2020-12345. Federal Home Loan Bank Housing Goals Amendments

AGENCY:

Federal Housing Finance Agency.

ACTION:

Final rule.

SUMMARY:

The Federal Housing Finance Agency (FHFA) is amending the existing Federal Home Loan Bank (Bank) Housing Goals regulation. The final rule replaces the existing regulation's four separate retrospective housing goals with a single prospective mortgage purchase housing goal with a target level of 20 percent. The final rule also establishes a separate small member participation housing goal with a target level of 50 percent. The final rule provides that a Bank may request FHFA approval of alternative target levels for either or both of the goals. The final rule also establishes that housing goals apply to each Bank that acquires any Acquired Member Assets (AMA) mortgages during a year, eliminating the existing $2.5 billion volume threshold that previously triggered the application of housing goals for each Bank. Enforcement of the final rule will phase in over three years.

DATES:

The final rule is effective August 24, 2020. Written requests from Banks proposing alternative target levels are due by September 15, 2020. The enforcement phase-in period applies to calendar years 2021, 2022, and 2023.

FOR FURTHER INFORMATION CONTACT:

Ted Wartell, Manager, Housing & Community Investment, [phone], [e-mail]; Ethan Handelman, Senior Policy Analyst, Housing and Community Investment, [phone], [e-mail]; or Marshall Adam Pecsek, Assistant General Counsel, Office of General Counsel, [phone], [e-mail]. These are not toll-free numbers. The telephone number for the Telecommunications Device for the Deaf is [phone]. The mailing address for each contact is: Federal Housing Finance Agency, 400 7th Street SW, Washington, DC 20219.

SUPPLEMENTARY INFORMATION:

I. Issuing This Rule During COVID-19 National Emergency

The COVID-19 national emergency is creating unprecedented economic disruption to the economy, including the mortgage market. FHFA recognizes the substantial public and private sector efforts in responding to the pandemic and considered the ongoing uncertainty and regulatory burden created by the existing, outdated Bank housing goals regulation. FHFA has concluded that issuing this final rule is an important priority to provide greater certainty for the Banks and other market participants. In addition, features of the final rule such as the three-year enforcement phase-in (discussed in Section IV.D.) and the option to propose alternative target levels (discussed in Section VIII.) will make the housing goals more adaptable during disruptions such as the COVID-19 national emergency.

II. Background

A. The Federal Home Loan Bank System

The eleven Federal Home Loan Banks are wholesale financial institutions organized under the Federal Home Loan Bank Act (Bank Act).^[1] The Banks are cooperatives; only members of a Bank may purchase the capital stock of a Bank, and only members or certain eligible housing associates (nonmember borrowers such as state housing finance agencies) may obtain access to secured loans, known as advances, or other products provided by a Bank.^[2] Any eligible institution (generally, a federally insured depository institution or state-regulated insurance company) may become a member of a Bank if it satisfies certain criteria and purchases a specified amount of the Bank's capital stock.^[3]

As government-sponsored enterprises, the Banks have certain privileges under federal law, which allow them to borrow funds at spreads over the rates on U.S. Treasury securities of comparable maturity that are narrower than those available to corporate borrowers generally. The Banks pass along their funding advantage to their members and housing associates—and ultimately to consumers—by providing advances and other financial services at rates that would not otherwise be available to these institutions.^[4] Among those financial services are the Banks' Acquired Member Assets (AMA) programs, under which the Banks provide financing for members' housing finance activities by purchasing mortgage loans.

B. AMA Programs

FHFA's AMA regulation authorizes the Banks to acquire eligible mortgages from their members and housing associates as a means of advancing their housing finance mission, and it prescribes the parameters within which the Banks may do so.^[5] Through the acquisition of AMA mortgages, the Banks provide a source of liquidity to their members and housing associates to further mission-related lending.

FHFA's AMA regulation authorizes each Bank, at its discretion, to purchase assets that qualify as AMA subject to the requirements of the AMA regulation. Currently, each of the Banks except the Atlanta Bank offers an AMA program for the purchase of single-family mortgages, though the size of their programs varies. As of December 31, 2018, the Banks' total outstanding AMA mortgages were $63 billion,^[6] representing less than 6 percent of their total assets. In contrast, the eleven Banks' total outstanding advances, their primary business line, represented 68 percent of total assets. Outstanding mortgages relative to total assets at the Banks offering AMA programs ranged from a high of 18 percent and 17 percent at the Indianapolis and Topeka Banks, respectively, to a low of 2 percent or less at the New York and Atlanta Banks. Further, as a point of comparison, in 2018, the mortgage purchases of the Federal National Mortgage Association (Fannie Mae) and the Federal Home Loan Mortgage Corporation (Freddie Mac) (collectively, the Enterprises), represented 63 percent of the secondary mortgage market comprising Fannie Mae, Freddie Mac, the Government National Mortgage Association (Ginnie Mae), and the Banks. The Banks' combined mortgage purchases represented less than 1 percent of that secondary market.

The AMA programs that the Banks currently offer are the Mortgage Purchase Program (MPP) and the Mortgage Partnership Finance (MPF) program. The Banks generally acquire 15- to 30-year conventional, conforming fixed-rate mortgage loans secured by 1- to 4-unit properties. The Banks also acquire single-family mortgage loans guaranteed or insured by a department or agency of the federal government (i.e., non-conventional mortgages).

C. Overview of the Existing Bank Housing Goals Regulation

The existing Bank housing goals regulation has been in effect since January 2011.^[7] The regulation implements section 10C(a) of the Bank Act, which requires the Director of FHFA to “establish housing goals with respect to the purchase of mortgages, if any, by the [Banks].” ^[8] Section 10C(b) requires that the Bank housing goals be “consistent with” the housing goals established by FHFA for the Enterprises under sections 1331 through 1334 of the Federal Housing Enterprises Financial Safety and Soundness Act of 1992 (Safety and Soundness Act), taking into consideration “the unique mission and ownership structure of the [Banks].” ^[9]

The regulation establishes three single-family owner-occupied purchase money mortgage goals and one single-family refinancing mortgage goal applicable to the Banks' purchases under their AMA programs. The goals for purchase money mortgages separately measure the percentage of purchase money mortgages acquired by a Bank that serve low-income families, families in low-income areas, and very low-income families.^[10] The goal for refinancing mortgages measures the percentage of refinancing mortgages acquired by a Bank that serve low-income families. The target levels of the housing goals are established retrospectively by FHFA in the year following the year of the Banks' AMA mortgage purchases, using Home Mortgage Disclosure Act (HMDA) data to calculate the percentage of single-family mortgage originations in the Bank's district that qualify for each of the housing goals.

The existing regulation provides that a Bank is subject to the housing goals if its AMA mortgage purchases in a given year exceed a volume threshold of $2.5 billion in unpaid principal balance. Each year, FHFA determines whether any Banks have exceeded the volume threshold. For each Bank that has exceeded the volume threshold, FHFA determines the Bank's performance under the housing goals by calculating the percentage share of the Bank's AMA mortgage purchases that qualify for each housing goal. A Bank meets a housing goal if its performance is equal to or greater than the target level of the housing goal established by FHFA based on HMDA data for that year.

III. Proposed Rule and Comments

A. Proposed Rule

On November 2, 2018, FHFA published in the Federal Register a proposed rule to amend the existing Bank housing goals regulation.^[11] The 90-day public comment period on the proposed rule ended January 31, 2019. FHFA proposed replacing the existing regulation's four single-family housing goals with a new, combined prospective single-family housing goal that would measure the affordable share of AMA mortgage purchases by each Bank. FHFA also proposed establishing a new, separate prospective housing goal that would measure the extent to which small members and housing associates sell loans to the Banks under the AMA programs. The prospective housing goals set forth in the proposed rule would provide certainty for the Banks by informing them of the housing goal target levels in advance and would provide clarity and flexibility for the Banks by consolidating multiple goals.

B. Overview of Comments on Proposed Rule

FHFA received 23 comment letters in response to the proposed rule. This number includes a joint comment letter from the presidents of the eleven Banks. One comment letter was unrelated to the proposed rule. FHFA also held a number of meetings, including webinars, with Bank representatives and other stakeholders to describe the contents of the proposed rule, discuss issues raised by the proposed rule, and obtain clarifications of specific comments made in the letters.

Overall, commenters supported FHFA's proposed rule because it would make the housing goals more effective at encouraging affordable housing and easier to implement for the Banks. The comments on particular provisions of the proposed housing goals rule are discussed in more detail starting in Section VI. below.

IV. Summary of Final Rule

A. Elimination of Volume Threshold

The final rule eliminates the current $2.5 billion volume threshold, such that all Banks are subject to the new housing goals, regardless of their AMA mortgage purchase volume, consistent with the proposed rule.

B. New Prospective Housing Goal for Mortgage Purchases

As proposed, the final rule establishes a new single combined prospective mortgage purchase housing goal in advance that replaces the four existing retrospective housing goals for home-purchase mortgages for low-income families, home-purchase mortgages for low-income areas, home-purchase mortgages for very low-income families, and refinancing mortgages for low-income families.^[12] The new housing goal includes each of these four categories, but does not include separate target levels for each category. The final rule establishes one overall target level for the new housing goal at 20 percent of the number of a Bank's total AMA mortgage purchases. The final rule also permits a Bank to request FHFA approval of an alternative target level for the goal. The final rule further provides that no more than 25 percent of the mortgages counted toward the housing goal may be mortgages for families with incomes above 80 percent of area median income (AMI).

C. New Prospective Small Member Participation Housing Goal

As proposed, the final rule establishes a new prospective housing goal that measures the extent of participation by small members and housing associates in a Bank's AMA program. Specifically, the final rule requires that at least 50 percent of a Bank's total AMA users must be community-based AMA users. “Community-based AMA user” is a new term defined to mean any user whose average total assets over the three-year period culminating in the year preceding the one being measured are no greater than the applicable community-based AMA user asset cap, which itself is a new term meaning $1,224,000,000, subject to annual adjustments by FHFA, beginning in 2021, to reflect any percentage increase in the preceding year's Consumer Price Index (CPI) for all urban consumers, as published by the U.S. Department of Labor. The final rule retains the proposed asset cap but incorporates it directly into the housing goals regulation and delinks the housing goals regulation from the Bank members regulation. Accordingly, these new terms replace “community financial institution or CFI” and “CFI asset cap,” which were proposed as new terms but neither of which appears in the final rule. “AMA user” is a newly defined term in § 1281.1 that means a participating financial institution from which the Bank purchased at least one AMA mortgage during the year for which the housing goals are being measured. If a Bank is unable to meet the 50 percent target level, the Bank may still meet the small member participation housing goal if the percentage of its total AMA users that are community-based AMA users is at least 3 percentage points greater than the percentage in the preceding year. The final rule permits a Bank to request FHFA approval of an alternative target level for this goal.

D. Phase-In Period for Enforcement of the New Housing Goals

Consistent with the proposed rule, the final rule establishes a three-year phase-in period for enforcement of the two new housing goals, starting with the effective date of the final rule and ending on December 31, 2023. During the phase-in period, FHFA will monitor and report the Banks' housing goals performance but will not impose a housing plan remedy in the event that a Bank fails to meet either or both of the housing goal target levels.

E. Other Changes

Consistent with the proposed rule, the final rule revises and simplifies the criteria and requirements under which mortgages are either included or excluded from FHFA's measurement of a Bank's performance under the housing goals. The final rule also revises the housing goals reporting requirements to reflect the new structure of the Bank housing goals.

V. Elimination of the Volume Threshold

The final rule eliminates the $2.5 billion volume threshold in the existing housing goals regulation. This means that all Banks acquiring any AMA mortgages are subject to the housing goals regardless of their AMA mortgage purchase volume. The elimination of the volume threshold is consistent with the proposed rule and addresses the fact that the threshold has operated as an upper limit on Bank AMA programs.

All eleven commenters who addressed the volume threshold supported its proposed elimination. Some commenters regarded the volume threshold as an artificial cap on liquidity. Others characterized it as a way for the Banks to avoid the goals entirely by keeping mortgage purchases below the threshold. No commenter opposed the proposed elimination of the threshold.

Thirty-six state or local advocacy and community development organizations, in a joint comment letter, supported the proposed elimination of the volume threshold, stating that it had served to effectively exempt the vast majority of Banks from the housing goals. A nonprofit consumer advocacy group commented that FHFA had made a strong argument for elimination of the volume threshold.

A trade association for credit unions, in support of the proposed elimination of the volume threshold, commented that the threshold operated as an upper limit on Bank AMA programs. A lender trade association asserted that as a result of removing the volume threshold, the Banks would be expected to either increase their total purchases, including affordable housing loans, or shift more of their total mortgage purchases towards affordable housing mortgages in order to comply with the housing goal, and characterized either result as desirable. This commenter urged close monitoring of the Banks to ensure they do not cease their AMA mortgage purchases to avoid noncompliance with the housing goal if the volume threshold were eliminated, although the commenter described that as an unlikely outcome.

A U.S. Senator stated that the volume threshold is inconsistent with the Banks' mission responsibility to meet the affordable housing needs in their districts. A trade association representing state housing finance agencies commented that the volume threshold has effectively exempted most Banks from the housing goals and served as a de facto upper limit on Banks' AMA purchases, resulting in less liquidity for affordable mortgage lending. Similarly, a nationwide nonprofit community development organization asserted that the volume threshold creates a “perverse incentive” for the Banks to limit their mortgage purchases, also claiming that removal of the threshold will encourage more AMA participation. An individual credit union commented that it supported all changes FHFA proposed.

In their joint comment letter, the Banks requested clarity on FHFA's intent concerning the final rule's effect on the Banks' overall AMA activity, noting that the proposed rule would allow a Bank to reduce its overall AMA activity as a means of increasing the percentage of its AMA purchases that qualify for the goal. FHFA notes that the final rule does not prohibit a Bank from managing its volume of AMA activity as a means of meeting the housing goal, nor does the existing regulation. FHFA recently issued an Advisory Bulletin ^[13] addressing FHFA's expectations for Banks' management of the risks of their AMA programs. FHFA also expects that the Banks will comply with the final rule by prioritizing purchases of AMA mortgages for low- and very low-income families, and for families in low-income areas.

As discussed in the proposed rule, the volume threshold was originally adopted to allow smaller Bank AMA programs to operate without meeting housing goal target levels, particularly programs focused on providing liquidity for smaller Bank members. Over time, however, the volume threshold has instead operated as an upper limit on Bank AMA programs. Banks below the volume threshold in effect avoid the housing goals, while Banks above the threshold face application of housing goals that AMA programs were not designed to, and typically did not, meet.

Housing goals will better serve their public purpose if they are flexible enough to be meaningful and achievable for a variety of Bank AMA programs. The final rule creates a mechanism for the housing goals to apply to all Banks while allowing flexibility to address unique situations for particular Banks if necessary. The new process for setting the target levels of the housing goals should help address issues faced by smaller AMA programs by allowing the Banks to propose meaningful and achievable alternative target levels based on the nature of the AMA program at each Bank.

Accordingly, as proposed, the final rule eliminates the existing $2.5 billion volume threshold, making the housing goals applicable to all Banks regardless of their AMA purchase volume. Banks may propose alternative target levels to FHFA for approval, as discussed below in Section VIII.

VI. Prospective Mortgage Purchase Housing Goal

A. Single Combined Goal With Multiple Categories for Eligibility

Section 1281.11(a) of the final rule replaces the four existing separate retrospective mortgage purchase housing goals with a single prospective mortgage purchase housing goal that includes all single-family, first lien AMA mortgages purchased by a Bank, with limited exceptions. Purchase money or refinancing mortgages meeting the income or geographic eligibility requirements for any of the existing four housing goals would count toward performance under this new combined goal. This includes mortgages for low- or very low-income borrowers and mortgages for borrowers living in low-income areas.

Refinancing mortgages for low-income borrowers, which previously counted only for the separate low-income families refinancing goal, are now included in the new combined goal. By combining the different categories into a single, combined goal, the final rule also counts refinancing mortgages for families of any income level who reside in low-income areas, subject to the cap on counting loans to higher-income borrowers in low-income areas, discussed in Section VI.D. below.

FHFA's proposal to consolidate the four existing housing goals into a single combined goal received support from a number of commenters. A bank trade association, a credit union trade association, and a credit union generally emphasized the additional flexibility the proposal would provide the Banks in serving specific needs within their districts. A lenders trade association commented that given the relatively small historical sizes of AMA programs, there is little value in segregating AMA mortgage purchases into more granular categories, and that a single, consolidated goal would reduce the Banks' compliance burden.

An organization representing state housing finance agencies also favored consolidating the goals, noting that, unlike the Enterprises' mortgage purchases, the Banks' AMA programs are limited in scope and resources, making it harder for the Banks to develop products or programs designed to serve each separate goal. It also commented that, unlike the Enterprises, the Banks generally operate with a regional focus, and that each Bank district has different needs.

Thirty-six state or local advocacy and community development organizations, in a joint comment letter, opposed combining the four existing housing goals into one goal. These commenters stated that the four housing goals align with the statutory requirement that the Banks' housing goals be consistent with the Enterprises' housing goals. The commenters also asserted that the four housing goals target different populations and neighborhoods, each with unique needs. The commenters expressed concern that the Banks could satisfy the new single combined housing goal by purchasing mortgages targeting certain easier-to-serve populations while providing little support for other populations, such as very low-income borrowers, that may prove more difficult to serve.

A consumer advocacy organization shared these concerns about service to very low-income borrowers under the proposal. The commenter also observed that the proposed rule preamble did not provide analysis of the past or trending distribution of purchase money and refinance loans in AMA mortgage acquisitions. It expressed concern about potential underrepresentation of refinance loans in the Banks' AMA programs if the final rule adopted the proposed combined goal for purchase money and refinance loans. The commenter stated that the combined goal is likely to have less impact in the current and immediate past rate environments where refinances were at very low levels, but that this may not always be the case. This commenter did not object to the proposal but expressed concern that a consolidated housing goal might reduce access to credit for some categories of low-income borrowers. The commenter urged FHFA to monitor Bank housing goals performance closely under the final rule and be prepared to return to more granular housing goals if AMA mortgage purchases skew away from serving very low-income borrowers.

After considering the comments, FHFA has concluded that the proposed new single, combined mortgage purchase housing goal is appropriate for the Banks' AMA programs. The Banks' AMA programs are voluntary and serve a small fraction of the secondary market. The new combined goal draws from the same categories used in the Enterprise and Bank housing goals regulations,^[14] while giving the Banks additional flexibility to meet the needs of each of their districts. FHFA also recognizes that, due to the voluntary nature of the Banks' AMA programs, requiring AMA programs to serve all the categories of borrowers at precise target levels, as under the existing Bank housing goals, may be more likely to discourage Bank participation in AMA programs than to increase service to underserved borrowers. In addition, concerns that combining the four existing housing goals categories would result in certain populations being neglected are at least partially addressed by the final rule's cap on counting loans to higher-income borrowers in low-income areas, discussed in Section VI.D. below.

Accordingly, for the foregoing reasons, and consistent with the proposed rule, the final rule establishes a single, combined mortgage purchase housing goal for low- and very low-income borrowers and borrowers in low-income areas and counts purchase money and refinance mortgages identically under the goal. Although the final rule does not establish specific target levels for each of the four categories under the new combined goal, FHFA will continue to require the Banks to report data allowing FHFA to identify and monitor AMA activity under each of the four categories.

B. Twenty (20) Percent Target Level for the New Mortgage Purchase Housing Goal

Section 1281.11 of the final rule sets the target level for the new combined mortgage purchase housing goal at 20 percent of the total number of AMA mortgage loans purchased by a Bank, consistent with the proposed rule. This means that a Bank meets the housing goal if 20 percent or more of the AMA mortgage loans it purchases serve some combination of low-income households, very low-income households, or households in low-income areas, subject to a 25 percent cap on loans purchased by the Bank that serve higher-income borrowers in low-income areas. Consistent with the approach in the existing regulation and the proposed rule, the mortgage purchase goal is denominated as a percentage of number of loans, not unpaid principal balance. Also consistent with the proposed rule, the final rule provides the Banks with the option to propose alternative Bank district-specific target levels, discussed further in Section VII. below.

1. Comments Received

Commenters provided mixed responses to the proposed 20 percent target level for the new mortgage purchase housing goal. A U.S. Senator supported the proposed target level, noting that many of the targeted borrowers have difficulty obtaining financing and that serving them is consistent with the mission of the Banks. A credit union also supported the proposed target level, stating that it would not appear to add any risk to the Banks or do anything other than improve on what already works well. A lenders trade association commented that the proposed target level would strike an appropriate balance between rigor and feasibility, but also recommended that, should any Banks discontinue their AMA programs as a result of the target level, FHFA be prepared to reevaluate the target level or work with those Banks on developing alternative Bank-specific target levels.

Several advocacy organizations recommended a different target level for the new mortgage purchase housing goal. Thirty-six state or local advocacy and community development organizations, in a joint comment letter, opposed the imposition of the same percentage target level on all Banks, instead recommending Bank district-specific target levels at levels unspecified in the comment letter. A nonprofit community revitalization organization regarded the proposed target level as too low, stating that eight of the eleven Banks already met this target level each year since 2011. A nonprofit consumer advocacy organization expressed concern that the proposed target level could encourage at least six Banks to reduce their AMA mortgage purchase efforts because their historical performance exceeded the proposed target level. The commenter favored setting the target level between 25 and 30 percent, which it noted several Banks had exceeded in the past and which it believed the Banks could meet in the future. A trade association representing state housing finance agencies similarly observed that several Banks have exceeded the proposed target level in recent years, suggesting that it could prompt some Banks to lower their purchases of mortgages for low- and moderate-income borrowers.

A credit union trade association took the opposite view, favoring an initial target level that all Banks have already met. The commenter stated that the Banks could then focus on building their AMA purchase volumes overall without fear of missing the goal. The commenter stated that this would help in assessing the impact of removing the $2.5 billion volume threshold, and that FHFA could then evaluate the data and set a new incremental target level at the end of the initial three-year period.

In their joint comment letter, the Banks stated that while a 20 percent target level might be achievable in favorable economic times, it might also be hard to sustain over the long term. The Banks indicated that such a level may be difficult to achieve during economic downturns, which might compromise Banks' prudent management of their AMA programs. A community bankers trade association similarly commented that a Bank's ability to reach a 20 percent target level likely depends on the macroeconomic climate at the time.

Several commenters offered alternative formulations of the goal that would change the target level. For instance, a nonprofit community revitalization organization suggested starting with a lower, unspecified target level and gradually increasing it each year, culminating in an unspecified target level above 20 percent. The commenter also suggested establishing specific subgoals based on Bank district needs and on the types of small member institutions or geographic areas, e.g., small members located in rural areas.

The Banks jointly recommended that the proposed 20 percent target level decrease to 10 percent in two situations. First, the 20 percent target level would drop if, in a single year, one or more AMA users who had individually or collectively provided 10 percent or more of the AMA mortgages ceased being a member of the Bank. Second, the 20 percent target level would drop upon a material increase in mortgage delinquencies (defined as delinquencies exceeding 90 days), as measured in either FHFA's published mortgage-backed securities statistics or in a Bank's AMA mortgage portfolio.

A community bankers trade association recommended that FHFA set an alternate floor that would be triggered upon an unexpected economic downturn that made the 20 percent target level too onerous, providing 10 percent as an example of such a floor. The commenter stated that members need to view the AMA programs as consistent and reliable, and doubts about the reliability of a Bank as an outlet for their mortgage production may result in members limiting their AMA activity.

The Banks jointly commented that any new housing goal should be established such that each Bank is in compliance as of the effective date. Accordingly, the Banks recommended having a separate percentage floor for any Bank performing below the 20 percent target level from the outset. Alternatively, the Banks suggested that a Bank that has submitted an alternative target level for FHFA approval should not be subject to the 20 percent target level while the review process is pending. The Banks further suggested, as an alternative to a percentage floor, that FHFA could establish a range of possible target levels in the regulation, and then periodically set a specific target level within the range by way of a supervisory letter to the Banks.

2. FHFA Determination on the Target Level for the Mortgage Purchase Housing Goal

The final rule establishes the target level for the prospective mortgage purchase housing goal at 20 percent, the same level as in the proposed rule. In determining the target level for the new mortgage purchase goal, FHFA considered the comments received, as well as national housing needs, the past performance of the Banks under the housing goal, the ability of the Banks to lead the industry in making mortgage credit available, and the size of the mortgage market for affordable loans relative to the overall mortgage market. The factors considered by FHFA in setting the Bank housing goal are similar to the factors that FHFA is required by statute to consider in setting the Enterprise housing goals.^[15] Those factors include: (1) National housing needs; (2) economic, housing, and demographic conditions; (3) past performance on the housing goals; (4) ability to lead the industry in making mortgage credit available; (5) the size of the affordable market relative to the overall market; and (6) the financial condition of the Enterprises.

a. National Housing Needs, Including Underserved Borrowers

In determining the target level for the new mortgage purchase housing goal, FHFA considered the nation's affordable housing needs, which affect both homeowners and renters, while focusing on homeownership as the policy area most directly connected to the Bank housing goals. The national homeownership rate declined every year from 2004 to 2017, with particularly sharp declines for younger households and African American households.^[16] Tight access to mortgage credit is an ongoing factor in the lack of access to homeownership, particularly in places with lower-cost homes.^[17] Workers in essential sectors like construction often cannot afford to purchase even modestly priced homes in most metropolitan statistical areas. As an example, a typical carpenter could afford the median home price with a three percent down payment in only 40 percent of metropolitan statistical areas in a recent analysis.^[18] Workers in less well-paid professions struggle even more. Improved financing opportunities can help mitigate homeownership difficulties for underserved borrowers. FHFA recognizes these affordable housing challenges and has considered them in selecting a target level for the prospective mortgage purchase housing goal.

b. Past Performance on the Prospective Mortgage Purchase Housing Goal

In setting the target level for the mortgage purchase housing goal at 20 percent, FHFA also analyzed what the Banks' past performance under the goal would have been if the goal had been in effect at that time (including all loans that would have been counted under the proposed rule). Chart 1 below replicates and extends the proposed rule preamble analysis to include 2018 performance and to reflect the formulation of the prospective mortgage purchase housing goal in the final rule, which excludes non-conventional loans sold by AMA users with assets above the community-based AMA user asset cap. A full color version of Chart 1 and the other charts below appear in this preamble to the final rule as published on FHFA's website. The tables and charts in this preamble mask the identity of individual Banks by using letters instead of names to maintain confidentiality of Bank data. The letters identifying the Banks have been randomized for each table and chart (e.g., Bank A may refer to different Banks in different tables).

Chart 1 shows generally that Bank AMA performance as measured by the new prospective mortgage purchase housing goal was above the 20 percent target level for most Banks in most years since 2011. Bank I would have met or exceeded the 20 percent target level in 2015-2017 if all non-conventional loans were included (as in the proposed rule), but Bank I performed below 20 percent in 2015-2017 based on exclusion of non-conventional loans from institutions exceeding the community-based AMA user asset cap (as in the final rule). Bank I's 2018 performance was back above 20 percent.

Three Banks—E, G and H—show performance of zero for some years in the chart. These Banks were not purchasing AMA loans at the beginning of the time series and began building their AMA programs over time. In addition, Bank H purchased only non-conventional AMA loans for several years in the series, so the exclusion of such loans sold by institutions above the community-based AMA user asset cap has large effects on both the numerator and denominator of the goals performance percentage. Hence, Bank H performance is zero in some years and very high in others. Bank H did not have an active AMA program in 2019.

c. Ability of the Banks To Lead the Industry in Making Mortgage Credit Available

FHFA has also considered the ability of the Banks to lead the industry in making mortgage credit available. To assess the ability of the Banks to lead the industry, FHFA started by comparing how the Banks would have performed under the prospective mortgage purchase housing goal in 2018 with how Fannie Mae and Freddie Mac would have performed if they had been subject to the same goal in 2018. Table 1 below shows performance percentages for each Enterprise, applying the same calculation method applicable under the Bank housing goals established in this final rule.

As shown in Table 1, the Enterprise performance under the Bank housing goals would have exceeded the performance of most Banks in 2018. Fannie Mae's performance under the prospective mortgage purchase housing goal would have been 36 percent, and Freddie Mac's performance would have been 33 percent. As discussed above, all but two Banks would have exceeded the 20 percent target level in 2018, but only four Banks would have had performance levels above 30 percent (including one Bank whose performance has varied widely). The average performance for all Banks combined in 2018 would have been 26.6 percent.

There are a number of factors that help explain the difference between the performance of the Banks and the performance of the Enterprises. FHFA, through the AMA regulation, requires that if the Banks establish AMA programs, AMA users must bear a substantial portion of the credit risk associated with the loans they sell by providing a credit enhancement obligation, which the PFI must fully secure. Additionally, some Banks, through their capital structure plans, have AMA stock purchase requirements for members selling loans to the Bank. Taken together, the credit enhancement obligation, collateral requirement and potential member stock purchase requirement result in loans for which the Bank AMA program is the best execution qualifying for the mortgage purchase housing goal at a lower rate than loans sold to the Enterprises.

Other factors that help explain the difference in performance include the vast difference between the size of the Enterprises' mortgage portfolios and the Banks' mortgage portfolios. The Enterprises together serve more than 60 percent of the single-family secondary market, while the Banks' AMA programs serve less than one percent. Further, Banks are permitted to purchase AMA mortgages only from members in their districts, while the Enterprises serve a national market. In 2018, Bank System members totaled 6,863, of which only 861 sold mortgages to the Banks. Additionally, the Enterprises are chartered to provide stability and liquidity in the secondary market for residential mortgages by purchasing and making commitments to purchase residential mortgages. The Banks, in contrast, operate AMA programs at their discretion.

In setting the target level for the mortgage purchase housing goal at 20 percent, FHFA recognizes that the Banks' AMA programs have little ability to lead the industry due to their limited size relative to the overall mortgage market. The mortgage credit enhancement requirements of AMA programs and the supervisory expectations in the AMA Advisory Bulletin address safety and soundness considerations. These factors further explain the difference between the historical performance of the Banks on the new mortgage purchase housing goal and the nationwide market level for the mortgage market as a whole.

d. Size of the Affordable Market Relative to the Overall Market

The Banks' AMA mortgage purchases represent less than one percent of secondary mortgage market purchases by Ginnie Mae, Fannie Mae, and Freddie Mac, as discussed above and in the proposed rule preamble. Using a target level based on a market-wide measure would not be appropriate for the housing goals applicable to the AMA programs, which are currently a set of niche programs for Bank members and housing associates ^[19] due to the Banks' unique structure and role in the market.

e. Target Level Set at 20 Percent in Final Rule

Based upon FHFA's consideration of the comments, the serious affordable housing challenges nationwide, the past performance of the Banks, the affordable market levels, and the other factors discussed above, § 1281.11(a)(1) of the final rule sets the target level for the new mortgage purchase housing goal at 20 percent of the total number of AMA mortgage loans purchased by a Bank in the year being measured. This target level will encourage the Banks to continue to make meaningful contributions to affordable housing while recognizing the limited ability of the Banks to affect the overall housing market.

In considering an appropriate target level, FHFA evaluated whether to set specific target levels by Bank district, as recommended by some commenters. Bank districts vary widely, ranging from two to sixteen states or territories in one district. Available national data do not enable FHFA to set district-specific target levels with sufficient accuracy to be effective for housing goals. However, the option requested by the Banks for FHFA approval of alternative district-specific target levels, which is included in the final rule, will allow Banks to propose alternative target levels supported by data and subject to public comment.

The final rule addresses concerns raised by several commenters about how the new mortgage purchase housing goal will function under unexpected market conditions or other adverse circumstances through a combination of the following mechanisms:

1. Enforcement period. Section 1281.15(a) of the final rule provides for an initial three-year period during which FHFA will not impose a housing plan if a Bank fails to meet a housing goal. During this period, FHFA will still measure the Banks' performance under both housing goals and make determinations of Bank compliance with those goals. This period will provide the Banks time to adjust their AMA programs as needed to ensure that purchase of affordable housing mortgages is an integral, ongoing part of their business plans. It will also provide FHFA with an opportunity to collect data on Bank performance under the housing goals to guide implementation going forward. The proposed rule described this period as a “three-year” period, but the proposed rule text would have provided for the imposition of a housing plan only for “any year after 2021,” rather than “2022.” The final rule corrects this drafting error and extends the phase-in period to account for the initial application of the final rule in 2021 by providing for the possible imposition of a housing plan for “any year after 2023.” FHFA received no comments specifically addressing the extension of a phase-in period through 2021 as opposed to any other year.

2. Alternative target levels. Section 1281.11(a)(1)(ii) of the final rule provides a Bank with the option to propose, for FHFA approval, an alternative district-specific target level for the new mortgage purchase housing goal. Banks concerned about their ability to meet the 20 percent target level may pursue this option.

3. Infeasibility determination. Because the target level is calculated as a percentage of a Bank's total AMA mortgage purchases, the same way it is calculated in the current regulation, the target level should remain feasible under a range of market conditions, particularly changes that affect the volume of mortgage activity overall. If unexpected market conditions arise that make achievement of the target level infeasible for a Bank, FHFA has the discretion to determine that the goal was infeasible under § 1281.15(a) of the regulation. If FHFA determines that the goal was infeasible, the Bank would not be required to submit a housing plan. The regulation requires FHFA to consider market and economic conditions and the financial condition of the Bank in determining whether a goal was infeasible. These factors are similar to the factors suggested in the Banks' comment as possible criteria for setting a lower target level.

C. Treatment of Non-Conventional Mortgages Under the New Mortgage Purchase Housing Goal

Section 1281.13(c) of the final rule allows single-family mortgages guaranteed or insured by a department or agency of the federal government (i.e., non-conventional mortgages) to count toward the mortgage purchase housing goal only if the mortgages were acquired by the Bank from a community-based AMA user (i.e., an AMA user with assets not in excess of the community-based AMA user asset cap). This is a change from the current regulation, which excludes all non-conventional loans from counting towards a Bank's housing goals performance regardless of the size of the selling institution. It is also a change from the proposed rule, which would have allowed all non-conventional mortgage loans to count towards a Bank's housing goal performance if the loans met the other applicable criteria.

A rural-focused nonprofit organization, a nonprofit national organization, a nonprofit consumer advocacy organization, a banker's trade association, and a U.S. Senator supported the proposal to count all non-conventional mortgage loans toward the mortgage purchase housing goal. The nonprofit consumer advocacy organization urged FHFA to monitor closely how this new housing goal treatment would affect the mix of AMA mortgages purchased by the Bank. No commenter opposed the proposal.

The final rule represents a middle ground between the current regulation (excluding all non-conventional loans) and the proposed rule (including all non-conventional loans). The current Bank housing goals regulation is consistent with the Enterprise housing goals regulation in excluding non-conventional single-family loans from counting toward the Enterprise housing goals. Loans backed by the federal government have been excluded from the Enterprise single-family housing goals for many years. The exclusion is intended to avoid giving housing goals credit to the Enterprises for loans where the primary form of support comes from the federal government rather than the Enterprises.

The final rule allows non-conventional loans to count towards the Bank housing goals only in limited circumstances. The Banks' unique mission and ownership structure address the limited liquidity available in particular to small AMA users. The final rule, therefore, allows non-conventional loans purchased from small AMA users to count toward the performance of the Banks under the housing goals. This approach will allow non-conventional loans from small AMA users to count without creating an incentive for the Banks to purchase a high volume of non-conventional loans from larger members in order to improve their performance under the housing goals.

Non-conventional loans will be excluded from the numerator and the denominator in the housing goal calculation unless purchased from a small AMA user. The final rule only addresses whether and how non-conventional loans count for purposes of the housing goals. It does not prevent the Banks from purchasing such loans through their AMA programs. Eligibility for purchase is governed by the AMA regulation.

D. Cap on Loans to Higher-Income Borrowers Counting Toward the Mortgage Purchase Goal

Section 1281.11(a)(2) of the final rule provides that no more than 25 percent of the mortgages purchased by a Bank that are counted towards the mortgage purchase housing goal may be for higher-income borrowers, defined as borrowers with incomes above 80 percent of area median income, in low-income areas. As discussed above, the final rule combines each of the four current Bank housing goals into a single housing goal incorporating the categories in the four goals. One of the four goals categories focuses on mortgages for families in low-income areas, which may include some mortgages for families with incomes above 80 percent of area median income. The 25 percent cap will limit the extent to which a Bank may rely on mortgages for such higher-income families in low-income areas to meet the mortgage purchase housing goal. The cap does not prohibit the purchase of such mortgages by a Bank, although purchases of loans to higher-income borrowers in low-income areas that exceed the cap will have the effect of lowering the housing goal performance number that is calculated for the Bank.

The definition of “families in low-income areas” remains unchanged in the final rule from the current regulation, other than one technical, conforming revision included in the proposed rule.^[20] The definition includes (1) families in low-income census tracts regardless of family income, (2) moderate-income families in minority census tracts (i.e., census tracts with minority population of at least 30 percent and a median income less than the area median income), and (3) moderate-income families in designated disaster areas. “Moderate income” is defined in the regulation as income not in excess of area median income. These criteria are summarized in Table 2 below:

The definition of “families in low-income areas” is different from the other components included in the housing goal because it does not include an income limitation for borrowers. For properties located in low-income census tracts, each mortgage purchase counts toward a Bank's achievement of the housing goal, regardless of family income. For properties in minority census tracts and for properties in designated disaster areas, mortgage purchases count if family income is less than or equal to the area median income, which would include families with incomes above 80 percent up to 100 percent of area median income.

As a result, it is possible for loans to higher-income households in low-income areas to count toward achievement of the housing goal. The final rule does not exclude such mortgages for higher-income families from counting entirely; rather, it limits the extent to which a Bank may rely on loans in low-income areas for families with incomes above 80 percent of area median income to be counted for purposes of meeting the housing goal. Loans to higher-income borrowers in low-income areas that exceed the cap will still be counted in the denominator when calculating the performance for the Bank, with the effect that loans above the cap will effectively lower the Bank's calculated housing goals performance. The 25 percent cap is intended to balance the need for homeownership investment in communities that have lacked consistent, large-scale homeownership investment, on the one hand (which would support counting all loans that meet the criteria for low-income areas), and concern about the impact of an influx of higher-income households on existing residents, on the other hand (which would support excluding all loans to higher-income borrowers in low-income areas).^[21]

The final rule therefore caps the percentage of mortgages to higher-income borrowers that a Bank can count toward the housing goal at 25 percent. Note that the cap does not prevent Banks from purchasing these loans to higher-income borrowers pursuant to the AMA rule. The cap limits the extent to which such loans count toward the goal, but includes all such loans in the denominator when calculating the Bank's housing goals performance. This means that if the number of mortgages purchased by a Bank that are for families with incomes exceeding 80 percent of area median income who are located in low-income areas would contribute more than 25 percent of a Bank's performance on the housing goal, then any such mortgages above the 25 percent cap will be excluded from the numerator in calculating a Bank's performance under the goal. Those mortgages above the 25 percent cap will still be included in the denominator.

Commenters generally favored the proposed 25 percent cap. A U.S. Senator supported the proposed cap, as did advocacy organizations, nonprofits, and a lenders trade association. The Banks did not oppose the establishment of a cap, but they requested that FHFA set it at 30 percent instead of 25 percent.

Thirty-six state or local advocacy and community development organizations, in a joint comment letter, supported the proposed 25 percent cap. They pointed to evidence of gentrification and displacement of lower-income borrowers in low-income areas and in high-minority census tracts, including in markets where housing has become very expensive and affordable housing is scarce, and stated that the proposed cap would preserve housing opportunities for low-income families in those areas without reducing lending for non-low-income families generally.

A nonprofit consumer advocacy group strongly supported the proposed 25 percent cap, stating that community revitalization of historically underserved areas requires support from households with a mix of incomes but that the goals should primarily benefit low- and moderate-income borrowers.

A nationwide nonprofit community development organization supported the proposed 25 percent cap, and also stated that higher-income borrowers in low-income areas are important for ensuring that new investment flows into historically disinvested communities. A trade association representing state housing finance agencies commented that the proposed 25 percent cap would ensure that AMA programs continue to support lending to low- and moderate-income borrowers while not neglecting underserved communities.

A lenders trade association commented that the proposed 25 percent cap would mitigate against the risk that AMA mortgage purchases would be concentrated among higher-income households in low-income areas, stating that allowing those mortgages to constitute the majority of a Bank's affordable loan purchases would limit the effectiveness of the housing goal.

A credit union trade association commented that the proposed 25 percent cap would help balance the benefit of new investment with the impact on existing residents, but also expressed concern that it could burden credit unions, which have defined fields of membership that limit the scope of localities and persons they may serve. This commenter also noted that the proposed rule preamble did not present data on the benefits that mortgages to higher-income borrowers living in low-income areas have on those areas and described the proposed 25 percent cap as arbitrary and in need of additional justification. Nevertheless, the commenter wrote that the cap would help to balance the benefit of new investment with the impact on existing residents.

FHFA recognizes that mortgages for higher-income borrowers in low-income areas may provide indirect benefits to neighboring communities and homeowners, but such effects, by their nature, are difficult to quantify. In addition, the market is already providing mortgages to borrowers with incomes exceeding 80 percent of AMI in low-income areas at increasing levels, as discussed in the proposed rule preamble.^[22] The potential positive effects from Bank purchases of mortgages for higher income borrowers are also limited by the relatively limited size of the Banks' AMA programs as a share of overall secondary market activity. FHFA has determined that concerns about the Banks relying excessively on loans to higher income borrowers outweigh the possible benefits, and therefore the final rule establishes the 25 percent cap on counting mortgages for higher-income borrowers in low-income areas.

To help determine the appropriate level for a cap on loans to borrowers with incomes above 80 percent of area median income counting toward the mortgage purchase housing goal, FHFA analyzed whether there would have been a difference in how many Banks met the mortgage purchase goal's target level under a 25 percent cap as opposed to a 30 percent cap. From 2011 to 2018, a 30 percent cap would have twice allowed an additional Bank (once in 2018 and once in 2015) to count more loans toward the goal and therefore exceed the 20 percent target level, as compared to a 25 percent cap. Table 3 below summarizes the results of the analysis and shows, for context, the number of Banks with active AMA programs each year.

FHFA also analyzed the number of Banks that would have had mortgages excluded from counting toward the mortgage purchase housing goal due to the cap. For this analysis, FHFA measured whether each Bank purchased more loans to higher-income borrowers in low-income areas than would have counted toward the goal at both the 25 percent cap and 30 percent cap levels. This analysis found that more Banks would have had loans excluded from counting toward the goal under the 25 percent cap than under the 30 percent cap. Under the 25 percent cap, most Banks in most years would have had mortgages to higher-income borrowers in low-income areas excluded from consideration for the mortgage purchase housing goal. Under the 30 percent cap, the number of Banks with mortgages excluded due to the cap would have been significantly lower. Table 4 below summarizes the analysis of Banks that would have had loans excluded due to the cap.

As illustrated by Table 3 and Table 4, the 25 percent cap would have had little impact on the performance of the Banks under the mortgage purchase housing goal, but it would have limited the incentive for Banks to purchase mortgages for borrowers with incomes in excess of 80 percent of AMI in low-income areas in order to meet the goal. FHFA therefore has concluded that the 25 percent cap is an appropriate level to encourage the Banks to focus efforts on meeting the goal by purchasing loans to low-income borrowers, while recognizing that loans to higher-income borrowers in low-income areas are valuable and will continue to occur.

VII. New Small Member Participation Housing Goal

Consistent with the proposed rule, § 1281.11(b) of the final rule establishes a new small member participation housing goal that requires each Bank annually to ensure that the percentage of total AMA users that are community-based AMA users meets at least one of the following: (1) 50 percent, (2) a percentage that is three percentage points higher than the percentage from the preceding year, or (3) an alternative target level approved by FHFA. This new small member participation housing goal reflects the cooperative structure of the Banks and the recognition that smaller lenders are well-positioned to reach borrowers with affordable housing needs.

A majority of the AMA users participating in AMA programs are small when measured by asset size, but a larger portion of the number of AMA mortgages purchased by the Banks come from AMA users of greater asset size. In 2018, 83 percent of individual AMA users had total assets below $1.173 billion, the applicable 2018 CFI asset cap, established pursuant to 12 U.S.C. 1422(10)(B). Those AMA users sold 51 percent of the total number of AMA mortgages purchased by the Banks. Charts 2 and 3 below show the distribution of each Bank's AMA users by asset size and share of the number of loans purchased by the Bank from them.

A. Purpose of the Small Member Participation Housing Goal

The new small member participation housing goal should encourage Banks to maintain a focus in their AMA programs on small AMA users. As discussed in the proposed rule preamble, loans purchased from community-based AMA users (referred to generally as “small AMA users” or “small members” in the proposed rule) are more likely to be affordable home loans to low-income households than loans purchased from large AMA users. Table 5 below illustrates that in 2018, small (i.e., community-based) AMA users sold low-income or very low-income AMA loans to the Banks at a rate four percentage points greater than large AMA users. The proposed rule preamble reported the same difference using 2017 data.

Small lenders often rely on selling loans to the Banks as their connection to the secondary mortgage market, whereas larger lenders may have multiple secondary market executions available.^[23] The small member participation goal should encourage the Banks to continue to support small AMA users that might otherwise have difficulty accessing national capital markets, rather than primarily to augment the financial results of large AMA users that have no such difficulty.^[24] Small lenders are also an important source of credit access for rural areas, places of persistent poverty, and other underserved populations.

The Banks already serve many small AMA users, so the small member participation housing goal should encourage the Banks to maintain that focus over time. FHFA anticipates that the working relationships between Banks and small AMA users will result in ongoing purchases of AMA mortgages to benefit borrowers in need of financing for affordable housing.

B. Target Level for the Small Member Participation Housing Goal

Section 1281.11(b)(1) of the final rule establishes the target level for the small member participation housing goal as 50 percent for community-based AMA users relative to total AMA users for each Bank, consistent with the proposed rule. Section 1281.11(b)(2) of the final rule, as proposed, also provides that a Bank may satisfy the goal by showing improvement in its community-based AMA user participation of 300 basis points (for example, from 36 percent to 39 percent) over the previous year's performance. In addition, as proposed, § 1281.11(b)(3) of the final rule allows a Bank to propose an alternative target level for FHFA approval.

Many commenters supported the proposed small member participation goal, and no commenters opposed it although, as discussed below, several commenters requested changes to the proposed target level, and some expressed concerns about whether certain types of AMA users would satisfy the proposed CFI asset cap-based size requirements. As noted in Section IV.C., the final rule retains the quantitative standard for the proposed cap but incorporates the standard directly to the Bank housing goals regulation. Several commenters supported, for purposes of determining community-based AMA user status, the use of the three-year-average standard used to determine CFI eligibility.^[25] The language in the final rule adopts that standard. Comments from a trade association representing state housing finance agencies, a lenders trade association, a credit union, a nonprofit, a community bankers trade association, and a U.S. Senator supported the proposed 50 percent target level on the basis that it would encourage small AMA user participation. The U.S. Senator specifically expressed support for the proposal allowing a Bank to meet the goal by demonstrating improvement over the previous year's performance.

The Banks recommended that the rule establish a range of specific target levels and that FHFA make periodic determinations of a specific target level within the range for the Banks to meet, based on data reported by a national trade organization. The Banks did not specify this range, but they suggested a target level of 40 percent, rather than 50 percent, on the basis that they expect declines in Bank membership.

In contrast, a consumer advocacy group and a community revitalization organization recommended setting a target level higher than 50 percent for the goal. The consumer advocacy group stated that the data in the proposed rule preamble suggest that a 50 percent target level is too low since 9 of the 11 Banks are far above that level. The community revitalization organization stated that a 50 percent target level would not motivate the Banks to do the outreach necessary to increase participation by small members, particularly community development financial institutions (CDFIs).

A credit union trade association requested more information to better understand the proposed 50 percent target level. It requested more data, including additional background on trends in the market and overall small member participation in the market.

FHFA found that the small member participation rates in 2018 were quite similar to the 2017 participation rates described in the proposed rule preamble. As shown in Table 6 below, most of the Banks had shares of small (i.e., community-based) AMA users well above the proposed 50 percent target level. Incremental progress for the two Banks below the 50 percent target level would require adding only a single new community-based AMA user.

The Enterprise housing goals regulation does not include any goal similar to the small member participation housing goal. FHFA is establishing the goal for the Banks, which are cooperatives owned by their members and which place high value on supporting small members. Maintaining and improving small member participation in the AMA programs supports liquidity for affordable housing. As discussed above, small members of the Banks have originated mortgages to low-income borrowers at a higher rate than larger members.

In response to the comment seeking additional data on small institutions, FHFA considered whether other data sources such as HMDA data and banking regulator-published data could be used to provide comparisons regarding small institution performance in lending to low-income and very low-income borrowers. FHFA did not find readily accessible market-wide data linking institution asset size to mortgage origination data that would allow FHFA to analyze market-wide trends.

For the above reasons, FHFA has concluded that 50 percent is an appropriate target level for the small member participation housing goal. It is demonstrably achievable for most Banks, while motivating them to maintain or expand efforts to recruit small member participation in AMA programs. Banks expecting a substantial decline in membership or other unusual circumstances may propose an alternative target level for FHFA approval, as further discussed in Section VIII. below. The provision allowing a Bank to meet the goal instead by demonstrating progress of 300 basis points in small AMA user participation addresses the needs of Banks with small AMA programs that are still building member participation.

C. Standard for AMA Users With Assets Not in Excess of the CFI Asset Cap/Community-Based AMA User Asset Cap

FHFA received comments from several commenters, including two credit union trade associations, that the final rule should not preclude credit unions, CDFIs, state housing finance agencies, and others from being counted as small AMA users for purposes of the small member participation housing goal. These comments appear to stem from a misreading of the proposed rule. Under the proposed and final rules, a certain percentage of each Bank's AMA users would need to be institutions with assets not in excess of the CFI asset cap. The final rule does not require, nor would the proposed rule have required, that any particular percentage of AMA users be CFIs. CFIs are treated no differently than other types of institutions for purposes of the small member participation housing goal. Further, as noted in Section IV.C. above, the final rule retains the proposed quantitative CFI asset cap standard but adopts the relevant standard directly rather than via cross-reference to the Bank membership regulation. The final rule adopts the three-year average of total assets and a determination of total assets relative to the cap once per year, as is used for the CFI asset cap.

A trade association for state housing finance agencies recommended that the final rule treat all state housing finance agencies as small AMA users regardless of their total asset size, to encourage interaction between the Banks and state housing finance agencies. Designation as a small AMA user would not provide any significant incentive to a state housing finance agency to sell mortgages through the AMA programs, as the designation would not change the terms of the AMA programs. Because the Banks generally meet the 50 percent target level already, the only incentive provided by such a designation would be a small incentive to a Bank to do outreach and solicit mortgages from such entities. Therefore, FHFA has concluded that providing a special designation of state housing finance agencies as community-based AMA users regardless of their total asset size is not warranted.

VIII. Alternative Target Levels for Housing Goals

Section 1281.11(c)(1) of the final rule provides each Bank, upon approval of its board of directors, the opportunity to submit for FHFA prior approval an alternative target level for either or both of the housing goals. A Bank's request must include proposed target levels for three consecutive years following the calendar year in which the proposal is submitted. A Bank is not required to propose the same target level for each of the three years. In the absence of FHFA's approval of a Bank's proposed alternative target level, the Bank is subject to the target level established in the final rule.

Section 1281.11(c)(2) requires that a Bank's submission include a detailed explanation of: (i) Why the applicable target level (20 percent for the mortgage purchase housing goal and 50 percent for the small member participation housing goal) is infeasible; (ii) why the Bank's proposed alternative target level is achievable; and (iii) how the Bank's proposed alternative target level will meaningfully further affordable housing mortgage lending in its district.

A significant number of commenters expressed support for allowing the Banks to submit requests for alternative target levels. These commenters included a credit union, the Banks, a bank trade association, a U.S. Senator, a credit union trade association, a consumer advocacy organization, and a nonprofit. No commenters opposed the proposal.

Thirty-six state or local advocacy and community development organizations, in a joint comment letter, recommended that FHFA require well-reasoned justification that a Bank's proposed alternative target level is a stretch for the Bank. A lenders trade association commented that Bank proposals for alternative target levels should be informed by thorough data analysis and compelling evidence in order to receive FHFA approval.

The Banks expressed concern that the lack of a detailed timeline in the proposed rule for FHFA's review and response would create uncertainty for Banks whose past performance was well below the target level for a housing goal and who propose an alternative target level. A credit union trade association expressed similar concern.

FHFA believes that the final rule's submission deadline of September 15 should allow sufficient time for review, discussions as needed, and response to the Bank before the start of the applicable calendar year. In addition, the Banks may propose alternative target levels at any time before the annual deadline of September 15, subject to the three-year waiting period.

A. Frequency of Bank Requests

Section 1281.11(c) of the final rule allows a Bank to request an alternative target level no more than once every three years, subject to two exceptions discussed below. The request must be submitted by September 15 of the year preceding the year in which the alternative target level would apply. The September 15 deadline is earlier than the October 31 deadline in the proposed rule. The earlier deadline will allow time for any comments to be submitted by the public on the proposed alternative target levels, as discussed further in Section VIII.C. below.

Section 1281.11(c)(1) of the final rule provides that each request for an alternative target level must be approved by the Bank's board of directors. The proposed rule was silent on the level of approval needed for a Bank's request for an alternative target level. The final rule clarifies that the Bank's board of directors must approve any request for an alternative target level, which is consistent with other regulatory requirements and established practice regarding similar Bank requests.

Also in contrast to the proposed rule, which would have allowed a Bank submission only in the year the final rule becomes effective, and every three years thereafter, the final rule does not restrict a submission to any particular calendar years. For example, under the final rule, a Bank could make a submission on or before September 15, 2022, and would then be prohibited from making an additional submission until 2025 (three years from the date of the previous submission). The Bank could submit its 2025 request at any time until the September 15, 2025 deadline. If a Bank made its first request for alternative target levels on or before September 15, 2023, the Bank would then be prohibited from making another submission until 2026, with a deadline of September 15, 2026.

A community bankers trade association suggested that Banks be allowed to propose alternative target levels every year to allow Banks to better adapt to market conditions. FHFA notes, however, that the housing goals, by their nature, are a long-term planning tool rather than a year-to-year steering mechanism. Their aim is to ensure that the Banks make affordable housing part of their ongoing business planning for their AMA programs. Too frequent steering adjustment could lead either to overly ambitious target levels attempting to maximize support for affordable home lending, or overly pessimistic target levels anticipating weak market activity. The target levels in the final rule are designed to be more stable over time and somewhat flexible to market volume. Should markets take a sudden turn, FHFA retains the flexibility to determine a goal infeasible for a particular Bank.

Accordingly, in light of the long-term nature of the housing goals and the Banks' desire for additional flexibility, the final rule allows a Bank to submit a proposed alternative target level once every three years and does not limit the submission to any particular calendar years.

B. Exceptions to Three-Year Waiting Period

Notwithstanding the three-year waiting period, § 1281.11(c)(3)(ii) of the final rule provides that FHFA may at any time require a Bank to submit a request for an alternative target level to address discontinuation of an AMA product or program or approval of a new AMA product or program. This provision is largely consistent with the proposed rule, with the addition of the reference to “product” in the final rule. Both terms “AMA program” and “AMA product” are defined in § 1268.1 of the AMA regulation.

In addition, § 1281.11(c)(3)(iii) of the final rule adds an exception to the three-year waiting period that allows a Bank's board of directors to submit a request to FHFA at any time for an alternative target level if warranted given particular economic, operational, or other circumstances. FHFA does not intend this provision to be used frequently or regularly.

C. Public Notice and Comment on Proposed Alternative Target Levels

Section 1281.11(c)(4) of the final rule provides that FHFA will make each request for alternative target levels available for public comment on FHFA's website for at least 30 days. This provision was not included in the proposed rule but has been added in the final rule in response to comments received on the proposed rule. Thirty-six state or local advocacy and community development organizations, in a joint comment letter, as well as a nonprofit consumer advocacy organization, suggested that requests for alternative target levels be subject to public comment before FHFA's approval. A bank trade association emphasized that proposals for alternative target levels should be based on thorough data analysis and compelling evidence. FHFA is persuaded by these comments. While public comment will add time to the review process, information beyond what a proposing Bank submits will aid FHFA in evaluating proposed alternative target levels, especially if that information is rooted in knowledge of district housing and economic conditions.

Materials posted for public comment will not include any confidential or proprietary information submitted by a Bank. The final rule requires that a Bank submit information that it considers to be confidential or proprietary as a separate document, clearly designated as confidential or proprietary, to facilitate posting for public comment.

IX. Participation Interests in AMA Mortgages

The final rule, as proposed, addresses participations under two different scenarios. Under the first scenario, a Bank purchases a mortgage and later sells a participation interest in the mortgage to another Bank. Section 1281.13(b)(1) provides that participations among Banks that are executed after the mortgage was first acquired by a Bank will not be counted as mortgage purchases by a Bank purchasing such a participation for purposes of the mortgage purchase housing goal. This is consistent with FHFA's practice under the current regulation. This exclusion applies even if the participation is executed on the same day as the original mortgage acquisition by a Bank.

Under the second scenario, two or more Banks each purchase participation interests in the same mortgage simultaneously. Section 1281.13(e) of the final rule provides that participations among Banks that are entered simultaneously pursuant to an existing participation agreement will be counted as mortgage purchases on a pro rata basis toward the mortgage purchase housing goal for each Bank according to each Bank's percentage interest. This provision codifies existing FHFA practice on the treatment of participations under this scenario. FHFA received no comments on this proposal and the final rule adopts this change as proposed.

X. Other Comments Received

FHFA received several other comments that are addressed below, grouped by topic.

A. Adjustment to Target Levels for Unexpected Adverse Events

Several commenters asked how FHFA could adjust the housing goals or its evaluations of Bank performance in case of unexpected adverse events. The final rule provides several flexibilities in the case of such events. First, as under the current regulation, the goal target levels are based on a percentage of total mortgage purchases, so they have some inherent ability to remain applicable even as overall market volume expands or contracts, unlike a numerical target level.

Second, the Banks may propose alternative target levels, no more frequently than every three years. FHFA may allow a Bank to submit more frequently if unexpected circumstances warrant.

Third, as under the current regulation, when FHFA makes its annual determination of housing goals performance, it takes into account the feasibility of achieving the housing goals. If FHFA determines that a housing goal was infeasible for a particular Bank, the Bank is not required to submit a housing plan to FHFA. Even if FHFA determines that a housing goal was feasible for a particular Bank, FHFA has the option to forego requiring a housing plan from the Bank if warranted.

B. Counting Rules

The Banks asked whether loans purchased through the MPF Government product while held on balance sheet for eventual deployment into an MPF Government MBS count toward the housing goals. Since these loans are purchased through an AMA program, they count if they otherwise meet the criteria for the mortgage purchase housing goal and were sold to the Bank by a community-based AMA user, i.e., an AMA user with assets not in excess of the community-based AMA user asset cap.

The Banks also asked how MPF loans facilitated by other Banks but technically purchased directly by the Chicago Bank from a member or housing associate of another Bank would count toward the mortgage purchase goal. For housing goals purposes, FHFA will count such loans toward the performance of the Bank of which the seller is a member or housing associate, continuing current practice.

C. Manufactured Housing

A nonprofit affordable housing advocacy organization with an interest in expanding access to manufactured housing noted that the proposed rule would allow chattel loans on manufactured housing to count for purposes of the housing goals and requested that FHFA ensure such loans do not originate from predatory lending practices. A nonprofit manufactured housing community trade association and a national nonprofit manufactured housing intermediary recommended that FHFA issue more detailed guidance on the purchase of loans secured by manufactured housing generally. To date, there have been few, if any, purchases of chattel loans by the Banks. Accordingly, the final rule does not add housing goals restrictions specific to the Banks' purchases of loans secured by manufactured housing.

Section 1281.13 of the final rule, consistent with the proposed rule, eliminates a provision of the regulation that precludes “HOEPA mortgages” and “mortgages with unacceptable terms and conditions” from counting towards the housing goals. As discussed in the proposed rule preamble, guidance issued by FHFA to the Banks generally on the purchases of mortgages with certain predatory features rendered this provision in the housing goals regulation redundant. None of the comments caused FHFA to determine that this guidance is inadequate. Moreover, one of the purposes of the final rule, as discussed in the proposed rule preamble, is to better align the housing goals with the AMA regulation so that limitations on the types of loans eligible for Bank purchase are specified in the AMA regulation, not in the Bank housing goals regulation. In addition, FHFA did not propose any amendments to the AMA regulation.

D. Monitoring

Many commenters requested that FHFA monitor various aspects of the Banks' housing goals activity closely, especially in the first years of activity under the revised housing goals, including:

• Monitoring for any Banks' disincentive to participate in AMA programs;
• Monitoring the AMA loan composition over time;
• Monitoring whether any Banks require members to meet the mortgage purchase goal individually in loans sold to the Bank; and
• Monitoring Bank performance on the small member participation housing goal.

FHFA will monitor for these and other factors to ensure that the housing goals function as intended.

XI. Other Provisions in the Final Rule

The final rule also revises other provisions of the Bank housing goals regulation, as discussed below.

A. Changes to Definitions—§ 1281.1

As proposed, § 1281.1 of the final rule adds, revises, or removes certain definitions of terms used in the current Bank housing goals regulation. Specifically, the final rule adds definitions of “AMA mortgage,” “AMA program,” and “AMA user.” The final rule revises the definitions of “dwelling unit,” “families in low-income areas,” “median income,” “metropolitan area,” “mortgage,” and “non-metropolitan area.” The final rule removes the definitions of “Acquired Member Assets (AMA) program,” “AMA-approved mortgage,” “conforming mortgage,” “HMDA,” “HOEPA mortgage,” “HUD,” “mortgage data,” “mortgage with unacceptable terms or conditions,” “owner-occupied housing,” “residential mortgage,” and “second mortgage. In contrast to the proposed rule, the final rule does not remove the definition of “conventional mortgage” for the reasons discussed under Section VI.D. above. Also in contrast to the proposed rule, the final rule does not add “CFI asset cap” or “community financial institution or CFI” as defined terms, instead adding new terms “community-based AMA user” and “community-based AMA user asset cap.” In response to the proposed revisions to § 1281.1, other than those addressed in Section VII.C. above regarding the proposed rule's use of “CFI asset cap” and “CFI,” FHFA did not receive any comments. The changes to these definitions not discussed elsewhere in the preamble are discussed below.

1. Definition of “AMA mortgage”

As proposed, the final rule replaces the term “AMA-approved mortgage,” with “AMA mortgage” as a technical, non-substantive change. The Bank housing goals regulation currently defines “AMA-approved mortgage” to mean a mortgage that meets the requirements of an AMA program, with cross-references to the AMA regulation and the New Business Activities regulation.^[26] Section 1281.1 of the final rule replaces the term “AMA-approved mortgage” with “AMA mortgage” and defines it to mean a mortgage that was purchased by a Bank under an AMA program.

2. Definition of “AMA Program”

The final rule replaces the term “Acquired Member Assets (AMA) program,” with “AMA program” as a technical change. The current Bank housing goals regulation defines “Acquired Member Assets (AMA) program” as a program that authorizes a Bank to hold assets acquired from a member or housing associate by a purchase or funding transaction subject to the requirements of the AMA regulation and New Business Activities regulation. At the time the current Bank housing goals regulation was adopted, the term “AMA program” was not a defined term in the AMA regulation. A definition for the term “AMA program” was subsequently added to the AMA regulation in 2016.^[27] There is no substantive difference between the definition of “Acquired Member Assets (AMA) program” in the Bank housing goals regulation and the definition of “AMA program” in the AMA regulation. Accordingly, for consistency in terminology between the two regulations, § 1281.1 of the final rule replaces the definition of “Acquired Member Assets (AMA) program” in the housing goals regulation to conform it to the definition of “AMA program” in the AMA regulation.

3. Definition of “AMA User”

As proposed, § 1281.1 of the final rule adds a number of new definitions to implement the small member participation housing goal. The final rule adds “AMA user” as a participating financial institution (which can be a member or housing associate) ^[28] under the AMA regulation ^[29] from which a Bank purchased at least one AMA mortgage during the year for which the housing goal is being measured.

4. Definition of “conforming mortgage”

The current Bank housing goals regulation defines “conforming mortgage” as a conventional, AMA-approved single-family mortgage with an original principal obligation that does not exceed the dollar limitation under the AMA regulation or under the Freddie Mac conforming loan limits. Only purchases of mortgages under AMA programs count for purposes of the housing goals, and the AMA programs include limits on the size of mortgages that may be purchased by a Bank. Thus, it is not necessary for the housing goals regulation to include a separate limit on the size of mortgages that may be counted for purposes of the housing goals. Accordingly, as proposed, the final rule removes the definition of “conforming mortgage” from the housing goals regulation as unnecessary.

5. Definition of “conventional mortgage”

The current Bank housing goals regulation defines “conventional mortgage” as any mortgage that does not include a guaranty, insurance or other obligation by the United States or any of its agencies or instrumentalities. This definition was included in the regulation because only conventional mortgages counted towards the Bank housing goals. The proposed rule would have expanded the coverage of the Bank housing goals to include both conventional mortgages and non-conventional mortgages. Therefore, under the proposed rule, there would have been no need to distinguish between conventional mortgages and non-conventional mortgages so the definition of “conventional mortgage” was no longer necessary.

However, because the final rule provides that non-conventional mortgages purchased from community-based AMA users will count towards the mortgage purchase goal, § 1281.1 of the final rule retains “conventional mortgage” as a defined term.

6. Definition of “dwelling unit”

The current Bank housing goals regulation defines “dwelling unit” to mean a room or unified combination of rooms intended for use, in whole or in part, as a dwelling by one or more persons, and includes a dwelling unit in a single-family property, multifamily property, or other residential or mixed-use property. In the 2015 final rule amending the Enterprise housing goals regulation, FHFA revised the analogous definition to exclude a combination of rooms that does not have plumbing or kitchen facilities.^[30] Accordingly, to align the definitions in the two regulations, as proposed, § 1281.1 of the final rule revises the definition of “dwelling unit” in the Bank housing goals regulation to exclude a combination of rooms that does not have plumbing or kitchen facilities.

7. Definition of “HOEPA mortgage”

The current Bank housing goals regulation defines “HOEPA mortgage” as a mortgage covered by the definition of “high-cost mortgage” under the Truth in Lending Act. This definition was included because the housing goals regulation excludes HOEPA mortgages from counting toward achievement of the Bank housing goals. However, the final rule removes the provision excluding HOEPA mortgages from counting for purposes of the Bank housing goals. Therefore, the final rule removes the definition of “HOEPA mortgage” as no longer necessary.

8. Definitions of “median income,” “metropolitan area,” “non-metropolitan area,” and “HUD”

The current Bank housing goals regulation defines “median income,” with respect to an area, as the unadjusted median family income for the area as determined by the Department of Housing and Urban Development (HUD). The current definition further provides that FHFA will provide the Banks annually with information specifying how the median family income estimates for metropolitan areas are to be applied for the purposes of determining median family income. FHFA's practice is to calculate the applicable median income figures for both metropolitan and non-metropolitan areas and to provide the median income information to the Banks. Accordingly, as proposed, § 1281.1 of the final rule aligns the definition of “median income” with FHFA's practice, by revising it to mean, with respect to an area, the unadjusted median family income for the area as determined by FHFA. The final rule also revises the definition to provide that FHFA will provide the Banks annually with information specifying how the median family income estimates for both metropolitan and non-metropolitan areas are to be applied for purposes of determining median income.

The current Bank housing goals regulation defines “metropolitan area” as a metropolitan statistical area (MSA), or a portion of such an area, including Metropolitan Divisions, for which median family income estimates are determined by HUD. The regulation defines “non-metropolitan area” as a county, or a portion of a county, including those counties that comprise Micropolitan Statistical Areas, located outside any metropolitan area for which median family income estimates are published annually by HUD. As proposed, § 1281.1 of the final rule aligns the definition of “metropolitan area” with FHFA's practice by revising it to mean an MSA, or a portion of such an area, including Metropolitan Divisions, for which median incomes are determined by FHFA. The final rule aligns the definition of “non-metropolitan area” with FHFA's practice by revising it to mean a county, or a portion of a county, including those counties that comprise Micropolitan Statistical Areas, located outside any metropolitan area, for which median incomes are determined by FHFA.

The current Bank housing goals regulation defines “HUD” as the United States Department of Housing and Urban Development. Because the term “HUD” was used only in the definitions of “median income,” “metropolitan area,” and “non-metropolitan area” and the final rule removes the references to “HUD” from those definitions, the definition of “HUD” is no longer necessary and is removed, as proposed.

9. Definition of “Mortgage”—Inclusion of Chattel Loans on Manufactured Housing

The current Bank housing goals regulation includes a detailed definition of “mortgage” which includes all loans secured by real estate and any interests in such mortgages. The definition is based on the definition of “mortgage” in the Enterprise housing goals regulation and excludes chattel loans on manufactured housing. As proposed, § 1281.1 of the final rule revises the definition of “mortgage” in the Bank housing goals regulation to include chattel loans on manufactured housing. While the Banks have purchased few, if any, chattel loans on manufactured housing, the AMA regulation does not prohibit such purchases. Adding chattel loans on manufactured housing to the definition of “mortgage” in the Bank housing goals regulation simplifies the Bank housing goals by removing a potential difference between the coverage of the Bank housing goals and the AMA regulation.

10. Definition of “mortgage with unacceptable terms or conditions”

The current Bank housing goals regulation defines “mortgage with unacceptable terms or conditions” as a mortgage that has one or more of a series of terms or conditions that FHFA determined to be harmful to borrowers. This definition was included in the regulation because the regulation excludes such mortgages from counting toward achievement of the Bank housing goals. Because the final rule removes the provision excluding mortgages with unacceptable terms or conditions from counting for purposes of the Bank housing goals, this definition is no longer necessary and is also removed, as proposed.

11. Definition of “owner-occupied housing”

The current Bank housing goals regulation defines “owner-occupied housing” as single-family housing in which a mortgagor resides, including two- to four-unit owner-occupied properties where one or more units are used for rental purposes. The definition of “owner-occupied housing” was included in the regulation because the Bank housing goals are currently limited to mortgages on owner-occupied housing. As proposed, the final rule expands the coverage of the Bank housing goals to include all AMA mortgages, including mortgages not only on owner-occupied single-family properties but also investor-owned single-family properties. The final rule does not establish separate criteria for evaluating whether a mortgage on an investor-owned property could be counted for purposes of the housing goals. Any such mortgages will be evaluated based on the income of the mortgagor in the same manner as the evaluation of a mortgage on an owner-occupied property. Because the regulation will no longer limit the Bank housing goals to mortgages on owner-occupied housing, the final rule removes the definition of “owner-occupied housing” from the Bank housing goals regulation as unnecessary.

12. Definition of “residential mortgage”

The current Bank housing goals regulation defines “residential mortgage” as a mortgage on single-family housing. The term “residential mortgage” is not used anywhere else in the regulation or in the final rule. Accordingly, as proposed, the final rule removes the definition of “residential mortgage” as unnecessary.

13. Definition of “second mortgage”

The current Bank housing goals regulation defines “second mortgage” as any mortgage that has a lien position subordinate only to the lien of the first mortgage. This term is used in § 1281.13(b)(8), which provides that “purchases of subordinate lien mortgages (second mortgages),” do not count for purposes of the housing goals. The final rule clarifies that this prohibition applies to all mortgages that are subordinate to the first mortgages, not only second mortgages. Because “second mortgage” will no longer appear in the regulation, as proposed, the final rule removes this definition as unnecessary.

B. General—§ 1281.10

Consistent with the proposed rule, the final rule revises § 1281.10 to reflect the new structure of the housing goals and removal of the volume threshold.

C. Changes to Bank Housing Goals—§§ 1281.11 and 1281.14

The final rule also adopts a proposed conforming change to § 1281.14(a) by eliminating the Bank volume threshold as a consideration in determining whether the Director evaluates annual performance of Bank performance under each housing goal.

In addition, the final rule requires that no more than 25 percent of the mortgages that are counted toward a Bank's achievement of the prospective mortgage purchase housing goal may be mortgages for families with incomes above 80 percent of area median income. This is consistent, in substance, with the proposed rule, which would have established the same requirement, but which would have characterized it as a requirement that at least 75 percent of the mortgages that are counted toward a Bank's achievement of the prospective mortgage purchase housing goal must be for low-income or very low-income families. The final rule also includes language clarifying that any purchases of mortgages for families with incomes above 80 percent of area median income in excess of the 25 percent cap shall be treated as a mortgage purchase for purposes of the housing goals and shall be included in the denominator for the housing goal, but such mortgages shall not be included in the numerator in calculating a Bank's performance under the housing goals.

D. General Counting Requirements—§ 1281.12

The final rule adopts all proposed revisions to § 1281.12. The current Bank housing goals regulation defines the “numerator” and “denominator” used to calculate performance under the current housing goals. The final rule deletes paragraph (a) as unnecessary in light of the mortgage goal calculation standards reflected in § 1281.11 of the final rule.

The current Bank housing goals regulation also provides that mortgages with missing data or information necessary for counting are included in the denominator when calculating a Bank's performance, but not in the numerator. This effectively penalizes a Bank's performance by treating mortgages with missing data or information as if they were loans that did not meet the applicable criteria. Accordingly, the final rule also removes paragraph (b)(1), so that mortgages with missing data or information are disregarded (i.e., not included in the numerator or denominator) for purposes of measuring a Bank's performance on the housing goals.

Finally, paragraph (c), which provides that a mortgage may only count once towards achievement of a housing goal even if it satisfies more than one goal, is redesignated as paragraph (b) and revised to permit each mortgage to be counted only once toward achievement of the prospective mortgage purchase housing goal, even if it satisfies multiple categories under the goal.

The final rule also makes conforming redesignations of paragraphs throughout the remainder of § 1281.12.

E. Special Counting Requirements—§ 1281.13

Paragraph (b) of § 1281.13 currently enumerates categories of transactions or activities that are not counted for purposes of the housing goals and are not included in the numerator or the denominator in calculating a Bank's housing goals performance. The proposed rule would have removed references to “numerator” and “denominator” as unnecessary in light of the simplified calculation methodology reflected in § 1281.11. However, the final rule retains clarifying language to specify that loans which are “not counted for purposes of the housing goals” are excluded from both the numerator and denominator.

F. Determination of Compliance With Housing Goals; Notice of Determination—§ 1281.14

The final rule adopts all proposed revisions to § 1281.14. The final rule amends § 1281.14(a) by removing the reference to the volume threshold, which is no longer applicable. The final rule also amends § 1281.14(a) to require that FHFA publish its annual determinations of Bank housing goals compliance and specifies the types of data to be included in the published determinations.

G. Housing Plans—§ 1281.15

The final rule revises § 1281.15 to provide that the Director may only require that a Bank submit a housing plan for any year after 2023. As discussed in Section IV.D. above, this is in contrast to the proposed rule, which would have extended this period only through 2021. This reflects the phase-in period for the new housing goals, eliminating possibility of a housing plan during the first three years in which the prospective mortgage purchase and small member participation housing goals are operative. Because a Bank may be required to submit a housing plan while awaiting FHFA's response to a proposal by the Bank for an alternative goal target level, the final rule amends § 1281.15 by adding new paragraph (b)(5) to require that the housing plan address any alternative target levels the Bank is requesting. This is generally consistent with the proposed rule, with certain technical revisions including replacing the reference to “Bank-specific housing goals” with “alternative target levels” for consistency and clarity.

H. Reporting Requirements—§§ 1281.1 and 1281.20

Consistent with the proposed rule, the final rule amends Subpart C of the current regulation to simplify and clarify the reporting requirements for the Banks under the new housing goals. The final rule, as proposed, revises the reporting requirements to reflect the new housing goals structure and to eliminate provisions that are either duplicative of, or potentially inconsistent with, the existing Bank reporting requirements in FHFA's Data Reporting Manual (DRM). The DRM, which is amended from time to time, includes detailed requirements about the data elements that the Banks must report and the timing and format of the required reporting.

The final rule, as proposed, consolidates the four sections that currently exist in Subpart C of the Bank housing goals regulation into a single section. Accordingly, §§ 1281.21, 1281.22 and 1281.23 are removed from the regulation. Section 1281.20 includes the new reporting requirements. Section 1281.20(a) requires the Banks to submit to FHFA any data that FHFA determines to be necessary to evaluate transactions and activities under the Bank housing goals. Section 1281.20(b) and (c) set out the data reporting requirements for the prospective mortgage purchase housing goal and the small member participation housing goal, respectively, and require such submissions to be made in accordance with the DRM. Section 1281.20(d) continues to permit FHFA to require a Bank to provide such additional reports, information, and data as FHFA may request from time to time.

Consistent with the proposed rule, the final rule also removes the provision in the current regulation that addresses errors, omissions or discrepancies in the data reported by a Bank. This provision is unnecessary in light of FHFA's existing supervisory and regulatory authorities and procedures.

Finally, consistent with the proposed rule, the final rule removes the definition of “mortgage data” from the regulation. The regulation defines “mortgage data” as data obtained from the Banks under the DRM. The final rule's revisions to the reporting requirements in Subpart C remove all references to the term “mortgage data,” making the definition unnecessary.

XII. Considerations of Differences Between the Banks and the Enterprises

When promulgating regulations relating to the Banks, section 1313(f) of the Safety and Soundness Act requires the Director of FHFA to consider the differences between the Banks and the Enterprises with respect to the Banks' cooperative ownership structure, mission of providing liquidity to members, affordable housing and community development mission, capital structure, and joint and several liability. FHFA requested comments from the public about whether these differences should result in any revisions to the proposed rule, but no significant, relevant comments were received. FHFA, in preparing this final rule, considered the differences between the Banks and the Enterprises as they relate to the above factors and determined these amendments to the Bank Housing Goal regulation to be appropriate and reflect the unique differences between the Banks and Enterprises. FHFA also considered these differences in light of section 10C of the Bank Act, which requires that the Bank housing goals be consistent with the Enterprise housing goals, with consideration of the unique mission and ownership structure of the Banks, and similarly determined these amendments to be appropriate in light of relevant factors.^[31]

XIII. Paperwork Reduction Act

The final rule does not contain any information collection requirement that would require the approval of OMB under the Paperwork Reduction Act (44 U.S.C. 3501 et seq.). Therefore, FHFA has not submitted any information to OMB for review.

XIV. Regulatory Flexibility Act

The Regulatory Flexibility Act (5 U.S.C. 601 et seq.) requires that a regulation that has a significant economic impact on a substantial number of small entities, small businesses, or small organizations must include an initial regulatory flexibility analysis describing the regulation's impact on small entities. Such an analysis need not be undertaken if the agency has certified that the regulation will not have a significant economic impact on a substantial number of small entities. (5 U.S.C. 605(b)). FHFA has considered the impact of the final rule under the Regulatory Flexibility Act. The General Counsel of FHFA certifies that the final rule is not likely to have a significant economic impact on a substantial number of small entities because the final rule applies to the Banks, which are not small entities for purposes of the Regulatory Flexibility Act.

XV. Congressional Review Act

In accordance with the Congressional Review Act (5 U.S.C. 801 et seq.), FHFA has determined that this final rule is a major rule and has verified this determination with the Office of Information and Regulatory Affairs of OMB.

List of Subjects in 12 CFR Part 1281

• Credit
• Federal home loan banks
• Housing
• Mortgages
• Reporting and recordkeeping requirements

Authority and Issuance

For the reasons stated in the SUPPLEMENTARY INFORMATION, under the authority of 12 U.S.C. 4526, 1430, 1430b, 1430c, and 1431, FHFA is amending part 1281 of Title 12 of the Code of Federal Regulations as follows:

CHAPTER XII—FEDERAL HOUSING FINANCE AGENCY

SUBCHAPTER E—HOUSING GOALS AND MISSION

PART 1281—FEDERAL HOME LOAN BANK HOUSING GOALS

1. Revise the authority citation for part 1281 to read as follows:

Authority: 12 U.S.C. 1430, 1430b, 1430c, 1431.

2. Amend § 1281.1 by:

a. Removing the definitions of “Acquired Member Assets (AMA) program” and “AMA-approved mortgage”;

b. Adding definitions for “AMA mortgage”, “AMA program”, and “AMA user” in alphabetical order;

c. Removing the definition of “Conforming mortgage”;

d. Adding in alphabetical order the definitions for “Community-based AMA user” and “Community-based AMA user asset cap”;

e. Revising the definition of “Dwelling unit” and paragraph (1) of the definition of “Families in low-income areas”;

f. Removing the definitions of “HMDA”, “HOEPA mortgage”, and “HUD”;

g. Revising the definitions of “Median income”, “Metropolitan area”, and “Mortgage”;

h. Removing the definitions of “Mortgage data” and “Mortgage with unacceptable terms or conditions”;

i. Revising the definition of “Non-metropolitan area”; and

j. Removing the definitions of “Owner-occupied housing”, “Residential mortgage”, and “Second mortgage”.

The revisions and additions read as follows:

§ 1281.1

Definitions.

* * * * *

AMA mortgage means a mortgage that was purchased by a Bank under an AMA program.

AMA program has the meaning set forth in § 1268.1 of this chapter.

AMA user means any participating financial institution, as defined in § 1268.1 of this chapter, from which the Bank purchased at least one AMA mortgage during the year for which the housing goals are being measured.

* * * * *

Community-based AMA user means any AMA user whose average total assets over the three-year period culminating in the year preceding the one being measured are no greater than the applicable community-based AMA user asset cap.

Community-based AMA user asset cap means $1,224,000,000, subject to annual adjustments by FHFA, beginning in 2021, to reflect any percentage increase in the preceding year's Consumer Price Index (CPI) for all urban consumers, as published by the U.S. Department of Labor.

* * * * *

Dwelling unit means a room or unified combination of rooms with plumbing and kitchen facilities intended for use, in whole or in part, as a dwelling by one or more persons, and includes a dwelling unit in a single-family property, multifamily property, or other residential or mixed-use property.

Families in low-income areas * * *

(1) Any family that resides in a census tract in which the median income does not exceed 80 percent of the area median income;

* * * * *

Median income means, with respect to an area, the unadjusted median family income for the area as determined by FHFA. FHFA will provide the Banks annually with information specifying how the median family income estimates for metropolitan and non-metropolitan areas are to be applied for purposes of determining median income.

Metropolitan area means a metropolitan statistical area (MSA), or a portion of such an area, including Metropolitan Divisions, for which median incomes are determined by FHFA.

* * * * *

Mortgage means a member of such classes of liens, including subordinate liens, as are commonly given or are legally effective to secure advances on, or the unpaid purchase price of, real estate under the laws of the State in which the real estate is located, or a manufactured home that is personal property under the laws of the State in which the manufactured home is located, together with the credit instruments, if any, secured thereby, and includes interests in mortgages. Mortgage includes a mortgage, lien, including a subordinate lien, or other security interest on the stock or membership certificate issued to a tenant-stockholder or resident-member by a cooperative housing corporation, as defined in section 216 of the Internal Revenue Code of 1986, and on the proprietary lease, occupancy agreement, or right of tenancy in the dwelling unit of the tenant-stockholder or resident-member in such cooperative housing corporation.

* * * * *

Non-metropolitan area means a county, or a portion of a county, including those counties that comprise Micropolitan Statistical Areas, located outside any metropolitan area, for which median incomes are determined by FHFA.

* * * * *

3. Amend § 1281.10 by revising paragraphs (a) and (b) to read as follows:

§ 1281.10

General.

* * * * *

(a) A prospective mortgage purchase housing goal;

(b) A small member participation housing goal;

* * * * *

4. Revise § 1281.11 to read as follows:

§ 1281.11

Bank housing goals.

(a) Prospective mortgage purchase housing goal— (1) Target levels. For each calendar year, the percentage of a Bank's AMA mortgages acquired during the calendar year that are for very low-income families, low-income families, or families in low-income areas must meet or exceed either:

(i) A target level of 20 percent; or

(ii) An alternative target level proposed by the Bank and approved by FHFA under paragraph (c) of this section.

(2) Cap on low-income areas loans counted toward goal. No more than 25 percent of the mortgages that are counted toward a Bank's achievement of the prospective mortgage purchase housing goal may be mortgages for families with incomes above 80 percent of area median income. Any purchases of mortgages for families with incomes above 80 percent of area median income in excess of the 25 percent cap shall be treated as mortgage purchases for purposes of the housing goals and shall be included in the denominator for the housing goal, but such mortgages shall not be included in the numerator in calculating a Bank's performance under the housing goal.

(b) Small member participation housing goal. For each calendar year, the percentage of a Bank's total AMA users that are community-based AMA users must meet or exceed one of the following:

(1) A target level of 50 percent;

(2) A percentage that is three percentage points greater than the percentage from the preceding calendar year; or

(3) An alternative target level proposed by the Bank and approved by FHFA under paragraph (c) of this section.

(c) Alternative target levels— (1) Submission of Bank requests. A Bank, upon approval of its board of directors, may submit a written request to FHFA for approval of different target levels for the prospective mortgage purchase housing goal, the small member participation housing goal, or both. A Bank's request under this paragraph must include proposed target levels for three consecutive years following the calendar year in which the request is submitted. A Bank is not required to propose the same target level for each of the three years.

(2) Content of Bank request. A Bank's request under paragraph (c)(1) of this section for an alternative target level must include a detailed explanation of:

(i) Why the target level for the goal in paragraphs (a) and (b) of this section, as applicable, is infeasible;

(ii) Why the Bank's proposed alternative target level is achievable; and

(iii) How the Bank's proposed alternative target level will meaningfully further affordable housing mortgage lending in its district.

(3) Frequency of Bank requests— (i) Three-year period. A Bank may not submit a request under paragraph (c)(1) of this section for an alternative target level more frequently than once every three years, except as provided in paragraphs (c)(3)(ii) or (c)(3)(iii) of this section. The deadline for submitting a request under paragraph (c)(1) of this section is September 15 of the calendar year preceding the calendar year in which the alternative target level would apply. FHFA will review each Bank request that is received by the deadline and will notify the Bank in writing if its request is approved. If FHFA does not notify a Bank that its request is approved, the Bank will remain subject to the target levels in paragraphs (a) and (b) of this section, as applicable.

(ii) Exception for changes in AMA products or programs. FHFA may require a Bank to submit a request under paragraph (c)(1) of this section for an alternative target level to address discontinuation of an AMA product or program or approval of a new AMA product or program.

(iii) Exception for special circumstances. A Bank may submit a request under paragraph (c)(1) of this section for an alternative target level more frequently than once every three years if warranted given economic, operational, or other circumstances.

(4) Public comment. FHFA will publish each request that is submitted under paragraph (c)(1) of this section for an alternative target level on FHFA's public website for a period of at least 30 days, to provide the public an opportunity to comment on the request. FHFA will publish each request without redactions or other changes, except that FHFA will not publish any confidential or proprietary material. A Bank must submit any material supporting its request under paragraph (c)(1) of this section that it considers to be confidential or proprietary as a separate document, clearly designated as confidential or proprietary.

5. Revise § 1281.12 to read as follows:

§ 1281.12

General counting requirements.

(a) General. Mortgage purchases financing single-family properties shall be evaluated based on the income of the mortgagors and the area median income at the time the mortgage was originated. To determine whether mortgages may be counted under a particular family income level (e.g., low- or very low-income), the income of the mortgagor is compared to the median income for the area at the time the mortgage was originated, using the appropriate percentage factor provided under § 1281.1.

(b) No double-counting. A mortgage may be counted only once toward the achievement of the prospective mortgage purchase housing goal, even if it satisfies multiple criteria for the prospective mortgage purchase housing goal.

(c) Application of median income. For purposes of determining an area's median income under § 1281.1, the area is:

(1) The metropolitan area, if the residence that secures the mortgage is in a metropolitan area; and

(2) In all other areas, the county in which the property is located, except that where the State non-metropolitan median income is higher than the county's median income, the area is the State non-metropolitan area.

(d) Sampling not permitted. Performance under the housing goals for each year shall be based on a tabulation of each mortgage during that year; a sampling of such purchases is not acceptable.

6. Amend § 1282.13 by:

a. Revising paragraph (b)introductory text, (b)(1) and (8);

b. Adding paragraph (c)(4);

c. Removing paragraph (d);

d. Redesignating paragraph (e) as paragraph (d); and

e. Adding new paragraph (e).

The revisions and additions read as follows:

§ 1281.13

Special counting requirements.

* * * * *

(b) Not counted. The following transactions or activities shall not be counted for purposes of the housing goals, meaning that in calculating the applicable percentage target level, they shall be excluded from both the numerator (i.e., AMA mortgages acquired during the calendar year that are for very low-income families, low-income families, or families in low-income areas) and the denominator (i.e., total AMA mortgages acquired during the calendar year), even if the transaction or activity would otherwise be counted under paragraph (c) of this section:

(1) Purchases of participation interests in AMA mortgages from another Bank, except as provided in paragraph (e) of this section;

* * * * *

(8) Purchases of subordinate lien mortgages;

* * * * *

(c) * * *

(4) Non-conventional mortgages. The purchase of a non-conventional single-family mortgage shall be treated as a mortgage purchase for purposes of the housing goals only if the mortgage was acquired from a community-based AMA user.

* * * * *

(e) Mortgage participation transactions. Where two or more Banks acquire a participation interest in the same mortgage simultaneously, the mortgage will be counted on a pro rata basis for the prospective mortgage purchase housing goal for each Bank with a participation interest.

7. Amend § 1281.14 by revising paragraph (a) to read as follows:

§ 1281.14

Determination of compliance with housing goals; notice of determination.

(a) Determination of compliance with housing goals. On an annual basis, FHFA will determine each Bank's performance under each housing goal and will publish the final determinations. FHFA will publish its final determination including the numbers and percentages for each Bank's AMA purchases that meet each of the housing goals criteria, including loans to low-income families, loans to very low-income families, and loans to families in low-income areas, including by each of the defined categories. FHFA's determination will include these numbers in total and separated into purchase money mortgages, refinancing mortgages, conventional mortgages, and non-conventional mortgages.

* * * * *

8. Amend § 1281.15 by revising paragraphs (a) and (b) to read as follows:

§ 1281.15

Housing plans.

(a) Housing plan requirement. For any year after 2023, if the Director determines that a Bank has failed to meet any housing goal and that the achievement of the housing goal was feasible, the Director may require the Bank to submit a housing plan for approval by the Director.

(b) Nature of plan. If the Director requires a housing plan, the housing plan shall:

(1) Be feasible;

(2) Be sufficiently specific to enable the Director to monitor compliance periodically;

(3) Describe the specific actions that the Bank will take to achieve the housing goal for the next calendar year;

(4) Address any additional matters relevant to the housing plan as required, in writing, by the Director; and

(5) Address any alternative target levels for which the Bank has submitted a request under § 1281.11(c)(1).

* * * * *

9. Revise Subpart C to read as follows:

Subpart C—Reporting Requirements

§ 1281.20

Reporting requirements.

(a) General. Each Bank must collect and submit to FHFA any data that FHFA determines to be necessary for FHFA to evaluate transactions and activities under the Bank housing goals.

(b) Reporting for prospective mortgage purchase housing goal. Each Bank must collect data on each AMA mortgage purchased by the Bank. The data must include any data elements specified by FHFA. On no less frequent than an annual basis, each Bank must submit such data to FHFA in accordance with the Data Reporting Manual.

(c) Reporting for small member participation housing goal. Each Bank must collect data on AMA user asset size. On no less frequent than an annual basis, each Bank must submit such data to FHFA in accordance with the Data Reporting Manual.

(d) Other reporting. Each Bank must provide to FHFA such additional reports, information, and data as FHFA may request from time to time.

Mark A. Calabria,

Director, Federal Housing Finance Agency.

Footnotes

1.  See 12 U.S.C. 1421 et seq.

2.  See 12 U.S.C. 1426(a)(4), 1430(a), 1430b.

3.  See 12 U.S.C. 1424; 12 CFR part 1263. The Bank Act also authorizes membership for community development financial institutions and non-federally-insured credit unions.

4.  Members are required to pledge specific types of eligible collateral, mainly mortgages or other real estate-related assets, to secure any advance taken down from a Bank. See 12 CFR 1266.7.

5.  See 12 CFR part 1268.

6.  See “Federal Home Loan Banks, Combined Financial Report for the Year Ended December 31, 2018,” 43 (Mar. 27, 2019), available at http://www.fhlb-of.com/​ofweb_​userWeb/​resources/​2018Q4CFR.pdf.

7.  12 CFR part 1281.

8.  12 U.S.C. 1430c(a).

9.  12 U.S.C. 1430c(b).

10.  “Low-income” and “very low-income” are defined in the final rule, as in the current regulation, as income not in excess of 80 percent and 50 percent of area median income, respectively. For a discussion of the definition of “families in low-income areas” see Section VI.D.

11.  83 FR 55114.

12.  Because the housing goals will no longer be based on retrospective HMDA data, the final rule, as proposed, removes the definition of “HMDA” from § 1281.1.

13.  “Acquired Member Assets Risk Management Advisory Bulletin,” AB 2020-01 (January 31, 2020).

14.  See 12 CFR part 1282.

15.  See 12 U.S.C. 4562(e).

16.  See Joint Center for Housing Studies of Harvard University, “The State of the Nation's Housing 2017,” 21 (2017), available at http://www.jchs.harvard.edu/​sites/​default/​files/​harvard_​jchs_​state_​of_​the_​nations_​housing_​2017_​0.pdf.

17.  Id. at 23.

18.  National Housing Conference, “Paycheck to Paycheck 2018,” 3 (Apr. 2018), available at https://www.nhc.org/​wp-content/​uploads/​2019/​04/​P2P2018_​Final.pdf.

19.  Eligible, non-member housing associates may participate in a Bank's AMA program. However, since 2000, no housing associate has sold an AMA loan to a Bank.

20.  In its 2015 final rule amending the Enterprise housing goals regulation, FHFA removed the reference to “block numbering areas” under the first prong of the “families in low-income areas” definition to conform to the terminology used by the U.S. Census Bureau. As proposed, § 1281.1 of the final rule makes a similar conforming revision to the definition of “families in low-income areas” in the Bank housing goals regulation by removing the reference to “block numbering areas.”

21.  For information on the effects of gentrification, see generally Federal Reserve Bank of Philadelphia, “Research Symposium on Gentrification and Neighborhood Change” (May 25, 2016), available at https://www.philadelphiafed.org/​community-development/​events/​2016/​research-symposium-on-gentrification;​; Diane K. Levy, Jennifer Comey & Sandra Padilla, “IN THE FACE OF GENTRIFICATION: Case Studies of Local Efforts to Mitigate Displacement” (Urban Institute 2006), available at https://www.urban.org/​sites/​default/​files/​publication/​50791/​411294-In-the-Face-of-Gentrification.PDF.

22.  See Federal Home Loan Bank Housing Goals Amendments, 83 FR 55114, 55120, Table 3 (Nov. 2, 2018), available at https://www.govinfo.gov/​content/​pkg/​FR-2018-11-02/​pdf/​2018-23890.pdf.

23.  See generally “Housing Finance Reform: Protecting Small Lender Access to the Secondary Mortgage Market, Hearing Before the S. Comm. on Banking, Housing and Urban Affairs,” 113th Cong., 1st Sess. (2013) (website), available at https://www.banking.senate.gov/​hearings/​housing-finance-reform-protecting-small-lender-access-to-the-secondary-mortgage-market.

24.  For this reason, FHFA grounds the small member participation housing goal not just in the housing goals section of the Bank Act, 12 U.S.C. 1430c, but also in the statutory basis for the AMA program more generally. See 12 U.S.C. 1430, 1430b, 1431; Texas Savings & Community Bankers Ass'n v. Federal Housing Finance Board, 201 F.3d 551 (5th Cir. 2000).

25.  See 12 U.S.C. 1422(10)(A)(ii); 12 CFR 1263.1 (par. (2), definition of “community financial institution or CFI”).

26.  12 CFR part 1272.

27.  12 CFR 1268.1.

28.  Although eligible housing associates may participate in a Bank's AMA program and the participation of a housing associate may count towards the small member participation housing goal, as noted above, since 2000, no housing associate has sold an AMA mortgage loan to a Bank.

29.  12 CFR 1268.1.

30.  See 80 FR 53392 (Sept. 3, 2015), codified at 12 CFR 1282.1.

31.  See 12 U.S.C. 1430c.