Part II in Our Series on US Bank Capital Requirements

• In this, the second of a series of blog posts on bank capital requirements and their impacts, we discuss forthcoming changes to U.S. bank capital requirements that are likely to occur as U.S. regulators implement the Basel III “Endgame” package of reforms.
• As we described in Part I of this series, current capital rules have ensured that U.S. banks’ capital levels are now strong and robust, as evidenced by their resilience during the COVID-19 induced downturn.
• The Basel “Endgame” package will significantly increase banks’ capital requirements over-and-above their already historically high levels in the past several years, resulting in significant additional funding costs for businesses, consumers, and investors.
• Policymakers should be transparent about the costs and benefits of implementing the Basel Endgame package, taking into account the totality of prudential reforms introduced since the Great Financial Crisis (GFC).

Background

In the first part of this series, we examined the post 2008-09 Great Financial Crisis (GFC) reforms to the U.S. prudential framework for banks, with a particular focus on capital requirements. As a result of those reforms, the quality and quantity of capital in the system was dramatically strengthened, particularly in the United States. Today, U.S. banks are subject to a panoply of capital requirements – up to nineteen separate requirements in fact. These include risk-based, leverage, stress testing, and loss-absorbing capital requirements – and as a result are better capitalized than at any point in recent history (see Figure 1). As the current Federal Reserve Vice Chair for Supervision, Michael Barr, stated during his confirmation hearing, “capital and liquidity in the system is very strong. The rules that Congress put in place after the financial crisis make it much less likely that such a financial firm could get itself into trouble and in a way that would cause problems for the broader economy.”^[1]

Figure 1. The Average Common Equity Tier 1 (CET1) Capital Ratios and Levels of All CCAR Firms Since 2009

Source: SIFMA Research Quarterly.^[2]

While the quality and quantity of bank capital is far stronger today than it was prior to the GFC, regulators have remained concerned that the current capital standards build in too much variability in the way risk-based capital requirements are calculated across banks.  To this end, the Basel Committee on Banking Supervision (BCBS) finalized revisions to its risk-based capital standards in 2019, commonly referred to as “Basel III Endgame” (hereafter referred to as the “Basel Endgame package”), which would promote greater standardization across risk-based capital requirements. Since then, national regulators have been working to implement the Basel Endgame package, with proposals issued in both the EU and the UK. The U.S. Notice of Proposed Rulemaking (NPR) implementing the reforms is due to be published by the federal banking agencies (the “Agencies”) in the coming months. Below, we take a closer look at what is in the Basel Endgame package, how it will change the U.S. capital framework and begin to discuss the potential impacts it could have on U.S. capital markets and the broader economy.

What Is In the Basel Endgame Package?

The Basel Committee has said that “[a] key objective of the [Basel Endgame package] revisions … is to reduce excessive [non-risk-based] variability of risk-weighted assets (RWAs).”^[3] The source of this variability is two-fold: it arises from the differences between the internal models used by the largest banks to calculate RWAs, as well as various ambiguities embedded in the existing Basel III framework.^[4] Consequently, the Basel Endgame package sets out stricter limits on the use and operation of internal models, and prescribes clear qualitative requirements to reduce ambiguities where possible. Figure 2 depicts the key components of the Basel Endgame package.

Figure 2. Key Components of the Basel Endgame Package.^[5]

In terms of the use and operation of internal models, the Basel Endgame package limits the ability of banks to use internal models as widely as they do today and creates stricter and more granular requirements for those internal models that are permitted. For example, the advanced internal ratings-based approaches will no longer be available for corporate and bank exposures. The package will also require or incentivize greater use of regulator-set standardized approaches to set capital requirements, which is achieved through a number of mechanisms, including by adding a minimum “aggregate output floor” for capital requirements calculated using the modeled approaches.

Take internal models for market risk capital requirements for example.  Historically, the adequacy and robustness of the market risk models are generally validated through back-testing (BT, a process that “involves the comparison of actual outcomes with model forecasts during a sample time period not used in model development at a frequency that matches the model’s forecast horizon or performance window.”^[6]). To set stricter criteria for the use of the internal models, the Fundamental Review of the Trading Book (FRTB), which reforms the way the market risk capital requirements are calculated under the Basel Endgame package, introduces an additional test – the Profit and Loss Attribution test (PLAT, which measures the adequacy of a bank’s internal models and prevents the bank from using materially inadequate internal models to calculate their capital requirements)^[7] – to complement the BT. A trading desk may use the internal models to calculate the market risk capital requirements only if it passes both the BT and the PLAT tests. Otherwise, the regulator-set standardized approaches must be used.

Constraining the use of internal models, placing greater reliance on the standardized approaches to set capital requirements, and adding more prescriptive qualitative criteria will result in significantly higher total risk-based capital requirements for almost all banks. For example, for large and internationally active banks, the FRTB could lead to over 63% increase in the market risk capital impacts and raise the Credit Valuation Adjustment (CVA) risk capital requirements by 7.7% according to the Basel September 2022 Quantitative Impact Study (QIS) report.^[8] These risk-based requirements capture the potential losses arising from banks’ capital market activities (e.g., trading, market making activities, underwriting, and meeting commercial end-users’ demand for hedging via derivatives).

In terms of reducing ambiguities embedded in the existing Basel III framework, the Basel Endgame package adds new, highly granular requirements designed to ensure consistent policy interpretation and implementation. A good example is the revised trading book and banking book boundary.  Historically, for purpose of the market risk capital rule, whether a financial instrument was considered part of the trading book (and hence subject to the market risk capital requirements) or the banking book (and hence subject to the credit risk capital requirements) depended on whether a bank had “trading intent” for this instrument, which could be interpreted differently by different entities.

During the GFC, some banks reclassified certain trading book instruments into their banking book as a result of the severe liquidity shocks and the concomitant reductions in the “fair value” (market value) of these instruments.  Such reclassifications were arguably permitted under the vague “trading intent” criteria. But they created non-risk-based variability of market risk RWAs across banks and the resulting capital requirements may not be commensurate with banks’ risks. To reduce the degree of arbitrariness and enhance clarity, reporting, disclosure and the integrity of capital requirements, the FRTB revises the trading book and banking book boundary, setting out a presumptive list detailing the instruments that must be considered as part of the trading book and those as part of the banking book.

In addition, the Basel Endgame package introduces an aggregate output floor to further limit potential capital benefits of using internal models. The output floor sets risk-based capital requirements calculated using modeled approaches at no lower than 72.5% of those required under the standardized approaches.^[9] And this only refers to risk-based capital requirements. There are also leverage ratio and other capital requirements, with the eventual binding capital requirements a bank must comply with determined by the largest of these different requirements.

How Will the Basel Endgame Package Impact the US Capital Markets?

As of end 2022, the U.S. capital markets funded nearly three quarters of all U.S. economic activity. This is a major contrast to other regions in the world, where most debt provided to non-financial corporations comes from bank loans. Figure 3 shows that the U.S. capital markets are the largest in the world, accounting for 41% of the $118 trillion in global equity market capitalization, or $48 trillion, and 39% of the $123 trillion securities outstanding across the globe, or $48 trillion. The U.S. capital markets continue to be among the deepest, most liquid and most efficient around the globe.

Figure 3.  The U.S. Capital Markets Highlights.

Source: SIFMA 2022 Capital Markets Outlook.^[10]

Even though they may not always be direct lenders, banks nonetheless play a critical role in facilitating capital formation and ensuring liquidity in the capital markets. The Basel Endgame package, and the FRTB in particular, will however significantly increase the capital requirements for banks’ capital markets activities and thus limit their capacity to provide liquidity support to a range of key funding markets and perform important capital markets-related activities (such as market making) on behalf of their clients.

In September 2022, the Agencies confirmed that they intended to implement “enhanced regulatory capital requirements that align with the [Basel Endgame package] issued by the Basel Committee on Banking Supervision …”^[11] While the Agencies are expected to implement the core elements of the Basel Endgame package, there is also an expectation, based on past experience, that they will impose super-equivalent requirements (often referred to as “gold-plating”), resulting in higher capital requirements than those contained in the Basel standards.

The expected material increase in capital requirements for U.S. banks’ capital market activities may arise from four potential sources – (1) the standalone impacts of the FRTB implementation; (2) the overlap between the FRTB and the Stress Capital Buffer (SCB) requirement; ^[12] (3) the interaction between the FRTB and the Collins Floor; and (4) additional constraints on the use of internal models.

The FRTB Standalone Impacts will be Significant

As discussed above, the capital requirement for market risk under the FRTB will be 63% higher than the current market risk rule.^[13] Additionally, the overlap between the FRTB and the SCB, the interaction between the Collins Floor and the FRTB, and any additional constraints on the use of the internal models will likely increase capital requirements even further.

The FRTB and the SCB Overlap will Lead to Double Counting Losses in a Bank’s Trading Operation

The SCB is the consequence of the Federal Reserve Board’s effort to simplify its capital rules for large banks by “integrat[ing] the Board’s stress test results with its non-stress capital requirements.”^[14] However, the FRTB is conceptually a stress test framework and “the revised internal models approach replaces VaR and stressed VaR with a single ES metric that is calibrated to a period of significant market stress”.^[15] The SCB replaced the Capital Conservation Buffer (CCB), which was fixed at 2.5% and remains in place in all other BCBS member jurisdictions.  In contrast to the CCB, the SCB floor is set at 2.5% – an example of the U.S. gold-plating of the Basel standards.  As the result of the 2022 stress test exercise, the SCB requirements for 21 of the 34 banks subject to the SCB rule are greater than 2.5%, and as high as 9.0%.^[16] Losses resulting from the stress test Global Market Shock (GMS) component, designed to stress trading positions held by banks with significant trading operations, are a key component of the stress losses and a major driver of excessive year-over-year variability of banks’ required SCBs for those banks subject to the GMS.^[17] Of the $612bn in aggregate hypothetical losses recorded in the 2022 stress test exercise, hypothetical GMS losses accounted for $100bn.^[18]

Both the FRTB and the GMS are designed to capture extreme potential losses arising from a bank’s trading operations. They share key conceptual similarities – both are stress test frameworks, both have risk factor shocks calibrated to equivalent deep market stresses, and both apply overly conservative correlation assumptions. For instance, risk factor shocks in both frameworks are calibrated based on a period of stress. In the 2023 stress test scenario, “shocks to risk factors in more-liquid markets, such as those for government securities, foreign exchange, or public equities, are calibrated to shorter horizons (such as three months), while shocks to risk factors in less-liquid markets, such as those for non-agency securitized products or private equities, have longer calibration horizons (such as 12 months)”.^[19] Similarly, the FRTB calibrates about 50% of the risk factors to a “liquidity horizon” of three-month or longer.^[20] Table 1 illustrates the conceptual overlaps between the FRTB and the GMS under the supervisory stress test. As a result, these potential extreme losses in a bank’s trading operations will effectively be counted twice (i.e., through both the FRTB and the GMS) for purpose of the risk-based capital requirements. Because of this double count, the SCB will likely exacerbate the capital impacts of the FRTB.

Table 1.  Key Overlaps Between the FRTB and the GMS

The FRTB and The Collins Floor

As our prior blog discussed, the current capital rules offer two approaches to calculate RWAs – the Advanced Approaches and the Standardized Approach.^[21] The Collins Floor requires the risk-based capital requirements calculated using the Advanced Approaches to be no less than 100% of the Standardized Approach. As a result, any increase in market risk capital requirement due to the FRTB will be fully reflected in increases to the Collins Floor.^[22] This stands in contrast to the Basel output floor, which provides a 5-year phase-in period with the final floor set at a much lower 72.5%.^[23] As of Q4 2022, the risk-based capital requirements for five out of the eight U.S. GSIBs were set by the Collins Floor.^[24]

The Internal Models Versus the Standardized Approaches

To ensure the robustness and adequacy of banks’ internal models, the Agencies issued the Guidance on Model Risk Management (SR 11-7) in 2011.^[25] The European Central Bank (ECB) embarked a Targeted Review of Internal Models (TRIM) at the beginning of 2016,^[26] and issued the ECB Guide to Internal Models in 2019.^[27] The EU’s CRR3 proposal[28] and the UK’s Basel 3.1 standards^[29] continue to allow the use of the internal models to set capital requirements. However, the Agencies have indicated that they are now intent on “replacing the Advanced Approaches with risk-based capital requirements based on the revised Basel standardized approaches for credit risk and operational risk”.^[30],[31] As a result, the Agencies will rely largely on the revised Basel standardized approaches to set the risk-based capital requirements. Such an outcome would disincentivize banks from enhancing and building expertise in the internal models which facilitate more accurate risk measurement and management overall, whereas “the standardized approaches [are] too crude and unrealistic to be useful tools for measuring and managing capital consumption and risk.”^[32]

This move away from internal models will not only lead to capital increases at the firm-wide level, but will affect how capital is allocated internally within firms, placing constraints on certain business lines more than others.^[33] The market’s experience with the Standardized Approach to Counterparty Credit Risk (SA-CCR) rule,^[34] which came into effect on January 1, 2022, provides a good real-world example of these sorts of impacts.^[35]

Conclusion

Capital requirements are a cornerstone of the prudential regulatory framework. The Basel Endgame package will bring substantial changes to the U.S. capital those requirements. These changes are expected to significantly increase banks’ capital requirements over-and-above their historically high levels. Higher capital levels have benefits (i.e., reduce the likelihood of a bank failure) but also costs (i.e., they limit banks’ ability to support capital markets and the broader economy, and make it harder and more expensive for businesses, consumers, and investors to obtain financing). Given policymakers chosen by both Republican and Democratic administrations have stated for some time that the current banks’ capital levels are strong and robust, the Agencies should carefully weigh the costs and benefits of implementing the Basel Endgame package and should be transparent about the rationale for further increasing capital levels.

Our next blog will focus on the costs/benefit analysis that should be applied to bank capital requirements and discuss the “optimal” level of bank capital.

 

Dr. Guowei Zhang is Managing Director and Head of Capital Policy at SIFMA.

Dr. Peter Ryan is Managing Director and Head of International Capital Markets and Strategic Initiatives at SIFMA.

 

---

References

^[1] https://www.congress.gov/117/chrg/CHRG-117shrg48337/CHRG-117shrg48337.pdf

^[2] https://www.sifma.org/resources/research/research-quarterly-us-financial-institutions/

^[3] https://www.bis.org/bcbs/publ/d424.pdf

^[4] https://www.bankingsupervision.europa.eu/ecb/pub/pdf/ssm.trim_project_report~aa49bb624c.en.pdf

^[5] Basic approach for CVA can be viewed as a much-simplified and non-risk sensitive alternative to the Standardized Approach.

^[6] https://www.federalreserve.gov/supervisionreg/srletters/sr1107.htm

^[7] PLAT compares a trading desk’s daily risk-theoretical P&L with the daily hypothetical P&L.  Material discrepancies in the distribution of the two P&Ls indicates the inadequacy of the desk internal models in accurately reflecting the market risk of the desk’s portfolios.

^[8] https://www.bis.org/bcbs/publ/d541.pdf

^[9] Given the stricter criteria for internal models, introducing the aggregate output floor will add undue complexity to the capital framework and create undue burdens for banks.

^[10] https://www.sifma.org/wp-content/uploads/2022/12/2022-Capital-Markets-Outlook-FINAL-FOR-WEB.pdf

^[11]https://www.federalreserve.gov/newsevents/pressreleases/bcreg20220909a.htm

^[12] https://bpi.com/on-the-overcapitalization-for-market-risk-under-the-u-s-regulatory-framework/

^[13] The revised CVA framework (i.e., the SA-CVA) is adapted from the FRTB sensitivity-based method and, as a result, it significantly increases CVA risk capital requirement.

^[14] https://www.federalreserve.gov/newsevents/pressreleases/bcreg20200304a.htm

^[15] https://www.bis.org/bcbs/publ/d457_note.pdf

^[16] https://www.federalreserve.gov/publications/files/large-bank-capital-requirements-20220804.pdf

^[17] https://www.sifma.org/resources/news/u-s-stress-test-capital-requirements-are-excessively-volatile-and-over-estimate-losses-identifying-the-problem-and-how-to-solve-it/

^[18] https://www.federalreserve.gov/publications/files/2022-dfast-results-20220623.pdf

^[19] https://www.federalreserve.gov/newsevents/pressreleases/files/bcreg20230209a1.pdf

^[20] Liquidity horizon refers to the time assumed to be required to exit or hedge a risk position without materially affecting market prices in stressed market conditions.

^[21] https://www.ecfr.gov/current/title-12/chapter-II/subchapter-A/part-252

^[22] This capital increase will be compounded should the SCB be applied to the Advanced Approaches because the stress test losses already reflect CVA and operational risk losses (along with market risk losses), creating another instances of potential double-counts.

^[23] Note that for the purpose of the Basel output floor, the standardized approach covers credit, market, CVA and op risks.

^[24] https://www.risk.net/risk-quantum/7955826/bny-mellon-goldman-join-citi-in-escaping-collins-floor

^[25] https://www.federalreserve.gov/supervisionreg/srletters/sr1107.htm

^[26] https://www.bankingsupervision.europa.eu/ecb/pub/pdf/ssm.trim_project_report~aa49bb624c.en.pdf

^[27] https://www.bankingsupervision.europa.eu/banking/tasks/internal_models/html/index.en.html

^[28] https://eur-lex.europa.eu/resource.html?uri=cellar:14dcf18a-37cd-11ec-8daf-01aa75ed71a1.0001.02/DOC_1&format=PDF

^[29] https://www.bankofengland.co.uk/prudential-regulation/publication/2022/november/implementation-of-the-basel-3-1-standards

^[30] https://www.govinfo.gov/content/pkg/FR-2019-11-01/pdf/2019-23800.pdf

^[31] https://www.occ.gov/news-issuances/federal-register/2019/fr8435234.pdf

^[32] https://www.risk.net/risk-management/7956021/us-credit-risk-modellers-prepare-for-life-after-irb

^[33] https://www.oliverwyman.com/our-expertise/insights/2016/aug/post-crisis-basel-reforms.html

^[34] https://www.govinfo.gov/content/pkg/FR-2020-01-24/pdf/2019-27249.pdf

^[35] In particular, the Risk Magazine reported that the SA-CCR “had an immediate effect on the status quo for banks and the cost of dealing in foreign exchange forwards and swaps.  According to data from transaction cost analysis firm BestX, the difference between US and European bid/offer spreads for G10 forwards jumped by 0.2 basis points between January and March” following the implementation of the rule (https://www.risk.net/our-take/7955926/living-with-sa-ccr-one-year-on).

Related Resources

• Pennsylvania + Wall Blog

  Blog Series on US Bank Capital Requirements