson of the financial crisis, and I cannot support it.

Stress testing ensures that large banks retain sufficient capital buffers to withstand severe recessions while continuing to provide vital credit to households and businesses. Despite criticism that the economic scenarios and market shocks in past stress tests were excessive and unrealistic, in fact, the actual COVID-19 shock to unemployment and gross domestic product has far exceeded the 2020 economic scenario. Moreover, the actual market shock in the first quarter of 2020 was similar in magnitude to the first quarter of the 2020 stress test scenario.

It is clear that recent changes in financial markets and the macroeconomic outlook could have a material impact on banks' risk profiles and financial conditions. So I support asking the banks to update and resubmit their capital plans. Under the rule, banks normally could not make any capital distributions until the Board provides banks notice of whether their stress capital buffer requirements are being recalculated. Temporarily halting shareholder payouts at large banks due to the COVID-19 shock would create a level playing field and allow all banks to preserve capital without suffering a competitive disadvantage relative to their peers.

Instead, the vote allows distributions to continue to be paid out, despite the material change in financial conditions. It implements a novel approach by authorizing third quarter dividends at a level equal to average net income over the prior four quarters—even though that past net income and more was already paid out in the prior quarters. The payouts will amount to a depletion of loss absorbing capital. This is inconsistent with the purpose of the stress tests, which is to be forward looking by preserving resilience, not backward looking by authorizing payouts based on net income from past quarters that had already been paid out.

The forward-looking sensitivity analysis indicates that many large banks are likely to need greater loss absorbing capital to avoid breaching their buffers in adverse circumstances next year. The wide range of loss rates and capital ratio outcomes in the sensitivity analysis reflects the unprecedented amount of uncertainty clouding the outlook. The banks themselves have noted the high level of uncertainty, and forecasters have placed a fair amount of weight on outcomes similar to the U- or W-shaped scenarios. Moreover, the sensitivity analysis results "do not account for the capital depletion that would result from common equity distributions over the projection horizon." Even without taking into account the authorized distributions, and only partially adjusting for the increased riskiness of the loans on bank balance sheets, the scenarios suggest that many banks could be operating within their stress capital buffers, and one quarter could be close to their minimum requirements.¹ Past experience shows that banks operating close to their regulatory minimums are much less likely to meet the needs of creditworthy borrowers, and the resulting tightening of credit conditions could impair the recovery.

Despite the substantial likelihood that banks will need larger capital buffers to absorb losses under plausible scenarios, the authorization permits distributions that will deplete capital buffers. Using backward-looking earnings as the basis for payouts in a forward-looking capital framework is problematic at a time when future earnings are likely to decline and required buffers are likely to rise. This action creates a significant risk that banks will need to raise capital or curtail credit at a challenging time.

1. Although the sensitivity analysis partially accounts for elevated draws on business credit lines and likely rating downgrades in certain industries, more widespread or steeper downgrades could lead to higher losses in future stress test analysis. In addition, the next round of stress tests could include a more complete picture of the credit quality of banks' commercial real estate (CRE) loans and fully incorporate changes in indexes of CRE prices, rents, and vacancy rates over the first half of 2020. Return to text