Enterprise Regulatory Capital Framework

Federal RegisterDec 17, 2020

Ask Donna

What actually matters in this document.

Text

FEDERAL HOUSING FINANCE AGENCY

12 CFR Parts 1206, 1225, and 1240

DEPARTMENT OF HOUSING AND URBAN DEVELOPMENT

Office of Federal Housing Enterprise Oversight

12 CFR Part 1750

RIN 2590-AA95

Enterprise Regulatory Capital Framework

AGENCY:

Federal Housing Finance Agency; Office of Federal Housing Enterprise Oversight.

ACTION:

Final rule.

SUMMARY:

The Federal Housing Finance Agency (FHFA or the Agency) is adopting a final rule (final rule) that establishes risk-based and leverage capital requirements for the Federal National Mortgage Association (Fannie Mae) and the Federal Home Loan Mortgage Corporation (Freddie Mac, and with Fannie Mae, each an Enterprise). The final rule also makes conforming amendments to definitions in FHFA's regulations governing assessments and minimum capital and removes the Office of Federal Housing Enterprise Oversight's (OFHEO) regulation on capital for the Enterprises.

DATES:

This rule is effective February 16, 2021.

FOR FURTHER INFORMATION CONTACT:

Naa Awaa Tagoe, Principal Associate Director, Office of Capital Policy, (202) 649-3140,

NaaAwaa.Tagoe@fhfa.gov;

Andrew Varrieur, Associate Director, Office of Capital Policy, (202) 649-3141,

Andrew.Varrieur@fhfa.gov;

or Mark Laponsky, Deputy General Counsel, Office of General Counsel, (202) 649-3054,

Mark.Laponsky@fhfa.gov.

These are not toll-free numbers. The telephone number for the Telecommunications Device for the Deaf is (800) 877-8339.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Introduction

II. The Proposed Rule

III. Overview of the Final Rule

A. Key Modifications to the Proposed Rule

B. Modifications to the 2018 Proposal

C. Regulatory Capital Requirements

D. Capital Buffers

E. Transition Period

IV. FSOC Review of the Secondary Mortgage Market

V. General Comments on the Proposed Rule

A. Access and Affordability and Other Aggregate Impacts

B. Similarities to the U.S. Banking Framework

C. Differences Between the Enterprises and Banks

D. Mortgage-Risk Sensitive Framework

E. Housing Finance Reform

VI. Definitions of Regulatory Capital

A. Guarantee Fees

B. Reserves

C. Subordinated Debt

VII. Capital Requirements

A. Risk-Based Capital Requirements

B. Leverage Ratio Requirements

1. Adjusted Total Assets

2. Sizing of the Requirements

C. Enforcement

VIII. Capital Buffers

A. Prescribed Capital Conservation Buffer Amount

1. Comments Applicable to Each Component Buffer

2. Stress Capital Buffer

3. Countercyclical Capital Buffer

4. Stability Capital Buffer

B. Prescribed Leverage Buffer Amount

C. Payout Restrictions

IX. Credit Risk Capital: Standardized Approach

A. Single-Family Mortgage Exposures

1. Base Risk Weights

2. Countercyclical Adjustment

3. Risk Multipliers

4. Credit Enhancement Multipliers

5. Minimum Adjusted Risk Weight

B. Multifamily Mortgage Exposures

1. Calibration Framework

2. Base Risk Weights

3. Countercyclical Adjustment

4. Risk Multipliers

5. Minimum Adjusted Risk Weight

C. PLS and Other Non-CRT Securitization Exposures

D. Retained CRT Exposures

1. Proposed Rule's Enhancements

2. Risk Weight Floor

3. Risk Weight Determination

4. Overall Effectiveness Adjustment

5. Loss-Timing Adjustment

6. Loss-Sharing Adjustment

7. Eligible CRT Structures

8. Other Comments and Issues

E. Other Exposures

X. Credit Risk Capital: Advanced Approach

XI. Market Risk Capital

XII. Operational Risk Capital

XIII. Impact of the Enterprise Capital Rule

XIV. Key Differences From the U.S. Banking Framework

XV. Transition Period

XVI. Temporary Increases of Minimum Capital Requirements

XVII. Administrative Law Matters

A. Regulatory Flexibility Act

B. Paperwork Reduction Act

C. Congressional Review Act

Introduction

On June 30, 2020, FHFA published in the

Federal Register

a notice of proposed rulemaking (proposed rule) seeking comment on a new regulatory capital framework for the Enterprises.

The proposed rule was a re-proposal of the regulatory capital framework set forth in the notice of proposed rulemaking published in the

Federal Register

on July 17, 2018 (2018 proposal).

While the 2018 proposal remained the foundation of the proposed rule, the proposed rule contemplated enhancements to establish a post-conservatorship regulatory capital framework that would ensure that each Enterprise operates in a safe and sound manner and is positioned to fulfill its statutory mission to provide stability and ongoing assistance to the secondary mortgage market across the economic cycle, in particular during periods of financial stress. FHFA is now adopting in this final rule the proposed regulatory capital framework, with certain changes to the proposed rule described below.

85 FR 39274.

83 FR 33312.

The Proposed Rule

Pursuant to the Federal Housing Enterprises Financial Safety and Soundness Act of 1992

(Safety and Soundness Act), as amended by the Housing and Economic Recovery Act of 2008

(HERA), the FHFA Director's principal duties include, among other duties, ensuring that each Enterprise operates in a safe and sound manner, that the operations and activities of each Enterprise foster liquid, efficient, competitive, and resilient national housing finance markets, and that each Enterprise carries out its statutory mission only through activities that are authorized under and consistent with the Safety and Soundness Act and its charter.

Pursuant to their charters, the statutory purposes of the Enterprises are, among other purposes, to provide stability in, and ongoing assistance to, the secondary market for residential mortgages.

Public Law 102-550, 106 Stat. 3941 (1992).

Public Law 110-289, 122 Stat. 2654 (2008).

12 U.S.C. 4513(a)(1).

12 U.S.C. 1451 note, 1716.

Consistent with these statutory duties and purposes, FHFA re-proposed the regulatory capital framework for the Enterprises for three key reasons. First, FHFA has begun the process to responsibly end the conservatorships of the Enterprises. This policy is a departure from the expectations of interested parties at the time of the 2018 proposal when the prospects for indefinite conservatorships informed comments and perhaps even the decision whether to comment at all.

Second, FHFA proposed to increase the quantity and quality of regulatory capital to ensure that each Enterprise operates in a safe and sound manner

and is positioned to fulfill its statutory mission to provide stability and ongoing assistance to the secondary mortgage market across the economic cycle, in particular during periods of financial stress. To achieve this objective, each Enterprise must be capitalized to be regarded as a viable going concern by creditors and counterparties both during and after a severe economic downturn. The importance of this going-concern standard was made clear by the Enterprises' funding difficulties and near failure during the 2008 financial crisis. The Enterprises fund themselves with a significant amount of short-term unsecured debt that must be regularly refinanced. Each Enterprise's funding needs are very likely to increase during an economic downturn, all else equal, as the Enterprise funds purchases of non-performing loans (NPLs) out of securitization pools and lenders increase their reliance on the Enterprise's cash window. These ordinary course and procyclical funding needs can be met only if the Enterprise continues to be regarded as a viable going concern by creditors throughout the duration of an economic downturn. Indeed, it was the increase in the Enterprises' borrowing costs and the associated difficulties that the Enterprises faced in refinancing their debt that were among the most immediate grounds for FHFA placing the Enterprises into conservatorship.

See Memorandum dated September 6, 2008 re: Proposed Appointment of the Federal Housing Finance Agency as Conservator for the Fannie Mae

at 29 (“The Enterprise's practice of relying upon repo financing of its agency collateral to raise cash in the current credit and liquidity environment is an unsafe or unsound practice that has led to an unsafe or unsound condition, given the unavailability of willing lenders to provide secured financing in significant size to reduce pressure on its discount notes borrowings.”); and

Memorandum dated September 6, 2008 re: Proposed Appointment of the Federal Housing Finance Agency as Conservator for the Freddie Mac

at 28 (“The Enterprise's prolonged reliance almost exclusively on 30-day discount notes is an untenable long-term source of funding and an unsafe or unsound practice that poses abnormal risk to the viability of the Enterprise. Operating without an adequate liquidity funding contingency plan is an unsafe or unsound condition to transact business.”); and Fin. Crisis Inquiry Comm'n, The Financial Crisis Inquiry Report: Final Report of the National Commission on the Causes of the Financial and Economic Crisis in the United States at 316 (2011) (the FCIC Report), available at

https://www.govinfo.gov/content/pkg/GPO-FCIC/pdf/GPO-FCIC.pdf;

(“In July and August 2008, Fannie suffered a liquidity squeeze, because it was unable to borrow against its own securities to raise sufficient cash in the repo market.”);

see id.

at 316 (“By June 2008, the spread [between the yield on the GSEs' long-term bonds and rates on Treasuries] had risen 65 percent over the 2007 level; by September 5, just before regulators parachuted in, the spread had nearly doubled from its 2007 level to just under 1 percent, making it more difficult and costly for the GSEs to fund their operations.”).

The 2008 financial crisis also established that credit, market, and other losses can be incurred quickly during a stress and that an Enterprise's capacity to absorb those losses as incurred while still timely performing its financial obligations defines creditors' and other counterparties' views as to whether the Enterprise remains a viable going concern. During a stress, creditors are unlikely to give much consideration to future revenue prospects in assessing whether an Enterprise can timely perform its financial obligations. Market confidence in the Enterprises waned in mid-2008 when Fannie Mae and Freddie Mac had total capital of, respectively, $55.6 billion and $42.9 billion, notwithstanding their rights to future guarantee fees.

It was in this historical context that HERA amended the Safety and Soundness Act to give FHFA greater authority to establish regulatory capital requirements for the Enterprises. OFHEO had previously been bound by the Safety and Soundness Act's prescriptive restrictions on the stress scenario used to calibrate the risk-based capital requirements. Under HERA's expanded authority, FHFA is required to prescribe by regulation risk-based capital requirements “to ensure that the enterprises operate in a safe and sound manner, maintaining sufficient capital and reserves to support the risks that arise in the operations and management of the enterprises.”

Importantly, the requirement that each Enterprise “maintain[] sufficient capital and reserves” applies before, during, and after a severe economic downturn, codifying in statute a going-concern standard.

12 U.S.C. 4611.

For the reasons given in Section IV.B.2 and elsewhere of the proposed rule, FHFA determined that the 2018 proposal's credit risk capital requirements were insufficient to ensure each Enterprise would continue to be regarded as a viable going concern during and after a severe economic downturn. Had the 2018 proposal been in effect at the end of 2007, Fannie Mae's and Freddie Mac's peak cumulative capital exhaustion would have left, respectively, capital equal to only 0.1 percent and 0.5 percent of their total assets and off-balance sheet guarantees. These amounts would not have sustained the market confidence necessary for the Enterprises to continue as going concerns, particularly given the prevailing stress in the financial markets at that time and given the uncertainty as to the potential for other write-downs and the adequacy of the Enterprises' allowances for loan and lease losses (ALLL).

Indeed, in October 2010, FHFA projected $90 billion in additional draws under the Senior Preferred Stock Purchase Agreements through 2013 under the baseline scenario. Only $34 billion in additional draws proved necessary.

See Fed. Hous. Fin. Agency, Projections of the Enterprises' Financial Performance

at 10 (Oct. 2010), available at

https://www.fhfa.gov/AboutUs/Reports/ReportDocuments/2010-10_Projections_508.pdf.

Reinforcing that point, the Enterprises' crisis-era cumulative capital losses, while significant, could have been greater. The Enterprises' losses were likely mitigated by unprecedented federal government support of the housing market and the economy during the crisis, including through the Home Affordable Modification Program, the Troubled Asset Relief Program, the 2009 stimulus package,

and the Federal Reserve System's purchases of more than $1.2 trillion of the Enterprises' debt and mortgage-backed securities (MBS) from January 2009 to March 2010. The Enterprises' losses also were likely dampened by the declining interest rate environment of the period, when the interest rates on 30-year fixed-rate mortgage loans declined by approximately 200 basis points through the end of 2011, facilitating refinancings and loss mitigation programs.

American Recovery and Reinvestment Act of 2009, Public Law 111-5, 123 Stat. 115 (2009).

In addition to ensuring each Enterprise would continue to be regarded as a viable going concern during and after a repeat of the 2008 financial crisis, FHFA also determined that enhancements to the quantity and quality of regulatory capital at the Enterprises were necessary to mitigate certain risks and limitations associated with the underlying historical data and models used to calibrate the 2018 proposal's credit risk capital requirements. Mitigation of model risk figured prominently in FHFA's design of the proposed rule. As discussed in Section IV.B.2 of the proposed rule, the calibration of the 2018 proposal's credit risk capital requirements attributed a significant portion of the Enterprises' crisis-era losses to the product characteristics of mortgage loans that are no longer eligible for acquisition.

The statistical methods used to allocate losses between borrower-related risk attributes and product-related risk

attributes pose significant model risk.

To ensure safety and soundness, the capital requirements should be sized to mitigate the risk of potential underestimation of credit losses that would be incurred in an economic downturn with national housing price declines similar to those observed in the 2008 financial crisis, even absent those ineligible loan types and even assuming a repeat of federal support of the economy and a declining interest rate environment. There also were some material risks to the Enterprises that were not assigned a risk-based capital requirement under either the 2018 proposal or the proposed rule—for example, risks relating to uninsured or underinsured losses from flooding, earthquakes, or other natural disasters or radiological or biological hazards. There also was no risk-based capital requirement for the risks that climate change could pose to property values in some localities.

These ineligible mortgage loan products included “Alt-A,” negative amortization, interest-only, and low or no documentation loans, as well as loans with debt-to-income ratio at origination greater than 50 percent, cash out refinances with total loan-to-value ratios (LTV) greater than 85 percent, and investor loans with LTV greater than or equal to 90 percent.

Reliance on static look-up grids and multipliers might also introduce additional model risk as borrower behavior, mortgage products, underwriting and collateral valuation practices, or the national housing markets continue to evolve.

The third reason FHFA re-proposed the Enterprises' regulatory capital framework was to make changes to mitigate the procyclicality of the aggregate risk-based capital requirements of the 2018 proposal. FHFA agreed with many of the commenters on the 2018 proposal that mitigating the procyclicality of the 2018 proposal's risk-based capital requirements would facilitate capital management and enhance the safety and soundness of the Enterprises. Mitigating that procyclicality was also critical, in FHFA's view, to position each Enterprise to fulfill its statutory mission to provide stability and ongoing assistance to the secondary mortgage market across the economic cycle.

The enhancements contemplated by the proposed rule, while important, preserved the 2018 proposal as the foundation of the Enterprises' regulatory capital framework. FHFA nonetheless determined to solicit comments on the revised framework in its entirety in light of the changed policy environment, the extent and nature of the enhancements, the technical nature of the underlying issues, the diverse range of interested parties, and the critical importance of the Enterprises' regulatory capital framework to the national housing finance markets.

Overview of the Final Rule

Key Modifications to the Proposed Rule

After carefully considering the comments on the proposed rule, and as described in this preamble, FHFA has determined to make a number of changes to the proposed rule to ensure that each Enterprise operates in a safe and sound manner and is positioned to fulfill its statutory mission across the economic cycle, in particular during periods of financial stress. Key modifications to the proposed rule include, among others:

• Changes to the approach to credit risk transfers (CRT) will better tailor the risk-based capital requirements to the risk retained by an Enterprise on its CRT. These enhancements include a change to the overall effectiveness adjustment for a CRT on a pool of mortgage exposures that has a relatively lower aggregate credit risk capital requirement, a change to the method for assigning a risk weight to a retained CRT exposure so as to increase the risk sensitivity of the risk weight, and a modification to the loss-timing adjustment for a CRT on multifamily mortgage exposures to better tailor the adjustment to the contractual term of the CRT and the loan terms of the underlying exposures. These changes will together generally increase the dollar amount of the capital relief for certain CRT structures commonly entered into by the Enterprises.

• The floor on the adjusted risk weight assigned to mortgage exposures will be 20 percent instead of 15 percent. This adjustment may increase to some extent the dollar amount of the capital relief provided by a CRT on a pool of mortgage exposures that, absent the 20 percent risk weight floor, would have had a smaller aggregate net credit risk capital requirement.

• The credit risk capital requirement for a single-family mortgage exposure that is or was in forbearance pursuant to the Coronavirus Aid, Relief, and Economic Security (CARES) Act or a program established by FHFA to provide forbearance for COVID-19-impacted borrowers will be assigned under an approach that is specifically tailored to these exposures. This approach will significantly reduce the credit risk capital requirement for a non-performing loan that is subject to a COVID-19-related forbearance and, following a reinstatement, will then disregard that period of non-performance.

• The framework for determining credit risk capital requirements will permit a modified re-performing loan to transition to a performing loan after a 5-year period of performance, treat a single-family mortgage exposure in a repayment plan (including following a COVID-19-related forbearance) as a non-modified re-performing loan instead of a modified re-performing loan, and apply a more risk-sensitive approach to single-family mortgage exposures with marked-to-market loan-to-value ratios between 30 and 60 percent.

• The combined risk multiplier for a single-family mortgage exposure will be capped at 3.0, as contemplated by the 2018 proposal.

• The countercyclical adjustment to the standardized credit risk capital requirement for a single-family mortgage exposure will be based on the national, not-seasonally adjusted expanded-data FHFA House Price Index® (expanded-data FHFA HPI) instead of the all-transaction FHFA HPI. The long-term HPI trend line will be subject to re-estimation according to a mechanism specified in the final rule. As of June 30, 2020, house prices were moderately greater than the 5 percent collar. As a result, the adjusted marked-to-market loan-to-value ratios of single-family mortgage exposures would be increased by the countercyclical adjustment, increasing the aggregate risk-based capital requirements for these exposures.

• The stress capital buffer will be periodically re-sized to the extent that FHFA's eventual program for supervisory stress tests determines that an Enterprise's peak capital exhaustion under a severely adverse stress would exceed 0.75 percent of adjusted total assets.

• The advanced approaches requirements will have a delayed effective date of the later of January 1, 2025 and any later compliance date provided by a transition order applicable to the Enterprise. During that interim period, an Enterprise's operational risk capital requirement will be 15 basis points of its adjusted total assets.

B. Modifications to the 2018 Proposal

With these modifications to the proposed rule, the final rule adopts most of the proposed rule's contemplated enhancements to the 2018 proposal, including:

• Simplifications and refinements of the grids and risk multipliers for the credit risk capital requirements for single-family mortgage exposures, including removal of the single-family risk multipliers for loan balance and the number of borrowers.

• A stability capital buffer tailored to the risk that an Enterprise's default or

other financial distress could pose to the liquidity, efficiency, competitiveness, and resiliency of national housing finance markets.

• A stress capital buffer that would, among other things, enhance the resiliency of the Enterprises, help ensure that each Enterprise would continue to be regarded as a viable going concern by creditors and other counterparties after a severe economic downturn, and dampen the procyclicality of the regulatory capital framework by encouraging each Enterprise to retain capital during periods of economic expansion while remaining able to provide stability and ongoing assistance to the secondary mortgage market during a period of financial stress by utilizing capital buffers to absorb losses as incurred.

• A countercyclical adjustment for single-family credit risk that would result in greater capital retention when housing markets may be vulnerable to correction, while better enabling the Enterprises to continue to support the secondary mortgage market during a period of financial stress.

• A prudential floor on the credit risk capital requirement assigned to mortgage exposures to mitigate the model and other risks associated with the methodology for calibrating the credit risk capital requirements and also provide further stability in the aggregate risk-based capital requirements through the economic cycle.

• A credit risk capital requirement on senior tranches of CRT held by an Enterprise to capitalize the retained credit risk, an adjustment to the CRT capital treatment to reflect that CRT is not equivalent in loss-absorbing capacity to equity financing, and operational criteria for CRT structures that together would help mitigate certain structuring, recourse, and other risks associated with these securitizations.

• Risk-based capital requirements for a number of exposures not expressly addressed by the 2018 proposal, including credit risk on commitments to acquire mortgage loans, counterparty risk on interest rate and other derivatives, and credit risk on an Enterprise's holdings or guarantees of the other Enterprise's MBS or debt.

• A revised method for determining operational risk capital requirements, as well as a higher floor.

• A requirement that each Enterprise maintain internal models for determining its own risk-based capital requirements that is intended to prompt each Enterprise to develop its own view of credit and other risks and not rely solely on the risk assessments underlying the standardized risk weights assigned under the regulatory capital framework.

• A 2.5 percent leverage ratio requirement and a 1.5 percent leverage buffer that together would serve as a credible backstop to the risk-based capital requirements and mitigate the inherent risks and limitations of any methodology for calibrating granular credit risk capital requirements.

C. Regulatory Capital Requirements

As implemented by this final rule, the regulatory capital framework will require each Enterprise to maintain the following risk-based capital:

• Total capital not less than 8.0 percent of risk-weighted assets, determined as discussed below;

• Adjusted total capital not less than 8.0 percent of risk-weighted assets;

• Tier 1 capital not less than 6.0 percent of risk-weighted assets; and

• Common equity tier 1 (CET1) capital not less than 4.5 percent of risk-weighted assets.

Each Enterprise also will be required to satisfy the following leverage ratios:

• Core capital not less than 2.5 percent of adjusted total assets; and

• Tier 1 capital not less than 2.5 percent of adjusted total assets.

Adjusted total assets will be defined as total assets under generally accepted accounting principles (GAAP), with adjustments to include certain off-balance sheet exposures. Total capital and core capital will have the meaning given in the Safety and Soundness Act. Adjusted total capital, tier 1 capital, and CET1 capital will be defined based on the definitions of total capital, tier 1 capital, and CET1 capital set forth in the regulatory capital framework (the Basel framework) developed by the Basel Committee on Bank Supervision (BCBS) that is the basis for the United States banking regulators' regulatory capital framework (U.S. banking framework). These supplemental regulatory capital definitions will fill certain gaps in the statutory definitions of core capital and total capital by making customary deductions and other adjustments for certain deferred tax assets (DTAs) and other assets that tend to have less loss-absorbing capacity during a financial stress.

To calculate its risk-based capital requirements, an Enterprise will determine its risk-weighted assets under two approaches—a standardized approach and an advanced approach—with the greater of the two used to determine its risk-based capital requirements. Under both approaches, an Enterprise's risk-weighted assets will equal the sum of its credit risk-weighted assets, market risk-weighted assets, and operational risk-weighted assets.

Under the standardized approach, the credit risk-weighted assets for mortgage loans secured by one-to-four residential units (single-family mortgage exposures) and mortgage loans secured by five or more residential units (multifamily mortgage exposures) will be determined using lookup grids and multipliers that assign an exposure-specific risk weight based on the risk characteristics of the mortgage exposure. These lookup grids and multipliers generally are similar to those of the 2018 proposal, with some simplifications and refinements.

This base risk weight would be equal to the adjusted total capital requirement for the mortgage exposure expressed in basis points and divided by 800, which is the 8.0 percent adjusted total capital requirement also expressed in basis points. For example, the credit risk capital requirement for a mortgage exposure with a base risk weight of 50 percent would be 400 basis points (800 multiplied by 50 percent).

Like the 2018 proposal, the base risk weight will be a function of the mortgage exposure's loan-to-value ratio with the property value generally marked to market (MTMLTV). For single-family mortgage exposures, the MTMLTV will be subject to a countercyclical adjustment to the extent that national house prices are 5.0 percent greater than or less than an inflation-adjusted long-term trend. For both single-family and multifamily mortgage exposures, this base risk weight will then be adjusted to reflect additional risk attributes of the mortgage exposure and any loan-level credit enhancement. To ensure an appropriate level of capital, this adjusted risk weight will be subject to a minimum floor of 20 percent.

As of June 30, 2020, under the final rule's standardized approach, the Enterprises' average risk weight for single-family mortgage exposures would have been 37 percent, and the Enterprises' average risk weight for multifamily mortgage exposures would have been 49 percent.

These average risk weights are determined based on the credit risk capital requirement for single-family and multifamily mortgage exposures after adjustments for mortgage insurance and other loan-level credit enhancement but before any adjustment for CRT.

While the standardized approach will utilize FHFA-prescribed lookup grids and risk multipliers, the advanced approach for determining credit risk-weighted assets will rely on each Enterprise's internal models. The advanced approach requirements will require each Enterprise to maintain its own processes for identifying and assessing credit risk, market risk, and operational risk. These requirements are

intended to ensure that each Enterprise continues to enhance its risk management system and also that neither Enterprise relies solely on the standardized approach's lookup grids and multipliers to define credit risk tolerances, measure its credit risk, or allocate capital. In the course of FHFA's supervision of each Enterprise's internal models for credit risk, FHFA also could identify opportunities to update or otherwise enhance the standardized approach's lookup grids and multipliers through a future rulemaking.

Under both the standardized and advanced approaches, an Enterprise will determine the capital treatment for eligible CRT by assigning risk weights to retained CRT exposures. Under the standardized approach, tranche-specific risk weights will be subject to a 10 percent floor. The risk-weighted assets of a retained CRT exposure will be subject to adjustments to reflect loss-sharing effectiveness, loss-timing effectiveness, and the differences between CRT and regulatory capital, ensuring that the capital relief afforded by the CRT appropriately reflects the credit risk retained by the Enterprise.

Each Enterprise also will determine a market risk capital requirement for spread risk. Market risks other than spread risk will not be assigned a market risk capital requirement, but FHFA continues to consider more comprehensive approaches for future rulemakings. Under the standardized approach, an Enterprise will determine its market risk-weighted assets using FHFA-specified formulas for some covered positions and its own models for other covered positions. An Enterprise will separately determine its market risk-weighted assets under an advanced approach that relies only on its own internal models for all covered positions.

The final rule also will require each Enterprise to determine its operational risk capital requirement utilizing the U.S. banking framework's advanced measurement approach, subject to a floor equal to 15 basis points of the Enterprise's adjusted total assets.

Each of these regulatory capital requirements will be enforceable by FHFA under its general authority to order an Enterprise to cease and desist from a violation of law, which would include the final rule and its regulatory capital requirements. Pursuant to that authority, FHFA may require an Enterprise to develop and implement a capital restoration plan or take other appropriate corrective action. FHFA also could elect to enforce the risk-based and leverage ratio requirements pursuant to its authority to require an Enterprise to develop a plan to achieve compliance with prescribed prudential management and operational standards, and FHFA also could enforce the core capital leverage ratio requirement or the risk-based total capital requirement pursuant to its separate authority to require prompt corrective action if an Enterprise fails to maintain certain prescribed regulatory levels.

D. Capital Buffers

To avoid limits on capital distributions and discretionary bonus payments, an Enterprise must maintain CET1 capital that exceeds its risk-based capital requirements by at least the amount of its prescribed capital conservation buffer amount (PCCBA). That PCCBA will consist of three separate component buffers—a stress capital buffer, a countercyclical capital buffer, and a stability capital buffer.

• The stress capital buffer will be at least 0.75 percent of an Enterprise's adjusted total assets. FHFA will periodically re-size the stress capital buffer to the extent that FHFA's eventual program for supervisory stress tests determines that an Enterprise's peak capital exhaustion under a severely adverse stress would exceed 0.75 percent of adjusted total assets.

• The countercyclical capital buffer amount initially will be set at 0 percent of an Enterprise's adjusted total assets. FHFA does not expect to adjust this buffer in the place of, or to supplement, the countercyclical adjustment to the risk-based capital requirements. Instead, as under the Basel and U.S. banking frameworks, FHFA will adjust the countercyclical capital buffer taking into account the macro-financial environment in which the Enterprises operate, such that the buffer would be deployed only when excess aggregate credit growth is judged to be associated with a build-up of system-wide risk. This focus on excess aggregate credit growth means the countercyclical buffer likely will be deployed on an infrequent basis, and generally only when similar buffers are deployed by the U.S. banking regulators.

• An Enterprise's stability capital buffer will be tailored to the risk that an Enterprise's default or other financial distress could pose to the liquidity, efficiency, competitiveness, or resiliency of national housing finance markets. The stability capital buffer will be based on an Enterprise's share of residential mortgage debt outstanding. As of June 30, 2020, Fannie Mae's and Freddie Mac's stability capital buffers would have been, respectively, 1.07 and 0.66 percent of adjusted total assets.

Finally, to avoid limits on capital distributions and discretionary bonus payments, the Enterprise also will be required to maintain tier 1 capital in excess of the amount required under its tier 1 leverage ratio requirement by at least the amount of its prescribed leverage buffer amount (PLBA). The PLBA will equal 1.5 percent of the Enterprise's adjusted total assets, such that the PLBA-adjusted leverage ratio requirement would function as a credible backstop to the PCCBA-adjusted risk-based capital requirements.

E. Transition Period

An Enterprise will not be subject to any requirement under the final rule until the compliance date for the requirement under the final rule. The compliance date for the regulatory capital requirements (distinct from the PCCBA or the PLBA) will be the later of the date of the termination of the conservatorship of the Enterprise (or, if later, the effective date of the final rule, which would be 60 days after publication in the

Federal Register

) and any later compliance date provided in a consent order or other transition order applicable to the Enterprise. In contrast, FHFA contemplates that the compliance dates for the PCCBA and the PLBA will be the date of the termination of the conservatorship of the Enterprise (or, if later, the effective date of the final rule), so as to provide additional authority to FHFA to restrict dividends and other capital distributions during the period in which the Enterprise raises regulatory capital to achieve compliance with the regulatory capital requirements. FHFA expects that this interim period could be governed by a capital restoration plan that would be binding on the Enterprise pursuant to a consent order or other transition order.

The final rule's advanced approaches requirements will be delayed until the later of January 1, 2025 and any later compliance date specific to those requirements provided in a consent order or other transition order applicable to the Enterprise. Regardless of the date of the termination of the conservatorship of an Enterprise, the Enterprise will be required to report its regulatory capital, PCCBA, PLBA, standardized total risk-weighted assets, and adjusted total assets beginning January 1, 2022.

IV. FSOC Review of the Secondary Mortgage Market

On September 25, 2020, the Financial Stability Oversight Council (FSOC) released a statement on its activities-based review of the secondary mortgage

market (FSOC Secondary Market Statement). FSOC found that any distress at the Enterprises that affected their secondary mortgage market activities could pose a risk to financial stability, if risks are not properly mitigated. Much of FSOC's analysis centered on the extent to which the proposed rule would adequately mitigate the potential stability risk of the Enterprises.

The FSOC Secondary Market Statement affirmed the overall quantity and quality of the regulatory capital required by the proposed rule. The FSOC Secondary Market Statement also indicated that greater capital requirements might be appropriate for some exposures. Notably, FSOC's analysis suggested that “risk-based capital requirements and leverage ratio requirements that are materially less than those contemplated by the proposed rule would likely not adequately mitigate the potential stability risk posed by the Enterprises.” FSOC also found that “it is possible that additional capital could be required for the Enterprises to remain viable concerns in the event of a severely adverse stress . . . .”

The FSOC Secondary Market Statement included other findings and recommendations that generally endorsed the objectives, rationales, and approaches of the proposed rule.

•

Going-concern standard.

Consistent with the proposed rule's objectives, FSOC “encourage[d] FHFA to require the Enterprises to be sufficiently capitalized to remain viable as going concerns during and after a severe economic downturn.” This recommendation should preclude a “claims-paying capacity” or similar framework that seeks only to ensure that an Enterprise has the ability to perform its guarantee and other financial obligations over time, perhaps subject to a stay or other pause in the payment of claims and other financial obligations during a resolution proceeding. Instead, each Enterprise should be capitalized not only to absorb losses as they are incurred in a severely adverse stress, but also so that the Enterprise would have sufficient regulatory capital after that stress to continue to be regarded as a viable going concern by creditors and other counterparties.

•

Enterprise-specific stability buffer.

In a significant departure from the 2018 proposal, the proposed rule contemplated an Enterprise-specific stability capital buffer tailored to the risk that an Enterprise's default or other financial distress could pose to the liquidity, efficiency, competitiveness, or resiliency of national housing finance markets. FSOC affirmed that “[a] stability capital buffer would mitigate risks to financial stability by reducing the expected impact of an Enterprise's distress on financial markets or other financial market participants and by addressing the potential for decreased market discipline due to an Enterprise's size and importance.” FSOC also recommended that “[t]he capital buffers should be tailored to mitigate the potential risks to financial stability.”

•

Quality of capital.

FSOC endorsed the proposed rule's use of the U.S. banking framework's definitions of regulatory capital to prescribe supplemental capital requirements. Specifically, FSOC “encourage[d] FHFA to ensure high-quality capital by implementing regulatory capital definitions that are similar to those in the U.S. banking framework.” This recommendation supports FHFA's determination in the proposed rule and in the 2018 proposal, consistent with the U.S. banking framework, not to include a measure of guarantee fees or other future revenues as an element of regulatory capital.

•

U.S. banking framework comparisons.

FSOC found that “[t]he Enterprises' credit risk requirements . . . 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.” This finding is consistent with FHFA's determination in the proposed rule that, as of September 30, 2019, the proposed rule's average credit risk capital requirements for the Enterprises' mortgage exposures generally were roughly half those of similar exposures under the U.S. banking framework. Those lower average credit risk capital requirements were before any adjustment for the capital relief afforded through CRT.

The FSOC Secondary Market Statement also identified potential opportunities to enhance the proposed rule and FHFA's regulatory framework more generally.

•

Buffer calibration.

FSOC “encourage[d] FHFA to consider the relative merits of alternative approaches for more dynamically calibrating the capital buffers.” The proposed rule contemplated a stress capital buffer sized as a fixed percent of an Enterprise's adjusted total assets, and FHFA sought comment on whether to adopt an alternative approach under which FHFA would periodically re-size the stress capital buffer, similar to the approach recently adopted by the U.S. banking regulators, to the extent that FHFA's eventual program for supervisory stress tests determines that an Enterprise's peak capital exhaustion under a severely adverse stress would exceed 0.75 percent of adjusted total assets. FHFA has adopted that alternative approach in this final rule.

•

Level playing field.

FSOC “encourage[d] 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 and other regulation of similar risks across market participants, consistent with the business models and missions of their regulated entities.” In the final rule, FHFA has adopted a risk weight floor on mortgage exposures that is equal to the smallest risk weight contemplated by the Basel framework for residential real estate exposures.

BCBS,

Basel III: Finalising post-crisis reforms

¶¶ 59-68 (Dec. 2017).

•

Other regulatory requirements.

FSOC noted that FHFA's “efforts to strengthen Enterprise liquidity regulation, stress testing, supervision, and resolution planning would help mitigate the potential risk to financial stability.” FSOC stated that it “support[ed] FHFA's commitment to developing its broader prudential regulatory framework for the Enterprises and encourage[d] FHFA to continue those efforts.”

FSOC also committed to continue to monitor the secondary mortgage market activities of the Enterprises and FHFA's implementation of the regulatory framework to ensure potential risks to financial stability are adequately addressed. Significantly, if FSOC later determines that such risks to financial stability are not adequately addressed by FHFA's capital and other regulatory requirements or other risk mitigants, FSOC may consider more formal recommendations or other actions, consistent with the interpretive guidance on nonbank financial company determinations issued by FSOC in December 2019.

If the activities-based approach contemplated by that guidance does not adequately address a potential threat to financial stability, FHFA understands that FSOC could consider a nonbank financial company, including an Enterprise, for potential designation for supervision and regulation by the Board of Governors of the Federal Reserve System (Federal Reserve Board).

V. General Comments on the Proposed Rule

FHFA received 128 public comment letters on the proposed rule from the Enterprises, trade associations, consumer advocacy groups, private individuals, and other interested parties.

Overall, most commenters supported FHFA's effort to establish a post-conservatorship regulatory capital framework that would ensure that each Enterprise operates in a safe and sound manner and is positioned to fulfill its statutory mission across the economic cycle. However, many commenters also expressed concern about the potential impacts, costs, and burdens of various aspects of the proposed rule.

See

comments on Enterprise Regulatory Capital Framework,

available

https://www.fhfa.gov/SupervisionRegulation/Rules/Pages/Comment-List.aspx?RuleID=674.

The comment period for the proposed rule closed on August 31, 2020.

A. Access and Affordability and Other Aggregate Impacts

Many commenters expressed concern about the potential aggregate impacts of the proposed rule, such as: Higher borrowing costs, including for first-time and low- and moderate-income borrowers and minority and rural communities; implications for the Enterprises' ability to satisfy their affordable housing goals or their duty to serve mandates or perform their countercyclical mission; greater cost of home ownership; an increased racial wealth gap; impacts on the affordability of multifamily housing; different pricing impacts on specific mortgage products; lower Enterprise returns on equity; reduced investor demand for the Enterprises' equity; shifts in market share from the Enterprises to banks, private-label securitization (PLS), or the Federal Housing Administration; limits on the ability of credit unions to serve their customers; incentives for the Enterprises to increase risk taking, retain mortgage credit risk, or engage in risk-based pricing of their guarantee fees; disincentives to engage in CRT; and greater compliance costs.

Some commenters urged that the Enterprises' charter mandate to serve the public interest should inform changes to the proposed rule. Other commenters challenged the perceived complexity of the proposed rule. Still other commenters requested that FHFA perform additional studies on the impact of all or parts of the proposed rule, while certain other commenters sought withdrawal or re-proposal of the proposed rule. Other commenters urged that any future changes to the Enterprises' guarantee fees should wait until there is additional clarity about the future regulatory and market structure.

Some commenters questioned whether the regulatory capital framework might impede an Enterprise's ability to raise capital, while some commenters thought that the Enterprises would still have an attractive return on equity under the proposed rule. A few commenters urged FHFA to consider that each Enterprise's existing books of businesses might have been priced assuming smaller required quantities of regulatory capital, which might be particularly relevant to the extent that recent refinancing volumes extend the expected life of the portfolio.

Many commenters generally supported FHFA's objective to establish a post-conservatorship regulatory capital framework that would ensure that each Enterprise operates in a safe and sound manner and is positioned to fulfill its statutory mission across the economic cycle. Some commenters argued that the interests of low- and moderate-income borrowers would be best served by capitalizing the Enterprises to support the secondary market during a period of financial stress, especially as these borrowers' access to credit tends to be most adversely affected by financial stress. Also, some commenters stated that appropriately capitalizing each Enterprise would mitigate risk to financial stability. A few commenters advocated that FHFA should protect taxpayers against future bailouts by requiring adequate loss-absorbing capacity.

FHFA carefully considered these comments in identifying and assessing potential changes to the proposed rule. As context for that discussion elsewhere in this preamble, FHFA notes that the Safety and Soundness Act requires FHFA to establish by regulation risk-based capital requirements for the Enterprises to ensure that each Enterprise operates in a safe and sound manner, maintaining sufficient capital and reserves to support the risks that arise in the operations and management of the Enterprise.

While FHFA has other mission-related mandates, this particular statutory mandate focuses only on safety and soundness.

12 U.S.C. 4611(a)(1). Safety and soundness is also the standard governing FHFA's authority to set a leverage ratio higher than the minimum prescribed by the statute. 12 U.S.C. 4612(c).

In addition to ensuring the Enterprises' safety and soundness, the proposed rule did still seek to ensure that each Enterprise will be positioned to fulfill its statutory mission across the economic cycle. This objective led to changes to the 2018 proposal to reduce the regulatory capital framework's procyclicality. The proposed rule also took specific steps to mitigate the potential impacts on higher risk exposures. These steps included setting the PCCBA as a fixed percent of adjusted total assets (not risk-weighted assets), removing the single-family risk multipliers for loan balance and number of borrowers, and reducing the risk-based capital requirements for low down-payment loans with private mortgage insurance. More generally, FHFA continues to believe that appropriately capitalizing each Enterprise is critical to ensuring that the secondary mortgage market supports access to affordable mortgage credit for low- and moderate-income borrowers and minority borrowers during periods of financial stress, when these borrowers are potentially most vulnerable to loss of access to affordable mortgage credit.

In FHFA's view, predictions of a material increase in mortgage credit borrowing costs as a result of the proposed rule are subject to scrutiny and significant uncertainty. Some economic theory and empirical evidence suggest that an increase in an Enterprise's equity financing would lead to some decrease in the Enterprise's cost of equity capital, mooting some, or perhaps much, of any such potential impact of increased regulatory capital requirements.

Evidencing that point, the significant increase in the U.S. banking framework's regulatory capital requirements following the 2008 financial crisis generally did not lead to significant increases in borrowing costs, contrary to the predictions of market participants at the time.

The Enterprises' cost of capital also might be affected by the pricing and availability of CRT over time. Further complicating the analysis, the Enterprises' pricing decisions will be influenced by a variety of regulatory and market considerations. The Enterprises' housing goals set by FHFA will be a particularly important consideration in each Enterprise's pricing decisions with respect to low- and moderate-income borrowers. As

discussed in Section V.D, an Enterprise's pricing decisions should be increasingly based on its own risk assessment as the Enterprise retains capital. An Enterprise's pricing decisions will also inevitably take into account the pricing and other economic decisions of the other Enterprise, with pricing equilibriums under a duopoly difficult to model and predict. To the extent that the Enterprises compete with other market participants, the cost of mortgage credit will depend on the pricing decisions of those competitors, with those competitors outside the scope of FHFA's regulatory capital framework. Finally, the proposed rule was intended to ensure each Enterprise could support the secondary market during a period of financial stress, and any assessment of the regulatory capital framework's impact on borrowing costs should evaluate borrowing costs over the course of the economic cycle. Commentary on the proposed rule generally did not address these complicating factors and should be considered in the context of similar concerns that post-crisis enhancements to the U.S. banking framework would significantly and adversely affect the cost of and access to credit.

Modigliani, F., and Miller, M.H. (1958),

The Cost of Capital, Corporation Finance and the Theory of Investment,

The American Economic Review, 48:3 (1958); BCBS,

The costs and benefits of bank capital—a review of the literature

(June 2019) at section 2.3; Jihad Dagher et al.,

IMF Staff Discussion Note: Benefits and Costs of Bank Capital

(March 2016) at Table 4.A; Federal Reserve Bank of Minneapolis,

The Minneapolis Plan to End Too Big to Fail

(November 2016).

See, e.g.,

Simon Firestone, Amy Lorenc, and Ben Ranish,

An Empirical Economic Assessment of the Costs and Benefits of Bank Capital in the US

(March 31, 2017).

B. Similarities to the U.S. Banking Framework

Some commenters supported the proposed rule's use of the Basel framework's regulatory capital definitions to prescribe supplemental capital requirements. Some commenters also supported the use of risk weights to define each mortgage exposure's risk-based capital requirement, the inclusion of the stress capital buffer, and the incorporation of other concepts from the Basel and U.S. banking frameworks. Some commenters advocated a general alignment of the credit risk capital requirements for similar mortgage exposures across the Enterprises and other market participants, which also was a recommendation in the FSOC Secondary Market Statement.

Other commenters criticized the extent to which the proposed rule incorporated concepts from the Basel and U.S. banking frameworks. Some commenters argued that the proposed rule inappropriately treated the Enterprises as banks and that “bank-like” quantities of required capital would be inappropriate for the Enterprises.

As discussed in Sections VIII.A.7 and VIII.B.6 of the proposed rule, as of September 30, 2019, and before adjusting for CRT or the buffers, the average credit risk capital requirements for the Enterprises' mortgage exposures generally were roughly half those of similar exposures under the U.S. banking framework.

The Enterprises together would have been required under the proposed rule's risk-based capital requirements to maintain $234 billion in risk-based adjusted total capital as of September 30, 2019 to avoid restrictions on capital distributions and discretionary bonuses. Had they been instead subject to the U.S. banking framework, the Enterprises would have been required to maintain approximately $450 billion, perhaps significantly more, in risk-based total capital (not including market risk and operational risk capital) to avoid similar restrictions.

In light of these facts, FHFA reiterates that the proposed rule would not have subjected the Enterprises to the same capital requirements that apply to U.S. banking organizations.

FHFA's mortgage risk-sensitive framework results in a more granular calibration of credit risk capital requirements for mortgage exposures, and some meaningful portion of the gap between the credit risk capital requirements of the Enterprises and large banking organizations under the proposed rule was due to the proposed rule's use of MTMLTV instead of OLTV, as under the U.S. banking framework, to assign credit risk capital requirements. Adjusting for the appreciation in the value of the underlying real property generally led to lower actual credit risk capital requirements at the Enterprises, and some of the gap between the credit risk capital requirements of the Enterprises and large U.S. banking organizations perhaps might be expected to narrow somewhat were real property prices to move toward their long-term trend.

These estimates are complicated and sensitive to important assumptions. There were several key drivers of the gap between the aggregate risk-based capital requirements under the proposed rule and under the U.S. banking framework. The lower underlying credit risk capital requirements contributed significantly to this gap. Different approaches to the capital relief for private mortgage insurance and CRT also contributed to some of the gap. The risk-weighted assets-based buffers of the U.S. banking framework also could increase the gap, depending on the assumptions made as to each Enterprise's buffer requirement. Some of the gap perhaps might be expected to narrow somewhat were real property prices to move toward their long-term trend.

C. Differences Between the Enterprises and Banks

Prompted in some cases perhaps by the comparisons in the proposed rule to the Basel and U.S. banking frameworks, many commenters emphasized the differences in the business models, statutory mandates, and risk profiles of the Enterprises and banking organizations. FHFA agrees with these commenters that there are important differences between the Enterprises and banking organizations. The proposed rule discussed those differences in several places, including Sections IV.B.2, VI.B.3, and XIII of the proposed rule, noting, for example, that while the Enterprises transfer much of the interest rate and funding risk on their mortgage exposures through their sales of guaranteed MBS, banking organizations generally fund themselves through customer deposits and other sources. The different interest rate risk profile of the Enterprises is one reason that the proposed rule's market risk capital requirements constituted a relatively small share of the aggregate risk-based capital requirement.

The differences between the business models, statutory mandates, and risk profiles of the Enterprises and banking organizations, however, should not preclude the proposed rule's comparison of the

credit risk

capital requirement of a large U.S. banking organization for a specific mortgage exposure to the

credit risk

capital requirement of an Enterprise for a similar mortgage exposure.

The different interest rate risk profiles do not preclude this comparison because the Basel and U.S. banking frameworks generally do not contemplate an explicit capital requirement for interest rate risk on banking book exposures, instead leaving interest rate risk capital requirements to bank-specific tailoring through the supervisory process.

Related to this comparison, the monoline nature of the Enterprises' mortgage-focused businesses suggests that the concentration risk of an Enterprise is generally greater than that of a diversified banking organization with a similar amount of mortgage credit risk. That heightened concentration risk would tend to suggest that greater credit risk capital requirements, relative to banking organizations, could be appropriate for the Enterprises for similar exposures, all else equal.

Comparisons of credit risk capital requirements can further safety and soundness by helping to identify and mitigate model and related risks relating to the calibration of the requirements. Comparisons of credit risk capital requirements can also further financial stability by identifying undue differences in regulatory requirements that might distort the market structure, as acknowledged by the FSOC Secondary Market Statement. According to the FSOC Secondary Market Statement, “[t]he alignment of market participants' credit risk capital requirements across similar credit risk exposures would mitigate risk to financial stability by minimizing market structure distortions.”

See

BCBS,

Interest Rate Risk in the Banking Book,

¶ 1 (April 2016), available at

https://www.bis.org/bcbs/publ/d368.pdf;

(“Interest rate risk in the banking book (IRRBB) is part of the Basel capital framework's Pillar 2 (Supervisory Review Process) and subject to the Committee's guidance set out in the 2004 Principles for the management and supervision of interest rate risk (henceforth, the IRR Principles).”).

The differences between the business models, statutory mandates, and risk profiles of the Enterprises and banking

organizations also should not be understood as inconsistent with capitalizing each Enterprise to remain a viable going concern both during and after a severe economic downturn. As discussed in Section II, each Enterprise has considerable funding risk even if it does not rely on customer deposits, and an Enterprise's ordinary course and procyclical funding needs can be met only if the Enterprise continues to be regarded as a viable going concern by creditors throughout the duration of a financial stress.

D. Mortgage-Risk Sensitive Framework

Many commenters expressed concern that those aspects of the proposed rule that tended to decrease the risk sensitivity of the regulatory capital framework could distort the pricing, risk transfer, or other economic decisions of the Enterprises. FHFA agrees with commenters that there are significant benefits to a mortgage risk-sensitive framework. There are, however, trade-offs associated with risk sensitivity. A more risk-sensitive framework tends to amplify the model and related risks associated with any methodology for calibrating a granular assessment of credit risk, which poses significant risk to safety and soundness. A more risk-sensitive framework can be significantly more procyclical, which was a concern of many commenters on the 2018 proposal. A more risk-sensitive framework also can adversely affect an Enterprise's ability to support access to affordable mortgage credit for higher risk borrowers, perhaps excessively so to the extent that the historical performance of these borrowers, which was used to determine the credit risk capital requirements, might not be predictive of future performance. FHFA believes that it has struck an appropriate balance between these competing policy considerations by preserving risk sensitivity while ensuring that each Enterprise operates in a safe and sound manner and is positioned to fulfill its statutory mission across the economic cycle.

FHFA also believes that those aspects of the final rule that might tend to decrease the regulatory capital framework's risk sensitivity will not unduly distort each Enterprise's pricing, credit, CRT, and other economic decisions. FHFA expects that each Enterprise, like other regulated financial institutions, will base its decisions on its own risk assessments, not solely or even primarily on the regulatory capital requirements. By capitalizing each Enterprise to remain a viable going concern without government support, the final rule will incentivize an Enterprise to continually enhance its own risk assessments so as to effectively manage its now-internalized risk. That incentive will be supplemented by the final rule's advanced approaches requirements, which will require each Enterprise to continually enhance its internal models. FHFA also anticipates that each Enterprise's decisions will be informed by other considerations, in particular the decisions of the other Enterprise and other market participants and also the statutory requirement to satisfy FHFA's housing goals. Evidencing this view that the regulatory capital framework generally will not define pricing decisions, the U.S. banking framework's standardized credit risk capital requirements for residential mortgage exposures have very limited risk sensitivity, and yet the pricing of mortgage credit risk varies widely across U.S. banking organizations and especially across borrowers. Mortgage insurers are subject to aligned Enterprise requirements to maintain minimum levels of financial strength, and yet the pricing of mortgage credit risk varies across mortgage insurers.

More generally, the regulatory capital framework should encourage decisions based on nuanced, dynamic, and diverse understandings of risk. A significant and perhaps underappreciated benefit of capitalizing each Enterprise so that its risks are internalized, rather than borne by taxpayers, is that each Enterprise will face market discipline and strong incentives to base its decisions more on its own understanding of the costs and benefits and less on that of its regulator. This is important because FHFA's risk-based capital requirements should not be regarded as the last or best view on risk. Other modeling approaches might consider the loss experiences of other market participants during the 2008 financial crisis, incorporate data from other economic downturns, both in the United States and abroad, take a different approach to the significant portion of the Enterprises' crisis-era losses that were attributed to product features that are no longer eligible for acquisition (approximately $108 billion), or employ different regularization techniques. The now apparent shortcomings of OFHEO's and the Enterprises' pre-crisis credit models, and other well-known failures of analytical models to accurately predict risk, reinforce the need for a meaningful degree of regulatory caution regarding any modeled estimate of risk. Reform should therefore provide incentives for each Enterprise to develop and act on its own view of risk.

Housing Finance Reform

Commenters raised a variety of issues relating to housing finance reform proposals. Some commenters urged FHFA to wait to finalize a regulatory capital framework for the Enterprises until Congress enacts housing reform legislation clarifying the extent of any federal government support of the Enterprises or their successors. Similarly, some commenters argued that the conservatorships should continue until Congress acts. Some commenters advocated for regulating the Enterprises' pricing or otherwise subjecting the Enterprises to utility-like regulation, while other commenters suggested other administrative or legislative reforms, for example, steps to ensure equitable access to the secondary market by lenders of all sizes and charter types.

Commenters also offered views on issues relating to the Enterprises' conservatorships, including the Enterprises' consent to conservatorship in 2008, subsequent actions by FHFA or the U.S. Department of the Treasury (Treasury), and FHFA's policy to responsibly end the conservatorships. Many commenters urged FHFA to end the conservatorships and recommended certain steps toward that end. Some commenters argued in favor of a resolution of the claims made by the Enterprises' legacy shareholders or that the liquidation preference of Treasury's senior preferred shares should be extinguished. Commenters advocated that FHFA should consider Treasury's commitment under the Senior Preferred Stock Purchase Agreements (PSPA) in designing the regulatory capital framework.

FHFA continues to believe that the regulatory capital framework should not assume extraordinary government support, whether under the PSPAs or otherwise. A central tenet of the reforms following the 2008 financial crisis is that the post-crisis regulatory framework should prevent future taxpayer rescues of financial institutions.

Expectations of government support increase risk to the Enterprises' safety and soundness and the stability of the national housing finance markets by undermining market discipline and encouraging excessive

risk taking.

Other regulatory capital frameworks generally would not treat a line of credit or similar arrangement, even one with a governmental actor, as a form of regulatory capital. Moreover, to the extent that there are existing arrangements under which the federal government could be exposed to the losses of a financial institution—for example, the Federal Deposit Insurance Corporation's Deposit Insurance Fund or its Orderly Liquidation Fund—those arrangements have motivated

greater

regulatory capital requirements to mitigate the risk to safety and soundness and to protect taxpayers. More practically, Treasury's commitment under the PSPAs is finite and cannot be replenished, and that commitment could be inadequate to ensure each Enterprise would remain a viable going concern during and after a severe economic downturn, particularly to the extent that an Enterprise's liabilities and other obligations were to grow relative to that fixed commitment.

The Dodd-Frank Act is an Act “[t]o promote the financial stability of the United States by improving accountability and transparency in the financial system, to end `too big to fail', to protect the American taxpayer by ending bailouts, to protect consumers from abusive financial services practices, and for other purposes.”

See

BCBS,

Global systemically important banks: revised assessment methodology and the higher loss absorbency requirement

¶ 3 (“[T]he moral hazard costs associated with implicit guarantees derived from the perceived expectation of government support may amplify risk-taking, reduce market discipline and create competitive distortions, and further increase the probability of distress in the future. As a result, the costs associated with moral hazard add to any direct costs of support that may be borne by taxpayers.”); Federal Reserve Board,

Calibrating the GSIB Surcharge

(2015) at 1 (“The experience of the crisis made clear that the failure of a SIFI during a period of stress can do great damage to financial stability, that SIFIs themselves lack sufficient incentives to take precautions against their own failures, that reliance on extraordinary government interventions going forward would invite moral hazard and lead to competitive distortions, and that the pre-crisis regulatory focus on microprudential risks to individual financial firms needed to be broadened to include threats to the overall stability of the financial system.”).

FHFA continues to support legislation to reform the flaws in the structure of the housing finance system that were at the root of the 2008 financial crisis and that continue to pose risk to taxpayers and financial stability. To that end, FHFA recommended specific legislative reforms in its last Annual Report to Congress. FHFA reiterates its recommendation that Congress authorize FHFA to charter competitors to the Enterprises and remove unnecessary statutory exemptions and other special treatments afforded the Enterprises. Chartering competitors to the Enterprises could reduce the size and importance of any single Enterprise, which could lead to a smaller stability capital buffer and therefore smaller aggregate capital requirements.

Pending legislation, FHFA, as conservator of each Enterprise, is required by statute to act “for the purpose of reorganizing, rehabilitating, or winding up the affairs of [the Enterprise].”

That definite and limited statutory purpose does not authorize an indefinite conservatorship. FHFA is in the process of preparing each Enterprise to responsibly exit conservatorship consistent with its statutory mandate and the FHFA Director's other duties. Finalization of the Enterprises' regulatory capital framework is a key step in that effort.

12 U.S.C. 4617(a)(2).

Finalization of the Enterprise's regulatory capital framework is also required by law. The Safety and Soundness Act not only authorizes, but affirmatively requires, FHFA to prescribe risk-based capital requirements by regulation.

FHFA has been subject to this statutory mandate for more than 12 years, and in FHFA's view, this final rule is long overdue.

12 U.S.C. 4611(a)(1) (“The Director

shall,

by regulation, establish risk-based capital requirements for the enterprises to ensure that the enterprises operate in a safe and sound manner, maintaining sufficient capital and reserves to support the risks that arise in the operations and management of the enterprises.”) (emphasis added). FHFA's predecessor agency, OFHEO, adopted a risk-based capital rule (12 CFR part 1750) that will not have been formally rescinded until the effective date of this final rule. That rule was suspended by FHFA at the inception of the conservatorships in 2008. That rule clearly failed to ensure the safety and soundness of each Enterprise.

VI. Definitions of Regulatory Capital

As discussed in Section VII, the proposed rule would have required each Enterprise to maintain specified amounts of core capital and total capital, as defined in the Safety and Soundness Act. The proposed rule would have supplemented the core capital and total capital requirements with risk-based and leverage ratio requirements based on the Basel framework's definitions of total capital, tier 1 capital, and CET1 capital. The supplemental definitions of regulatory capital would have made deductions and other adjustments for certain DTAs, ALLL, goodwill, intangibles, and other assets that might tend to have less loss-absorbing capacity during a financial stress. The tier 1 and CET1 capital requirements also would have ensured that retained earnings and other high-quality capital are the predominant form of regulatory capital.

Some commenters supported the proposed rule's use of the Basel framework's regulatory capital definitions to prescribe supplemental capital requirements, potentially as a means to better align credit risk capital requirements across market participants and also to facilitate comparability across regulatory capital frameworks. Some commenters suggested that CRT should be treated as an element of regulatory capital, while a few commenters argued that tier 1 capital was the best basis for both leverage ratio and risk-based capital requirements. Commenters otherwise generally focused on the proposed rule's treatment of guarantee fees, reserves, and subordinated debt.

A. Guarantee Fees

Consistent with the 2018 proposal, neither the statutory definitions nor the supplemental definitions of regulatory capital in the proposed rule would have included a measure of future guarantee fees or other future revenues. FHFA instead gave consideration to the loss-absorbing capacity of future revenues in calibrating the stress capital buffer.

Many commenters argued that a measure of guarantee fees should be included in one or more of the definitions of regulatory capital. That measure, for example, could be limited to guarantee fees that have been received by an Enterprise but not yet recognized as revenue for accounting purposes. These commenters generally contended that future revenues are available to absorb future losses or pay future claims, as reflected in the estimates of capital exhaustion produced by the Enterprises' annual stress tests. A few commenters noted that the proposed rule could incentivize an Enterprise to create interest-only strips of guarantee fee revenue to recognize assets that could count toward regulatory capital. Commenters also suggested that the proposed rule's approach could have a relatively greater impact on higher risk mortgage exposures.

After considering these comments, FHFA has determined to not include a measure of future revenues in any of the final rule's definitions of regulatory capital. Future revenues instead would continue to be considered in sizing the stress capital buffer, as discussed in Section VIII.A.2. Like the proposed rule, the final rule seeks to ensure that each Enterprise would be capitalized to remain a viable going concern both during and after a severe economic downturn. The 2008 financial crisis established that credit, market, and other losses can be incurred quickly during a stress, and it is an Enterprise's capacity to absorb those losses as incurred while still timely performing its financial obligations that defines creditors' and other counterparties' views as to whether the Enterprise is a viable going concern. During a stress, creditors are unlikely to give much

consideration to future revenue prospects in assessing whether an Enterprise can timely perform its financial obligations. Market confidence in the Enterprises waned in mid-2008 when Fannie Mae and Freddie Mac had total capital of, respectively, $55.6 billion and $42.9 billion, notwithstanding their right to future guarantee fees. Moreover, as discussed in Section IV, the FSOC Secondary Market Statement endorsed the proposed rule's use of the U.S. banking framework's definitions of regulatory capital to prescribe supplemental capital requirements, and these definitions do not include a measure of future revenues.

B. Reserves

The statutory definition of total capital includes a general allowance for foreclosure losses. As for advanced approaches banking organizations under the U.S. banking framework, the proposed rule would have permitted an Enterprise to include in the supplemental definition of tier 2 capital only the excess of its eligible credit reserves over its total expected credit loss, provided the amount does not exceed 0.6 percent of its credit risk-weighted assets. A few commenters suggested that it might be appropriate to include some portion of ALLL in the supplemental definitions of regulatory capital, particularly if the U.S. banking regulators were in the future to adjust their approach to ALLL after considering the implications of the current expected credit losses methodology (CECL) for estimating allowances for credit losses.

The final rule adopts the proposed rule's approach to ALLL. The limited inclusion of ALLL in tier 2 capital was an outgrowth of FHFA's calibration methodology for mortgage exposures under which the base risk weights and risk multipliers are intended to require credit risk capital sufficient to absorb the lifetime

unexpected

losses incurred on mortgage exposures experiencing a shock to house prices similar to that observed during the 2008 financial crisis. The same is also true for non-mortgage exposures. FHFA will continue to monitor the implications of CECL implementation for this issue and could consider adjustments in the future.

C. Subordinated Debt

The proposed rule would have treated some subordinated debt instruments as tier 2 capital. Some commenters supported the proposed rule's approach. One commenter thought that each Enterprise should be financed primarily through term unsecured debt rather than equity because debt can lock in a structured schedule of funding to meet liquidity needs. Other commenters urged FHFA not to treat subordinated debt instruments as a capital element. In the view of some commenters, the historical record has led to a market expectation that subordinated debt is not actually at risk of absorbing losses. A few commenters expressed concern that, unlike equity instruments, an Enterprise would not be able to suspend debt service on subordinated debt.

FHFA has adopted the proposed rule's approach to subordinated debt in the final rule, and certain subordinated debt instruments will continue to be treated as tier 2 capital. To ensure tier 2 capital actually provides loss-absorbing capacity, an Enterprise would be permitted to include an instrument in its tier 2 capital only if FHFA has determined that the Enterprise has made appropriate provision, including in any resolution plan of the Enterprise, to ensure that the instrument would not pose a material impediment to the ability of an Enterprise to issue common stock instruments following any future appointment of FHFA as conservator or receiver under the Safety and Soundness Act.

VII. Capital Requirements

A. Risk-Based Capital Requirements

The proposed rule would have required each Enterprise to maintain the following risk-based capital:

• Total capital not less than 8.0 percent of risk-weighted assets;

• Adjusted total capital not less than 8.0 percent of risk-weighted assets;

• Tier 1 capital not less than 6.0 percent of risk-weighted assets; and

• CET1 capital not less than 4.5 percent of risk-weighted assets.

As discussed in Section III.B.3 of the proposed rule, a lesson of the 2008 financial crisis is that the Enterprises' safety and soundness depends not only on the quantity but also on the quality of their capital. To that end, FHFA proposed to supplement the risk-based capital requirement based on statutorily defined total capital with additional risk-based capital requirements based on the Basel framework's definitions of total capital, tier 1 capital, and CET1 capital.

FHFA noted in the 2018 proposal and the proposed rule that the Enterprises' DTAs, which are included in total capital and core capital by statute, may provide minimal to no loss-absorbing capability during a period of financial stress as recoverability (via taxable income) may become uncertain. The 2018 proposal addressed this issue by establishing a risk-based capital requirement for DTAs. However, the 2018 proposal did not include adjustments for other capital elements that tend to have less loss-absorbing capacity during a financial stress (

e.g.,

ALLL, goodwill, and intangibles), although FHFA did request comment on how best to compensate for the loss-absorbing deficiencies of ALLL and preferred stock within the framework of the 2018 proposal. The 2018 proposal also requested comment on, but did not adjust for, accumulated other comprehensive income (AOCI), leaving open the possibility that an Enterprise could have positive total capital and core capital despite being insolvent under GAAP. By incorporating deductions and other adjustments, the supplemental risk-based capital requirements for adjusted total capital, tier 1 capital, and CET1 capital would have addressed these safety and soundness issues. The supplemental risk-based capital requirements also would have ensured that retained earnings and other high-quality capital would be the predominant form of regulatory capital.

The shift to a terminology of risk-weighted assets in the proposed rule was a change from the 2018 proposal. The addition of three new risk-based capital requirements raised the need for a straightforward mechanism to specify the aggregate regulatory capital required for each. Also, this approach and its associated terminology are well-understood by those familiar with the U.S. banking framework. Expressing the risk-based capital requirement for an exposure as a risk-weight would facilitate transparency and comparability with the U.S. banking framework and other regulatory capital frameworks. Because these concepts are well-understood, this approach also should facilitate market discipline over each Enterprise's risk-taking by its creditors and other counterparties.

As discussed in Section V.A, many commenters expressed concern about the potential impacts of the proposed rule's regulatory capital requirements on borrowing costs, the Enterprises' ability to satisfy their affordable housing goals or other statutory mandates, the incentives for the Enterprises to increase risk taking or engage in CRT, among other concerns. As discussed in Sections VII.B and VIII.B, many commenters contended that the PLBA-adjusted leverage ratio requirement (

i.e.,

the sum of the leverage ratio requirement and the PLBA) likely would often exceed the PCCBA-adjusted risk-based capital requirements.

Commenters also offered related views on the definitions of regulatory capital and the risk weights and other approaches to assigning risk-based capital requirements for the purpose of determining compliance with these required ratios, as discussed in Sections VI and IX.

Specifically, with respect to the required ratios of risk-based capital, commenters offered views on the relative mix of capital instruments contemplated by the risk-based capital requirements. A few commenters argued that tier 1 capital was the best basis for both leverage ratio and risk-based capital requirements. Some commenters urged FHFA to not treat subordinated debt instruments as a capital element because, in their view, the historical record has led to a market expectation that subordinated debt is not actually at risk of absorbing losses.

After considering these comments, FHFA has determined to adopt each of the required risk-based capital ratios as proposed. FHFA continues to believe it is important to supplement the risk-based capital requirement based on statutorily defined total capital with additional risk-based capital requirements based on the Basel framework's definitions of total capital, tier 1 capital, and CET1 capital. The supplemental risk-based capital requirements will reflect customary deductions and other adjustments for assets that might tend to have less loss-absorbing capacity during a financial stress. The tier 1 and CET1 capital requirements will ensure that retained earnings and other high-quality capital are the predominant form of regulatory capital. The use of the U.S. banking framework's required ratios of risk-based capital will foster comparability and enhance market discipline. As discussed in Section IV, the FSOC Secondary Market Statement endorsed the proposed rule's use of the U.S. banking framework's definitions of regulatory capital to prescribe supplemental capital requirements.

While the final rule adopts required ratios of risk-based capital based on the U.S. banking framework, FHFA reiterates that this approach does not result in each Enterprise having the same risk-based capital requirements as U.S. banking organizations. Under the final rule, the credit risk capital requirement for an exposure is determined by multiplying the risk weight assigned to the exposure by 8 percent. The risk weight of an exposure is the key driver of its credit risk capital requirement, and as of June 30, 2020, the risk weight assigned to single-family mortgage exposures under the final rule would have been roughly three-quarters that of similar exposures under the U.S. banking framework. The Enterprises together would have been required under the final rule's risk-based capital requirements to maintain $283 billion in risk-based adjusted total capital as of June 30, 2020 to avoid restrictions on capital distributions and discretionary bonuses. Had they been instead subject to the U.S. banking framework, the Enterprises would have been required to maintain approximately $450 billion, perhaps significantly more, in risk-based total capital (not including market risk and operational risk capital) to avoid similar restrictions.

B. Leverage Ratio Requirements

1. Adjusted Total Assets

The proposed rule's leverage ratio requirements would have been based on an Enterprise's adjusted total assets. Adjusted total assets would have been defined as total assets under GAAP, with adjustments to include many of the off-balance sheet and other exposures that are included in the supplemental leverage ratio requirements of the U.S. banking framework.

Commenters generally supported basing the supplemental leverage ratio requirement on tier 1 capital. Commenters also generally supported basing the leverage ratio requirements on adjusted total assets, although a few preferred total assets as defined under GAAP. Some commenters suggested the leverage ratio should be adjusted to exclude credit risk that had been transferred to third parties through mortgage insurance or CRT. Another commenter advocated including CRT as an element of capital for purposes of calculating the leverage ratio.

FHFA is adopting the definition of adjusted total assets as proposed.

2. Sizing of the Requirements

The primary purpose of the proposed rule's leverage ratio requirements was to provide a credible, non-risk-based backstop to the risk-based capital requirements to safeguard against model risk and measurement error with a simple, transparent, independent measure of risk. From a safety-and-soundness perspective, each type of requirement offsets potential weaknesses of the other, and well-calibrated risk-based capital requirements working with a credible leverage ratio requirement is more effective than either would be in isolation. The proposed rule's leverage ratio requirements would have had the added benefit of dampening some of the procyclicality inherent in the aggregate risk-based capital requirements.

Under the proposed rule, each Enterprise would have been required to maintain capital sufficient to satisfy two leverage ratio requirements:

• Core capital not less than 2.5 percent of adjusted total assets; and

• Tier 1 capital not less than 2.5 percent of adjusted total assets.

As discussed in Section V.A, many commenters expressed concern about the potential impacts of the proposed rule's regulatory capital requirements on borrowing costs, the Enterprises' ability to satisfy their affordable housing goals or other statutory mandates, the incentives for the Enterprises to increase risk taking or engage in CRT, among other concerns. Commenters also offered related views on the definitions of regulatory capital for the purpose of determining compliance with the leverage ratio requirements, as discussed in Sections VI and IX.

Commenters criticized FHFA's method for sizing the proposed rule's two leverage ratio requirements, with many focusing on FHFA's consideration of the Enterprises' historical loss experience. Some commenters urged FHFA to adopt the 2018 proposal's bifurcated alternative that would have prescribed different leverage ratio requirements for trust and non-trust assets. Other commenters described rationales for lower leverage ratio requirements or for not adopting a leverage ratio requirement at all. Some commenters contended that the model risk, measurement error, and related risks mitigated by the leverage ratio requirements were already mitigated by other aspects of the proposed rule. Other commenters indicated that they did not have sufficient information to assess the relationship between the proposed rule's risk-based capital requirements and the leverage ratio requirements and urged FHFA to make additional information available to the public.

Commenters also offered related views on the proposed rule's PLBA-adjusted leverage ratio requirement, and some of those comments have implications for these leverage ratio requirements. The PLBA-adjusted leverage ratio requirement prescribed the tier 1 capital necessary to avoid restrictions on capital distributions and discretionary bonuses. Many of these commenters contended that the PLBA-adjusted leverage ratio requirement likely would often exceed the PCCBA-adjusted risk-based capital requirements. A binding PLBA-adjusted leverage ratio requirement, in the view of many of these commenters, could

reduce the risk sensitivity of the regulatory capital framework, decrease an Enterprise's incentive to engage in CRT, incentivize an Enterprise to increase risk taking, or reduce an Enterprise's ability to offset lower returns on higher risk exposures with higher returns on lower risk exposures. Some commenters, on the other hand, argued that the PLBA-adjusted leverage ratio requirement was inadequate given the Enterprises' historical loss experience and the risk that each Enterprise poses to financial stability. One commenter thought that the PLBA-adjusted leverage ratio requirement should be the primary measure for setting the Enterprises' regulatory capital requirements because the risk-based capital requirements are complex, less transparent, and perhaps subject to manipulation. Some commenters suggested sizing the PLBA-adjusted leverage ratio requirement based on the pre-CRT risk-based capital requirements. Commenters' views specific to the PLBA are further discussed in Section VIII.B.

FHFA has determined to finalize the leverage ratio requirements as proposed. FHFA continues to believe that the proposed rule's calibration methodology for the leverage ratio requirements was fundamentally sound. First, the leverage ratio requirements are generally aligned with the analogous leverage ratio requirements of U.S. banking organizations, after adjusting for the difference in the average risk weight on their exposures.

The monoline nature of the Enterprises' mortgage-focused businesses suggests that the concentration risk of an Enterprise is greater than that of a diversified banking organization with a similar amount of mortgage credit risk, perhaps meriting a leverage ratio requirement greater than 2.5 percent, all else equal. Related to that concentration risk, the leverage ratio requirements are roughly aligned with, if not below, the 4 percent total leverage ratio requirement of the Federal Home Loan Banks, which also have mortgage-focused businesses.

Second, the leverage ratio requirements are broadly consistent with the Enterprises' historical loss experiences. The Enterprises' crisis-era cumulative capital losses peaked at the end of 2011 at $265 billion, approximately 4.8 percent of their adjusted total assets as of December 31, 2007.

Third, the risks and limitations associated with the underlying historical data and models used to calibrate the credit risk capital requirements reinforce the importance of leverage ratio requirements that safeguard against model risk and measurement error.

The U.S. banking framework's leverage ratio requirement requires banking organizations to maintain tier 1 capital no less than 4.0 percent of total assets. Insured depository institutions subsidiaries of certain large U.S. bank holding companies also must maintain tier 1 capital no less than 6.0 percent of total assets to be “well capitalized.” Using data for the 18 bank holding companies subject to the Federal Reserve Board's supervisory stress testing program in 2018, FHFA determined that the average risk weight on the assets of these banks was 61 percent in the fourth quarter of 2018. Under the U.S. banking framework, the Enterprises' mortgage assets generally would be assigned a 50 percent risk weight under the standardized approach. This suggests that the average risk weight on the assets of the Enterprises would have been approximately 81 percent (50 percent divided by 61 percent) of that of these large bank holding companies. That in turn implies a risk-adjusted analogous leverage ratio requirement for the Enterprises of 3.3 percent (81 percent of the 4.0 percent leverage ratio requirement for U.S. banking organizations).

That 4.0 percent leverage ratio requirement should be considered in the context of the safety and soundness benefits of the statutory requirement that each Federal Home Loan Bank advance be fully secured. Related to that, the safety and soundness benefits of that collateral might be furthered by law, as any security interest granted to a Federal Home Loan Bank by a member (or affiliate of a member) is entitled to special protections under the Federal Home Loan Bank Act.

FHFA's view is that substantially all of each Enterprise's valuation allowances on its DTAs should not be deducted from the estimate of peak capital exhaustion. First, a substantial portion of the Enterprises' DTA valuation allowances were on DTAs first recognized under GAAP during the stress period. As such, these valuation allowances had no net impact on adjusted total capital exhaustion during the stress period because the initial GAAP recognition was offset by the subsequent valuation allowance. Second, had the Enterprises been more appropriately capitalized as of December 31, 2007, much of the DTAs that were already recognized under GAAP at the beginning of the stress period would not have been deducted from adjusted total capital, with the effect that the valuation allowance during the stress period would have contributed to adjusted total capital exhaustion. In other words, there was only a relatively small amount of DTAs that (i) was recognized under GAAP as of the beginning of the stress period, (ii) would have already been deducted from adjusted total capital at the time of the beginning of the stress period, and (iii) were subject to a valuation allowance during the stress period. Despite this, given the complexity of the issue, the considerable attention to the issue by interested parties, and the somewhat different impacts of DTA valuation allowances on different measures of regulatory capital, the proposed rule also noted that the sizing of the regulatory capital requirements was consistent with historical loss experiences even if all of the DTA valuation allowances were deducted from the estimate of peak capital exhaustion.

As discussed in Section IV.B.2 of the proposed rule, a disproportionate share of the Enterprises' crisis-era losses arose from certain single-family mortgage exposures that are no longer eligible for acquisition by the Enterprises. The calibration of the credit risk capital requirements attributed a significant portion of the Enterprises' crisis-era losses (approximately $108 billion) to these products. The statistical methods used to allocate losses between borrower-related risk attributes and product-related risk attributes pose significant model risk. It is possible that the calibration understates the credit losses that would be incurred in an economic downturn with national housing price declines of similar magnitude, even assuming a repeat of crisis-era federal support of the economy and the declining interest rate environment.

The FSOC Secondary Market Statement affirmed the sizing of these leverage ratio requirements. FSOC's analysis suggested that “leverage ratio requirements that are materially less than those contemplated by the proposed rule would likely not adequately mitigate the potential stability risk posed by the Enterprises.” FSOC also found that “it is possible that additional capital could be required for the Enterprises to remain viable concerns in the event of a severely adverse stress . . . .”

FHFA has considered commenters' views that the Enterprises' historical loss experience was an inappropriate consideration in calibrating the proposed rule's leverage ratio requirements because it did not reflect the changes to the Enterprises' acquisition criteria since the 2008 financial crisis. Some commenters suggested that the Enterprises' historical loss experiences should be adjusted to remove the Enterprises' valuation allowances on DTAs, the dividends paid to Treasury, and other deductions from capital that were subsequently reversed.

As discussed in the proposed rule, a portion of the crisis-era losses arose from single-family loans that are no longer eligible for acquisition by the Enterprises. However, the sizing of the leverage ratio requirements must guard against potential future relaxation of underwriting standards and regulatory oversight over those underwriting standards. The sizing of leverage ratio requirements also must take into account the model risk posed by the attribution of such losses to specific product characteristics.

The Enterprises' historical loss experience actually might tend to understate the regulatory capital that would be necessary to remain a viable going concern. The Enterprises' crisis-era losses likely were mitigated to at least some extent by the unprecedented support by the federal government of the housing market and the economy and also by the declining interest rate environment of the period. The calibration of the leverage ratio requirements cannot assume a repeat of those loss mitigants. Also, there are some material risks to the Enterprises that are not assigned a risk-based capital requirement—for example, risks relating to uninsured or underinsured losses from flooding, earthquakes, or other natural disasters or radiological or biological hazards. There also is no risk-

based capital requirement for the risks that climate change could pose to property values in some localities.

FHFA also considered commenters' views that the proposed rule's leverage ratio requirements were disproportionate to the capital exhaustion estimated by the Enterprises' annual stress tests. FHFA believes that the Enterprises' stress tests are not an appropriate consideration in calibrating the leverage ratio requirements. The leverage ratio requirements are calibrated to be a credible backstop to the risk-based capital requirements, which are themselves calibrated to absorb the lifetime unexpected losses incurred in a shock similar to that observed during the 2008 financial crisis. The capital exhaustion projected by the Enterprises' past stress tests is different in key respects from the projected lifetime unexpected losses in a severely adverse stress. The Enterprises' stress tests use a nine-quarter loss horizon, whereas much of the projected lifetime unexpected losses would be recognized after the end of that horizon. The Enterprises' stress tests then offset those limited losses with the revenues recognized in the horizon, yielding a projection of capital exhaustion considerably lower than lifetime unexpected losses. Furthermore, the capital exhaustion projected by an Enterprise's stress test results could change significantly across the economic cycle, with projected capital exhaustion following a long period of house price appreciation being considerably less than the projections produced by a stress test at a different point in the economic cycle.

FHFA agrees with commenters that the risk-based capital requirements should, as a general rule, exceed the regulatory capital required under the leverage ratio requirements. At the same time, if the leverage ratio requirements are to be an independently meaningful and credible backstop, there will inevitably be some exceptions in which the leverage ratio requirements exceed the risk-based capital requirements. In FHFA's view, the measurement period of September 30, 2019 was, in fact, consistent with the circumstances under which a credible leverage ratio would be binding, given the exceptional single-family house price appreciation since 2012, the strong credit performance of both single-family and multifamily mortgage exposures, the significant progress by the Enterprises to materially reduce legacy exposure to NPLs and re-performing loans, robust CRT market access enabling substantial risk transfer, and the generally strong condition of key counterparties, such as mortgage insurers.

Some commenters' analysis suggested that the leverage ratio requirements generally would exceed the risk-based capital requirements over most of the economic cycle. That could evidence flaws in FHFA's method for calibrating the leverage ratio requirements, the risk-based capital requirements, or both. After taking into account the views of commenters, and also after considering the FSOC Secondary Market Statement's affirmation of the sizing of the leverage ratio requirements and its suggestion that additional capital could be required, FHFA has adopted adjustments to the risk-based capital requirements that generally should reduce the likelihood that the leverage ratio requirements would exceed the risk-based capital requirements.

C. Enforcement

Under the proposed rule, FHFA stated that it may draw upon several authorities to address potential Enterprise failures to meet the risk-based capital requirements and leverage ratio requirements. An Enterprise failure to meet a capital threshold that is required by regulation may be addressed through enforcement mechanisms for regulatory violations including procedures for cease and desist and consent orders.

FHFA may also use the enforcement tools available under its authority to prescribe and enforce prudential management and operations standards (PMOS).

The prompt corrective action (PCA) framework set out in the Safety and Soundness Act

also provides for enforcement tools when a shortfall occurs in capital requirements that are set forth in the statute, using the statute's prescribed capital concepts.

12 U.S.C. 4581, 12 CFR part 1209.

12 U.S.C. 4513b; 12 CFR part 1236.

12 U.S.C. 4614

et seq.

Commenters generally did not comment on the proposed rule's enforcement framework for the risk-based capital requirements and leverage ratio requirements. After taking into account any implications posed by the changes adopted in the final rule, FHFA is adopting the proposed rule's enforcement framework as proposed.

VIII. Capital Buffers

A. Prescribed Capital Conservation Buffer Amount

Under the proposed rule, to avoid limits on capital distributions and discretionary bonus payments, an Enterprise would have had to maintain regulatory capital that exceeds each of its adjusted total capital, tier 1 capital, and CET1 capital requirements by at least the amount of its PCCBA. The proposed rule's PCCBA would consist of three separate component buffers—a stress capital buffer, a countercyclical capital buffer, and a stability capital buffer.

1. Comments Applicable to Each Component Buffer

Each component buffer of the proposed rule's PCCBA was tailored to achieve its own policy objective and had its own rationale and sizing considerations. Many commenters, however, offered criticisms and other views on the PCCBA as a whole or that could be relevant to one or more of the component buffers. FHFA considered these cross-cutting comments in identifying and assessing potential changes to each of these buffers.

Commenters generally supported the flexibility that the PCCBA afforded the Enterprises in their capital planning and to continue to support the secondary market during a period of financial stress. Many commenters criticized the overall size of the proposed rule's PCCBA, particularly its sizing relative to the risk-based capital requirements. These commenters expressed concern that the PCCBA could adversely affect the availability of mortgage credit or the Enterprises' ability to fulfill their statutory mission. Some commenters recommended eliminating the PCCBA, capping the PCCBA as a share of the underlying risk-based capital requirements, or otherwise reducing the PCCBA. A few commenters thought that the PCCBA added unnecessary complexity. Other commenters offered alternatives to the PCCBA based on the PSPA or reinsurance arrangements. A few commenters thought that the PCCBA should not have to be composed solely of CET1 capital.

Some commenters noted that even with the PCCBA, the Enterprises likely would need support from the federal government to remain viable during a severe economic downturn. Some commenters observed that the PCCBA would mitigate the procyclicality of the aggregate risk-based capital requirements. A few commenters argued that the PCCBA could be replaced with a stress testing program that informs regulatory approvals of capital distributions and bonuses. At least one commenter suggested that FHFA should periodically reassess and solicit public comment on the sizing of the PCCBA or its component buffers.

A recurring comment related to the risk sensitivity of the PCCBA. Each of the PCCBA component buffers would

have been determined as a percent of an Enterprise's adjusted total assets. While some commenters supported this approach, many commenters advocated assessing the PCCBA or one or more of its component buffers as a percent of an Enterprise's risk-weighted assets. Related to this concern, the FSOC Secondary Market Statement found that, “[b]ecause the proposed buffers change based on adjusted total asset size and market share, an Enterprise's capital buffers could decline on a risk-adjusted basis in response to deteriorating Enterprise asset quality or during periods of stress.” While acknowledging that a more risk-sensitive approach could increase the procyclicality of the aggregate risk-based requirements, FSOC “encourage[d] FHFA to consider the relative merits of alternative approaches for more dynamically calibrating the capital buffers.”

The final rule adopts the proposed rule's approach to assess each of the PCCBA component buffers as a specified percent of an Enterprise's adjusted total assets. This is a notable departure from the Basel and U.S. banking frameworks, and it is a departure that does reduce the risk-sensitivity of the framework. FHFA continues to believe that the balance of considerations weighs in favor of this approach. In FHFA's view, a fixed-percent PCCBA is important, among other reasons, to reduce the impact that the PCCBA potentially could have on higher risk exposures, avoid amplifying the secondary effects of any model or similar risks inherent to the calibration of granular risk weights for mortgage exposures, and further mitigate the procyclicality of the aggregate risk-based capital requirements. While the Basel and U.S. banking framework assess the analogous buffers against risk-weighted assets, FHFA's underlying credit risk capital requirements for mortgage exposures are considerably more risk sensitive than the analogous requirements of those frameworks. As discussed in Section V.D, that heightened risk sensitivity engenders more procyclicality than the Basel and U.S. banking frameworks, at least with respect to the aggregate risk-based capital required on mortgage exposures, and that procyclicality is in tension with FHFA's objective to ensure the safety and soundness of each Enterprise and that each Enterprise can fulfill its statutory mission to provide stability and ongoing assistance to the secondary mortgage market across the economic cycle. This tension is heightened by the concentration risk associated with the monoline nature of the Enterprises' mortgage-focused businesses. Notwithstanding the final rule's approach, however, FHFA has taken steps to enhance the risk sensitivity of the stress capital buffer.

2. Stress Capital Buffer

Under the proposed rule, an Enterprise's stress capital buffer would have equaled 0.75 percent of the Enterprise's adjusted total assets. The proposed stress capital buffer was similar in amount and rationale to the 0.75 percent going-concern buffer contemplated by the 2018 proposal. For the reasons elaborated in Section III.B.2 of the proposed rule, and as also contemplated by the Basel and U.S. banking frameworks,

FHFA continues to believe that each Enterprise should be capitalized to remain a viable going concern both during and after a severe economic downturn. While the regulatory capital requirements are sized to ensure an Enterprise would be regarded as a viable going concern by creditors and other counterparties, the stress capital buffer is sized to ensure that the Enterprise would, in ordinary times, maintain regulatory capital that could be drawn down during a financial stress and still maintain regulatory capital sufficient to satisfy the regulatory capital requirements after that stress.

78 FR at 51105 (“In calibrating the revised risk-based capital framework, the BCBS identified those elements of regulatory capital that would be available to absorb unexpected losses on a going-concern basis. The BCBS agreed that an appropriate regulatory minimum level for the risk-based capital requirements should force banking organizations to hold enough loss-absorbing capital to provide market participants a high level of confidence in their viability. The BCBS also determined that a buffer above the minimum risk-based capital requirements would enhance stability, and that such a buffer should be calibrated to allow banking organizations to absorb a severe level of loss, while still remaining above the regulatory minimum requirements.”).

Some commenters thought that the stress capital buffer was appropriately sized at 0.75 percent of an Enterprise's adjusted total assets. Other commenters argued that the stress capital buffer was excessive or should be eliminated. Some commenters suggested that each Enterprise needs to be capitalized only to absorb losses incurred in a severely adverse stress, not to be regarded as a viable going concern by creditors and other counterparties after that stress. One commenter suggested that FHFA consider calibrating a buffer based on an actuarial model for minimum capital, perhaps after considering the Federal Housing Administration's process for determining the minimum economic net worth and soundness of its Mutual Mortgage Insurance Fund.

Many commenters advocated increasing the risk sensitivity of the stress capital buffer. Some of these commenters suggested that the stress capital buffer should be assessed against risk-weighted assets, not adjusted total assets. A few commenters suggested that it was inappropriate to assess the same stress capital buffer on each Enterprise because each has a different risk profile. Some commenters urged FHFA to adopt the proposed rule's alternative that would rely on FHFA's eventual program for supervisory stress tests, although one commenter thought that should be implemented only after FHFA's supervisory stress testing capabilities have been developed.

After considering these comments, FHFA has determined to adopt the proposed rule's alternative approach under which FHFA would periodically re-size the stress capital buffer to the extent that FHFA's eventual program for supervisory stress tests determines that an Enterprise's peak capital exhaustion under a severely adverse stress would exceed 0.75 percent of adjusted total assets. Pending FHFA's implementation of its supervisory stress testing program, or in any year in which FHFA does not assign a greater stress capital buffer, an Enterprise's stress capital buffer would be 0.75 percent of its adjusted total assets.

FHFA is adopting the alternative approach because a dynamically re-sized stress capital buffer would be more risk-sensitive than a fixed-percent stress capital buffer, potentially varying in amount across the economic cycle and also varying in response to changes in the risk of the Enterprise's mortgage exposures. By leveraging a supervisory stress test, this approach could also incorporate nuanced assumptions, such as with respect to the continued availability and pricing of CRT during a period of financial stress. The final rule's approach is also consistent with the FSOC Secondary Market Statement's recommendation that “encourage[d] FHFA to consider the relative merits of alternative approaches for more dynamically calibrating the capital buffers.”

3. Countercyclical Capital Buffer

Under the proposed rule, the countercyclical capital buffer for the Enterprises would have initially been set at 0 percent of adjusted total assets. The proposed rule's countercyclical capital buffer was similar in purpose and rationale to the analogous buffer of the U.S. banking framework.

Many commenters argued that FHFA should not adopt a countercyclical capital buffer. One commenter thought

the value of the countercyclical capital buffer was unclear, as the concept was still theoretical and yet to be modeled and vetted. One commenter argued the countercyclical capital buffer should be more predictable and have a phase-in period and time limitation. Another commenter suggested that FHFA should include a buffer that was triggered when home prices moved a specified amount above the long-term trend. Other commenters suggested that FHFA should clarify the degree of alignment with the U.S. banking framework. Some commenters noted that the U.S. banking regulators have been reluctant to adjust the countercyclical capital buffer. A few commenters advocated adjusting the countercyclical capital buffer based on excessive credit growth in the national housing finance markets. Some commenters were concerned that the method for sizing the countercyclical capital buffer was overly subjective. Several commenters suggested that the countercyclical capital buffer was unnecessary because of stress testing or because the Safety and Soundness Act already authorizes FHFA to temporarily increase regulatory capital requirements.

The final rule adopts the countercyclical capital buffer as proposed. FHFA continues to believe that the countercyclical capital buffer serves an important purpose to the extent that it facilitates FHFA's exercise of its existing authorities to temporarily increase regulatory capital requirements when excess aggregate credit growth poses heightened risk to the safety and soundness of the Enterprises. As discussed in the proposed rule, FHFA does not expect to adjust this buffer as a means to replace or supplement the countercyclical adjustment to the risk-based capital requirements for single-family mortgage exposures. Instead, as under the Basel and U.S. banking frameworks, FHFA would adjust the countercyclical capital buffer taking into account the macro-financial environment in which the Enterprises operate, such that it would be deployed only when excess aggregate credit growth is judged to be associated with a build-up of system-wide risk. This focus on excess aggregate credit growth would have meant that the countercyclical capital buffer likely would be deployed on an infrequent basis and generally only when similar buffers are deployed by the U.S. banking regulators. FHFA also affirms that any adjustment to the countercyclical capital buffer would be made in accordance with applicable law and after appropriate notice to the Enterprises.

4. Stability Capital Buffer

a. Proposed Rule's Approach

As discussed in Section III.B.4 of the proposed rule, the lessons of the 2008 financial crisis have established that the failure of an Enterprise could result in significant harm to the national housing finance markets, as well as the U.S. economy more generally. The Enterprises remain the dominant participants in the housing finance system, owning or guaranteeing 45 percent of residential mortgage debt outstanding as of June 30, 2020. The Enterprises also continue to control critical infrastructure for securitizing and administering $5.8 trillion of single-family and multifamily MBS. Because of the interconnectedness between the Enterprises, distress at one Enterprise could cause distress at the other Enterprise. The Enterprises' imprudent risk-taking and inadequate capitalization led to their near collapse and were among the proximate causes of the 2008 financial crisis. The precipitous financial decline of the Enterprises was also among the most destabilizing events of the 2008 financial crisis, leading to their taxpayer-backed rescue in September 2008. Even today, a perception persists that the Enterprises are “too big to fail.” This perception reduces the incentives of creditors and other counterparties to discipline risk-taking by the Enterprises. This perception also produces competitive distortions to the extent that it enables the Enterprises to fund themselves at a lower cost than other market participants.

Pursuant to the Safety and Soundness Act, as amended by HERA, the FHFA Director's principal duties are, among other duties, to ensure that each Enterprise operates in a safe and sound manner and that the operations and activities of each Enterprise foster liquid, efficient, competitive, and resilient national housing finance markets.

FHFA proposed to incorporate into each Enterprise's PCCBA an Enterprise-specific stability capital buffer that would be tailored to the risk that the Enterprise's default or other financial distress could have on the liquidity, efficiency, competitiveness, or resiliency of the national housing finance markets (housing finance market stability risk).

FHFA cited several reasons for the proposed rule's stability capital buffer.

12 U.S.C. 4513(a)(1).

FHFA's proposed stability capital buffer should not be construed to imply or otherwise suggest that a similar capital surcharge would necessarily be appropriate for the Enterprises' counterparties or other market participants in the housing finance system. Some of these market participants do not pose much, if any, risk to the liquidity, efficiency, competitiveness, or resiliency of national housing finance markets.

First, an Enterprise-specific stability capital buffer would foster liquid, efficient, competitive, and resilient national housing finance markets by reducing the expected impact of the Enterprise's failure on the national housing finance markets. Under a regulatory capital framework in which each Enterprise is subject to the same capital requirements and has the same probability of default, a larger Enterprise's default would nonetheless still pose a greater expected impact due to the greater magnitude of the effects of its default on the national housing finance markets. As a result, a probability of default that might be acceptable for a smaller Enterprise might be unacceptably high for a larger Enterprise. By subjecting a larger Enterprise to a larger capital surcharge, an Enterprise-specific stability capital buffer would reduce the probability of a larger Enterprise's default, aligning the expected impact of its default with that of a smaller Enterprise.

Second, an Enterprise-specific stability capital buffer also would foster liquid, efficient, competitive, and resilient national housing finance markets by creating incentives for each Enterprise to reduce its housing finance market stability risk by curbing its market share and growth in ordinary times, with the possibility of an expanded role during a period of financial stress.

Third, an Enterprise-specific stability capital buffer could offset any funding advantage that an Enterprise might have on account of being perceived as “too big to fail.” That, in turn, would remove the incentive for counterparties to shift risk to the Enterprise, where that incentive not only increases the housing finance market stability risk posed by the Enterprise but also undermines the competitiveness of the national housing finance markets.

Fourth, a larger capital cushion at an Enterprise could afford the Enterprise and FHFA more time to address emerging weaknesses at the Enterprise that could adversely impact the national housing finance markets. In addition to mitigating national housing finance market risk, the additional time afforded by a larger capital cushion could help FHFA ensure that each Enterprise operates in a safe and sound manner.

Finally, with respect to safety and soundness, any perception that an

Enterprise is “too big to fail” leads to moral hazard that undermines market discipline by creditors and other counterparties over the risk taking at an Enterprise. By increasing the regulatory capital at an Enterprise, the stability capital buffer would shift more tail risk back to the Enterprise's shareholders, which should have the added benefit of offsetting any “too big to fail” funding advantage arising from unpriced tail risk. The resulting enhanced market discipline should enhance safety and soundness by increasing each Enterprise's incentives to effectively manage its risks.

FHFA proposed a stability capital buffer based on a market share approach. Under FHFA's market share approach, an Enterprise's stability capital buffer would have depended on an Enterprise's share of total residential mortgage debt outstanding that exceeds a threshold of 5.0 percent market share. The stability capital buffer, expressed as a percent of adjusted total assets, would have increased by 5 basis points for each percentage point of market share exceeding that threshold. FHFA also solicited comment on an alternative approach that would have the Enterprises compute their stability capital buffer in a manner analogous to the U.S. banking approach for determining the surcharge for global systemically important bank holding companies (GSIB).

b. FSOC Secondary Market Statement

The proposed rule's stability capital buffer was a significant departure from the 2018 proposal. That proposal did not contemplate an Enterprise-specific capital surcharge or other buffer that was tailored to the Enterprise's size or importance, any funding advantage that the Enterprise might have on account of being perceived as “too big to fail,” or the risk that the Enterprise's default could pose to the national housing finance markets. The FSOC Secondary Market Statement generally affirmed the merit of this enhancement to the 2018 proposal, and in particular the importance of a separate capital buffer that is expressly intended to mitigate an Enterprise's stability risk.

FSOC found that any distress at the Enterprises that affected their secondary mortgage market activities could pose a risk to financial stability, if risks are not properly mitigated. This important, if perhaps obvious, finding was echoed by the statements made by several of the FSOC principals in connection with FSOC Secondary Market Statement.

This finding also confirmed a premise of the proposed rule's stability capital buffer.

See

Statement of CFTC Chairman Heath P. Tarbert on FSOC's Activities-Based Review of Secondary Mortgage Market Activities, available at:

https://www.cftc.gov/PressRoom/SpeechesTestimony/tarbertstatement092520

(“The good news is that for the first time, the FSOC is formally acknowledging that any distress that affects the secondary market activities of the GSEs could pose a risk to the financial stability of the United States if not properly mitigated.”); Statement by FDIC Chairman Jelena McWilliams on FSOC Activities-Based Review of Secondary Mortgage Market Activities, available at:

https://www.fdic.gov/news/speeches/spsep2520.html

(“Prior to the global financial crisis, Fannie Mae and Freddie Mac were two of the largest, most highly leveraged financial companies in the world. Since being placed into conservatorship in September of 2008, their role in the mortgage market has only grown.”); Statement by the Acting Comptroller of the Currency Regarding FSOC's Consideration of Secondary Mortgage Market Activities, available at:

https://www.occ.gov/news-issuances/news-releases/2020/nr-occ-2020-128.html

(“I support the FSOC's activities-based review of the secondary mortgage market and the thoughtful analysis of the Government Sponsored Enterprises' contribution to financial stability risks as well as of the efforts to address them, . . . .”); CFPB Director Kraninger's Remarks at the Financial Stability Oversight Council Meeting, available at:

https://www.consumerfinance.gov/about-us/newsroom/director-kraningers-remarks-financial-stability-oversight-council-meeting/

(“As the dominant participants in the secondary mortgage market, [the GSEs] provide the liquidity needed by lenders to provide affordable housing options to consumers. Financial stability and access to credit may be imperiled if the GSEs cannot perform this role effectively. It therefore is critical that we take steps to mitigate that risk.”).

FSOC recommended that the regulatory capital requirements should be an important mitigant of the Enterprises' potential stability risk. Specifically, the FSOC Secondary Market Statement stated that “[a] stability capital buffer would mitigate risks to financial stability by reducing the expected impact of an Enterprise's distress on financial markets or other financial market participants and by addressing the potential for decreased market discipline due to an Enterprise's size and importance.” Even more importantly, FSOC also recommended that the capital buffers should be intentionally tailored to that potential stability risk, stating “[t]he capital buffers should be tailored to mitigate the potential risks to financial stability.”

After the FSOC Secondary Market Statement, and given the historical record as to the significant harm an Enterprise's failure could have on the financial system and the economy more generally, it is clear that not only FHFA, but also the other federal regulators, expect that a meaningful stability capital buffer that is specific to each Enterprise's stability risk is a critical feature of the Enterprises' regulatory capital framework.

c. Comments on the Proposed Rule

Many commenters criticized the overall size of each Enterprise's stability capital buffer. Some commenters thought that the stability capital buffer was excessive or even unnecessary given the sizing of the risk-based capital requirements or because of Treasury's commitment under the PSPA. One commenter suggested capping the stability capital buffer at a fixed percent. Other commenters urged eliminating the stability capital buffer because, in their view, it conflicts with the Enterprises' countercyclical mission, while others questioned its applicability because the Enterprises transfer much of the interest rate risk and funding risk on the mortgage exposures that secure their guaranteed MBS. One commenter remarked that the Enterprises' failures in the 2008 financial crisis were due to their underwriting practices, not their market shares.

A few commenters thought that the Enterprises' stability capital buffers were insufficient. Some commenters emphasized the necessity of the stability capital buffer in light of Treasury's rescue of the Enterprises during the 2008 financial crisis. One commenter thought that the stability capital buffer reflects the lessons learned from past crises and the Enterprises' effects on the economy.

Many commenters criticized the proposed rule's market share approach. Some commenters were concerned that the market share approach would be procyclical, increasing an Enterprise's stability capital buffer during a period of financial stress as the Enterprise increased its acquisition share. Some commenters thought that the market share approach might not be well-tailored to an Enterprise's housing finance market stability risk. Many commenters expressed support for either or both of the U.S. banking framework's GSIB surcharge methods, perhaps with adjustments. Other commenters viewed each of the U.S. banking framework's GSIB surcharge methods as inapplicable to the Enterprises due to the different business models.

d. Final Rule's Approach

FHFA is adopting the stability capital buffer as proposed. Consistent with the findings and recommendations of the FSOC Secondary Market Statement, FHFA continues to believe that the stability capital buffer is a critical feature of the Enterprises' regulatory capital framework. An Enterprise-specific stability capital buffer will foster liquid, efficient, competitive, and

resilient national housing finance markets by reducing the expected impact of the Enterprise's failure on the national housing finance markets. It also will create incentives for each Enterprise to reduce its housing finance market stability risk by curbing its market share and growth in ordinary times, preserving room for a larger role during a period of financial stress. An Enterprise-specific stability capital buffer could offset any funding advantage that an Enterprise might have on account of being perceived as “too big to fail,” which would remove the incentive for counterparties to shift risk to the Enterprise and thereby increase the housing finance market stability risk posed by the Enterprise. A larger capital cushion at an Enterprise could afford the Enterprise and FHFA more time to address emerging weaknesses at the Enterprise that could adversely impact the national housing finance markets. By increasing the regulatory capital at an Enterprise, the stability capital buffer also will shift more tail risk back to the Enterprise's shareholders, which should have the added benefit of offsetting any “too big to fail” funding advantage arising from unpriced tail risk and thereby enhance market discipline over excessive risk taking.

As urged by many commenters, FHFA carefully considered the proposed rule's alternative that would have had each Enterprise compute its stability capital buffer in a manner analogous to the U.S. banking approach for determining the GSIB surcharge. However, limits on available data preclude, at least at this time, the adjustments that would be necessary to ensure that a modified U.S. banking framework approach yields an Enterprise-specific stability capital buffer that is reasonably tailored to each Enterprise's housing finance market stability risk.

While the U.S. banking framework's GSIB surcharge methods might appear adaptable to financial institutions other than banking organizations, adopting an analogous approach for calibrating the Enterprises' stability capital buffer is not practicable for at least two reasons. First, the U.S. banking framework determines some of the systemic risk indicators using data specific to banking organizations, which presents data limitations that would need to be overcome. For example, each of the U.S. banking framework's systemic indicators is a relative measure determined by dividing the banking organization's applicable measure by the aggregate measure for a set of large banking organizations. The Enterprises' measures are not included in such aggregate measures, and the GSIB surcharge tiers were calibrated based on the bank-only aggregate measure. Therefore, each Enterprise's measure cannot simply be added to that aggregate measure.

Second, FHFA has not identified reliable alternative systemic risk indicators for the Enterprises. For example, the U.S. banking framework's systemic indicators for substitutability relate to measures of payments activity, assets under custody, and underwritten transactions in debt and equity markets. Using the data inputs specified by the U.S. banking framework, the systemic indicator for substitutability would have produced an exceedingly small measure for each Enterprise, perhaps even zero. That measure is clearly inconsistent with any reasonable understanding of the substitutability of the Enterprises, which currently have a near absence of private-sector market participants that could quickly fill the role of the Enterprises in supporting the secondary market.

Without considerable adjustments that are not practicable with existing data, applying the U.S. banking framework's GSIB surcharge methods to the Enterprises would produce results having little, if any, correspondence with a commonsense understanding of each Enterprise's housing finance market stability risk. Consistent with this conclusion, the U.S. banking framework's GSIB framework does not apply to any nonbank financial companies supervised by the Federal Reserve Board, and instead the Federal Reserve Board contemplates a tailored approach to these financial institutions.

80 FR 49084.

With respect to the market share approach, FHFA continues to believe that the sizing of each Enterprise's stability capital buffer is reasonably tailored to the Enterprise's housing finance market stability risk. As of June 30, 2020, Fannie Mae and Freddie Mac would have had stability capital buffers of, respectively, 1.07 and 0.66 percent of adjusted total assets. Under the 33 percent average risk weight on their exposures at that time, Fannie Mae and Freddie Mac's stability capital buffers would have been 3.3 and 2.0 percent of risk-weighted assets, respectively, which would have been a somewhat less than U.S. GSIBs of similar size. Notably, were the average risk weight on the Enterprises' exposures to increase to 35 percent, Fannie Mae's and Freddie Mac's stability capital buffers would be equivalent to 3.1 and 1.9 percent of risk-weighted assets, respectively, considerably below the capital surcharges of U.S. GSIBs of similar size.

FHFA acknowledges that the market share approach could increase the procyclicality of the aggregate risk-based capital requirements. There is inherently some tension between tailoring the stability capital buffer to an Enterprise's housing finance market stability risk, which generally would increase when it expands its role, and mitigating the procyclicality of the regulatory capital framework. To strike an appropriate balance, the final rule adopts the approach of the proposed rule, which provided that an increase in an Enterprise's stability capital buffer would in effect apply two years after an increase in the Enterprise's market share.

B. Prescribed Leverage Buffer Amount

Under the proposed rule, to avoid limits on capital distributions and discretionary bonus payments, an Enterprise would have been required to maintain tier 1 capital in excess of the amount required under the tier 1 leverage ratio requirement by at least the amount of a PLBA equal to 1.5 percent of the Enterprise's adjusted total assets. The primary purpose of the PLBA was to serve as a non-risk-based supplementary measure that provides a credible backstop to the combined PCCBA and risk-based capital requirements. From a safety-and-soundness perspective, each of the risk-based and leverage ratio requirements offsets potential weaknesses of the other. Taken together, well-calibrated risk-based capital requirements working with a credible leverage ratio requirement are more effective than either would be in isolation. FHFA deemed it important that the buffer-adjusted risk-based and leverage ratio requirements are also closely calibrated to each other so that they have an effective complementary relationship.

Many commenters criticized the sizing of the PLBA. Some of these commenters suggested reducing the PLBA to 0.5 percent or 0.75 percent of adjusted total assets. Some commenters argued the PLBA should be removed entirely. A few commenters did support the proposed rule's PLBA of 1.5 percent of adjusted total assets. Other commenters suggested that payout restrictions should be based only on the PCCBA-adjusted risk-based capital requirements.

As discussed in Section VII.B.2, commenters also offered related views on the proposed rule's PLBA-adjusted leverage ratio requirement, and those comments have some implications for the PLBA itself. The PLBA-adjusted

leverage ratio requirement prescribed the tier 1 capital necessary to avoid restrictions on capital distributions and discretionary bonuses. Many of these commenters contended that the PLBA-adjusted leverage ratio requirement likely would often exceed the PCCBA-adjusted risk-based capital requirements. A binding PLBA-adjusted leverage ratio requirement, in the view of many of these commenters, could reduce the risk sensitivity of the regulatory capital framework, decrease an Enterprise's incentive to engage in CRT, incentivize an Enterprise to increase risk taking, or reduce an Enterprise's ability to offset lower returns on some exposures with higher returns on other exposures. Some commenters, on the other hand, argued that the PLBA-adjusted leverage ratio requirement was inadequate given the Enterprises' historical loss experience and the risk that each Enterprise poses to financial stability. Some commenters suggested sizing the PLBA-adjusted leverage ratio requirement based on the pre-CRT risk-based capital requirements.

After considering these comments, FHFA has determined to adopt the PLBA as proposed. FHFA continues to believe that the proposed rule's calibration methodology for the PLBA was fundamentally sound. The 1.5 percent PLBA is calibrated to ensure that the PCCBA and PLBA have an effective complementary relationship such that each is independently meaningful. The PLBA for Fannie Mae and Freddie Mac would have been, respectively, $53 billion and $38 billion as of September 30, 2019 and would have been $58 billion and $41 billion as of June 30, 2020. For Fannie Mae, the PLBA would have been less than its PCCBA, while for Freddie Mac the reverse would have been true. Moreover, the relative sizing of the PLBA is generally consistent with the relative sizing of similar buffers under the U.S. banking framework. A 1.5 percent PLBA for the Enterprises is 37.5 percent of the 4.0 percent PLBA-adjusted leverage ratio requirement to avoid payout restrictions. The 2.0 percent supplementary leverage ratio requirement of the U.S. banking framework is 40 percent of the 5.0 percent buffer-adjusted leverage ratio requirement to avoid payout restrictions. Finally, FHFA notes that the Federal Home Loan Banks are subject to a 4.0 percent total leverage ratio requirement. While the Federal Home Loan Banks might have greater interest rate risk profiles than the Enterprises, the Federal Home Loan Banks also have the safety and soundness benefits of the statutory requirement that each advance be fully secured, and that security interest has special protection under the Federal Home Loan Bank Act.

FHFA agrees with commenters that the PCCBA-adjusted risk-based capital requirements should, as a general rule, exceed the regulatory capital required under the PLBA-adjusted leverage ratio requirement. Some commenters' analysis suggested that the PLBA-adjusted leverage ratio requirement generally would exceed the PCCBA-adjusted risk-based capital requirements over most of the economic cycle. That could evidence flaws in FHFA's method for calibrating the PLBA-adjusted leverage ratio requirements, the PCCBA-adjusted risk-based capital requirements, or both. After taking into account the views of commenters, and also after considering the FSOC Secondary Market Statement's affirmation of the sizing of the leverage ratio requirements and its suggestion that additional capital could be required, FHFA has adopted adjustments to the risk-based capital requirements that generally should reduce the likelihood that the PLBA-adjusted leverage ratio requirements would exceed the PCCBA-adjusted risk-based capital requirements.

C. Payout Restrictions

Under the proposed rule, an Enterprise would have been subject to limits on its capital distributions and discretionary bonus payments if either its capital conservation buffer was less than its PCCBA or its leverage buffer was less than its PLBA. An Enterprise's maximum payout ratio would have determined the extent to which it is subject to limits on capital distributions and discretionary bonuses. An Enterprise also would not have been permitted to make distributions or discretionary bonus payments during the current calendar quarter if, as of the end of the previous calendar quarter: (i) The eligible retained income of the Enterprise was negative; and (ii) either (A) the capital conservation buffer of the Enterprise was less than its stress capital buffer, or (B) the leverage buffer of the Enterprise was less than its PLBA.

Some commenters supported the payout restrictions as proposed. A few commenters suggested that restrictions on discretionary bonuses would be unfair to employees. Other commenters argued against payout restrictions when an Enterprise is profitable. Some contended that an Enterprise should not be permitted to make any capital distribution at all if it maintained regulatory capital less than its PCCBA-adjusted risk-based capital requirements or its PLBA-adjusted leverage ratio requirements. Other commenters sought clarification as to the circumstances under which an Enterprise would be subject to enforcement action for maintaining regulatory capital less than its PCCBA-adjusted risk-based capital requirements or its PLBA-adjusted leverage ratio requirements. A few commenters suggested changes to the proposed rule's maximum payout ratios.

The final rule adopts the payout restrictions as proposed. FHFA continues to believe that the payout restrictions are appropriately tailored to ensure each Enterprise will maintain safe and sound levels of regulatory capital in the ordinary course while also being able to draw down its regulatory capital during a period of financial stress.

With respect to commenters' suggested clarifications, FHFA continues to expect that each Enterprise generally will seek to avoid any payout restriction by maintaining regulatory capital in excess of its buffer-adjusted risk-based and leverage ratio requirements during ordinary times. FHFA also expects that, consistent with its statutory mission to provide stability and ongoing assistance to the secondary mortgage market across the economic cycle, each Enterprise might draw down its buffers during a period of financial stress. However, it would not be consistent with the safe and sound operation of an Enterprise for the Enterprise to maintain regulatory capital less than its buffer-adjusted requirements in the ordinary course except for some reasonable period after a financial stress, pending the Enterprise's efforts to raise and retain regulatory capital.

Nothing in the final rule limits the authority of FHFA to take action to address unsafe or unsound practices or violations of law, including actions inconsistent with an Enterprise's charter. FHFA could, depending on the facts and circumstances, determine that it is an unsafe or unsound practice, or that it is inconsistent with the Enterprise's statutory mission, for an Enterprise to maintain regulatory capital that is less than its buffer-adjusted requirements during ordinary times. If FHFA were to make that determination, FHFA would have all of its enforcement and other authorities, including its authority to issue a cease-and-desist order, to require the Enterprise to remediate that unsafe or unsound practice—for example, by developing and implementing a plan to raise additional regulatory capital.

IX. Credit Risk Capital: Standardized Approach

A. Single-Family Mortgage Exposures

Much like the proposed rule, the standardized credit risk-weighted assets for each single-family mortgage exposure will be determined under the final rule using grids and risk multipliers that together will assign an exposure-specific risk weight based on the risk characteristics of the single-family mortgage exposure. The base risk weight will be a function of the single-family mortgage exposure's MTMLTV, among other things. The MTMLTV will be subject to a countercyclical adjustment to the extent that national house prices are 5.0 percent greater or less than an inflation-adjusted long-term trend. This base risk weight will then be adjusted based on other risk attributes, including any mortgage insurance or other loan-level credit enhancement and the counterparty strength on that enhancement. Finally, this adjusted risk weight will be subject to a floor.

1. Base Risk Weights

In general, FHFA calibrated the proposed rule's base risk weights and risk multipliers for single-family mortgage exposures to require credit risk capital sufficient to absorb the lifetime unexpected losses incurred on single-family mortgage exposures experiencing a shock to house prices similar to that observed during the 2008 financial crisis. Lifetime unexpected losses are the difference between lifetime credit losses in such conditions (also known as stress losses) and expected losses. The proposed rule would have required an Enterprise to determine a base risk weight for each single-family mortgage exposure using one of four single-family grids (each, a single-family grid) based on performance history:

•

Non-performing loan (NPL):

A single-family mortgage exposure that is 60 days or more past due.

•

Modified re-performing loan (modified RPL):

A single-family mortgage exposure that is not an NPL and has previously been modified or entered a repayment plan.

•

Non-modified re-performing loan (non-modified RPL):

A single-family mortgage exposure that is not an NPL, has not been previously modified or entered a repayment plan, and has been an NPL at any time in the last 48 calendar months.

•

Performing loan:

A single-family mortgage exposure that is not an NPL, a modified RPL, or a non-modified RPL. A non-modified RPL generally would have transitioned to a performing loan after not being an NPL at any time in the prior 48 calendar months.

Many commenters generally supported the proposed rule's base risk weights, which resulted in exposure-specific credit risk capital requirements generally similar to those of the 2018 proposal, subject to some simplifications and refinements. Several commenters suggested that FHFA should establish a process for reviewing the base risk weights every few years that includes soliciting public input from interested parties.

FHFA also received comments on the framework for calibrating the proposed rule's base risk weights. Some commenters advocated greater transparency into, and justification of, the calibration framework, particularly the increase in base risk weights relative to the 2018 proposal.

One commenter argued that the house price shock and recovery assumptions underlying the calibration framework were inappropriate given the changes in the national housing finance markets since the 2008 financial crisis, including the enhanced consumer protections and greater capital requirements for mortgage insurers and other market participants. Another commenter recommended a separate capital requirement of 50 basis points of adjusted total assets to mitigate the model risk associated with the calibration framework. Several commenters argued that FHFA should acknowledge that accounting losses comprised a substantial portion of the Enterprises' crisis-era loss experience. Some commenters suggested that the credit risk capital requirements were motivated by an intent to drive changes to the structure of the national housing finance markets. Commenters also suggested that the final rule should permit flexibility to allow the Enterprises to adapt to an evolving market and for their partners to innovate.

FHFA previously published a white paper on its calibration framework available at

https://www.fhfa.gov/PolicyProgramsResearch/Research/Pages/FHFA-Mortgage-Analytics-Platform-Whitepaper-V2.aspx.

Commenters suggested that the base risk weights for high MTMLTV loans were excessive and could adversely impact lending by state housing finance agencies. Some commenters argued that the base risk weight should be assigned based on original loan-to-value (OLTV) instead of MTMLTV for the first few years because, among other things, the change would reduce procyclicality. One commenter recommended splitting each single-family grid's band for single-family mortgage exposures with MTMLTV between 30 percent and 60 percent into three equally sized bands to increase the risk sensitivity of the base risk weights. Some commenters argued that the base risk weights for some higher MTMLTV single-family mortgage exposures were excessive. One commenter suggested using a national house price index instead of state-level house prices to calculate the MTMLTV for a single-family mortgage exposure.

A few commenters advocated the use of a borrower's original credit score instead of the refreshed credit score because the refreshed credit score could materially impact a borrower's access to credit and might increase procyclicality.

Commenters urged changes to the proposed rule's treatment of modified RPLs and non-modified RPLs. Some commenters suggested permitting a modified RPL to transition to a performing loan after several years of performance because these modified RPLs perform much like single-family mortgage exposures that had never been delinquent. One commenter proposed that single-family mortgage exposures subject to repayment plans and other loss mitigation programs that do not modify the required payments should be treated as non-modified RPLs so as to not discourage use of these plans and programs.

Many commenters advocated changes for single-family mortgage exposures in COVID-19-related forbearance. Commenters argued that these exposures (and other single-family mortgage exposures in similar disaster-related forbearance programs) should not be treated as NPLs or modified RPLs for purposes of assigning a basis risk weight and instead generally should be assigned a lower base risk weight. Commenters also suggested that these exposures should be assigned a different performance classification only after the forbearance period ends.

After considering these comments, FHFA has adopted the following changes to the proposed rule's base risk weights.

• The final rule adopts a revised definition of modified RPL that provides that a modified RPL will become a performing loan after 60 calendar months of performance. This treatment is similar to the treatment afforded to non-modified RPLs. In its analysis supporting the proposed rule, FHFA found a material difference in loan performance for modified RPLs that re-performed for four years and performing loans that were never modified. However, FHFA also found this difference began to diminish after five years of re-performance. In light of the commenters' recommendation and upon

re-examining the available information, the final rule allows for modified RPLs that perform for five years to be reclassified as performing loans.

• Each single-family grid's band for single-family mortgage exposures with an MTMLTV between 30 percent and 60 percent has been divided into three separate, equally-sized bands. This change will moderately enhance the regulatory capital framework's risk sensitivity without materially increasing its complexity.

• A single-family mortgage exposure in a repayment plan will be treated as a non-modified RPL instead of a modified RPL. This change will avoid discouraging the use of these programs, which are important means of mitigating the Enterprises' losses. If after the forbearance the borrower elects a payment deferral instead of a reinstatement or a repayment plan, the single-family mortgage exposure will still be treated as a modified RPL.

The final rule also implements a tailored approach to any single-family mortgage exposure that is in a forbearance pursuant to the CARES Act or a forbearance program for COVID-19-impacted borrowers. During the forbearance (and pending negotiations or other steps reasonably expected to result in a modification), the base risk weight for an NPL will be equal to the product of 0.45 and the base risk weight that would otherwise be assigned to the NPL. After the forbearance, any period of time during which the single-family mortgage exposure was past due will be disregarded for the purpose of assigning a risk weight if the entire amount past due was repaid upon the termination of the forbearance. In effect, a single-family mortgage exposure will, after a reinstatement, return to the classification it had before the COVID-19-related forbearance. As discussed above, because a repayment plan will not be treated as a modification, a single-family mortgage exposure that is subject to a repayment plan after a COVID-19-related forbearance will be treated as a non-modified RPL instead of a modified RPL.

With respect to commenters' concerns about the perceived increase in the base risk weights, FHFA notes that, while the proposed rule's base risk weights generally were greater than the base risk weights implicit in the single-family grids of the 2018 proposal, that change generally would not result in greater aggregate credit risk capital requirements after taking into account offsetting changes to the risk multipliers. The proposed rule eliminated the 2018 proposal's risk multipliers for number of borrowers and loan size, and reallocated the associated unexpected losses across the base risk weights. The practical effect of this change was that the base risk weights in the single-family grids are greater than they otherwise would have been if the two risk multipliers had not been eliminated.

2. Countercyclical Adjustment

Under the proposed rule, the MTMLTV used to assign a base risk weight to a single-family mortgage exposure would have been subject to a countercyclical adjustment that an Enterprise would have been required to make when national house prices increased or decreased by more than 5.0 percent from an estimated inflation-adjusted long-term trend (MTMLTV adjustment). The proposed rule's MTMLTV adjustment would have been based on FHFA's U.S. all-transactions FHFA HPI.

Several commenters generally supported the MTMLTV adjustment as an effective means of mitigating the procyclicality of the aggregate risk-based capital requirements. One commenter suggested that the MTMLTV adjustment was duplicative of the countercyclical capital buffer and therefore unnecessary. A commenter argued that, while the MTMLTV adjustment functioned effectively when applied to historical datasets, it might not function as expected in the future and could, under certain circumstances, reduce the Enterprises' incentives to acquire high OLTV single-family mortgage exposures. Other commenters thought that the procyclicality could be addressed by increasing reliance on OLTV and credit scores at origination instead of MTMLTV and refreshed credit scores. Some commenters thought that CRT could play a role in mitigating procyclicality.

Many commenters recommended changes to the MTMLTV adjustment. Some commenters suggested that the MTMLTV adjustment should be regionalized by using home prices in each state or metropolitan statistical area to avoid distorting regional lending based on national house price trends. Another commenter advocated using a purchase-only HPI instead of the all-transactions FHFA HPI. That commenter also advocated using data from 1975 to 2001 to specify the long-term trend. Commenters also proposed periodically reevaluating the MTMLTV adjustment.

Some commenters focused on the 5.0 percent collar. A few commenters advocated not using a collar and instead applying the MTMLTV adjustment regardless of the extent to which national house prices had departed from the long-term trend. Other commenters suggested a wider collar or an asymmetrical collar that set thresholds at different levels of deviation above and below the long-term trend. One commenter suggested applying the MTMTLTV adjustment to only half the incremental house price appreciation above the collar.

After considering the views of commenters, FHFA has determined to adopt the proposed rule's MTMLTV adjustment with two changes. First, FHFA agrees with commenters that an expanded-data HPI, for example the recently published national, not-seasonally adjusted, expanded-data FHFA House Price Index®, provides a better basis for identifying departures from the inflation-adjusted long-term national house price trends. The expanded-data FHFA HPI excludes the potential valuation biases associated with refinancing transactions, which generally assign a house valuation through an appraisal. The expanded-data FHFA HPI also more accurately reflects market activity by supplementing the Enterprises' acquisitions with data from Federal Housing Administration mortgages and real property records. The additional data provide sufficient sample sizes to ensure robust estimation of the HPI back to 1975.

To estimate the long-term trend using the expanded-data FHFA HPI, FHFA employed the same trough-to-trough methodology used in the proposed rule. The parameters of the long-term trend are estimated using a linear regression on the natural logarithm of real HPI from the trough in the first quarter of 1976 to the trough in t

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.