Importance of Capital
Capital is skin in the game. With more capital, shareholders are more likely to bear the consequences of their bank’s decisions. That is basic fairness. That also mitigates moral hazard and fosters market discipline over excessive risk taking.
Capital is a bulwark against the unknown. The COVID-19 pandemic reminds us that shocks are not always, or perhaps ever, anticipated.
Capital promotes financial stability. By reducing the likelihood of a large bank’s insolvency, capital reduces the risk of contagion. That in turn can help prevent bailouts.
Capital furthers banks’ mission. Capital positions banks to perform their vital functions in our economy even through a severe stress, as they did during the recent pandemic.
In light of the importance of capital, I hoped to support our effort to enhance the capital framework. But I must dissent for two main reasons.
First, I have serious reservations about important aspects of the so-called “Basel III” standards that this proposal would implement. These Basel III standards do not exactly inspire confidence. Notably, there is an absence of any stated rationale for some key decisions made by the Basel Committee. It would take a big leap of faith to support this proposal.
Second, while departures from these international standards can be appropriate to serve American interests, this proposal generally deviates from these standards with a singular focus: pushing capital levels yet higher and higher. This reverse engineering of higher capital pays little heed to the associated economic costs and, if finalized, would result in a capital framework unmoored from the principles that have historically rationalized the framework.
Faith in Basel
The Basel Committee serves an important role in developing the underlying theories for calibrating the capital framework, pooling data, coordinating research, developing international consensus, and fostering a level regulatory playing field for internationally active banks. The Basel Committee should have a continuing role to these ends.
What has changed is the extent to which we are asked to put our faith in the Basel Committee. As the complexity of the capital framework mounts, we are asked to defer more and more to the technical work of, and the backroom deals made at, the Basel Committee. In the case of the Basel III standards, the Basel Committee has made some key decisions with little or no explanation. That then leaves the U.S. bank regulators unable to defend or perhaps even understand important aspects of the Basel III standards that we are now proposing to implement.
Take for example the business-indicator approach to operational-risk capital. The first Basel consultative document acknowledged that this approach “does not lend itself to accurate application in the case of banks engaged predominantly in fee-based activities.”1 The second consultative document reiterated that the approach resulted in “overcapitalization of banks with high fee revenues and expenses.”2 It also proposed a fix.3 But that fix was then quietly dropped from the final Basel III standards without public explanation. That leaves this proposal to take an approach that its own Basel Committee authors have said does not work.
Another example: The specification of the operational-risk-capital formula and the sizing of the scaling factor and other coefficients have a significant impact on capital requirements. Helpfully, the annexes to the two consultative papers provide good insight into the early thinking of the subject-matter experts who participated in the Basel process.4 Unhelpfully, the Basel Committee quietly made considerable changes to the formula’s coefficients between the last consultative paper and the final Basel III standards without public explanation. That leaves this proposal unable to offer any rationale for the sizing of these coefficients.
A final example: To be eligible for the reduced credit-risk-capital requirement for investment-grade corporate exposures, the company (or its parent) must have securities outstanding on a recognized securities exchange. The Basel Committee has offered no rationale for concluding that having a publicly listed security correlates strongly with a company’s capacity to meet financial commitments. Understandably, the European Union and United Kingdom regulators’ implementation proposals have dropped the concept. Our proposal keeps it, but is stuck offering only a cursory rationale.
These are just a few examples of key design decisions that the public and I are asked to accept on faith.5 That leap of faith would abdicate my responsibility to come to an informed judgment. It is simply not enough to do something just because the Oracles at Basel so ordained.6 A better approach would be for the U.S. bank regulators to independently assess the Basel III standards with particular scrutiny of each aspect finalized without explanation. Where the Basel Committee has not articulated a convincing rationale, we should fill that gap by laying out our own well developed rationale. Where no convincing rationale is possible, we instead should consider our own approach that has a good grounding in theory and evidence. To those ends, attached are some additional questions on the proposal.
A Singular Focus on Higher Capital
The second reason I must dissent is that key aspects of this proposal are driven by a singular focus: pushing capital levels yet higher and higher.
Take for example the approach to mortgages. Despite the U.S. bank regulators’ role at the Basel Committee in developing a more empirically informed risk-sensitive approach, we have now reversed course and decided not to adopt the Basel III credit-risk-capital requirements for residential real estate exposures. Instead, the proposal adds 20 percentage points to each of the corresponding Basel III risk weights (a 160-basis-point surcharge7), such that the credit-risk-capital requirements for a residential real estate exposure would be up to twice as large as that contemplated by the Basel III standards.
|Loan-to-value (“LTV”) Bucket||Basel III||Proposal||Percentage Increase|
|LTV ratio ≤ 50%||20%||40%||100%|
|50% < LTV ratio ≤ 60%||25%||45%||80%|
|60% < LTV ratio ≤ 80%||30%||50%||67%|
|80% < LTV ratio ≤ 90%||40%||60%||50%|
|90% < LTV ratio ≤ 100%||50%||70%||40%|
|LTV ratio > 100%||70%||90%||29%|
Basing an exposure’s risk-based-capital requirement on that exposure’s risk should be, as the term suggests, the key principle for calibrating that requirement.8 Unlike some other aspects of the Basel III standards, we have grounds to believe that the Basel III credit-risk-capital requirements for residential real estate are better aligned with the underlying credit risk than the current risk-insensitive requirements adopted in 1989. These Basel III requirements are generally consistent with a proposal and analysis submitted by the U.S. bank regulators to the Basel Committee to support the development of the Basel III standards. Also, the Federal Housing Finance Agency (“FHFA”), the regulator for the housing government-sponsored enterprises (each, a “GSE”), recently finalized a capital framework for the GSEs9 that assigns credit-risk capital requirements for single-family mortgage loans that are broadly consistent with the Basel III requirements.10 Notably, FHFA disclosed much of the underlying data and its calibration methodology for single-family mortgage loans.11
The proposal offers no loss history or other evidence to support the sizing of the surcharge. The proposal also does not acknowledge that we have rejected a recommendation made by the Financial Stability Oversight Council (“FSOC”) that the U.S. bank regulators coordinate with FHFA to harmonize capital requirements across the banks and the GSEs to mitigate risks to financial stability driven by capital arbitrage.12
Instead, the entire discussion offering any modicum of rationale for this significant—and even controversial—decision is a sparse three sentences buried in the impact analysis.13 The purported rationale is to avoid putting smaller banks, which would not be subject to the proposal, at a competitive disadvantage to large banks. That is certainly a critically important policy objective. But to achieve both competitive parity and risk sensitivity, one alternative worth exploring is to extend the modernized set of Basel III credit-risk-capital requirements for residential real estate exposures to all banks, both large and small, while perhaps affording each smaller bank an option to keep its current, generally larger, capital requirements for those exposures.
But that approach is not explored in the proposal, likely because extending the modernized set of Basel III requirements for residential real estate exposures to all banks would be contrary to our effort to drive capital higher, as these Basel III requirements are generally less than today’s requirements. And so, in reverse engineering a significant increase in capital, we have backed ourselves into this surcharge without evidence or real rationale.
There likely will be real economic costs. Large banks generally would see an increase in the capital requirement on mortgage loans to borrowers who cannot afford a 20% down payment.14 The increased capital requirements could lead to an increase in interest rates for low- and moderate-income and other historically underserved borrowers who cannot always afford a 20% down payment, making it that much harder for these families to achieve homeownership.15
The proposal’s surcharge also would perpetuate an existing capital arbitrage that drives mortgage lending out of the banking system and concentrates it in the monoline GSEs.16 In September 2020, FSOC conducted a review of the secondary mortgage market to identify potential risks to financial stability. That review found that credit-risk-capital requirements at the GSEs that are less than that of banks and other credit providers would concentrate risk at the GSEs.17 According to FSOC, alignment of credit-risk-capital requirements across the GSEs and banks would reduce this capital arbitrage and mitigate the associated risks to financial stability.18 To that end, FSOC “encourages FHFA and other regulatory agencies to coordinate and take other appropriate action to avoid market distortions that could increase risks to financial stability by generally taking consistent approaches to the capital requirements…”19
We certainly have not done that here. While lower capital requirements outside the banking system should not lead to a race to the bottom, how could FHFA or any other regulator align its credit-risk-capital requirements with those contemplated here when the proposal is so at odds with the available evidence? And so, here we are, perpetuating a capital arbitrage that poses risk to financial stability and risks future bailouts. Adopting the Basel III credit-risk-capital requirements for residential real estate exposures for all banks could have reduced that risk—but at the expense of a singular focus on yet higher and higher capital.
The Dual-Requirement Structure
Another example of this singular focus on higher capital is the proposal’s so-called “dual-requirement structure.” Under this structure, a large bank would determine each risk-based capital ratio using the existing standardized approach (revised to include the new market-risk approach) and then separately determine that ratio using the new expanded risk-based approach, with the smaller of the two ratios being the binding constraint. As of the end of 2021, more than one-third of the large banking organizations covered by this proposal would have had risk-weighted assets under the standardized approach that were greater than, or quite close to, the risk-weighted assets under the expanded risk-based approach.
This dual-requirement structure forgoes an opportunity to simplify an already complicated capital framework. The dual-requirement structure also introduces internal inconsistencies that compound into incoherence. Some large banks would have one capital requirement for a securitization exposure, while other large banks would have a different capital requirement for the same exposure. Some large banks would use one set of criteria for determining when an exposure is in default, while other large banks would use a different set of criteria. Some large banks would apply one haircut to certain collateral, while other large banks would apply a different haircut. Some large banks would apply one credit-conversion factor to a commitment, while other large banks would apply a different one. And so on. Some large banks would find their holding company subject to one approach and their bank subsidiary subject to the other approach. Some large banks would find themselves alternating between the two approaches across the business cycle. Extraneous events could even lead to bizarre outcomes. Take for example a bank bound by the standardized approach. Assume that the bank incurs a large penalty under an enforcement action relating to consumer compliance issues, which would increase the bank’s operational-risk capital under the expanded risk-based approach. If that increase were large enough, the bank would become bound by the expanded risk-based approach. The result? Reduced credit-risk-capital requirements for some residential real estate and retail exposures. Why should an enforcement action lead to reduced credit-risk-capital requirements?
As with residential real estate exposures, the rationale for this dual-requirement structure is to avoid putting smaller banks at a competitive disadvantage to large banks. Again, that is certainly a critically important policy objective; but again, there would seem to be better fixes. One alternative worth exploring is to make the expanded risk-based approach the generally applicable approach, but then give each smaller bank the option to keep its current standardized approach.20 Again, the problem with that approach is that it would result in a less than targeted increase in capital. And so, in reverse engineering a significant increase in capital, we have backed ourselves into the complexities and inconsistencies of this dual-requirement structure.
There are other examples of this push toward higher capital. The proposal would add a surcharge to the Basel III credit-risk-capital requirements for retail exposures. The proposal would not adopt the reduced Basel III credit-risk-capital requirements for exposures to small businesses, securities firms and other nonbank financial institutions, or highly capitalized banking organizations; or for short-term exposures to banking organizations. The internal-loss multiplier for operational-risk capital would be floored at one. Banks using the models-based measure for market risk would be required to use the standardized approach for default-risk capital. The proposal would not adopt the Basel III standards’ approach to simple, transparent, and comparable securitizations. And so on.
Maybe this is the right approach in some cases, maybe not. But the proposal offers little rationale. Count me skeptical, especially given the approach to mortgages.
The Importance of Giving Reasons
To close, it is perhaps worth unpacking what is something of a theme of this dissent: it is very important that the U.S. bank regulators offer well developed rationales for our decisions.
While capital has many benefits, it also has costs.21 Any change should endeavor to strike an appropriate balance. In doing so, we should remain humbly aware of the limits on our own knowledge and cautiously cognizant of the risks of unintended consequences. We are more likely to get that balance right—or more aptly, to get it less wrong—if we explicitly state our approach to calibrating each component of the capital framework and then explain how our changes would better implement those policy objectives.
Well developed rationales check the understandable inclination of stakeholders and policymakers to work backward from some gut sense as to the right level of capital, a gut sense that might be motivated less by evidence and more by parochial interests, recency bias, or other extraneous concerns. Well developed rationales add legitimacy by dispelling any notion that the changes are arbitrary and by perhaps even fostering consensus. Where consensus fails, well developed rationales clarify where we disagree, which in turn focuses efforts to bridge disagreement.
Perhaps for these reasons, the law even requires us to disclose our reasons so as to give the public a meaningful opportunity to comment on what we have wrong.22 Reasoned explanations can be particularly important where, as here, there is a significant change in policy.23
The Basel Committee technical working groups generally have done a commendable job laying out the theory, data, and rationales underlying their initial thinking. That often cannot be said for the changes quietly made by the Basel Committee between the last consultative paper and the final Basel standard. This proposal does not put enough effort into remedying that lack of transparency. The proposal also provides cursory rationales for some of our other more significant decisions. Going forward, whether we are changing the capital framework or other aspects of the regulatory framework, we should do a better job of giving our reasons.
Question 1. Dual-requirement structure. As of the end of 2021, more than one-third of the large banking organizations covered by this proposal would have had risk-weighted assets under the standardized approach that were greater than, or quite close to, the risk-weighted assets under the expanded risk-based approach. All capital buffer requirements, including the stress capital buffer requirement, would apply regardless of whether the expanded risk-based approach or the standardized approach produces the lower ratio. How might the mix of banking organizations bound by the expanded risk-based approach and the standardized approach evolve over the business cycle or otherwise change over time (e.g., as operational-loss history rolls off)? Are there any issues posed by having a holding company being subject to a different approach than its bank subsidiary? What are the advantages and disadvantages of requiring large banking organizations to determine capital under two different approaches (e.g., added complexity, internal inconsistencies, and cliff effects)? Are there alternative structures that would ensure competitive parity between large and smaller banking organizations?
Question 2. Agency authorities. What requirements or restrictions on each agency’s authorities are implicated by the proposal? Does the proposal tailor or otherwise differentiate among banking organizations to the extent required by law?
Question 3. Administrative law. Are the rationales offered by the proposal sufficient to give the public a meaningful opportunity to comment, particularly given the changes in policy contemplated by this proposal? Are there any specific studies or data that should be disclosed so that the public has a meaningful opportunity to comment?
Regulatory Capital Deductions
Question 4. Mortgage servicing assets (“MSAs”). Category III and IV banking organizations would be subject to the same requirements and restrictions applicable to Category I and II banking organizations governing the capital treatment of MSAs. Are these requirements and restrictions consistent with the available evidence as to the risk associated with exposures to MSAs? Would other approaches be more aligned with that risk? Would the proposal’s approach tend to shift mortgage intermediation to nonbanks?
Question 5. Surcharges. The expanded risk-based approach would increase the Basel III risk weights for residential real estate exposures (by 20 percentage points), other real estate exposures not dependent on cash flows generated by the real estate (by 25 percentage points for exposures to individuals, 15 percentage points for exposures to small- or medium-sized entities (“SMEs”)), and retail exposures (by 10 percentage points) to ensure competitive parity between the large banking organizations covered by this proposal and the smaller banking organizations that are not. To what extent are the resulting credit-risk-capital requirements aligned with the available evidence on the loss history of residential real estate and other retail exposures? What roughly would be the operational-risk-capital requirement associated with residential real estate and retail exposures? To what extent would these credit-risk- and operational-risk-capital requirements shift intermediation of mortgage loans and retail credit to nonbanks (e.g., the GSEs or the Federal Housing Administration)? What might be the impact on cost of credit for low- and moderate-income borrowers and other historically underserved borrowers? Are there alternative approaches that would better achieve both risk sensitivity and competitive parity?
Question 6. Other notable Basel III departures. The expanded risk-based approach does not adopt Basel III’s reduced risk weight for exposures to corporate SMEs, securities firms and other nonbank financial institutions, or highly capitalized banking organizations; or for short-term exposures to banking organizations. What considerations should inform the agencies’ treatment of these exposures?
Question 7. Investment-grade corporate publicly traded requirement. The expanded risk-based approach would assign a 65% risk weight to a corporate exposure only if that exposure is to a company that is investment grade and that has a publicly traded security outstanding or that is controlled by a company that has a publicly traded security outstanding. Does any data or other evidence suggest a correlation between an investment-grade company’s creditworthiness and whether that company (or its parent) has a publicly traded security outstanding? What would be the advantages and disadvantages of other approaches, such as eliminating the publicly traded requirement or replacing it with a requirement that the company or its parent be required to provide the banking organization with information sufficient to conduct adequate due diligence, whether by contract or regulation?
Question 8. Private mortgage insurance. The expanded risk-based approach would not recognize private mortgage insurance in the calculation of the LTV ratio for the purpose of assigning a risk weight to a residential real estate exposure. What would be the advantages and disadvantages of recognizing private mortgage insurance in that calculation? Would any suggested approach for recognizing private mortgage insurance be consistent with the Basel III standards, including its parameters on the range of guarantors and other protection providers (CRE22.76)?
Question 9. Minimum haircut floors. When undertaking certain repo-style transactions and eligible margin loans with unregulated financial institutions, the expanded risk-based approach would condition recognition of collateral on receipt of a minimum amount of collateral. What are the advantages and disadvantages of addressing risks associated with nonbank leverage in this way through the capital framework, especially as compared to addressing those risks through market regulation? Should the agencies consider the different approaches taken by the European Union or United Kingdom’s implementation proposals?
Question 10. Credit-linked notes. Under the Basel III standards, cash-funded credit-linked notes issued by a bank against exposures in the banking book that fulfill the criteria for credit derivatives may be treated as cash-collateralized transactions. Should the agencies consider changes to clarify the treatment of credit-linked notes under either the standardized approach or the expanded risk-based approach? If so, to what aspects of the capital framework should the agencies consider changes?
Question 11. Listing requirement for certain collateral. Under the expanded risk-based approach, a banking organization could recognize the risk-mitigating benefits of a corporate debt security that meets the definition of financial collateral only if the corporate issuer of the debt security has a publicly traded security outstanding or is controlled by a company that has a publicly traded security outstanding. Having a publicly traded security outstanding is not required for collateral recognition under the standardized approach. What degree of burden and operational issues would a banking organization confront in having different sets of instruments recognized as eligible financial collateral in the standardized approach and the expanded risk-based approach? What would be the advantages and disadvantages of other approaches, such as eliminating the publicly traded requirement or replacing it with a requirement that the company or its parent be required to provide the banking organization with information sufficient to conduct adequate due diligence, whether by contract or regulation?
Question 12. Calibration of supervisory parameter. Under the expanded risk-based approach, a banking organization would apply a supervisory parameter p of 1.0 to securitization exposures (or 1.5 to resecuritization exposures). Does the public domain information regarding the methodology and rationale for the calibration of the supervisory parameter provide the public with a meaningful opportunity to comment on its sizing? Is the sizing of the supervisory parameter appropriate to mitigate any structuring, recourse, or other risks that might be posed by securitizations?
Question 13. Simple, transparent, and comparable (“STC”) securitizations. The expanded risk-based approach would not adopt the Basel III approach to STC securitizations. The proposal does not offer any rationale. What would be the advantages and disadvantages of adopting the Basel III approach to STC securitizations?
Question 14. Impact of surcharges on underlying exposures. The expanded risk-based approach would add surcharges to the Basel III risk weights for residential real estate exposures and retail exposures. How would these surcharges affect a banking organization’s incentives to securitize pools of these exposures? How would the surcharges interact with the proposed risk-weight floor and the supervisory parameter to affect a banking organization’s incentives to retain more junior tranches?
Question 15. High-fee businesses. The proposal does not propose a fix to address the apparent issue, which was acknowledged in the Basel consultative documents, that the business-indicator approach overcapitalizes banking organizations with high fee revenue and expense. Does the business-indicator approach overcapitalize banking organizations with high fee revenue and expense? What specific approaches should the agencies consider to address this issue?
Question 16. Floored internal-loss multiplier. The proposal would floor the internal-loss multiplier at one to ensure “a robust minimum amount of coverage.” Other implementing authorities have set the internal-loss multiplier equal to one, as permitted by the Basel III standards. What are the advantages and disadvantages of flooring the internal-loss multiplier at one? How would flooring the internal-loss multiplier at one affect banks’ incentives to enhance their operational risk management processes? If the internal-loss multiplier were floored at one, should the various coefficients of the operational risk capital formula be re-estimated by fitting the loss history data to a model specification that includes a floored internal-loss multiplier? If so, would an updated formula that included those re-estimated coefficients tend to result in an increase or decrease in operational-risk capital?
Question 17. Business-indicator coefficients. If the business-indicator coefficients were estimated by fitting the Basel Committee’s loss-history data to a model specification that did not include the internal-loss multiplier, would the introduction of the internal-loss multiplier to the model merit a re-estimation of the business-indicator coefficients? If so, would an updated formula that included those re-estimated coefficients tend to result in an increase or decrease in operational-risk capital?
Question 18. Internal-loss-multiplier coefficients. There is little discussion in the public Basel documents regarding the calibration of the internal-loss-multiplier coefficients (i.e., the 15 multiplier and the 0.8 dampener). What does the available evidence suggest about the appropriateness of the sizing of these coefficients? Does the appropriateness of the sizing of these coefficients depend on whether the internal-loss multiplier is floored at one?
Question 19. Static coefficients. As the Basel Committee acknowledged, there is relatively limited loss-history data on operational-risk losses. Should the agencies revisit the calibration of the coefficients of the operational-risk-capital formula from time to time? If so, what frequency or events should trigger a revisiting of these static coefficients?
Question 20. Application to large banking organizations with small trading books. Regardless of the size of its trading book, a holding company subject to Category I, II, III, or IV standards would be subject to the proposed market-risk-capital requirements, as would a subsidiary of such a holding company if that subsidiary engaged in any trading activity over any of the four most recent quarters. Several of those banking organizations have limited trading activity and are not subject to the current market-risk-capital requirements. The proposal retains the authority for the primary Federal supervisor to exclude a banking organization on a case-by-case basis. What considerations should inform whether all Category I, II, III, and IV banking organizations should be subject to the new market risk capital requirements? Are case-by-case exemptions an appropriate way to tailor application of these new market-risk-capital requirements to Category I, II, III, and IV banking organizations?
Question 21. Diversification benefit. Under the sensitivities-based method, the capital requirements for delta, vega, and curvature risk would be summed across risk classes, respectively, with no recognition of diversification benefits “because in stress diversification across different risk classes may become less effective.” Under the models-based measure, to constrain the empirical correlations and provide an appropriate balance between perfect diversification and no diversification between risk-factor classes, the internally modelled capital calculation for modellable risk factors would equal half of the entity-wide liquidity-horizon-adjusted expected-shortfall-based measure across all risk classes plus half of the sum of the liquidity-horizon-adjusted expected-shortfall measures for each risk class. The stressed expected-shortfall capital requirement for non-modellable risk factors would be similar to the models-based approach for modellable risk factors, except that it would provide significantly less recognition for hedging and portfolio diversification relative to the internally modelled capital calculation for modellable risk factors. What does the available data or other evidence suggest regarding the benefits of diversification across risk classes during periods of stress? Does each of these approaches strike an appropriate balance in recognizing the benefits of diversification during periods of stress?
Question 22. Default-risk-capital requirement under the internal-models approach. The proposal would require each covered banking organization to use the standardized default-risk-capital requirement regardless of whether it uses an approved models-based measure for market risk. The rationale is “to reduce the operational burden for a banking organization and to further promote consistency in risk-based capital requirements across banking organizations and within the capital rule.” Does the associated operational burden for banking organizations merit this approach? Would concerns about operational burden be better addressed by affording a banking organization an option to use the standardized default-risk-capital requirement?
Question 23. Sensitivities-based-method risk weights and correlation parameters. Is the data available for interested parties to estimate or otherwise assess the calibration of the sensitivities-based-method risk weights and correlation parameters? If so, are the sensitivities-based-method risk weights and correlation parameters generally consistent with the available data? Should the agencies consider re-calibrating the sensitivities-based-method risk weights and correlation parameters from time to time? Should the sensitivities-based-method risk weights and correlation parameters be re-estimated following the significant recent increase in U.S. interest rates? Do static sensitivities-based-method risk weights or correlation parameters pose risk to safety and soundness or financial stability?
Question 24. Sensitivities-based-method granularity. Does the sensitivities-based method provide for sufficient granularity in the risk weights and correlation parameters to ensure sufficient risk sensitivity?
Question 25. Profit-and-loss attribution-test thresholds. The Basel Committee’s 2016 consultative document committed to revisiting the preliminary calibration of the profit-and-loss attribution-test thresholds for amber zone and red zone.24 Those thresholds have not been updated in the Basel III standards, and there was no public-domain suggestion that the Basel Committee technical working group revisited those thresholds. Is actual profit-and-loss attribution data available for interested parties to assess the calibration of the profit-and-loss attribution-test thresholds for the green zone, amber zone, and red zone? Are the assumptions of the Basel Committee technical working group (e.g., as to data distributions) adequately disclosed for interested parties to assess the calibration of those profit-and-loss attribution-test zones? Are the profit-and-loss attribution-test zones appropriately calibrated?
Question 26. Cyclicality of market-risk-capital requirements. To what extent would aggregate market-risk-capital requirements be expected to vary across the business cycle? Does the cyclicality vary across the standardized measure and the models-based measure? What aspects of the framework might drive those variations? If the aggregate market-risk-capital requirements might be procyclical, would that procyclicality be consistent with the underlying risks or should the agencies consider any changes to mitigate that procyclicality while preserving risk sensitivity?
Credit Valuation Adjustment (“CVA”) Risk
Question 27. Scope and alternative treatment. The proposed capital requirements for CVA risk would apply to all Category I, II, III, and IV banking organizations. What would be the advantages and disadvantages of adopting a materiality threshold below which a banking organization may choose an alternative treatment, as contemplated by the Basel III standards (MAR50.9)?
Question 28. Commercial end-users. The capital requirements for CVA risk would not include a tailored approach to commercial end-users. Some other implementing authorities have proposed a commercial end-user exemption for CVA-risk-capital requirements. What considerations should inform whether a commercial end-user exemption is appropriate? Is the absence of an alpha factor under the standardized approach for counterparty credit risk (“SA-CCR”) for uncleared derivatives with commercial end-users sufficient to address any issues under the proposed capital requirements for CVA risk?
Question 29. Financials. The supervisory risk weights for each counterparty would reflect the potential variability of credit spreads based on a combination of the sector and credit quality of the counterparty. The supervisory risk weights would be the same for all financials. What would be the advantages and disadvantages of varying the supervisory risk weights for different types of financials? What types of financials should have different risk weights (e.g., based on whether a financial is subject to comprehensive regulation or is well capitalized)?
1 Basel Comm. on Banking Supervision, Consultative Document: Operational risk – Revisions to the simpler approaches ¶ 46, at 16 (2014) [hereinafter First Operational-Risk Consultation].
2 Basel Comm. on Banking Supervision, Consultative Document: Standardised Measurement Approach for operational risk ¶ 16(d), at 4 (2016) (cleaned up) [hereinafter Second Operational-Risk Consultation]. The Basel Committee further explains: “banks with a high fee component in respect to the overall business-indicator amount have a very high business-indicator value which results in capital requirements that are too conservative relative to the operational risk faced by these banks.” Id. (cleaned up).
3 Id. ¶ 20, at 4.
5 This list of examples could go on, especially with respect to the proposal’s hyper-technical market-risk-capital requirements. For most large banks, their market-risk-capital requirement would be determined largely by the sensitivities-based method, which applies dozens of static risk weights and correlation parameters to a bank’s trading positions. The underlying data and calibration methodology used to estimate these numbers are not in the public domain. Similarly, there is little discussion in the public domain of the calibration methodology, and no public domain data, that interested parties can use to assess the calibration of the profit-and-loss attribution-test thresholds for amber zone and red zone, which play important roles in determining the market-risk-capital requirement under the internal-models approach.
6 Concern with transparency and accountability at the Basel Committee is nothing new. Forty years ago, as part of legislation mandating that the U.S. bank regulators establish minimum levels of risk-based capital and encourage their international counterparts to do the same, Congress acknowledged concerns about international forums like the Basel Committee and authorized the Government Accountability Office to audit the U.S. bank regulators’ participation in these international forums. International Lending Supervision Act of 1983, Pub. L. No. 98-181, § 908(b)(3), 97 Stat. 1278, 1280 (codified at 12 U.S.C. § 3907(b)(3)); id. § 911(a)(2), 97 Stat. at 1282–83 (codified at 12 U.S.C. § 3910(a)(2)).
7 The total capital requirement for an exposure is equal to the product of the risk weight for the exposure, the exposure amount, and the 8% total capital requirement. Expressed as a percent of the exposure amount, the total capital requirement is equal to the product of the risk weight for the exposure and the 8% total capital requirement. This means that a 20-percentage-point increase in a risk weight is equivalent to an increase in the total capital requirement (expressed as a percent of the exposure amount) equal to 160 basis points.
8 See, e.g., NPR at 13 (“The expanded risk-based approach would be more risk-sensitive than the current U.S. standardized approach by incorporating more credit-risk drivers (for example, borrower and loan characteristics) and explicitly differentiating between more types of risk (for example, operational risk, credit valuation adjustment risk)…. The proposed changes would also enhance the alignment of capital requirements to the risks of banking organizations’ exposures and increase incentives for prudent risk management.”).
9 See Enterprise Regulatory Capital Framework, 85 Fed. Reg. 82,150 (Dec. 17, 2020) (codified as amended at 12 C.F.R. pts. 1206, 1225, 1240, & 1750).
10 For a single-family performing loan with a 750 FICO credit score at origination, the Basel III risk weights and the FHFA base risk weights at origination are around 20% for a loan with an LTV at origination of less than 50%, and generally range between 20% and 30% for loans with LTVs at origination of between 50% and 80%. See 12 C.F.R. § 1240.33(c), Table 2; id. § 1240.33(b)(2) (imposing a 20% risk-weight floor); Basel Comm. on Banking Supervision, The Basel Framework, CRE20.82 (Table 11) (last updated Dec. 8, 2022). Comparisons are complicated with respect to single-family performing loans with LTVs at origination greater than 80%, as FHFA’s framework affords capital relief to private mortgage insurance, unlike the U.S. bank capital framework, and also generally gives more capital relief to credit risk transfers than the U.S. bank capital framework, which result in meaningful reductions to these base risk weights. Fed. Hous. Fin. Agency, Fact Sheet: Final Rule on Enterprise Capital 6 (2020) (“The average risk weight for single-family mortgage exposures was 43 percent before credit enhancements, 37 percent before credit risk transfer and 31 percent post-credit risk transfer.” (cleaned up)). As of December 31, 2022, Fannie Mae and Freddie Mac’s weighted average FICO at origination were, respectively, 752 and 750. Fed. Nat’l Mortg. Ass’n (Fannie Mae), Annual Report (Form 10-K) 94 (Feb. 14, 2023); Fed. Home Loan Mortg. Corp. (Freddie Mac), Annual Report (Form 10-K) 63 (Feb. 22, 2023).
11 See Enterprise Regulatory Capital Framework, 85 Fed. Reg. 39,274, 39,303 n.64 (proposed Jun. 30, 2020).
12 See Fin. Stability Oversight Council, Statement on Activities-Based Review of Secondary Mortgage Market Activities 2 (2020) [hereinafter, FSOC, Secondary Mortgage Market Statement]. This statement was approved unanimously by FSOC, whose voting members include the Chairman of the FDIC, the Chairman of the Board of Governors of the Federal Reserve System, the Comptroller of the Currency, and the Director of the Consumer Financial Protection Bureau. See Financial Stability Oversight Council Issues Statement on Activities-Based Review of Secondary Mortgage Market Activities, Fin. Stability Oversight Council (Sept. 25, 2020). The U.S. bank regulators’ staffs participated in that review by, among other things, providing quantitative analysis and other analytical support.
13 NPR at 501–02 (“In addition, the proposal attempts to mitigate potential competitive effects between U.S. banking organizations by adjusting the U.S. implementation of the Basel III reforms, specifically by raising the risk weights for residential real estate and retail credit exposures. Without the adjustment relative to Basel III risk weights in this proposal, marginal funding costs on residential real estate and retail credit exposures for many large banking organizations could have been substantially lower than for smaller organizations not subject to the proposal. Though the larger organizations would have still been subject to higher overall capital requirements, the lower marginal funding costs could have created a competitive disadvantage for smaller firms.”).
14 Today, the standardized credit risk capital requirement for most residential mortgage loans is 4% (50% risk weight multiplied by the 8% total capital requirement). Under the proposal, loans with LTVs greater than 80% and less than or equal to 90% would have an expanded risk-based approach credit-risk-capital requirement of 4.8% (60% risk weight multiplied by 8%). There would also be a separate operational-risk-capital requirement.
15 The proposal does acknowledge this impact, although as an afterthought, with the discussion of the surcharge’s effects on historically underserved borrowers being added just before our vote. Related to this, the impact on borrowing costs could depend on whether smaller banks or nonbanks step in to fill the gap, and the extent to which large banks might adjust their return targets to reflect any associated reduction in the covariance of each residential real estate exposure’s returns with the market returns, among many other factors.
16 Although the first GSE was established in 1938, the GSEs’ footprint remained relatively small even until the 1980s. The GSEs did not begin to grow rapidly until the U.S. bank regulators began increasing capital requirements in the 1980s. While those increases in banks’ capital requirements were generally appropriate, the GSEs’ capital requirements remained unchanged. This gap gave the GSEs a competitive advantage in holding mortgage-related risk. The natural incentive was for banks to take a conventional mortgage loan that had a 4% capital requirement, pay a guarantee fee to a GSE, and get back a GSE-guaranteed mortgage-backed security that had a 1.6% capital requirement. This capital arbitrage drove rapid growth in the GSEs. The GSEs’ market share increased from 10% in the early 1980s to 25% by the end of that decade and then to 44% by the end of 2003, culminating in the $191.5 billion 2008 bailout of the GSEs. See generally U.S. Dep’t of Treas., Housing Reform Plan 4–11 (2019). Today, following the increase in banks’ capital requirements after the 2007–08 financial crisis, the GSEs now account for more than half the market. See Fed. Nat’l Mortg. Ass’n (Fannie Mae), Annual Report (Form 10-K) 84 (Feb. 14, 2023) (showing Fannie Mae had approximately 28% of single-family mortgage debt outstanding as of December 31, 2022); Fed. Home Loan Mortg. Corp. (Freddie Mac), Annual Report (Form 10-K) 12 (Feb. 22, 2023) (showing Freddie Mac’s single-family mortgage portfolio as of December 31, 2022 was $2.986 billion, or approximately 23% of the $13.195 billion of single-family mortgage debt outstanding).
17 FSOC, Secondary Mortgage Market Statement, supra note 12, at 2 (“The Enterprises’ credit risk requirements, however, likely would be lower than other credit providers across significant portions of the risk spectrum and during much of the credit cycle, which would create an advantage that could maintain significant concentration of risk with the Enterprises.”).
18 Id. (“The alignment of market participants’ credit-risk capital requirements across similar credit risk exposures would mitigate risk to financial stability by minimizing market structure distortions.”).
20 My understanding is that the so-called Collins Amendment would not preclude this approach.
21 See, e.g., Basel Comm. on Banking Supervision, Working Paper 38: Assessing the impact of Basel III: Evidence from macroeconomic models: literature review and simulations (2021); Basel Comm. on Banking Supervision, Working Paper 37: The costs and benefits of bank capital – a review of the literature (2019); Basel Comm. on Banking Supervision, An assessment of the long-term economic impact of stronger capital and liquidity requirements (2010).
22 See Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29 (1983).
23 See Perez v. Mortg. Bankers Ass’n, 575 U.S. 92, 106 (2015) (“As we held in Fox Television Stations, and underscore again today, the Administrative Procedure Act requires an agency to provide more substantial justification when its new policy rests upon factual findings that contradict those which underlay its prior policy; or when its prior policy has engendered serious reliance interests that must be taken into account. It would be arbitrary and capricious to ignore such matters.” (cleaned up) (citing F.C.C. v. Fox Television Stations, Inc., 556 U.S. 502, 515 (2009))).
24 Basel Comm. on Banking Supervision, Consultative Document: Revisions to the minimum capital requirements for market risk § 2.1.3, at 8 (2014) (“Upon finalization of the traffic light approach into the market risk standard, the Basel Committee will continue to monitor the effectiveness of the finalized calibration of the thresholds to ensure their appropriateness.”).