BPI and Coalition of Trades Assert Basel Endgame Proposal Violates APA

Ladies and Gentlemen:

The Bank Policy Institute the Financial Services Forum, the Securities Industry and Financial Markets Association and the U.S. Chamber of Commerce1 are filing this comment on the joint notice of proposed rulemaking issued by the Board of Governors of the Federal Reserve System, the Federal Deposit Insurance Corporation, and the Office of the Comptroller of the Currency that would amend the capital requirements applicable to large banking organizations and those with significant trading activity.2 As described below, any final rule adopted based on this proposal would violate the Administrative Procedure Act,3 and the proposal’s defects can only be cured through proposal of a new rule.

Risk-based capital requirements are based on ratios that take a bank’s regulatory capital4 as their numerator and risk-weighted assets as their denominator; the proposal largely focuses on the latter and would fundamentally change how risk-weighted assets are calculated for nearly every type of asset and exposure. The risk weight attached to each asset has a direct and significant effect on whether a bank chooses to hold and how it prices that asset; collectively, those risk weights drive the bank’s overall capital requirement and its ability to compete for capital against other companies. Simply put, the stakes of how bank assets are risk-weighted are extremely high for banks, consumers, businesses and other bank customers, and the U.S. economy.

The proposal would result in a complete overhaul of how assets are risk weighted. First, the proposal would require all banking organizations with total consolidated assets of $100 billion or more to meet all applicable risk-based capital requirements as calculated under both (i) the existing U.S. standardized approach, as modified by the proposal and (ii) an entirely new “expanded risk-based approach” (“ERBA”) that includes entirely new standardized measures for credit risk, operational risk, and credit valuation adjustment risk (“CVA”).5 Second, the proposal would eliminate entirely the existing U.S. internal models-based approach to calculating risk-weighted assets for credit risk. Third, the proposal would introduce a new approach to market risk into both the U.S. standardized approach and ERBA. Through these revisions, the proposal would substantially increase the regulatory costs of about $22 trillion in assets held by banks subject to the proposal, which represent 80 percent of all U.S. banking assets.

Given the consequences of the proposal’s sweeping changes, one would reasonably expect the agencies to have undertaken a policymaking process that was deliberate, transparent, and based on all available data, and to have carefully considered all relevant aspects of the problems the proposal purports to address and potential alternative approaches to those problems. Unfortunately, the agencies failed to do so. The agencies provide no empirical justification or support for the weight assigned to numerous assets, even where they possess voluminous historical data that enables the evaluation of the risk of those assets. In the few cases where the agencies do state that they are relying on actual data, they have chosen not to disclose that data secret or to assert reliance on their collective “supervisory experience.” And, not surprisingly given its deficit of data and analysis, the proposal makes no serious attempt at weighing its costs and benefits. Instead, the agencies simply purport to implement agreements reached in 2017 and 2019 by the Basel Committee on Banking Supervision, though even then with some significant deviations, almost always in a more stringent direction.6 The ultimate result is a proposed rule that, if finalized, would be arbitrary, capricious, and an abuse of agency discretion. As such, the agencies should repropose the rules before finalizing them.

In a separate, forthcoming companion comment letter, we analyze the proposal in detail, and identify as a matter of substantive capital policy what revisions to capital requirements based on sound data and sensible analysis would look like. That letter effectively treats the proposal as an advance notice of proposed rulemaking, and we hope that it will assist the agencies in producing a new proposed rule that employs data and analysis to calculate risk, appropriately analyzes the costs and benefits of potential changes to their capital rules, and provides the public with adequate information on which to comment. As both a procedural and substantive matter, the current proposal fails that test.

In this letter, we focus on the proposal’s legal deficiencies, and organize our discussion as follows: Part I provides an executive summary of the proposal’s legal problems; Part II describes applicable legal requirements under the APA; Part III describes how the proposal violates both the procedural and substantive standards of the APA because it lacks a sufficient evidentiary basis, ignores evidence that is available, fails to explain the methodology and assumptions that underlie the proposal, and improperly fails to disclose underlying data and analysis from the public; Part IV describes how the proposed rule’s reliance on Basel Committee agreements as a floor for any U.S. requirement is improper and arbitrary; Parts V through VIII identify the numerous reasons why the proposal is arbitrary and capricious and thus violates the APA’s requirement of reasoned decision-making; Part IX explains why, as applied by the agencies in the proposal, the bank capital statutes would violate the non-delegation doctrine; and Part X describes why the agencies should issue a new proposal to amend the capital rules in a manner consistent with APA standards should they choose to move forward with any changes to those rules.

To read the full comment letter, please click here, or click on the download button below.

[1] See Appendix for more information on the Associations.

[2] See Regulatory Capital Rule: Large Banking Organizations and Banking Organizations With Significant Trading Activity, 88 Fed. Reg. 64028 (Sept. 18, 2023) [hereinafter the “Proposal”].

[3] 5 U.S.C. § 551 et seq.

[4] Regulatory capital primarily consists of shareholder equity in the bank.

[5] A “standardized” approach to calculating RWAs involves the creation of separate categories of exposures, with all exposures within a category receiving uniform risk weights across all banking organizations. This stands in contrast to internal models-based approaches to calculating RWAs, in which banks use models based on their own loss history, calibrated to a certain standard mandated by the regulator, to assign bespoke risk weights to their exposures.

[6] In 2017, the Basel Committee on Banking Supervision (the “Basel Committee”) adopted significant changes to the standard for calculating RWAs associated with credit risk and operational risk. With respect to credit risk, the revised standard permits firms to continue to use internal models to calculate RWAs, but placed a floor on the output of those internal models based on standardized measures of RWAs. With respect to operational risk, the revised standard eliminated internal models in their entirety and introduced a standardized approach to calculating RWAs. As explained below, policymakers intended that these changes would not result in meaningful increases to overall capital requirements. To address perceived shortcomings with the previous international standard (primarily, under accounting for tail risk), the Basel Committee also adopted a new standard for calculating RWAs for market risk in 2016, which was then revised in 2019 to address certain calibration concerns.