Board of Governors of the Federal Reserve System

I cannot support today's proposal to weaken the enhanced supplementary leverage ratio (eSLR). Enhancing the resilience of the U.S. Treasury market is an important objective that I share with my colleagues, but this proposal unnecessarily and significantly reduces bank-level capital by $210 billion for global systemically important banking organizations (GSIBs) and weakens the eSLR as a backstop. I am skeptical that it will achieve the stated objective of improving the resiliency of the Treasury market. Despite my reservations, I could support a much more modest adjustment to the eSLR were it to be accompanied by prompt, full, and effective implementation of the Basel III Endgame reforms to risk-based capital.

The proposal significantly reduces bank capital
The proposal would reduce tier 1 capital requirements by 27 percent at GSIBs' depository institution subsidiaries, resulting in a $210 billion decline in bank capital. At the holding company, the decline is 1.4 percent ($13 billion). The proposal would also reduce total loss absorbing capacity by 5 percent ($73 billion), and long-term debt requirements by 16 percent ($132 billion). Taken together, these changes would significantly increase the risk that a GSIB bank would fail, orderly resolution would not be possible, and the Deposit Insurance Fund would incur higher loses.

The decline in capital at the holding company level would be much worse if not for risk-based requirements; but if the eSLR is reduced, the risk-based capital requirements could go down as well. This can occur not only as a result of policy changes, but also by actions that banks can take to reduce their risk-based requirements. Looking solely at the leverage ratio without considering the limiting effect of risk-based capital requirements, the aggregate reduction in the eSLR requirement at the holding company level would be about $210 billion for GSIBs and $280 billion for their depository institution subsidiaries. Significantly lowering the eSLR in this manner increases the incentives firms have to game their risk-based requirements by lowering their risk-weighted asset density. This kind of gaming of risk-based requirements is precisely what leverage ratios are designed to block.

The proposal downplays the concern about the decline in bank-level capital by pointing to the expectation that holding companies serve as a source of strength to their banks. In practice, however, there are lots of examples of firms failing without supporting their bank subsidiary, as happened with Silicon Valley Bank in 2023. During the Global Financial Crisis, the Federal Reserve granted numerous exemptions in order to allow depository institutions to upstream capital to bail out their broker dealer affiliates—the opposite of the source of strength doctrine.

The reduction in capital is significantly larger than under the 2018 proposal
The declines in capital are much deeper than what the agencies have previously considered. This proposal would reduce the supplementary leverage ratio requirement by nearly three times as much for GSIB holding companies, in percentage terms, as the never-finalized 2018 proposal would have done. It also would reduce tier 1 capital requirements for GSIBs' depository institutions by 10 percentage points or $65 billion more than the 2018 proposal.

The proposal is unlikely to significantly enhance Treasury market intermediation, especially in times of stress
Treasury market intermediation primarily happens at the broker-dealer, but the overwhelming bulk of the capital depletion under the proposal happens in the bank. While firms could, in theory, use the additional headroom provided under the proposal to increase their participation in Treasury market intermediation, it is not clear that result would occur. Firms could just as easily shift to other activities with low risk-based capital requirements and significantly higher returns than Treasury market intermediation. Moreover, much of the capital that is freed up at the holding company level, where not otherwise constrained, is likely to be diverted to returning equity to shareholders, rather than intermediation.

Even if some further Treasury market intermediation were to occur in normal times, this proposal is unlikely to help in times of stress. If banks use up their excess capital in normal times, there will not be excess capital in stressful times. Moreover, firms' internal stress models measuring value at risk will likely limit Treasury intermediation when volatility increases, as they have in the past.

In short, firms will likely use the proposal to distribute capital to shareholders and engage in the highest return activities available to them, rather than to meaningfully increase Treasury intermediation.

The proposal erodes the transparency of the capital backstop.
In addition, the proposal takes a relatively straightforward leverage ratio and lowers it by using a risk-based GSIB surcharge metric, making it more difficult for the market to understand and rely on in times of stress. The way the proposal integrates the GSIB surcharge calculation is also at odds with the GSIB surcharge framework we use in the United States, where the Method 2 GSIB surcharge is generally the binding calculation. This proposal relies instead on the Method 1 calculation for no other reason than that it allows for a larger reduction in capital.

An alternative option to exclude Treasuries should not be pursued.
Lastly, I want to touch on the alternative options contemplated by the proposal that would exclude exposures to the Treasury and Federal Reserve. While there may be some superficial appeal to completely excluding exposures to the United States government because of its risk-free status, such a change would be ill-advised. Such a decision easily leads to a slippery slope—an erosion in global capital levels as other jurisdictions apply a similar logic to their own sovereign debt.

Conclusion
In conclusion, this proposal puts our banking system at risk by weakening capital of the largest banking organizations. Despite my reservations, I remain open to working towards a much more modest eSLR reform if paired with Basel III implementation.

Thank you.