Basel Finalization: The History and Implications for Capital Regulation – Part II

This is the second of a three-part series on Basel Finalization. To access Part I, please click here. To access Part III, please click here.

Structure and Scope

This is the second in a series of posts to prepare readers to assess the proposed implementation of the Basel Finalization package suggested by U.S. regulators, which we expect in the first half of 2023. This post focuses on the most likely potential options for the overall structure and scope of U.S. implementation. The choices made could result in significant differences in overall bank capital requirements, depending on the proposal’s structure and scope of application. The first post in the series described the purpose and principles underlying the Basel framework and its evolution over the last several decades, culminating in the 2017 Basel Finalization revisions.

Aside from the resulting level of overall capital requirements, the most fundamental questions regarding U.S. implementation of the Basel Finalization revisions are how they would affect the overall structure of the existing U.S. capital framework and which banks would be subject to those changes. As detailed in our initial post, the largest banks must comply with two “stacks” of risk-based capital requirements. These combinations of capital minimum and buffer ratios relative to specific RWA calculations, taken together, make up the total ratio of capital to risk-weighted assets a bank must comply with to avoid negative consequences, such as limitations on capital distributions. In addition to the risk-based capital requirements, banks also remain subject to leverage capital requirements, which are not the focus of this post. [1]

Whether the U.S. regulators retain the dual RWA approach, and its resulting two-stack set of capital requirements—and, if so, which stack (or stacks) it will assess using the Basel Finalization revisions to calculate RWAs—will significantly influence firms’ capital requirements. The first part of this post outlines the two most likely structural options and highlights potential advantages and disadvantages of each. The second describes a few of the potential options for the scope of U.S. implementation.

The Two-Stack Structure of the Existing U.S. Capital Framework

The Advanced Approaches Stack

Under the U.S. capital rules, U.S. global systemically important banks (GSIBs) and firms with at least $700 billion in total assets or $75 billion in cross-jurisdictional activity (Category I and II firms, per the U.S. regulatory tailoring categories) must comply with a comprehensive set of minimum capital requirements and buffers, as measured against their advanced approaches RWA.

The required ratios of common equity tier 1 (CET1) capital to RWAs have certain minimum requirements and buffers:

  • A minimum requirement of 4.5 percent;
  • A capital conservation buffer (CCB) of 2.5 percent;
  • A countercyclical capital buffer (CCyB), set at the discretion of national regulators from 0 to 2.5 percent; and
  • A GSIB surcharge buffer, which only applies to Category I firms, varies by firm, and is determined annually.

In accordance with the internationally agreed-upon Basel standards, the U.S. has set minimum requirements and buffers for GSIBs; however, there are several deviations. One major deviation is that the U.S. requires each GSIB to calculate its GSIB surcharge through two methods: (1) the method agreed upon by the Basel Committee, known as “Method 1” in the United States, and (2) a U.S.-specific method, referred to as “Method 2”. The latter method leads to higher GSIB surcharges, since it replaces the substitutability indicator of Method 1 with a measure of weighted short-term wholesale funding. This results in a higher calculation compared with Method 1. As a result, GSIB capital surcharges in the United States are nearly double the amount they would be if the Basel standard were followed purely.

The Standardized Approach Stack

All firms with at least $100 billion in total assets—that is, firms in the U.S. regulatory tailoring Categories I through IV—must comply with a separate set of minimum and buffer requirements measured against their standardized approach RWA. For banks in Categories I and II, this set of minimum requirements and buffers is in addition to the advanced approaches stack we described. So for these firms, they must meet both sets of requirements and are effectively bound by the more stringent of the two.

Crucially, in addition to using a different methodology to calculate RWA, the standardized approach stack also uses somewhat different minimum requirements and buffers. Although the minimum ratio of 4.5 percent and the GSIB surcharge buffer (where applicable) do not vary, the standardized approach stack takes a very different approach to the CCB. Instead of simply applying the Basel framework’s static 2.5-percent CCB buffer, the standardized approach stack replaces it with the stress capital buffer (SCB), which can fluctuate annually based on the Fed’s stress test results and is floored at the 2.5-percent CCB requirement. Currently, the weighted-average SCB of Category I and II firms is 4 percent, or 1.5 pp above the Basel’s CCB requirement.

In addition, for firms with less than $250 billion in total assets and less than $75 billion in nonbank assets, wholesale short-term funding, and off-balance-sheet exposures (i.e., Category IV firms), the standardized approach stack does not include the CCyB.

The Current Two-Stack Structure of Capital Requirements

The resulting two “stacks” of CET1 capital requirements applicable to Category I and II banks are illustrated in this exhibit.

As of the third quarter of 2022, Category I and II banks face CET1 capital requirements of $819 billion under the standardized approach, 20 percent higher than the advanced approaches.[2] Moreover, none of those banks are bound by CET1 capital requirements under the advanced approaches, since the standardized approach results in higher requirements for each firm. This is mainly due to the use of a variable SCB in the standardized approach in place of the static 2.5-percent CCB.

Reconciling the Basel Finalization Package with a Two-Stack Framework: Options for U.S. Implementation

As noted in our introductory post, the Basel Finalization revisions significantly limit the use of firms’ internal models, the key element of the advanced approaches, and they increase the risk sensitivity of the standardized approach. The Basel Committee also announced, for the very first time, that each jurisdiction would be deemed “compliant” with the Basel capital framework if they adopt a framework that required their internationally active banks to meet the Basel framework’s minimum and buffer requirement, based on the Basel framework’s standardized approaches alone. This aspect of the Basel Finalization package, widely understood to be included at the request of U.S. regulators, suggests that U.S. regulators may consider the option of eliminating the advanced approaches stack altogether. As a result, it is an open question as to whether U.S. regulators will retain the two-stack approach. If so, it is also unclear whether and how they will modify each stack in terms of how RWA is calculated and what minimum and buffer requirements are affected.

Although U.S. regulators might resolve these questions in various ways, we next describe two potential approaches.

Option 1: Adopt a One-Stack Approach by Eliminating Advanced Approaches and Replacing the U.S. Standardized Approach with the Revised Basel Standardized Approaches

One option would be to eliminate the advanced approaches stack altogether and modify the standardized approach stack to reflect the Basel Finalization package’s full set of standardized approaches for purposes of assessing the capital levels of firms under the U.S. stack. The Option 1 exhibit assumes the Basel Finalization revisions would result in a 15-percent increase in risk-weighted assets and is set mainly for illustrative purposes.[3]

This approach has the advantage of simplicity, at least with respect to the requirements for Category I and II firms. These firms would comply with a single stack of capital requirements and calculate RWAs using the Basel Finalization package’s standardized approach. In addition, this approach would be clearly compliant with the Basel framework, which now expressly permits it.

However, the single-stack approach would also entail a highly significant increase in capital requirements due to the inclusion of operational risk and credit valuation adjustment (CVA) risk in RWA, without the offset of the 72.5-percent output floor afforded in the Basel Finalization package.[4] This increase, combined with a higher GSIB surcharge under Method 2 and a higher SCB relative to Basel’s 2.5-percent CCB, will further raise capital requirements for U.S. banks, which are already elevated. And, as noted in a previous post, further increases in capital requirements will result in economic costs that are expected to outweigh the benefits.

Another complication is that some of the shortcomings of Basel III are already addressed in the U.S. capital framework through the stress tests. Implementing the Basel Finalization changes without a corresponding recalibration of the assumptions in the Federal Reserve’s stress tests would result in a significant overcapitalization for market risk and operational risk. For example, the FRTB improves the current market risk framework that it replaces by better capturing tail risk. However, the global market shock in the Federal Reserve’s stress tests effectively does the same thing. So, this approach would also require adjustments to the stress test assumptions for avoiding overcapitalization of these risks.[5]

The single-stack approach could also affect the modeling of RWA in the stress tests. For example, given the higher risk sensitivity of Basel’s standardized approach, we could see significant increases in RWA over the projection horizon, due to the increased riskiness of loans to corporates and mortgages under stressed conditions.

Finally, the single-stack approach could present challenges with respect to compliance with the Collins Amendment, which is discussed in more detail below.

Option 2: Retain a Two-Stack Approach by Replacing the Advanced Approaches Stack with a New Stack Based on the Full Set of Basel III Standardized Approaches

A second option would be to retain a two-stack approach, but to replace the advanced approaches stack with an alternative stack based on the full set of standardized approaches including in the Basel Finalization package. The current “U.S.” standardized stack would remain as is.[6] This approach would be most similar to the existing framework, because it would maintain two stacks and only apply the SCB to the U.S. standardized approach stack.

From a regulatory perspective, this approach offers several potential advantages. First, it would avoid an increase in capital requirements that results from the first approach for most large firms. This is because the SCB, which may be higher than the static 2.5-percent CCB, and the GSIB surcharge under Method 2 would only be applied in the context of the U.S. standardized stack.[7] Even with the addition of the operational risk and CVA risk capital charges to the Basel standardized stack, most firms would nonetheless remain bound by the U.S. standardized stack instead. Second, the U.S. would avoid an increase in capital requirements and still maintain a 100-percent output floor in the Basel Finalization stack.

In addition, the U.S. SCB framework would remain untouched. Therefore, it would foster continuity in stress testing practices and promote consistency across U.S. firms’ requirements, since the Federal Reserve could continue to subject all Category I–IV firms to the same supervisory stress tests and use the same models to project capital levels under stress. Option 2 would still require an adjustment to the global market shock in the stress tests, to avoid double-counting for market risk in the capital requirements. U.S. regulators would also apply the FRTB to the current U.S. standardized approach, which would be subject to stress testing.

Finally, implementation of the Basel Finalization package would not disrupt firms not already subject to a two-stack approach (i.e., Category III and IV firms, as well as all smaller banks). It would not require the introduction of new elements of the Basel standardized approaches into “their” stack, thereby minimizing compliance costs associated with implementing the revised Basel requirements.

However, maintaining two separate capital stacks would pose important challenges as well. As we noted, RWAs calculated using the full set of revised Basel standardized approaches are likely to be higher than RWAs calculated using the current advanced approaches, due primarily to the introduction of the operational risk and CVA charges. So, although the advanced approaches stack only rarely binds Category I and II firms today, this approach would increase the likelihood that the new stack, calculated using the full set of revised Basel standardized approaches, would in fact bind firms. The capital requirements of those firms would therefore increase—particularly those with a SCB close to Basel’s 2.5-percent CCB and similar Method 1 and 2 capital surcharges.

The Scope of the Basel Finalization Package

For the largest U.S. banks (those in Categories I and II, per the U.S. regulatory tailoring categories), the key question regarding Basel Finalization is how it will apply. For slightly smaller banks, primarily those with at least $250 billion in assets (Category III), the key question is whether it will apply.

By its terms, the Basel framework applies “on a consolidated basis to internationally active banks.” However, current U.S. capital rules are not based on any such standard. Rather, these rules gradate the scope and scale of capital requirements that apply, based on a complex tailoring framework mandated by Congress and implemented by U.S. regulators in recent years. Currently, only GSIBs and firms with $700 billion or more in total assets or $75 billion or more of cross-jurisdictional activity (Category I and II firms) are subject to an advanced approaches stack of capital requirements. (This is discussed in greater detail in our previous post, which is based on Basel’s Advanced Internal Ratings-Based Approach.)

Firms outside Categories I and II are subject only to a standardized stack of minimum and buffer requirements based on RWA calculated under the current U.S. standardized approach. However, as we noted, the current U.S. standardized approach does not include separate RWAs for operational risk or CVA risk, two key elements of the Basel Finalization package’s revised standardized approaches. Although it seems likely these standardized capital charges will apply to Category I and II firms as part of whatever Basel Finalization implementation U.S. regulators pursue, it remains quite unclear whether U.S. regulators will also incorporate them into the capital requirements that apply to other firms (e.g., those in Categories III and IV).

It is worth noting that the introduction of standardized capital charges for operational risk and CVA are by no means the only changes that the Basel Finalization package made to standardized risk-weighting. As we describe in our first post, other elements of that package introduce substantially more risk sensitivity in the standardized risk-weighting of certain assets (e.g., certain commercial loans). Another crucial question for U.S. implementation then is whether U.S. regulators will incorporate these changes that increase the risk sensitivity of the standardized approach to credit risk into the capital framework that applies to Category III or IV banks (or other smaller banks), either apart from or in addition to any changes they might make in operational risk or CVA charges.

Applying Basel Finalization Revisions Only to Categories I and II

One approach would be for regulators to apply the Basel Finalization revisions only to firms in Categories I and II, as we discussed. Regardless of the structure chosen, however, the key implication for all other firms outside these categories would be the continued calculation of their RWAs using the current U.S. standardized approach, with no explicit operational risk and CVA charges.

Narrowly scoping application in this way would require the fewest firms to revise their capital planning and compliance processes, while maintaining U.S. compliance with the Basel framework. However, it would mean that the capital standards applicable to unaffected banks would not reflect those aspects of the Basel Finalization package that improve the risk sensitivity of certain standardized risk weights. It is also possible that such a narrow approach to implementation might perpetuate and/or exacerbate differences in the capital charges faced by U.S. banks of different sizes for the same activities or businesses.

In addition, as detailed in our prior post, the Collins Amendment places a floor on capital requirements equal to (1) the minimum requirements generally applicable to insured depository institutions under the agencies’ prompt corrective action regulations (i.e., the U.S. standardized approach stack of minimum requirements); and (2) those that applied to insured depository institutions as of July 2010. Should the Basel Finalization package result in a single stack of requirements, there would no longer be a simple, ongoing method for demonstrating compliance with the Collins Amendment. Given the inherently greater stringency of the revised Basel III standardized approaches for these firms, U.S. regulators might decide that a one-time assessment of compliance at the time of transition would suffice. There would certainly be substantial justification for doing so. This concern would be alleviated through the dual stack structure we described.

Expanding Application of Basel Finalization Revisions to Category III and IV Firms

Alternatively, U.S. regulators could effectively extend implementation of Basel Finalization to a broader set of banks by incorporating some or all of the revisions included in that package into the standardized stack that applies to them. This would further unify the capital standards applicable to a broader range of U.S. banks. But, as we noted, it could both carry high implementation costs and significantly increase the capital requirements applicable to some of the relevant banks.

Whether and to what extent any expanded application of the Basel Finalization package would do so will depend on precisely which Basel Finalization revisions were introduced. Some are likely to increase capital requirements (e.g., the introduction of a standardized capital charge of operational risk), while others are likely to decrease those requirements (e.g., greater risk sensitivity in corporate risk weights). Again, several permutations are possible in resolving such questions. Each involves different tradeoffs in terms of overall impact on capital requirements, risk sensitivity of the framework, compliance burdens and variation/consistency in requirements among different bank cohorts.

Compliance with the Collins Amendment would also be a significant consideration if the scope of application is expanded to include Category III and IV firms. If smaller firms must also comply with the Basel Finalization package’s full set of standardized approaches, the more lending-focused firms could see a decrease in capital requirements. This would occur primarily because the anticipated reduction in credit risk RWA would partially offset the inclusion of an operational risk charge. As a result, if the scope of application includes Categories III and IV firms and contains only a single stack of requirements, it could raise questions about compliance with the Collins Amendment for traditional lending banks in Categories III and IV.

Conclusion

The choice of exactly how (and where) to implement the Basel Finalization revisions within the overall structure of the U.S. capital framework will have significant secondary effects for other outstanding questions relating to implementation. These include scope of application, interplay with other elements of the U.S. capital framework and implementation of Basel framework elements other than the Basel III endgame revisions. All these follow-on decisions could significantly affect the overall capital requirements U.S. firms would face.


[1] The global systemically important banks are subject to 12 separate capital requirements: six risk-based capital requirements, two leverage-based requirements and four requirements for total loss-absorbing capacity.

[2] The analysis uses the stress capital buffers based on the 2022 stress test results and GSIB surcharges effective as of January 1, 2023.

[3] This estimate is based on an average estimate from Vice Chair of Supervision Randal Quarles on the largest banks (up to 20 percent), and the effect of the implementation of the Basel 3.1 package on U.K. banks by the PRA (13 percent).

[4] The output floor requires firms to calculate their RWAs under both the advanced approaches and the standardized approach. It puts a floor of 72.5 percent of the standardized requirement for the capital requirements determined under the advanced approaches.

[5] For example, BPI has previously discussed how FRTB implementation could require firms to overcapitalize for market risk. On the Overcapitalization for Market Risk Under the U.S. Regulatory Framework (April 21, 2022), available at https://bpi.com/on-the-overcapitalization-for-market-risk-under-the-u-s-regulatory-framework/.

[6] The U.S. regulators would likely require banks to replace the current Basel 2.5 approach for market risk under the current U.S. standardized approach with FRTB, to reduce the burden in maintaining two separate approaches for the measurement of RWA for market risk.

[7] The Basel Committee’s Method 1 GSIB surcharge is about half of the capital surcharge under the U.S.’s Method 2 GSIB surcharge.