GAO-19-702, REVERSE MORTGAGES: FHA Needs to Improve Monitoring and Oversight of Loan Outcomes and Servicing

REVERSE

MORTGAGES

FHA Needs to

Improve Monitoring

and Oversight of

Loan Outcomes

and Servicing

Report to Congressional Requesters

September 2019

GAO-19-702

United States Government Accountability Office

______________________________________ United States Government Accountability Office

September 2019

REVERSE MORTGAGES

FHA

Needs to Improve Monitoring and Oversight of

Loan

Outcomes and Servicing

What GAO Found

The vast majority of reverse mortgages are made under the Federal Housing

Administration’s (FHA) Home Equity Conversion Mortgage (HECM) program. In

recent years, a growing percentage of HECMs insured by FHA have ended

because borrowers defaulted on their loans. While death of the borrower is the

most commonly reported reason why HECMs terminate, the percentage of

terminations due to borrower defaults increased from 2 percent in fiscal year

2014 to 18 percent in fiscal year 2018 (see figure). Most HECM defaults are due

to borrowers not meeting occupancy requirements or failing to pay property

charges, such as property taxes or homeowners insurance. Since 2015, FHA has

allowed HECM servicers to put borrowers who are behind on property charges

onto repayment plans to help prevent foreclosures, but as of fiscal year-end

2018, only about 22 percent of these borrowers had received this option.

Reported Home Equity Conversion Mortgage Termination Reasons, Fiscal Years 2014–2018

FHA’s monitoring, performance assessment, and reporting for the HECM

program have weaknesses. FHA loan data do not currently capture the reason

for about 30 percent of HECM terminations (see figure). FHA also has not

established comprehensive performance indicators for the HECM portfolio and

has not regularly tracked key performance metrics, such as reasons for HECM

terminations and the number of distressed borrowers who have received

foreclosure prevention options. Additionally, FHA has not developed internal

reports to comprehensively monitor patterns and trends in loan outcomes. As a

result, FHA does not know how well the HECM program is serving its purpose of

helping meet the financial needs of elderly homeowners.

FHA has not conducted on-site reviews of HECM servicers since fiscal year 2013

and has not benefited from oversight efforts by the Consumer Financial

Protection Bureau (CFPB). FHA officials said they planned to resume the reviews

in fiscal year 2020, starting with three servicers that account for most of the

market. However, as of August 2019, FHA had not developed updated review

procedures and did not have a risk-based method for prioritizing reviews. CFPB

conducts examinations of reverse mortgage servicers but does not provide the

results to FHA because the agencies do not have an agreement for sharing

confidential supervisory information. Without better oversight and information

sharing, FHA lacks assurance that servicers are following requirements,

including those designed to help protect borrowers.

Why GAO Did This Study

Reverse mortgages allow seniors to

convert part of their home equity into

payments from a lender while still living

in their homes. Most reverse mortgages

are made under FHA’s HECM program,

which insures lenders against losses on

these loans. HECMs terminate when a

borrower repays or refinances the loan

or the loan becomes due because the

borrower died, moved, or defaulted.

Defaults occur when borrowers fail to

meet mortgage conditions such as

paying property taxes. These borrowers

risk foreclosure and loss of their homes.

FHA allows HECM servicers to offer

borrowers foreclosure prevention

options. Most HECM servicers are

supervised by CFPB.

GAO was asked to review HECM loan

outcomes and servicing and related

federal oversight efforts. Among other

objectives, this report examines (1) what

FHA data show about HECM

terminations and the use of foreclosure

prevention options, (2) the extent to

which FHA assesses and monitors the

HECM portfolio, and (3) the extent to

which FHA and CFPB oversee HECM

servicers. GAO analyzed FHA loan data

and FHA and CFPB documents on

HECM servicer oversight. GAO also

interviewed agency officials, the five

largest HECM servicers (representing

99 percent of the market), and legal aid

groups representing HECM borrowers.

What GAO Recommends

GAO makes eight recommendations to

FHA to, among other things, improve its

monitoring and assessment of the

HECM portfolio and oversight of HECM

servicers, and one recommendation to

CFPB to share HECM servicer

examination information with FHA. FHA

and CFPB generally agreed with the

recommendations.

View GAO-19-702. For more information,

contact

Alicia Puente Cackley at (202) 512-8678

[email protected]

Highlights of GAO-19-702, a report to

congressional requesters

Page i GAO-19-702 Reverse Mortgages

Letter 1

Background 5

HECM Defaults Have Increased, and Use of Foreclosure

Prevention Options Is Limited or Unknown 18

Weaknesses Exist in HECM Termination Data, Performance

Assessment, and Portfolio Monitoring 28

FHA’s Oversight of Servicers and Collaboration on Oversight

between FHA and CFPB Are Limited 37

CFPB Collects and Analyzes Consumer Complaints on Reverse

Mortgages, but FHA Does Not Use All Available Data 45

Housing Market Conditions, Exit of Large Bank Lenders, and

Policy Changes Help Explain the Decline in Reverse Mortgages

since 2010 54

Conclusions 60

Recommendations for Executive Action 61

Agency Comments and Our Evaluation 63

Appendix I Objectives, Scope, and Methodology 66

Appendix II Description of and Results for GAO’s Econometric

Model of Home Equity Conversion Mortgage Take-Up Rates 77

Appendix III Reported Home Equity Conversion Mortgage

Termination Reasons 92

Appendix IV Reported Home Equity Conversion Mortgage

Terminations and Defaults 93

Appendix V Comments from the Department of Housing

and Urban Development 95

Contents

Page ii GAO-19-702 Reverse Mortgages

Appendix VI Comments from the Consumer Financial Protection Bureau 99

Appendix VII GAO Contact and Staff Acknowledgments 100

Tables

Table 1: Summary of Selected Federal Housing Administration

(FHA) Foreclosure Prevention Options for Home Equity

Conversion Mortgage (HECM) Borrowers 23

Table 2: Summary of Mortgagee Optional Election Assignments,

Fiscal Years 2015–2018 24

Table 3: Number of FHA On-Site Reviews of Home Equity

Conversion Mortgage Servicers, Fiscal Years 2010–2018 38

Table 4: Number of Reverse Mortgage Complaints Received by

the Consumer Financial Protection Bureau (CFPB),

Calendar Years 2015–2018 47

Table 5: Reported Home Equity Conversion Mortgage

Terminations, Fiscal Years 2014–2018 68

Table 6: Descriptive Statistics, List of Variables Used in

Regression Analysis, 2004–2016 86

Table 7: Fixed-Effects Regression Estimates of HECM Take-Up

Rates, 2004–2016 87

Table 8: Reported Home Equity Conversion Mortgage (HECM)

Termination Reasons, Number and Percentage of Loans,

Fiscal Years 2014–2018 92

Table 9: Reported Home Equity Conversion Mortgage

Terminations and Defaults, by State, Number and

Percentage of Loans, Fiscal Years 2014–2018 93

Figures

Figure 1: Difference between Forward and Reverse Mortgages 6

Figure 2: Home Equity Conversion Mortgage Termination

Reasons and Repayment Alternatives 9

Figure 3: Status of Home Equity Conversion Mortgages, Fiscal

Years 1989–2018 10

Figure 4: Reported Home Equity Conversion Mortgage

Termination Reasons, Fiscal Years 2014–2018 19

Page iii GAO-19-702 Reverse Mortgages

Figure 5: Total Reported Servicer Advances for Terminated Home

Equity Conversion Mortgages, Fiscal Years 2014–2018 22

Figure 6: Types of Reverse Mortgage Complaints Received by the

Consumer Financial Protection Bureau, Calendar Years

2015–2018 49

Figure 7: Home Equity Conversion Mortgage (HECM) Originations

and Take-up Rates, Calendar Years 1989–2018 55

Abbreviations

CFPB Consumer Financial Protection Bureau

Fannie Mae Federal National Mortgage Association

FHA Federal Housing Administration

HECM Home Equity Conversion Mortgage

HERMIT Home Equity Reverse Mortgage Information

Technology

HUD Department of Housing and Urban Development

IPUMS NHGIS Integrated Public Use Microdata Series National

Historical Geographic Information System

OMB Office of Management and Budget

This is a work of the U.S. government and is not subject to copyright protection in the

United States. The published product may be reproduced and distributed in its entirety

without further permission from GAO. However, because this work may contain

copyrighted images or other material, permission from the copyright holder may be

necessary if you wish to reproduce this material separately.

Page 1 GAO-19-702 Reverse Mortgages

441 G St. N.W.

Washington, DC 20548

September 25, 2019

The Honorable Maxine Waters

Chairwoman

Committee on Financial Services

House of Representatives

The Honorable Denny Heck

House of Representatives

The aging of the U.S. population, the large share of seniors’ wealth held

in home equity, and the preference of many older adults to age in place

underscore the importance of knowing more about reverse mortgages

and the role they play for some senior homeowners. A reverse mortgage

is a type of loan that allows eligible seniors to convert part of their home

equity into payments from a lender while still living in their homes.

Congress authorized the Department of Housing and Urban Development

(HUD) to insure reverse mortgages to help meet the financial needs of

elderly homeowners.

The vast majority of reverse mortgages are made

under the Home Equity Conversion Mortgage (HECM) program

administered by HUD’s Federal Housing Administration (FHA).

HECMs

are originated and serviced by private FHA-approved lenders and

servicers. FHA insures these entities against losses on the loans and

charges borrowers premiums to help cover the potential cost of insurance

claims. While not involved in administering the HECM program, the

Consumer Financial Protection Bureau (CFPB) collects consumer

complaints about reverse mortgages and supervises nonbank

(nondepository institution) reverse mortgage lenders and servicers for

compliance with, and enforces violations of, federal consumer financial

protection laws.

Congress authorized the Home Equity Conversion Mortgage (HECM) program in 1988 by

adding Section 255 to Title II of the National Housing Act. See Housing and Community

Development Act of 1987, Pub. L. No. 100-242, § 417 (1988) (codified as amended at 12

U.S.C. § 1715z-20).

The reverse mortgage market includes some proprietary (non-FHA) products but is

currently dominated by HECMs. Proprietary reverse mortgages were outside the scope of

our review.

As discussed later in this report, the HECM market is currently dominated by nonbank

lenders and servicers.

Letter

Page 2 GAO-19-702 Reverse Mortgages

Reverse mortgage servicing involves a range of activities, such as

making payments to borrowers, providing monthly account statements,

monitoring loan balances, and responding to borrower inquiries. If a

borrower falls behind on property charges (for example, taxes and

homeowners insurance), servicers must generally temporarily pay them

on the borrower’s behalf (referred to in this report as servicer advances)

but may ultimately initiate foreclosure proceedings if the borrower does

not catch up.

In recent years, FHA has made program changes allowing

servicers to offer foreclosure prevention options—options for distressed

HECM borrowers to help delay or avoid foreclosure.

HECMs terminate—that is, the loan balance is paid off and the loan

ends—for a variety of reasons. For example, borrowers may choose to

repay the loan or refinance into a new HECM. Additionally, events such

as the borrower dying, moving, or defaulting—that is, not meeting

mortgage conditions—result in the loan becoming “due and payable.” In

some cases, generally when a borrower defaults, the lender may

foreclose on the borrower to obtain title to the property and sell the home

to satisfy the debt.

In these circumstances, the borrower may end up

being displaced from his or her home. Consumer advocacy organizations

have expressed concerns about an observed increase in HECM

foreclosures and servicing problems.

You asked us to review HECM loan outcomes and servicing and related

federal oversight efforts. This report examines (1) what FHA data show

about HECM terminations, servicer advances, and the use of foreclosure

prevention options; (2) FHA’s assessment and monitoring of HECM

portfolio performance, servicer advances, and foreclosure prevention

In addition to property taxes and homeowners insurance premiums, borrowers must pay

other property charges such as homeowners’ association or condominium fees and any

other special assessments that may be levied by municipalities or state law on the

property. See 24 C.F.R. § 206.205. As discussed later in this report, servicers offer some

borrowers repayment plans to help pay unpaid property charges. According to FHA policy,

no borrower may be given more than 60 months to repay the servicer advances. See

Department of Housing and Urban Development, Mortgagee Letter 2015-11 (Apr. 23,

2015).

In some cases, the borrower may be able to deed the property to the servicer to satisfy

the debt and avoid foreclosure, known as a deed-in-lieu of foreclosure.

Center for NYC Neighborhoods, Protecting Senior Homeowners from Reverse Mortgage

Foreclosure: Policy Brief (August 2017) and National Consumer Law Center, Examples of

Senior Homeowners Struggling with Ineffective and Inconsistent Servicing of HECM

Loans (October 2017).

Page 3 GAO-19-702 Reverse Mortgages

options; (3) FHA’s and CFPB’s oversight of HECM servicers; (4) how

FHA and CFPB collect, analyze, and respond to consumer complaints

about HECMs; and (5) how and why the market for HECMs has changed

in recent years.

To address the first and second objectives, we analyzed FHA data to

determine the number of and reasons for HECM terminations in fiscal

years 2014 through 2018.

We also analyzed FHA data on servicer

advances for unpaid property charges and other costs for HECMs in a

due and payable status. Additionally, we analyzed information from FHA

on HECM borrowers approved for selected foreclosure prevention

options. To assess the reliability of FHA’s data, we reviewed FHA

documentation, performed electronic testing on the data to check for

missing values and obvious errors, corroborated the data with other

available sources (such as published FHA reports), and interviewed

agency officials and FHA’s data system contractor about interpreting data

fields. We determined the data were sufficiently reliable for characterizing

loan terminations, servicer advances, and use of foreclosure prevention

options in the HECM program. We also reviewed FHA data and reports

and interviewed FHA officials to determine how the agency monitors and

analyzes the HECM portfolio, including the use of any performance

indicators or program evaluations. We compared FHA’s practices against

leading practices for assessing program performance, federal internal

control standards, and Office of Management and Budget (OMB) policies

and procedures on managing federal credit programs (OMB Circular A-

129).

To address the third objective, we reviewed FHA and CFPB policies and

procedures for overseeing HECM servicers and interviewed agency

officials with oversight responsibilities. We reviewed completed

examinations of HECM servicers to determine the extent to which the

The federal fiscal year begins on October 1 and ends on September 30 of each year.

We have previously identified performance goals and measures as important

management tools that can serve as leading practices for planning at lower levels within

federal agencies, such as individual programs or initiatives. For example, see GAO,

Veterans Justice Outreach Program: VA Could Improve Management by Establishing

Performance Measures and More Fully Assessing Risks, GAO-16-393 (Washington, D.C.:

Apr. 28, 2016); Performance Measurement and Evaluation: Definitions and Relationships,

GAO-11-646SP (Washington, D.C.: May 2, 2011). See also Standards for Internal Control

in the Federal Government, GAO-14-704G (Washington, D.C.: September 2014). See

also Office of Management and Budget, Policies for Federal Credit Programs and Non-

Tax Receivables, OMB Circular No. A-129 (revised January 2013).

Page 4 GAO-19-702 Reverse Mortgages

agencies have assessed and taken steps to enforce compliance with

servicing and consumer protection requirements. We compared FHA’s

oversight of HECM servicers to relevant parts of OMB Circular A-129.

Additionally, we interviewed FHA and CFPB officials to determine the

extent to which the agencies collaborate and share information on

oversight of HECM servicers. We compared their efforts to approaches

federal agencies use to enhance collaboration toward joint goals that we

identified in prior work.

To address the fourth objective, we analyzed all reverse mortgage

consumer complaints received by CFPB through its Consumer Complaint

Database from calendar years 2015 through 2018 to determine the

number of complaints by year, state, submission method (for example,

internet, phone, or email), and company. We also analyzed a random

generalizable sample of 100 reverse mortgage consumer complaints to

identify patterns in consumer-described issues about reverse mortgages.

We determined the CFPB data were sufficiently reliable for the analysis

we conducted by reviewing CFPB documentation, performing electronic

testing of the data, and interviewing CFPB officials about our

interpretation of data fields. To determine the extent to which FHA

collects and analyzes complaints, we reviewed the nearly 105,000

HECM-related calls received by the National Servicing Center from

calendar years 2015 through 2018. We also reviewed the 147 HECM-

related calls received by the FHA Resource Center during the same time.

However, we did not perform an analysis on these data similar to that

conducted on the CFPB data because of limitations in how the data were

collected. For example, data did not include information on whether the

call was a complaint or inquiry, and a large majority of the data did not

include information on who the caller was (e.g., a borrower, servicer, or

lender). Additionally, we reviewed CFPB and FHA policies and

procedures for collecting and addressing consumer complaints and for

incorporating consumer complaints into their oversight of HECM

servicers. Further, we interviewed officials from both agencies to better

understand their complaint processes. We compared CFPB’s and FHA’s

efforts against federal internal control standards and against criteria we

developed previously on leveraging related agency resources.

GAO, Managing for Results: Key Considerations for Implementing Interagency

Collaborative Mechanisms, GAO-12-1022, (Washington, D.C.: Sept. 27, 2012).

GAO-14-704G and GAO-12-1022.

Page 5 GAO-19-702 Reverse Mortgages

To address the fifth objective, we analyzed FHA data from calendar years

1989 (the start of the HECM program) through 2018 to identify trends in

the volume of HECM originations. Additionally, using FHA and Census

Bureau data, we calculated HECM take-up rates—the ratio of new HECM

originations to senior homeowners—from calendar years 2000 through

2017 to measure the extent to which senior homeowners participate in

the program. We also developed an econometric model using FHA,

Census Bureau, and other data to examine the relationship between

HECM take-up rates and a number of explanatory variables. We also

reviewed relevant research and interviewed academic and HUD

economists about other factors (for example, consumers’ perception of

reverse mortgages) that are difficult to directly include in the model but

that may influence HECM take-up rates. Appendix I contains a more

detailed description of our objectives, scope, and methodology. Appendix

II contains a description and results of our econometric model of factors

associated with HECM take-up rates.

We conducted this performance audit from July 2018 to September 2019

in accordance with generally accepted government auditing standards.

Those standards require that we plan and perform the audit to obtain

sufficient, appropriate evidence to provide a reasonable basis for our

findings and conclusions based on our audit objectives. We believe that

the evidence obtained provides a reasonable basis for our findings and

conclusions based on our audit objectives.

A reverse mortgage is a nonrecourse loan against home equity that does

not require mortgage payments as long as the borrower meets certain

conditions. In contrast to traditional forward mortgages, reverse

mortgages typically are “rising debt, falling equity” loans (see fig. 1). As

the borrower receives payments from the lender, the lender adds the

principal and interest to the loan balance, reducing the borrower’s home

equity. Also unlike traditional forward mortgages, reverse mortgages have

no fixed term.

Background

HECM Program

Page 6 GAO-19-702 Reverse Mortgages

Figure 1: Difference between Forward and Reverse Mortgages

Page 7 GAO-19-702 Reverse Mortgages

Prospective borrowers must meet a number of requirements to be eligible

for a HECM (see sidebar). The amount of money a borrower can receive

from a HECM—called the principal limit—depends on three things: (1) the

age of the youngest borrower or eligible nonborrowing spouse, (2) the

lesser of the appraised value of the home or the FHA mortgage limit as of

the date of loan closing (for calendar year 2019, $726,525), and (3) the

expected average interest rate.

The borrower can receive funds in a variety of ways—for example, as

monthly payments, a line of credit, a combination of the two, or a single

lump sum.

A large majority of borrowers choose the line of credit option.

The interest rate lenders charge is typically an adjustable rate, although

the lump sum option can be chosen at a fixed interest rate.

HECMs can terminate for a variety of reasons. For example, a borrower

may choose to repay the loan, refinance into a new HECM, or be required

to satisfy the debt because the loan has become due and payable. A

HECM becomes due and payable when a borrower dies, fails to retain

ownership of the home, or does not meet his or her mortgage obligations

such as paying property charges, meeting occupancy requirements, or

maintaining the home.

In these cases, the borrowers or heirs must

satisfy the debt or correct the condition that resulted in the due and

payable loan status.

In this report, we use the term default to refer to

According to Mortgagee Letter 2014-07, a nonborrowing spouse is defined as the

spouse, as determined by the law of the state in which the borrower and spouse reside or

the state of celebration, at the time of closing and who is not listed on the mortgage as a

borrower. In the case of a HECM for Purchase, the principal limit is based on the lesser of

the appraised value of the home or the sale price of the property being purchased. HECM

for Purchase is a program in which seniors may use a HECM to buy a new home. Unlike a

traditional HECM, a HECM for purchase is made against the value of the home to be

purchased, rather than against the value of a home the borrower already owns.

The monthly payments can be for as long as the borrower has the loan (tenure

payments) or for a fixed period (term payments). Borrowers cannot receive more than the

greater of 60 percent of the principal limit amount or the sum of mandatory obligations

plus 10 percent of the principal limit in the first year of the loan.

See 24 C.F.R. § 206.27(c). An example of failing to retain ownership of the home is

when one individual conveys title to the home to another individual, such as an aging

parent transferring ownership to an adult child.

According to FHA, servicers are required to obtain HUD approval prior to calling loans

due and payable for certain reasons, such as failure to meet occupancy requirements and

property charge defaults.

Home Equity Conversion Mortgage (HECM)

Eligibility Requirements

Key borrower requirements

• Be 62 years of age or older

• Own your home outright or have sufficient

equity in your home to secure the HECM

• Occupy the property as your principal

residence

• Participate in a consumer information

session given by a HECM counselor

approved by the Department of Housing

and Urban Development (HUD)

Key financial requirements

• Have financial resources to continue to

make timely payment of ongoing property

charges (e.g., taxes and insurance)

• Verification of income, assets, financial

obligations, and credit history

Source: GAO analysis of HUD information. | GAO-19-702

Page 8 GAO-19-702 Reverse Mortgages

HECMs that are due and payable because the borrower has not paid

property charges, met occupancy requirements, or maintained the home.

HECM borrowers (or their heirs) satisfy the debt by (1) paying the loan

balance using their own funds, (2) selling the home and using the

proceeds to pay off the loan balance, (3) providing a deed-in-lieu of

foreclosure (which transfers title for the property to the lender to satisfy

the debt), or (4) selling the home for at least the lesser of the loan

balance or 95 percent of the property’s appraised value (also known as a

short sale). According to FHA regulations, the borrowers or their heirs

generally have 30 days after being notified that the loan is due and

payable to satisfy the debt or bring the loan out of due and payable

status.

Servicers generally have 6 months to take first legal action to

initiate foreclosure from the date that they, as applicable, notified, should

have notified, or received approval from FHA that the HECM is due and

payable. According to FHA regulations, the borrower is generally allowed

to correct the condition that resulted in the due and payable loan status

and reinstate the loan, even after foreclosure proceedings have begun.

Figure 2 illustrates the reasons why HECMs terminate and how borrowers

typically satisfy the debt under various termination scenarios.

See 24 C.F.R. § 206.125(a)(2). If a loan becomes due and payable due to a reason

other than the death of the borrower, such as if a mortgage condition has not been met,

the lender has 30 days to notify FHA. In the case of borrower death, the lender has 60

days to notify FHA. After notifying and receiving approval (as applicable) from FHA that

the HECM can be called due and payable, the lender has 30 days to notify the borrower

(or their heirs). The borrower (or the heirs) has 30 days from the date of notice to engage

in one of the options noted or the lender may proceed with foreclosure.

24 C.F.R. § 206.125(a)(3).

Page 9 GAO-19-702 Reverse Mortgages

Figure 2: Home Equity Conversion Mortgage Termination Reasons and Repayment Alternatives

If the servicer experiences a loss because the loan balance exceeds the

recovery from selling the property, the lender can file a claim with FHA for

the difference. Additionally, when the loan balance reaches 98 percent of

the maximum claim amount (the lesser of the appraised value of the

home at origination or FHA’s loan limit), the lender can “assign” the loan

to FHA and file a claim for the full amount of the loan balance, up to the

maximum claim amount. Lenders can only assign HECMs in good

standing to FHA (that is, assignments can only be for HECMs not in a due

and payable status). FHA continues to service the assigned loans using a

contractor until the loans become due, either due to the death of the

borrower or for other reasons. Additionally, the FHA insurance

guarantees borrowers will be able to access their loan funds, even if the

loan balance exceeds the current value of the home or if the lender

experiences financial difficulty. Further, if the borrower or heir sells the

Page 10 GAO-19-702 Reverse Mortgages

home to repay the loan, he or she will not be responsible for any loan

amount above the value of the home.

As of the end of fiscal year 2018, FHA had insured over 1 million HECMs.

According to FHA data, these include an active HECM portfolio of

approximately 551,000 loans serviced by various FHA-approved

servicers, 79,000 FHA-assigned loans serviced by an FHA contractor,

and about 468,000 terminated loans (see fig. 3). HECM terminations have

exceeded new originations every year since fiscal year 2016, and the

number of HECMs assigned to FHA has grown substantially since fiscal

year 2014.

Figure 3: Status of Home Equity Conversion Mortgages, Fiscal Years 1989–2018

Throughout this report, we use the term “originations” to refer to HECMs originated by

lenders and subsequently approved for mortgage insurance (endorsed) by FHA.

Page 11 GAO-19-702 Reverse Mortgages

Note: The active Home Equity Conversion Mortgage portfolio consists of both privately owned and

FHA-assigned loans.

As of the end of fiscal year 2018, FHA’s total insurance-in-force for

HECMs (total insured mortgage balances outstanding) was roughly $100

billion. HECMs are held in two FHA insurance funds. HECMs originated

prior to fiscal year 2009 are in the General Insurance and Special Risk

Insurance Fund (roughly 27 percent of all HECMs), and those originated

in fiscal year 2009 and later are in the Mutual Mortgage Insurance Fund

(roughly 73 percent of all HECMs).

When the post-2008 HECM portfolio

became part of FHA’s Mutual Mortgage Insurance Fund, it also was

included in the fund’s capital ratio assessment and became subject to

annual actuarial reviews.

As we found in a November 2017 report,

subjecting HECMs to the annual actuarial review requirements has

improved the transparency of the program’s financial condition and has

highlighted the financial risks of the HECM portfolio to FHA.

According to FHA, the financial performance of the HECM portfolio has

been historically volatile, largely due to uncertainty in future home prices,

interest rates, and other factors. In recent years, FHA has responded with

several policy changes to help strengthen the portfolio’s financial

performance and mitigate risks. Because FHA’s projected losses on

HECMs depend on factors such as maximum claim amount, the length of

time the borrower stays in the home, changes in home prices, and

interest rates, most of FHA’s policy changes have been aimed at better

aligning expected revenues (charging borrowers premiums) with

expected costs (cash outflows due to paying insurance claims). For

example, FHA has made changes to insurance premiums and principal

Beginning with the fiscal year 2009 loan cohort, the Housing and Economic Recovery

Act of 2008 placed new HECMs in FHA’s Mutual Mortgage Insurance Fund.

The capital ratio is the fund’s economic value divided by the amortized insurance-in-

force. The National Housing Act requires an annual independent actuarial review of the

fund’s financial position. See 12 U.S.C. § 1708(a)(4). Each year, an independent actuarial

contractor conducts two separate actuarial reviews—one for forward mortgages and one

for HECMs—to estimate the economic value of the two portfolios. In a separate annual

report to Congress, FHA combines the findings of the forward mortgage and HECM

actuarial reviews to determine the capital ratio for the fund as a whole.

For additional information, see GAO, Federal Housing Administration: Capital

Requirements and Stress Testing Practices Need Strengthening, GAO-18-92

(Washington, D.C.: Nov. 9, 2017).

Page 12 GAO-19-702 Reverse Mortgages

limits, the most recent of which took effect in fiscal year 2018.

Effective

in fiscal year 2019, FHA also revised property appraisal practices for new

HECMs to guard against inflated property valuations.

According to

agency officials, FHA made this change to address appraisal bias

concerns identified in research by an economist in HUD’s Office of Policy

Development and Research.

The HECM market includes various participants. After a lender originates

a HECM, the loan must be serviced until it terminates. HECM lenders and

servicers must be FHA-approved and can be the same entity but often

are not. HECM lenders often sell the mortgage to another entity, which

FHA refers to as an investor, and this entity has the right to enforce the

mortgage agreement.

HECM servicers are typically third parties that

contract with lenders or investors but do not have ownership in the loans

they service. As previously discussed, HECM servicers perform a number

of functions, such as making payments to the borrowers and providing

monthly account statements. Servicers also must monitor borrower

compliance with various mortgage conditions and, if necessary,

For new HECMs, FHA currently charges an initial mortgage insurance premium of 2

percent at loan closing and, over the life of the loan, an annual mortgage insurance

premium of 0.5 percent of the outstanding mortgage balance. See Mortgagee Letter 2017-

12. Prior to that mortgagee letter, the initial mortgage insurance premium was 0.5 percent

and the annual insurance premium was 1.25 percent for amounts 60 percent or less of the

principal limit, and the initial mortgage insurance premium was 2.5 percent for amounts

greater than 60 percent of the principal limit.

Inflated property appraisals can negatively affect the financial performance of the HECM

program because the appraised value is used in determining the amount of home equity

available to the borrower and the amount of any insurance claim paid to a lender. Since

October 1, 2018, FHA has performed a property appraisal risk assessment on all HECMs

submitted for approval of insurance. Lenders are required to provide a second

independent property appraisal in cases where FHA determines there may be inflated

property valuations. See Department of Housing and Urban Development, Mortgagee

Letter 2018-06 (Sept. 28, 2018).

Kevin Park, “Reverse Mortgage Collateral: Undermaintenance or Overappraisal?”

Cityscape: A Journal of Policy Development and Research, vol. 19, no. 1 (2017): pp.7-28.

Servicing accompanies all mortgages, but the right to service a mortgage becomes a

distinct asset when contractually separated from the loan when the loan is sold or

securitized. Originators can service mortgages that they originate or purchase, or they can

sell the mortgages but retain the mortgage servicing rights. Servicers other than the

originator may also purchase mortgage servicing rights on securitized loans or may be

hired to service loans for others. For more information on nonbank mortgage servicers,

see GAO, Nonbank Mortgage Servicers: Existing Regulatory Oversight Could Be

Strengthened, GAO-16-278 (Washington, D.C.: Mar. 10, 2016).

HECM Market Participants

Page 13 GAO-19-702 Reverse Mortgages

communicate with borrowers about any violation of these conditions

(defaults) and, as appropriate, ways they can avoid being foreclosed on.

HECM servicers also transfer up-front and annual insurance premiums to

FHA each month and file claims with FHA for losses on insured HECMs.

In carrying out these duties, servicers are responsible for complying with

various requirements, including FHA regulations, policies, and

procedures, as well as federal consumer financial laws.

Historically, commercial banks, thrifts, and credit unions were the primary

lenders and servicers of mortgage loans. Following the 2007–2009

financial crisis and subsequent revisions to regulatory bank capital

requirements, banks reevaluated the benefits and costs of being in the

mortgage lending market, as well as retaining mortgages and the right to

service them. Since the financial crisis, some banks have exited or

reduced their mortgage lending and servicing businesses. This

development, among others, created an opportunity for nonbank

servicers to increase their presence in the mortgage market. Nonbank

issuers such as mortgage originators and servicers are not subject to the

same comprehensive federal safety and soundness standards as banks.

While banks offer a variety of financial products to consumers, nonbank

servicers are generally involved only in mortgage-related activities and do

not take deposits from consumers.

Almost all HECMs are originated, owned, and serviced by nonbank

entities:

• Lenders. According to FHA, in fiscal year 2018, 54 lenders originated

HECMs, including 49 nonbank entities and five banks.

• Investors. As of the end of fiscal year 2018, six investors (all nonbank

entities) and the Federal National Mortgage Association (Fannie Mae)

owned roughly 92 percent of the privately owned (non-FHA-assigned)

HECM portfolio, while the remaining 8 percent was owned by a

mixture of bank and nonbank entities.

• Servicers. Five nonbank entities serviced over 99 percent of the

privately owned HECM portfolio as of the end of fiscal year 2018. As

Fannie Mae is a government-sponsored enterprise that purchases mortgages from

lenders and either pools the loans into securities or holds them in its portfolio.

Page 14 GAO-19-702 Reverse Mortgages

previously noted, FHA has a contractor (also a nonbank entity) that

services FHA-assigned HECMs.

A number of federal agencies have roles in overseeing the reverse

mortgage market, including the following:

• FHA. Insures HECMs and administers the HECM program, including

issuing program regulations and enforcing program requirements.

FHA supplements regulations through additional policies, procedures,

and other written communications for the HECM program. For

example, FHA officials said the agency utilizes its Single Family

Housing Handbook, HECM handbook, and mortgagee letters to

communicate changes about the HECM program.

In 2013,

Congress enacted a law that allowed FHA to make changes to HECM

program requirements by notice or mortgagee letter in addition to

regulation.

Since then, FHA has made several policy changes to the

HECM program through mortgagee letters.

• CFPB. Supervises nonbank reverse mortgage lenders and servicers

for compliance with, and enforces violations of, federal consumer

financial protection laws.

CFPB can also issue regulations under the

federal consumer protection laws addressed specifically to protecting

consumers considering reverse mortgages.

Additionally, CFPB

examines entities for compliance with federal consumer financial laws

to obtain information about an institution’s compliance management

systems and procedures and to detect and assess risks to consumers

A mortgagee letter is a written communication to lenders and servicers about changes in

FHA operations, policies, and procedures. Mortgagee letters for the HECM program are

accessible at https://www.hud.gov/program_offices/housing/sfh/hecm/hecmml.

The Reverse Mortgage Stabilization Act of 2013, Pub. L. No. 113-29, § 2, 127 Stat. 509

(2013) (codified at 12 U.S.C. § 1715z-20(h)). According to the act, the HUD Secretary has

the discretion to make any additional or alternative requirements that are necessary to

improve the fiscal safety and soundness of the HECM program.

CFPB also has supervisory authority over insured depository institutions and credit

unions with more than $10 billion in assets for compliance with requirements of federal

consumer financial law. See 12 U.S.C. § 5515(a).

CFPB was created by the Dodd-Frank Wall Street Reform and Consumer Protection Act

and has rulemaking authority to implement provisions of federal consumer financial law

and enforcement authority to assess servicers’ compliance with various mortgage

servicing rules. See Consumer Financial Protection Act of 2010, Pub. L. No. 111-203, §§

1021, 1024, 124 Stat. 1376, 1979, 1987 (2010) (codified as amended at 12 U.S.C. §§

5511, 5514).

Federal Entities Involved

in Reverse Mortgage

Oversight

Page 15 GAO-19-702 Reverse Mortgages

and markets. Further, CFPB collects consumer complaints regarding

consumer financial products or services (including reverse mortgages)

and educates consumers about their rights under federal consumer

financial protection laws.

• Federal depository institution regulators. These regulators monitor

compliance with relevant laws and regulations, such as provisions of

the Federal Trade Commission Act and the Truth in Lending Act,

primarily through periodic examinations, for federally regulated

lenders that originate HECMs.

Several features and requirements of the HECM program provide

consumer protections to borrowers. For example, borrowers must

undergo preloan counseling, the program limits costs and fees lenders

can charge, and lenders must provide certain disclosures. In addition,

FHA has made several changes to the HECM program in recent years to

help borrowers who have defaulted due to unpaid property charges. As

previously discussed, if a HECM borrower does not pay his or her

property charges, FHA regulations generally require the servicer to pay

the property charges on the borrower’s behalf to help avoid a tax

foreclosure by the local authority and protect the investor’s and FHA’s

interest in the home. FHA regulations also allow servicers to charge

certain fees once a loan is called due and payable. These are typically

amounts related to attorney or trustee fees, property preservation, and

appraisal fees during the foreclosure process. The payments and fees

that servicers make on behalf of borrowers—referred to as servicer

advances in this report—are added to the loan balance and accrue

interest.

In 2010, HUD’s Office of Inspector General reported that HUD was not

tracking borrower defaults or servicer advances for the HECM program

Depository institutions include institutions chartered as commercial banks, savings

associations (or thrifts), or credit unions. The federal depository institution regulators are

the Board of Governors of the Federal Reserve System, Federal Deposit Insurance

Corporation, Office of the Comptroller of the Currency, and National Credit Union

Administration.

Consumer Protections and

Foreclosure Prevention

Options

Page 16 GAO-19-702 Reverse Mortgages

and made several recommendations to FHA.

To address these

recommendations, FHA took several steps. For example, in 2011, FHA

stopped the practice of allowing servicers to defer foreclosing on loans

that were in default due to unpaid property changes and issued a

mortgagee letter addressing how to handle these loans.

Additionally, in

September 2012, FHA announced the launch of a new data system for

the HECM program, the Home Equity Reverse Mortgage Information

Technology (HERMIT) system which would be used starting in October

2012.

With this new system, FHA combined former legacy systems that

had been used to collect insurance premiums, service FHA-assigned

loans, and process claims. According to FHA, adopting the HERMIT

system allowed FHA to better monitor and track the HECM portfolio in

real time and to automate insurance claim processing.

Finally, FHA modified program features to help minimize potential

borrower defaults and help strengthen borrower eligibility requirements.

For example, in 2013, FHA reduced the amount of equity borrowers could

generally withdraw during the first year from 100 to 60 percent of the

principal limit.

According to FHA, this change was designed to

encourage borrowers to access their equity slowly over time rather than

all at once to reduce risks to borrowers and FHA’s insurance fund. In

2015, the financial requirements for HECMs changed to include a

financial assessment of the prospective borrower prior to loan approval.

Department of Housing and Urban Development, Office of the Inspector General, HUD

Was Not Tracking Almost 13,000 Defaulted HECM Loans With Maximum Claim Amounts

of Potentially More Than $2.5 Billion, 2010-FW-0003 (Aug. 25, 2010). The report

recommended that FHA (1) discontinue the practice of allowing servicers to defer

foreclosure on loans in default due to nonpayment of taxes and insurance; (2) issue formal

guidance to servicers regarding loans in default due to nonpayment of taxes and

insurance; (3) develop and implement a plan to minimize the risk of future defaults due to

nonpayment of taxes and insurance; and (4) develop a data system to better track

defaults and amounts.

See Department of Housing and Urban Development, Mortgagee Letter 2011-01 (Jan.

3, 2011). Mortgagee Letter 2011-01 was rescinded and replaced by Mortgagee Letter

2015-11 (Apr. 23, 2015).

See Department of Housing and Urban Development, Mortgagee Letter 2012-17 (Sept.

11, 2012).

See Department of Housing and Urban Development Mortgagee Letter 2013-27 (Sept.

3, 2013).

See Department of Housing and Urban Development, Mortgagee Letter 2014-21 (Nov.

10, 2014) and Mortgagee Letter 2014-22 (Nov. 14, 2014). The financial assessment

requirements in these letters were effective as of April 27, 2015.

Page 17 GAO-19-702 Reverse Mortgages

FHA began requiring HECM lenders to look at the prospective borrower’s

credit history, income, assets, and financial obligations. Based on the

results of the financial assessment, the lender may require a set-aside for

the payment of property charges.

Additionally, FHA made several program changes to help distressed

HECM borrowers by allowing servicers to offer options to help borrowers

delay or in some cases avoid foreclosure if they are behind on paying

property charges. These foreclosure prevention options include

repayment plans, at-risk extensions, and extensions for low-balance

arrearage, as described later in this report.

FHA also has taken steps to

help nonborrowing spouses stay in their homes after a borrowing spouse

dies by deferring repayment of the HECM as long as the nonborrowing

spouse fulfills certain conditions.

In these cases, the servicer can assign

According to FHA regulations, HECM lenders may require what is referred to as a “life

expectancy set-aside” for the payment of certain property charges. See 24 C.F.R. §

206.205.

See Department of Housing and Urban Development Mortgagee, Letter 2015-11 (Apr.

23, 2015) and Mortgagee Letter 2016-07 (Mar. 30, 2016). According to Mortgagee Letter

2015-11, servicers may offer borrowers in default due to unpaid property charges

repayment plans to satisfy outstanding servicer advances made for unpaid property

charges or an extension of foreclosure time frames due to a borrower being “at-risk.”

According to the mortgagee letter, the servicer must determine if the borrower is eligible

for a repayment plan under FHA’s requirements. If a repayment plan is insufficient or

unsuccessful, borrowers may request an “at-risk” extension if the youngest living borrower

is at least 80 years old and has critical circumstances such as a terminal illness,

substantiated long-term physical disability, or a “unique” occupancy need (e.g., terminal

illness of family member receiving care in the residence). According to Mortgagee Letter

2016-07, servicers may delay submitting a due and payable request (and proceeding with

foreclosure) if the borrower owes less than $2,000 in unpaid taxes and insurance.

According to the letter, servicers must document either that (1) they contacted the

borrower, the borrower has expressed a willingness to repay, and the borrower is currently

attempting to make payments or (2) they were unable to contact the borrower but the

borrower is current on occupancy requirements and there is no indication the borrower

has vacated the property.

In 2013, a federal district court found that HUD had interpreted the HECM authorizing

statute incorrectly when it required loans to be due and payable upon a borrower’s death

even when a nonborrowing spouse was present in the home. See Bennett v. Donovan, 4

F.Supp. 3d 5 (D.D.C. 2013). Following the court decision, HUD has issued various

mortgagee letters establishing a process known as the mortgagee optional election

assignment, which allows nonborrowing spouses to avoid foreclosure and defer paying off

the loan balance.

Page 18 GAO-19-702 Reverse Mortgages

the HECM to FHA under what FHA refers to as the mortgagee optional

election assignment process.

Our analysis of FHA data found that about 272,155 HECMs terminated

from fiscal years 2014 through 2018.

Over that period, the number of

terminations rose from about 24,000 in fiscal year 2014 to a peak of

roughly 82,000 in fiscal year 2016, before declining to about 60,000 in

fiscal year 2018, as previously shown in figure 3.

As shown in figure 4, death of the borrower was the most common

reported reason for HECM terminations, followed by borrower defaults.

See Department of Housing and Urban Development, Mortgagee Letter 2015-15 (June

12, 2015). The mortgagee letter established the process, as well as time frames, for when

servicers need to submit information to FHA in order for the loan to be assigned to FHA.

The time frames were extended for certain assignments in an additional mortgagee letter.

See Department of Housing and Urban Development, Mortgagee Letter 2016-05 (Feb. 12,

2016).

See fig.3 for a comparison of active and terminated HECMs during this period.

We report on data starting in fiscal year 2014 because, as previously discussed, FHA

adopted a new system of record (HERMIT) for the HECM program in fiscal year 2013.

Unless otherwise noted, we analyzed data for the HECM program for the 5-year period

spanning fiscal years 2014–2018.

The HERMIT system records borrower death and conveyed title as a type of default for

administrative reasons related to the claims filing process. We did not categorize borrower

deaths or conveyed titles as defaults in order to differentiate those cases from loan

terminations resulting from living borrowers not meeting their mortgage obligations. We

treated them as separate loan termination reasons. We removed HECMs that had

previously been assigned to FHA and that terminated in fiscal years 2014 through 2018

from our termination analysis. We did not include them because, as discussed later in this

report, FHA’s practice is to not foreclose on FHA-assigned HECMs that default.

Accordingly, the denominator for our terminations analysis was 256,147 loans. For more

information on our termination analysis methodology, see app. I. For detailed information

on the number of loans and percentages by termination reason for fiscal years 2014–

2018, see app. III.

HECM Defaults Have

Increased, and Use of

Foreclosure

Prevention Options Is

Limited or Unknown

Death of the Borrower Is

the Most Common Reason

for HECM Terminations,

but Defaults Have

Increased in Recent Years

Page 19 GAO-19-702 Reverse Mortgages

The relative size of each termination category varied from fiscal years

2014 through 2018, with borrower defaults accounting for an increasing

proportion of terminations in recent years. In fiscal year 2018, borrower

defaults made up 18 percent of terminations.

Figure 4: Reported Home Equity Conversion Mortgage Termination Reasons, Fiscal Years 2014–2018

Note: Due to rounding, figures may not sum to 100 percent.

Specific results for all major termination categories over the 5-year period

were as follows:

 Death. About 34 percent of terminations (approximately 87,000 loans)

were due to the death of the borrower. Borrower deaths ranged from

roughly 29 percent to 40 percent of annual terminations over the 5-

year period.

 Default. About 15 percent of terminations (approximately 40,000

loans) were due to borrower defaults. As discussed in appendix IV,

this percentage varied widely by location and was highest in Michigan

(36 percent) and lowest in the District of Columbia (1 percent). About

29,000 defaults were for noncompliance with occupancy or residency

requirements, about 11,000 were for nonpayment of property charges,

Page 20 GAO-19-702 Reverse Mortgages

and about 200 were for not keeping the property in good repair. The

borrowers of these loans likely lost their homes through foreclosure or

a deed-in-lieu of foreclosure.

However, it is possible that some of

these borrowers would have ultimately lost their homes even if they

had not taken out a HECM. For example, as noted in CFPB’s 2012

report to Congress on reverse mortgages, some borrowers may have

taken out a HECM to help pay off their traditional mortgage rather

than as a way to pay for everyday expenses. In these cases, the

money borrowers received from their HECMs may have helped them

temporarily but may ultimately have been prolonging an unsustainable

financial situation.

In addition, some borrowers who did not meet

occupancy or residency requirements may have permanently moved

out of their homes—for example, to assisted living or nursing home

facilities.

Borrower defaults as a percentage of annual HECM terminations grew

from 2 percent of terminations in fiscal year 2014 to 18 percent in

fiscal year 2018. Noncompliance with occupancy requirements was

the primary cause of defaults each year, but unpaid property charges

represented a growing share. From fiscal years 2014 through 2018,

property charge defaults as a percentage of all defaults grew from 26

percent to 45 percent, and property charge defaults as a percentage

of all terminations grew from less than 1 percent to 8 percent.

• Loan balance repaid. About 9 percent of terminations (approximately

23,000 loans) were due to the borrower repaying the loan balance.

This category accounted for a declining share of terminations over the

5-year period, falling from 24 percent in fiscal year 2014 to 4 percent

in 2018.

Due to limitations in how short sales are coded in the HERMIT system, we could not

readily distinguish between short sales used after a borrower default and those used after

a borrower’s death. According to FHA, short sales are used in both circumstances.

See Consumer Financial Protection Bureau, Reverse Mortgages: Report to Congress

(Washington, D.C.: June 28, 2012).

Representatives from some legal aid organizations have cited instances where HECM

borrowers were reportedly residing in their homes but were foreclosed on for not

completing and returning annual occupancy certificates—a task that might become more

difficult for some borrowers as they age. The five HECM servicers we spoke with

described methods they use in addition to occupancy certificates to remind borrowers

about or verify compliance with occupancy requirements. These include phone calls,

reminder letters, and property inspections. We did not attempt to independently verify

servicers’ use of these methods or claims that borrowers residing in their homes were

foreclosed on for not returning occupancy certificates.

Page 21 GAO-19-702 Reverse Mortgages

• Refinanced. About 8 percent of terminations (about 20,000 loans)

were due to the borrower refinancing into a new HECM. This category

remained relatively stable over the 5-year period, accounting for about

5 percent to 10 percent of terminations each year.

• Borrower moved or conveyed title. About 3 percent of terminations

(approximately 8,000 loans) were due to the borrower either moving

or conveying title to the property to someone else. The percentage of

terminations in this category declined from 6 percent in fiscal year

2014 to 2 percent in fiscal year 2018.

• Unknown. For about 30 percent of terminations (roughly 78,000

loans), we were unable to readily determine a termination reason from

FHA’s data. Over the 5-year period, this category accounted for over

25 percent of terminations each year and reached a high of 39

percent in fiscal year 2018. We discuss challenges related to

determining termination reasons later in this report.

For HECMs that terminated in fiscal years 2014 through 2018, servicers

advanced almost $3 billion on behalf of borrowers for unpaid property

charges or various other costs that are charged once a loan becomes due

and payable.

The advances increased from $508 million in fiscal year

2014 to a peak of $731 million in fiscal year 2016, before declining to

$453 million in fiscal year 2018 (see fig. 5). This pattern aligns with the

overall trend in terminations, which also peaked in fiscal year 2016. Over

the 5-year period, advances for property charges made up 58 percent of

the total. The remaining 42 percent consisted of advances for other costs,

many of them foreclosure-related, such as attorney fees and appraisal

costs.

According to the HERMIT User Guide, corporate advances are any expense incurred by

the lender or servicer for any disbursement made on a borrower’s behalf after the HECM

becomes due and payable. These include advances for taxes, insurance, or condominium

or homeowners’ association dues (property charges); attorney and trustee fees; title,

recording, or sheriff fees; state and other taxes on deeds; special assessment liens; bank

attorney fees for bankruptcy proceedings; and appraisal fees.

HECM Servicers

Advanced Almost $3

Billion on Behalf of

Borrowers for Unpaid

Property Charges or Other

Costs

Page 22 GAO-19-702 Reverse Mortgages

Figure 5: Total Reported Servicer Advances for Terminated Home Equity

Conversion Mortgages, Fiscal Years 2014–2018

From fiscal years 2014 through 2018, HECM servicers advanced a total

of $567 million on behalf of living borrowers who defaulted on their

HECMs due to unpaid property charges.

For these loans, the median

advance was $7,007.

FHA officials told us HECM servicers continue to pay property charges after a borrower

dies to avoid a tax foreclosure on the property.

Additionally, the mean advance was $12,653. We also found that the amount advanced

for these loans varied widely. For example, advances at the 25th and 75th percentiles

were approximately $2,923 and $15,445, respectively.

Page 23 GAO-19-702 Reverse Mortgages

From April 2015 (the effective date of FHA’s current repayment plan

policy) through the end of fiscal year 2018, 22 percent of HECM

borrowers with overdue property charges had received repayment plans,

and FHA’s information on the use of other foreclosure prevention options

was limited.

As previously noted, property charge defaults and issues

surrounding nonborrowing spouses not being included on the mortgage

have been long-standing problems in the HECM program.

Since 2015,

FHA has made program changes to allow servicers to offer different types

of foreclosure prevention options to distressed HECM borrowers and

nonborrowing spouses of deceased borrowers (see table 1). These

options can help delay and, in some cases, avoid foreclosure.

Table 1: Summary of Selected Federal Housing Administration (FHA) Foreclosure Prevention Options for Home Equity

Conversion Mortgage (HECM) Borrowers

Foreclosure prevention

option

Description

Effective date

Mortgagee optional

election assignment

Allows a HECM servicer to assign the loan to FHA so that an eligible nonborrowing

spouse can stay in the home after the borrower’s death.

June 2015

Repayment plan

Allows a borrower in default due to unpaid property charges to repay them over a

maximum of 60 months, as long as the borrower meets certain financial criteria.

During this time, the HECM servicer delays foreclosure by requesting an extension of

foreclosure time frames from FHA.

April 2015

At-risk extension

Allows a HECM servicer to delay foreclosure for borrowers at least 80 years old who

are in default due to unpaid property charges and are experiencing a critical

circumstance such as a terminal illness or long-term disability. HECM servicers must

request from FHA an extension of foreclosure time frames, and supporting

documentation for extensions must be updated annually.

April 2015

Low-balance extension

Allows a HECM servicer to delay calling a loan due and payable when the borrower

owes less than $2,000 for unpaid property taxes or hazard insurance.

March 2016

Source: GAO analysis of FHA mortgagee letters. | GAO-19-702

According to officials from HUD’s Office of General Counsel, HUD does

not have the statutory authority to require servicers to provide HECM

As described later in this report, we conducted this analysis using data from April 23,

2015—the effective date of FHA’s current policy on repayment plans—to September 30,

2018.

See Department of Housing and Urban Development, Office of Inspector General, 2010-

FW-0003. See also Bennett v. Donovan, 4 F.Supp.3d 5 (D.D.C. 2013).

About One-Quarter of

HECM Borrowers with

Overdue Property

Charges Received

Repayment Plans, and

Use of Other Foreclosure

Prevention Options Is

Limited or Unknown

Page 24 GAO-19-702 Reverse Mortgages

borrowers foreclosure prevention options.

Our analysis of FHA data

found that servicers’ use of selected foreclosure prevention options for

HECM borrowers was limited or that FHA did not have readily available

information to assess the extent of use, as follows:

Mortgagee optional election assignments. According to information

generated by FHA, HECM servicers submitted 1,445 requests for

mortgagee optional election assignments from June 2015 (when FHA

made this option available) through September 30, 2018 (see table 2). In

total, FHA approved roughly 70 percent (1,013) of the requests and

denied the remaining 30 percent (432).

Table 2: Summary of Mortgagee Optional Election Assignments, Fiscal Years 2015–2018

Mortgagee optional election

assignment status June–September 2015

2016

2017 2018

Total

Percentage

Requested

429

434

293

289

1,445

100

Approved

411

176

211

215

1,013

Denied

258

432

Source: Federal Housing Administration. | GAO-19-702

Note: The mortgagee optional election assignment process went into effect in June 2015.

According to FHA officials, the top two reasons for denying mortgagee

optional election assignments were HECM servicers not meeting the

deadline for electing to pursue the assignment and not meeting the

Officials from HUD’s Office of General Counsel said that, in contrast, HUD does have

statutory authority to mandate foreclosure prevention options for forward mortgage

borrowers because the options mitigate financial losses to FHA. According to the officials,

section 230 of the National Housing Act provides FHA the statutory authority to require

loss mitigation activities for forward mortgages, such as forbearance or loan modification,

as an alternative to foreclosure. See 12 U.S.C. § 1710(a)(2); 12 U.S.C. § 1715u(a). The

officials said the difference in statutory authority stems partly from differences in how

forward and reverse mortgages work. Officials from HUD’s Office of General Counsel

noted that traditional loss mitigation options focus on helping borrowers make payments to

lenders and continue paying down their loan balances, which helps mitigate losses to FHA

by reducing foreclosures, insured loan amounts, and claim payments. The officials said

that because reverse mortgage borrowers generally do not make payments to their lender

and have increasing loan balances, options that help HECM borrowers delay or avoid

foreclosure generally do not mitigate losses to FHA.

Page 25 GAO-19-702 Reverse Mortgages

deadline to initiate the assignment.

FHA officials told us the third most

common reason for denial was a nonborrowing spouse not submitting

evidence of marketable title to the property or the legal right to remain in

the property for life within required time frames.

With respect to the 432

denials, FHA provided information indicating that as of May 31, 2019, 79

percent (342) of the associated loans had not terminated; 14 percent (62

loans) terminated because the loan balance had been paid off; and the

remaining 7 percent ended in foreclosure (22 loans), deed-in-lieu of

foreclosure (four loans), or short sale (two loans).

Estimating the universe of HECMs potentially eligible for mortgagee

optional election assignments is difficult because nonborrowing spouses

were not listed on loan documentation for HECMs originated prior to

August 4, 2014.

As a result, FHA does not know how many eligible

nonborrowing spouses could have, but did not, apply for the mortgagee

optional election assignment, or how many are potentially eligible to apply

for it in the future. FHA officials told us they have relied on an industry

association and HECM servicers to estimate how many nonborrowing

spouses may be associated with pre-August 2014 HECMs.

Mortgagee Letter 2015-15 (effective June 12, 2015) details HECM servicer requirements

for the mortgagee optional election assignment process and specifies two main time

frames servicers must meet. First, HECM servicers must elect to pursue the assignment

within 120 days of the borrower’s death. Second, if the servicer elects to pursue

assignment and determines that the nonborrowing spouse and the HECM meet eligibility

requirements, it must initiate the assignment to FHA within 120 days of the election

decision. Mortgagee Letter 2016-05 (effective Feb. 12, 2016) allows servicers to request a

60-day extension for the nonborrowing spouse to demonstrate that legal title or the legal

right to remain in the house has been secured and for the servicer to complete its

assessment.

Mortgagee Letter 2015-15 states that in order to be an eligible surviving nonborrowing

spouse, an individual, among other things, must have or obtain this information within 90

days following the death of the borrowing spouse.

According to officials, FHA has required lenders to collect nonborrowing spouse

information for HECMs originated on or after August 4, 2014. See Department of Housing

and Urban Development, Mortgagee Letter 2014-07 (Apr. 25, 2014).

In September 2018, representatives of the National Reverse Mortgage Lenders

Association told us the association estimated that there were about 8,600 outstanding

HECMs that potentially had an associated nonborrowing spouse not originally named on

the loan. The association estimated this number by asking its members to identify loans

originated before August 2014 that had only one borrower but reported being married at

the time of loan origination. The 8,600 figure is a rough estimate because, among other

things, it is not known whether the borrowers are still married and not all reverse mortgage

servicers are members of the association.

Page 26 GAO-19-702 Reverse Mortgages

FHA officials told us they sent letters to borrowers with FHA-assigned

HECMs that were originated prior to August 4, 2014, to inform them of the

mortgagee optional election process and ask them to self-identify whether

there was a nonborrowing spouse associated with their loan. FHA officials

also noted they were drafting a similar letter for servicers to send to

borrowers with HECMs not assigned to FHA. As of August 2019, the

servicer letter was still in draft form, pending completion of an ongoing

internal review of FHA’s mortgagee optional election assignment

processes and the related time frames. FHA officials said once the

ongoing review is complete, they anticipated that FHA would issue a new

mortgagee letter with revised time frames that would afford both HECM

servicers and borrowers more time to meet FHA requirements for

mortgagee optional election assignments.

Repayment plans. Our analysis of FHA data showed that 22 percent of

borrowers with property charge defaults were granted a repayment plan

from April 2015 (the effective date of FHA’s current repayment plan

policy) through the end of fiscal year 2018. All five legal aid organizations

we interviewed said the availability of repayment plans was a top

concern. For example, for some of their clients, repayment plans were

unavailable because the borrowers did not meet certain financial

requirements. In contrast, representatives of the top five HECM servicers

told us they generally do offer repayment plans when feasible to help

borrowers delay or avoid foreclosure. Servicers we interviewed noted that

while repayment plans can delay or avoid foreclosure, they are rarely

successful in the long-run and borrowers in such plans often miss

payments. Servicers said the same reasons that typically contribute to

initial defaults also explain why repayment plans are rarely successful.

For example, borrowers on limited incomes may struggle to pay

increasing property tax and insurance costs or may fall behind on

property charges when the death of a spouse reduces their income.

At-risk extensions. Our analysis of FHA data found that from April 2015

(the effective date of FHA’s at-risk extension policy) through the end of

fiscal year 2018, about 2 percent of borrowers with property charge

defaults received an at-risk extension. To grant an at-risk extension, FHA

requires HECM servicers to provide valid documentation that the

youngest living borrower is at least 80 years of age and has critical

circumstances such as a terminal illness, long-term physical disability, or

a unique occupancy need (for example, terminal illness of family member

Page 27 GAO-19-702 Reverse Mortgages

living in the home).

Representatives from one legal aid organization told

us that some HECM servicers have straightforward requirements for the

documentation borrowers must submit to obtain an at-risk extension,

while others do not. Representatives from another legal aid organization

said that meeting FHA’s annual renewal requirement for at-risk

extensions was challenging for some borrowers because they have to

submit documentation to HECM servicers every year as they age and

continue to struggle with serious health issues or disabilities.

Low-balance extensions. FHA officials told us they do not track how

often HECM servicers use the option to delay calling a loan due and

payable if the borrower has unpaid property charges of less than $2,000.

Our analysis of FHA data on servicer advances found that approximately

8,800 HECMs that terminated in fiscal years 2014 through 2018 had

unpaid property charges of less than $2,000 at the time of termination.

Some of these HECMs may have been eligible for a low-balance

extension when they terminated. Representatives from one legal aid

organization said they represented a HECM borrower who was at risk of

foreclosure for having 27 cents in unpaid property charges. HECM

servicers told us they use the low-balance extension option to varying

degrees. For example, representatives from one servicer said the servicer

follows instructions from the entity that owns the HECM and, in some

cases, the owners of the loan do not want to offer the low-balance

extension to the borrower. In these cases the servicer calls the loan due

and payable for any amount in unpaid property charges and initiates the

foreclosure process in accordance with FHA regulations. Another HECM

servicer told us it tries to use the low-balance extension every time a

borrower owes less than $2,000 in unpaid property charges.

See Department of Housing and Urban Development, Mortgagee Letter 2015-11 (Apr.

23, 2015).

To conduct this analysis we requested data on total advances, including data on unpaid

property charges, from FHA’s contractor that administers the HERMIT system.

Additionally, we requested borrower death dates to determine unpaid property charges

before and after HECM borrowers’ deaths. We calculated the total number of loans with

unpaid property charges of less than $2,000 for living borrowers by removing any loans for

which the borrower had died.

Page 28 GAO-19-702 Reverse Mortgages

Since fiscal year 2013, FHA has used the HERMIT system to collect data

on the servicing of HECMs, but the system does not contain

comprehensive and accurate data about the reasons why HECMs

terminate, a key servicing event. According to the HERMIT User Guide,

servicers should provide a reason in HERMIT when they terminate a

HECM.

However, as noted previously in figure 4, for about 30 percent of

HECM terminations from fiscal years 2014 through 2018 (roughly 78,000

loans), we were unable to determine the reason for termination.

Specifically, for these loans we could not identify in HERMIT any

associated borrower death or default, or evidence that the borrower

repaid, refinanced, moved, or conveyed title. Instead, these loans were

coded as terminating for “other reasons” or coded based on how the debt

was satisfied rather than an actual termination reason.

The HERMIT User Guide provides a list of termination codes available in

the system, but the list and guide have shortcomings that limit analysis of

HECM terminations. First, the list includes codes servicers can use to

indicate that a loan terminated for “other reasons,” but the guide does not

specify what these other reasons are. However, servicers have been

using the “other reasons” code increasingly over the past 5 years. We

asked servicers how they used the “other reasons” code and found

inconsistency in and uncertainty about its use. For example, servicers’

responses ranged from not using it at all, to using it when they did not

Home Equity Conversion Mortgage Servicer Provider HERMIT User Guide, Servicing

Module, version 2.17 (September 2018). The HERMIT User Guide refers to termination

transactions, and servicers are required to input a termination transaction when the loan

ends. For simplicity, we use the phrase termination codes in this report.

Of these approximately 78,000 loans, HECM servicers coded about 55,000 (71 percent)

as “terminate-other” in HERMIT. For the remaining 23,000 loans (29 percent), HECM

servicers used codes that could be associated with more than one termination reason,

such as deed-in-lieu of foreclosure, foreclosure, or short sale—actions which could apply

to terminations resulting from the borrower dying, defaulting, or moving. For more

information on our termination analysis methodology, see app. I.

Weaknesses Exist in

HECM Termination

Data, Performance

Assessment, and

Portfolio Monitoring

FHA Lacks

Comprehensive Data on

Reasons for HECM

Terminations

Page 29 GAO-19-702 Reverse Mortgages

intend to file an insurance claim with FHA, to not being sure under what

circumstances they used it.

Second, the list of termination codes consists of both reasons for

termination and descriptions of how the debt was satisfied.

As a result,

the final status code of some loans in HERMIT shows only the way in

which the debt was satisfied—for instance, a deed-in-lieu of foreclosure,

foreclosure, or short sale. These codes could apply to terminations

resulting from the borrower dying, defaulting, or moving and do not

ultimately provide a specific reason for loan termination.

FHA officials were unaware of any proxy variables that we could use to

help identify the underlying termination reasons for these loans. The

officials said the termination reasons are available on an individual loan

basis in the HERMIT system but not in an extractable form.

discussed later in this report, FHA does not regularly track and report on

HECM termination reasons, due partly to this system limitation.

The limitations in FHA’s data are inconsistent with federal internal control

standards, which require agencies to use quality information to achieve

their objectives.

To meet this internal control standard, agencies can

obtain relevant data that are reasonably free from error and bias and

evaluate sources of data for reliability. FHA’s annual report to Congress

states that the HECM program helps seniors remain in their homes and

A servicer may not need to file an insurance claim for a terminated HECM if, for

example, the borrower or the heirs sell the home for an amount that fully pays off the loan

balance.

For example, the list includes termination codes for borrower deaths, moves,

repayments, and refinances. While there is no termination code for defaults, defaulted

loans have other identifiers and can be cross-matched to terminated loans. The list also

includes a “terminate-other” code to be used when servicers initiate termination for “other

reasons.” Finally, the list includes codes for deed-in-lieu of foreclosure, foreclosure, and

short-sale, which indicate how the debt was satisfied and do not represent actual

termination reasons.

FHA officials told us that they do not believe these terminated HECMs were the result of

borrower defaults because defaults must be recorded in HERMIT. We used separate

default reports from HERMIT to identify all loans with a default as part of our termination

analysis; for more information see app. I.

GAO-14-704G.

Page 30 GAO-19-702 Reverse Mortgages

age in place.

However, without comprehensive and accurate data on

HECM terminations, FHA does not have a full understanding of loan

outcomes—information FHA and Congress need in order to know how

well the HECM program and FHA’s policies are working to help seniors

age in place.

While FHA has taken steps to improve the performance of the HECM

program in recent years, it has not incorporated key elements of

performance assessment into its management of the program. We have

previously reported that a program performance assessment contains

three key elements: program goals, performance metrics, and program

evaluations.

Performance assessment can provide important

information about whether, and why, a program is working well or not.

Additionally, OMB Circular A-129 states that agencies must establish

appropriate performance indicators for federal credit programs, such as

the HECM program, and that such indicators should be reviewed

periodically.

It states further that agency management structures should

clearly delineate accountability and responsibility for defining performance

indicators and monitoring and assessing program performance.

We found limitations in FHA’s performance assessment of the HECM

program, specifically a lack of performance indicators and recent program

evaluations:

Lack of HECM performance indicators. According to HUD’s strategic

plan for fiscal years 2018–2022 and the agency’s most recent annual

performance report, the HECM program falls under the strategic goal of

advancing economic opportunity and the strategic objective of supporting

See Department of Housing and Urban Development, Annual Report to Congress

Regarding the Financial Status of the FHA Mutual Mortgage Insurance Fund, Fiscal Year

2018 (Washington, D.C.: Nov. 15, 2018).

Program goals communicate what the agency proposes to accomplish and allow

agencies to assess or demonstrate the degree to which those desired results were

achieved. Performance metrics are concrete, objective, observable conditions that permit

the assessment of progress made toward the goals. Program evaluations are individual

systematic studies conducted periodically or on an ad hoc basis to assess how well a

program is working, typically relative to its objectives. For example, see GAO-16-393 and

GAO-11-646SP.

Office of Management and Budget, Policies for Federal Credit Programs and Non-Tax

Receivables, OMB Circular No. A-129 (revised January 2013).

FHA’s Performance

Assessment Has

Limitations

Page 31 GAO-19-702 Reverse Mortgages

fair, sustainable homeownership and financial viability.

The strategic

plan and annual performance report include some strategies for achieving

this objective, such as modernizing FHA underwriting guidelines, lending

standards, and servicing protocols to serve the needs of borrowers,

protect taxpayers, and ensure the sustainability of FHA’s program.

However, none of the six performance indicators associated with this

strategic objective and discussed in the strategic plan or corresponding

performance report are HECM-specific. Four of the indicators focus on

FHA-insured forward mortgages.

Another indicator focuses on

construction of manufactured housing.

The remaining indicator—

maintaining a capital reserve ratio for FHA’s Mutual Mortgage Insurance

Fund that meets or exceeds the statutory minimum requirement—

encompasses both forward mortgages and HECMs but does not provide

specific information about HECM loan outcomes, risk factors, or loan

characteristics.

Additionally, FHA’s annual reports to Congress on the financial status of

the insurance fund contain multiple tables of HECM data but limited

information on loan outcomes. For example, among other things, the

fiscal year 2018 report provides the number of new HECM originations,

the average age of new borrowers, the amount of HECM insurance

See Department of Housing and Urban Development, Strategic Plan 2018–2022

(Washington, D.C.: May 2019) and Fiscal Year 2020 Annual Performance Plan and Fiscal

Year 2018 Annual Performance Report (Washington, D.C.: Mar. 22, 2019).

The four indicators are (1) early payment default rate, (2) percentage of new home-

purchase mortgages to first-time homebuyers, (3) FHA’s market share of single-family

mortgage originations, and (4) percentage of mortgages with higher-risk features

(including cash-out refinance loans and home purchase loans with down-payment

assistance or debt-to-income ratios greater than 50 percent). According to FHA, an early

payment default occurs when a borrower becomes 90 days delinquent within the first six

mortgage payments. FHA considers the percentage of loans in early payment default

status to be an important indicator of the credit quality of new mortgage originations. The

debt-to-income ratio represents the percentage of a borrower’s gross monthly income that

goes toward monthly debt payments.

The indicator is the number of alternative construction letters issued by HUD’s Office of

Manufactured Housing Programs.

As previously noted, since 2009, HECMs have been subject to FHA’s annual actuarial

review requirements and are included in estimates of the Mutual Mortgage Insurance

Fund’s capital ratio. The actuarial reviews provide estimates of the HECM portfolio’s

economic value and information on the program’s financial condition. They do not provide

detailed information on HECM loan outcomes.

Page 32 GAO-19-702 Reverse Mortgages

claims paid, and estimates of the HECM portfolio’s capital position.

However, the report does not include other information that would provide

insight into loan outcomes, such as the percentage of HECM terminations

due to borrower defaults, the proportion of active HECMs with delinquent

property charges, or the percentage of distressed HECM borrowers who

have received foreclosure prevention options.

Limited program evaluations. The last comprehensive evaluations of

the HECM program were done in 2000 and 1995.

Officials said they

were in the planning phase for a new evaluation of the HECM program

but had not set a start date and did not expect the evaluation to include

an analysis of reasons for HECM terminations or the use of foreclosure

prevention options for borrowers in default. Instead, the officials told us

the evaluation would focus on the impact of an FHA policy change

implemented in 2015 that requires prospective HECM borrowers to

undergo a financial assessment to evaluate their ability to pay ongoing

property charges. While financial assessments could help reduce tax and

insurance defaults, and ultimately foreclosures, they only apply to new

HECMs issued on or after the effective date of the policy (April 27, 2015)

and are not relevant to other HECMs within the portfolio. Therefore, for

most of the HECM portfolio, an equally important consideration is the

impact of FHA’s policy changes that created foreclosure prevention

options for distressed borrowers. As previously noted, borrower defaults

have accounted for an increasing proportion of terminations in recent

years, and in fiscal year 2018, borrower defaults made up 18 percent of

terminations. Expanding the program evaluation to include the impact of

foreclosure prevention options would provide a more complete picture of

how well FHA is reducing defaults in the HECM portfolio and helping

HECM borrowers.

FHA officials acknowledged the need for more performance assessment

of the HECM program. The officials said their recent focus has been on

See Department of Housing and Urban Development, Annual Report to Congress

Regarding the Financial Status of the FHA Mutual Mortgage Insurance Fund, Fiscal Year

2018 (Washington, D.C.: Nov. 15, 2018).

See Abt Associates, Evaluation Report of FHA’s Home Equity Conversion Mortgage

Insurance Demonstration, a report prepared for the Department of Housing and Urban

Development (Washington, D.C.: Mar. 31, 2000) and Department of Housing and Urban

Development, Office of Policy Development and Research, Evaluation of the Home Equity

Conversion Mortgage Insurance Demonstration: A Report to Congress (Washington, D.C.:

Mar. 15, 1995).

Page 33 GAO-19-702 Reverse Mortgages

financial aspects of the program, in particular losses associated with

insurance claims. According to the FHA Commissioner, a key challenge

for the HECM program is that FHA has historically administered it without

a designated program head. The 2000 program evaluation noted that

lenders and servicers found it frustrating that FHA did not have one

person with responsibility for the HECM program. Further, the 2000

program evaluation noted that the division of responsibility for the

program fell across many offices and that it was hard to find senior

managers with a sense of ownership for the HECM program. In January

2019, an economist from HUD’s Office of Policy Development and

Research transferred to the Office of Housing (which includes FHA) to

serve as a Senior Advisor to the Deputy Assistant Secretary for Single

Family Programs, with a focus on the HECM program.

Without more comprehensive performance indicators and program

evaluations, FHA lacks information that could be useful for monitoring the

effects of recent policy changes and may be missing evidence of the

need for further program improvements. Additionally, in the absence of

performance indicators and reporting, FHA and Congress lack insight into

how well the HECM program is helping senior homeowners.

According to OMB Circular A-129, agencies must have monitoring,

analysis, and reporting mechanisms in place to provide a clear

understanding of a program’s performance.

The circular says these

mechanisms should be sufficiently flexible to perform any analysis

needed to respond to developing issues in the loan portfolio. However,

we found shortcomings in FHA’s internal reporting. We also found that

FHA had not analyzed the implications of its foreclosure prioritization

process for FHA-assigned loans.

Internal reporting for the HECM program is limited. Although FHA

adopted the HERMIT system to improve oversight of the HECM portfolio,

it has not used program data to regularly report key loan performance

information—for example, HECM termination reasons, servicer advances,

Office of Management and Budget, Policies for Federal Credit Programs and Non-Tax

Receivables, OMB Circular No. A-129 (revised January 2013).

FHA’s Internal Reporting

and Analysis for the

HECM Portfolio Have

Shortcomings

Page 34 GAO-19-702 Reverse Mortgages

and use of foreclosure prevention options.

FHA officials said they have

been more focused on the analysis and reporting of claims and other

financial data for the HECM program. However, according to OMB

Circular A-129, effective reporting provides accurate, timely information

on program performance, early warnings of issues that may arise, and

analytics to drive decision-making.

FHA has generated some reports from HERMIT to help oversee the

HECM portfolio, but it has been slow to develop regular and

comprehensive reporting mechanisms. FHA officials told us that while

data on defaults and use of foreclosure prevention options have generally

been available in HERMIT since 2015, FHA was unable to obtain reports

on these topics until the summer of 2018 because of contract funding

limitations. FHA officials said that starting in September 2018, FHA began

receiving regular reports from its HERMIT system contractor on issues

such as HECMs assigned to FHA; HECM origination, assignment, and

termination activity by month; summary information on the number and

dollar amount of HECMs originated each year; and HECMs with a default

date. Additionally, around the same time, FHA requested and received ad

hoc reports (one-time reports created for specific purposes) from the

contractor that included spreadsheets of all active HECMs with a

repayment plan and all active HECMs for which there was an identified

nonborrowing spouse.

FHA officials said the purpose of the reports generated from HERMIT is

to help FHA better manage HECM program performance. However, our

review of these regular and ad hoc reports found that many are lists of

loans that meet certain criteria and do not provide summary statistics that

could be used to readily identify patterns or trends in metrics, such as the

number of or reasons for HECM terminations or use of different

foreclosure prevention options. The reports require additional analysis to

generate meaningful management information. According to OMB

Circular A-129, graphics, tables, and trend analysis that compare

performance over time and against expectations and other information

can provide critical context for understanding program performance.

FHA implemented the HERMIT system, which captures data on defaults and advances,

partly in response to a 2010 Office of Inspector General recommendation, but it has not

regularly tracked and reported this information. Tracking defaults and servicer advances is

critical to understanding how many borrowers are unable to meet their mortgage

obligations and how far behind they are on property charges, both of which can affect the

amount of insurance claims FHA may need to pay in the future.

Page 35 GAO-19-702 Reverse Mortgages

Further, the circular says dashboards (easy-to-comprehend summaries of

key quantitative and qualitative information) and watch lists are tools that

can help all levels of the organization receive appropriate information to

inform proactive portfolio management and ensure program decisions are

informed by robust analytics.

FHA’s lack of analysis and internal reporting on HECM termination

reasons hampered the agency’s ability to respond to a 2017 Freedom of

Information Act request about the number of and reasons for HECM

foreclosures.

FHA’s response contained data showing that over 99

percent of HECM foreclosures occurring from April 2009 through

December 2016 resulted from the death of the borrower. However, FHA

officials told us they subsequently looked more closely into the issue and

redid the analysis using more reliable and updated information from

January 2013 through December 2017. The revised analysis showed that

61 percent of foreclosures over that period were due to borrower deaths,

37 percent were due to borrower defaults, and 2 percent were due to

conveyance of title. If FHA had regular and meaningful management

information about HECM terminations, it could have initially responded to

the 2017 request with more reliable information.

FHA officials told us that HERMIT is an accounting system to process

HECM claims and has limitations as a broader portfolio monitoring tool.

However, our analysis of HERMIT data and reports generated by FHA’s

HERMIT contractor suggest that the system can be used for this broader

purpose. Without more robust program analysis and internal reporting,

FHA is not well positioned to detect and respond to any emerging issues

and trends in the HECM portfolio. As previously discussed, these trends

include growing numbers of HECMs entering default and an increasing

number of loans being assigned to FHA.

FHA has not evaluated its foreclosure prioritization process for FHA-

assigned HECMs. As previously noted, FHA-assigned loans are a

growing part of the HECM portfolio. According to FHA officials, the

agency generally does not foreclose on borrowers whose HECMs have

been assigned to FHA and who are in default due to unpaid property

The 2017 request came from a consumer group and a legal aid organization. The

Freedom of Information Act was enacted in 1966 and requires federal agencies to provide

the public with access to government records and information to facilitate an informed

public and accountable government. See 5 U.S.C. § 552.

Page 36 GAO-19-702 Reverse Mortgages

charges.

According to FHA, the properties associated with these loans

are typically occupied. FHA officials said the agency prioritizes processing

foreclosures on assigned HECMs for which the property is vacant

(because the borrower passed away, for example). FHA officials said that

prioritizing foreclosure processing for those loans and delays by the

Department of Justice in completing those foreclosures has effectively

resulted in few foreclosures on assigned loans with property charge

defaults. However, FHA regulations state that servicers generally must

initiate foreclosure within 6 months of calling a loan due and payable due

to a death or default (if the borrower or heirs have not yet paid the debt

off).

FHA’s prioritization of processing vacant properties for foreclosure and

generally not foreclosing on FHA-assigned HECMs with a property

charge default raises issues and potential risks that FHA has not fully

analyzed. First, defaulted borrowers whose loans are privately owned

(that is, have not been assigned to FHA) face a greater risk of foreclosure

than defaulted borrowers with FHA-assigned loans. According to a

representative from one HECM servicer we interviewed, FHA’s practice is

unfair because it treats HECM borrowers inconsistently. Second, FHA’s

foreclosure prioritization processing may create a financial incentive for

HECM borrowers with assigned loans to not pay their property charges,

which, in turn, can have negative financial consequences for FHA,

localities, and taxpayers. For example, because FHA does not foreclose

on assigned loans in tax and insurance default, FHA advances tax and

insurance payments on behalf of the borrower and adds them to the loan

balance to secure and maintain its first-lien position on the mortgaged

property.

This makes it more likely that the loan balance will increase to

a point that it exceeds the value of the home. When the borrower dies or

vacates such a property, FHA may not be able to recoup the loan balance

in a foreclosure sale, resulting in a loss to the insurance fund.

The 2017 actuarial review of FHA’s Mutual Mortgage Insurance Fund noted this as well.

The actuary’s report states that FHA does not foreclose on assigned HECMs in tax and

insurance default. See Pinnacle Actuarial Resources, Fiscal Year 2017 Independent

Actuarial Review of the Mutual Mortgage Insurance Fund, a report prepared for the

Department of Housing and Urban Development (Bloomington, Ill.: Nov. 10, 2017).

See 24 C.F.R. § 206.125(d).

See Pinnacle Actuarial Resources, Fiscal Year 2017 Independent Actuarial Review of

the Mutual Mortgage Insurance Fund.

Page 37 GAO-19-702 Reverse Mortgages

As of August 2019, FHA had not evaluated the various risks of generally

not foreclosing on assigned HECMs with property charge defaults. As a

result, FHA does not know how its process for prioritizing foreclosures for

assigned loans affects the HECM portfolio, HECM borrowers,

neighborhoods, and FHA’s insurance fund.

FHA’s oversight of HECM servicers is limited. FHA requires HECM

servicers, among other things, to inform borrowers of their loan status,

including any conditions resulting in a loan becoming due and payable; to

notify struggling borrowers of the availability of housing counseling and

foreclosure prevention options; to inform surviving nonborrowing spouses

of conditions and requirements for the deferral period; and to manage the

transfer of loan servicing from one entity to another. These requirements

are identified in FHA regulations, handbooks, and mortgagee letters.

properly implemented, these requirements can help ensure that HECM

borrowers and nonborrowing spouses are aware of their mortgage

responsibilities, options for resolving situations that can result in

foreclosure, and who to contact with loan servicing questions. FHA

officials said they maintain communication with HECM servicers,

including through an industry working group, about their compliance with

FHA requirements. The officials also noted that FHA conducts reviews of

due and payable requests and insurance claims, which can include

checks for some of the requirements discussed above. However, FHA

has not performed comprehensive on-site reviews of HECM servicers’

compliance with program requirements since fiscal year 2013 and does

not have current procedures for conducting these reviews.

See e.g. 24 C.F.R. § 206.125(a). Other servicing requirements can be found in HUD

Handbook 4000.1 (FHA’s Single Family Housing Policy Handbook), HUD Housing

Handbook 4235.1 (the HECM Handbook), and various mortgagee letters accessible at

https://www.hud.gov/program_offices/housing/sfh/hecm/hecmml.

FHA’s Oversight of

Servicers and

Collaboration on

Oversight between

FHA and CFPB Are

Limited

FHA Has Not Performed

On-Site Reviews of HECM

Servicers for More Than 5

Years and Lacks Current

Review Procedures

Page 38 GAO-19-702 Reverse Mortgages

The lack of on-site reviews of HECM servicers is inconsistent with OMB

requirements for managing federal credit programs.

OMB Circular A-

129 states that agencies should conduct on-site lender and servicer

reviews biennially where possible and annually for lenders and servicers

with substantial loan volumes or those with other risk indicators such as

deterioration in their credit portfolio, default rates above acceptable levels,

or an abnormally high number of reduced or rejected claims. The purpose

of these reviews is to evaluate and enforce lender and servicer

performance and identify any noncompliance with program requirements.

The circular encourages agencies to develop a risk-rating system for

lenders and servicers to help establish priorities for on-site reviews and to

monitor the effectiveness of required corrective actions. The circular also

says that agencies should summarize review findings in written reports

with recommended corrective actions.

FHA previously conducted on-site reviews of HECM servicers. However,

according to agency data, FHA has not performed on-site reviews since

fiscal year 2013. From fiscal years 2010 through 2013, FHA’s Quality

Assurance Division (a component of the Office of Lender Activities and

Program Compliance) conducted 14 on-site reviews of HECM servicers

(see table 3). These reviews examined compliance with FHA servicing

requirements and included detailed reviews of samples of loans.

Table 3: Number of FHA On-Site Reviews of Home Equity Conversion Mortgage

Servicers, Fiscal Years 2010–2018

Fiscal year

2010

2011

2012

2013

2014–2018

Total

Number of reviews

Source: Federal Housing Administration (FHA). | GAO-19-702

Office of Management and Budget, Policies for Federal Credit Programs and Non-Tax

Receivables, OMB Circular No. A-129 (revised January 2013).

Page 39 GAO-19-702 Reverse Mortgages

FHA provided us three examples of HECM servicing reviews conducted

in fiscal year 2013. While not representative of all reviews, the three

reviews identified multiple violations of FHA requirements, as follows:

• Quality control plans. Two of the three reviews found that the

servicers’ quality control plans—an internal control mechanism to help

ensure compliance with FHA requirements—were missing required

elements. For example, one review found that the servicer’s plan

lacked 13 required elements, including those intended to ensure

compliance with fair lending laws and immediate reporting of fraud or

other serious violations. Another review found deficiencies with the

servicer’s plan, including in the areas of customer service, servicing

transfers, and fees and charges.

• Communication with borrowers. In these same two reviews, FHA

found that the servicers did not always provide borrowers with a

designated contact person or timely and accurate information about

their loan status. For both servicers, FHA’s reviews of files for a

sample of active loans found no evidence that the servicer had

provided the borrower a contact person to handle inquiries. FHA

requires servicers to designate for borrowers a contact person

knowledgeable about servicing and provide the name of the person

annually and whenever the contact person changes. Additionally, both

reviews found that the servicers’ annual loan statements to borrowers

were missing critical information, such as the net principal limit (total

loan funds available), and that the servicers did not provide borrowers

with statements after every loan disbursement, as required.

• Filing claims. In two of the three reviews, FHA found deficiencies in

the servicers’ filing of insurance claims. For example, in one review,

FHA identified multiple cases where the servicer submitted claims that

were greater than the amounts warranted, including excess attorney

and appraisal fees, property preservation and protection expenses,

and interest costs. In another review, FHA found numerous instances

where the servicer missed various deadlines—including for submitting

claims, commencing foreclosure, and obtaining appraisals—and

therefore was not entitled to the full claim amounts it received.

• Loan disbursements. One of the three reviews found numerous

instances in which the servicer did not respond to borrowers’ requests

for payment plan changes within the required time frame of 5

business days, and therefore did not make timely loan disbursements

to borrowers.

Page 40 GAO-19-702 Reverse Mortgages

FHA required these servicers to take corrective actions, including

updating quality control plans, revising policies and procedures,

reimbursing FHA for unwarranted claim amounts, indemnifying FHA for

losses on a loan, and paying late charges to borrowers who did not

receive timely loan disbursements.

FHA has the option of referring

violations of FHA requirements to HUD’s Mortgagee Review Board, which

can take administrative actions such as issuing letters of reprimand,

suspending or withdrawing approval to participate in FHA programs,

entering into settlement agreements to bring an entity into compliance,

and imposing civil money penalties.

FHA officials said they had not

referred any HECM servicers to the board as a result of findings from on-

site reviews.

According to FHA’s current Director of the Quality Assurance Division,

under previous leadership, the division suspended on-site reviews of

HECM servicers after fiscal year 2016 because of servicers’ concerns

about the clarity and consistency with which FHA was conducting the

reviews and applying enforcement remedies.

He said the Quality

Assurance Division had intended to revise its guidance for conducting the

reviews and then resume them, but the effort had stalled during a change

in leadership. The current Director said he was not aware that HECM

servicing reviews had been suspended until the fall of 2017, when the

division began targeting on-site reviews for fiscal year 2018, and noticed

that HECM servicers were not included in the prior year’s targeting

methodology.

The lack of recent HECM servicer reviews is problematic for a number of

reasons. First, as previously noted, the number of HECM borrowers

In some circumstances, the lender or servicer must indemnify—or repay—FHA for

losses that it incurs after a loan has gone into default and the property has been sold.

Officials from HUD’s Office of General Counsel said that enforcement actions against

HECM lenders and servicers have been rare, but they noted False Claims Act settlements

against two companies that service HECMs. In 2017 and 2018, HUD’s Office of General

Counsel and Office of Inspector General worked with the Department of Justice to

coordinate investigations into and settlements with the servicers. According to the

Department of Justice officials, in both cases, the servicers allegedly did not disclose

noncompliance with various deadlines—such as for property appraisals and

commencement of foreclosure proceedings—in order to claim interest payments from

FHA they were not entitled to receive. One servicer agreed to pay $89 million to resolve

the allegations, and the other agreed to pay $4.25 million.

The current Director said he was in an acting position starting in July 2017 and became

permanent in January 2018.

Page 41 GAO-19-702 Reverse Mortgages

defaulting on their loans has grown in recent years. As a result, knowing

whether servicers are providing borrowers with accurate and timely

communications about their mortgage obligations and the status of their

loans has become increasingly critical. Second, FHA has recently made

program changes and implemented foreclosure prevention options, such

as at-risk extensions and mortgagee optional election assignments, to

help struggling borrowers and nonborrowing spouses delay or avoid

foreclosure. But FHA does not know how effectively servicers inform

borrowers of these options and use these tools due to its lack of

oversight. Third, as discussed earlier, the majority of HECM servicers are

nonbank entities that may pose risks because they are not subject to the

same comprehensive federal safety and soundness regulations as banks

and rely on funding sources, such as lines of credit, that may be less

stable than deposits.

The Director of the Quality Assurance Division said FHA plans to begin

conducting HECM servicer reviews in fiscal year 2020 but will first need to

revise its procedures for reviewing HECM servicers, which were last

updated in 2009. However, the Director told us the division decided not to

develop criteria for selecting HECM servicers for review. Instead, he said

FHA plans to review all HECM servicers with significant portfolios at least

once every 3 years, starting with the three servicers that account for 96

percent of the HECM portfolio.

While FHA’s plan to review HECM servicers with significant portfolios

captures one aspect of portfolio risk (loan volume), it does not account for

other risk indicators that OMB Circular A-129 says agencies should

consider.

The circular also encourages agencies to develop risk-rating

systems that incorporate these indicators. While the current HECM

servicing market is dominated by a small number of companies, the ability

to prioritize on-site reviews based on risk ratings will be important if the

market becomes less concentrated in the future. Additionally, some

HECM servicers may warrant review more frequently than once every 3

years if their business volume or performance poses substantial risks to

FHA or to borrowers. FHA’s plans do not account for these contingencies.

In contrast, FHA uses risk-based criteria to prioritize on-site reviews for HECM lenders

and forward mortgage lenders and servicers.

Page 42 GAO-19-702 Reverse Mortgages

CFPB oversees reverse mortgage servicers through examinations

designed, among other things, to identify whether servicers engage in

acts or practices that violate federal consumer financial laws. CFPB

issued its Reverse Mortgage Examination Procedures in 2016 and began

conducting examinations in 2017.

CFPB’s procedures include reviewing

servicers’ compliance with the Real Estate Settlement Procedures Act of

1974 and its implementing regulations (which, among other things,

contain requirements for notifying borrowers of servicing transfers,

responding to borrowers’ written information requests and notices of

error, and disclosures relating to force-placed insurance);

the Truth In

Lending Act and its implementing regulations (which impose requirements

on servicers governing the use of late fees and delinquency charges,

provisions for payoff statements, and disclosures regarding rate changes

for adjustable-rate mortgages); and other consumer protection laws.

Additionally, CFPB’s procedures include a review of whether a HECM

servicer is following selected elements of FHA’s HECM program

requirements. For example, CFPB’s examiners are directed to determine

whether information provided to the borrowers about life expectancy set-

aside accounts (an FHA requirement) is clear, prominent, and readily

understandable, and whether the borrower incurred penalties or

unnecessary charges in the event the servicer failed to make

disbursements of set-aside funds for insurance, taxes, and other charges

with respect to the property in a timely manner. CFPB examiners also are

directed to determine whether the servicer referred a HECM to

foreclosure improperly after the death of a borrower, such as when an

eligible nonborrowing spouse still occupies the home. If CFPB’s reverse

mortgage examinations identify violations, CFPB may require the

Consumer Financial Protection Bureau, Reverse Mortgage Servicing Examination

Procedures (Washington, D.C.: October 2016). CFPB’s oversight of reverse mortgage

servicers is not limited to those participating in the HECM program.

According to CFPB, loan requirements may include that borrowers have insurance on

their property. If the borrower fails to obtain insurance or lets the insurance lapse, the loan

contract may allow for the lender or servicer to provide insurance to cover the property

(force-placed insurance). CFPB notes that force-placed insurance protects only the lender

from losses, but the lender will charge the borrower for the insurance costs. Force-placed

insurance is usually more expensive than what borrowers could obtain by purchasing

policies themselves.

General servicing requirements regarding policies and procedures, early interventions,

continuity of contract, and loss mitigation procedures under the Real Estate Settlement

Procedures Act of 1974, as amended, are generally inapplicable to reverse mortgage

servicers. See 12 C.F.R. § 1024.30(b)(2).

CFPB Conducts Oversight

of Reverse Mortgage

Servicers, but It Has Not

Completed Steps to Share

Examination Results with

FHA

Page 43 GAO-19-702 Reverse Mortgages

examined entity to take corrective actions, which are recorded in the

examination results as matters requiring attention.

CFPB examinations of reverse mortgage servicers have found

deficiencies in monitoring of servicing actions, compliance with consumer

protection laws, and communications with consumers. For example,

CFPB reported in the March 2019 edition of its Supervisory Highlights

that one or more reverse mortgage servicing examinations found cases

where the servicer did not provide the heirs of deceased borrowers a

complete list of the documents needed to evaluate their case for a

foreclosure extension.

(Extensions can give heirs additional time to sell

or purchase the property and delay or avoid foreclosure.) As a result, in

some instances, one or more servicers foreclosed rather than seeking a

foreclosure extension from FHA. According to CFPB, in response to the

examinations, one or more servicers planned to improve communications

with borrowers’ heirs, including specifying the documents needed for a

foreclosure extension and the relevant deadlines. CFPB officials said they

plan to continue examining reverse mortgage servicers.

In addition to conducting examinations and issuing matters requiring

attention, CFPB officials said the bureau has other options—including

issuing warning letters and taking enforcement actions—to stop unlawful

practices or promote future compliance by supervised entities. Warning

letters advise companies that certain practices may violate federal

consumer financial law.

Enforcement actions are legal actions against

an entity initiated through federal district court or by an administrative

CFPB matters requiring attention are corrective actions that result from examination

findings that require the attention of a supervised institution’s board of directors or

principals and that are directly related to violations of consumer financial law, compliance

program deficiencies, or control weaknesses. CFPB also has the option to forward

examination findings to its Action Review Committee for further input and consideration.

According to CFPB officials, the Action Review Committee can take additional

enforcement action if the committee finds concerns about consumer protections.

Consumer Financial Protection Bureau, Supervisory Highlights, Winter 2019, Issue 18,

(Washington, D.C.: March 2019).

Warning letters are not accusations of wrongdoing, but instead are meant to help

recipients review certain practices to ensure that they comply with federal law. CFPB may

make aspects of these letters public without identifying the recipient if the bureau

determines that other entities might benefit from a reminder to review certain practices or

thinks the public should be aware of particular activities.

Page 44 GAO-19-702 Reverse Mortgages

adjudication proceeding.

CFPB officials told us the bureau had not

issued any warning letters or enforcement actions against HECM

servicers as of August 2019.

While CFPB has examined reverse mortgage servicers and plans to

continue doing so, CFPB officials said the bureau and FHA do not have

an agreement in place to share supervisory information, which inhibits

sharing of examination results. Information-sharing agreements may

address topics such as what and how information will be shared and

handling of sensitive information. CFPB officials said that an agreement

with FHA would be needed to ensure that supervisory information in the

bureau’s examinations is kept confidential. Under the Dodd-Frank Wall

Street Reform and Consumer Protection Act, CFPB must share results of

the examination of a supervised entity with another federal agency that

has jurisdiction over that entity, provided that CFPB received from the

agency reasonable assurances as to the confidentiality of the information

disclosed.

In addition, in previously issued work, we noted that interagency

collaboration can serve a number of purposes, including, among other

things, policy development, oversight and monitoring, and information

sharing and communication.

CFPB officials said CFPB and FHA had taken initial steps in 2017 toward

developing an information-sharing agreement. However, as of August

2019, an information-sharing agreement had not been completed. CFPB

officials told us there were existing ways for the two agencies to share

examination findings, but that an information-sharing agreement would

According to CFPB officials, CPFB enforcement actions are typically the result of formal

investigations that include hearings and subpoenas against companies, involve

settlements, and are typically time intensive for CFPB because of the investigative

process.

However, CFPB has issued warning letters and taken enforcement actions related to

reverse mortgage advertising.

See Pub. L. No. 111-203, § 1022(c)(6)(C) (2010) (codified at 12 U.S.C. § 5512(c)(6)(C)).

GAO-12-1022.

Page 45 GAO-19-702 Reverse Mortgages

facilitate the process.

CFPB officials said developing information-

sharing agreements can be a lengthy process and that both agencies had

other competing priorities. However, because of the limited information

sharing between CFPB and FHA, FHA is not benefiting from oversight

findings about servicers it could rely on to help implement the HECM

program. Having this information is particularly important given that FHA

does not comprehensively review HECM servicers itself and CFPB’s

examinations address a number of FHA requirements. Access to CFPB’s

examination results could enhance FHA’s oversight of HECM servicers

and potentially help it respond to consumer protection issues facing

HECM borrowers.

CFPB collects, analyzes, and reports on consumer complaints related to

reverse mortgages. The bureau began collecting reverse mortgage

consumer complaints in December 2011 and has collected about 3,600

complaints since then.

CFPB collects complaints through an online

forum on its website called the Consumer Complaint Database, as well as

via email, mail, phone, fax, or referral from another agency.

CFPB’s

authority to collect complaints comes from the Dodd-Frank Wall Street

Reform and Consumer Protection Act, which states that one of the

FHA can request access to relevant CFPB examination results under 12 C.F.R. §

1070.43. Additionally, the agencies have the ability to have high-level conversations about

confidential supervisory information under 12 C.F.R. § 1070.45(a)(5). According to CFPB

officials, CFPB’s Office of Enforcement has used both of these processes with FHA to

share information on reverse mortgage issues.

Of the 3,600 complaints, CFPB officials said about 3,000 were forwarded to companies

and were posted publicly on CFPB’s website. The remaining 600 complaints were referred

to other agencies, such as federal regulators. CFPB officials said they do not publicly post

complaints that are referred to other agencies.

CFPB’s Consumer Complaint Database is available through its website at

https://www.consumerfinance.gov/data-research/consumer-complaints/.

CFPB Collects and

Analyzes Consumer

Complaints on

Reverse Mortgages,

but FHA Does Not

Use All Available Data

CFPB Has Received

About 3,600 Reverse

Mortgage Complaints

since 2011

Page 46 GAO-19-702 Reverse Mortgages

bureau’s primary functions is collecting, investigating, and responding to

consumer complaints.

CFPB officials told us the bureau uses consumer complaints as part of its

criteria for selecting entities to examine, including reverse mortgage

servicers, and to inform its educational publications. For example, in June

2015, CFPB released a report on reverse mortgage advertising and

consumer risks.

In August 2017, CFPB released an issue brief on the

costs and risks of using a reverse mortgage to delay collecting Social

Security benefits.

100

In February 2015, CFPB issued a report on reverse mortgage consumer

complaints it received from December 2011 through December 2014.

101

CFPB found that consumer complaints indicated frustration and confusion

over the terms and requirements of reverse mortgages. CFPB also found

that many complaints were about problems with loan servicing. For

example, some consumers complained that they were at risk of

foreclosure due to nonpayment of property taxes or homeowners

insurance and that they faced obstacles when trying to prevent a

foreclosure. CFPB officials told us they did not currently have plans to

publish additional reports on reverse mortgage complaints, but that CFPB

would continue to produce educational materials on reverse mortgages

and internally review the data on a routine basis.

For this report, we performed a high-level analysis of roughly 2,500

reverse mortgage complaints received by CFPB from calendar years

2015 through 2018.

102

We analyzed patterns in the number of complaints

by year, state, submission method, and company.

Pub. L. No. 111-203, §1021(c)(2) (2010) (codified at 12 U.S.C. § 5511(c)(2)).

Consumer Financial Protection Bureau, Office for Older Americans, A Closer Look at

Reverse Mortgage Advertisements and Consumer Risks (Washington, D.C.: June 2015).

100

Consumer Financial Protection Bureau, Office for Older Americans, Issue Brief: The

Costs and Risks of Using A Reverse Mortgage to Delay Collecting Social Security

(Washington, D.C.: August 2017).

101

Consumer Financial Protection Bureau, Office for Older Americans, Snapshot of

Reverse Mortgage Complaints: December 2011–December 2014 (Washington, D.C.:

February 2015).

102

We selected this period because, as previously mentioned, CFPB’s February 2015

publication included summary findings on reverse mortgage complaints received prior to

2015.

Page 47 GAO-19-702 Reverse Mortgages

• By year. Complaint volumes varied across the 4 years, with the most

complaints received in 2016 and the least received in 2018 (see table

4).

Table 4: Number of Reverse Mortgage Complaints Received by the Consumer

Financial Protection Bureau (CFPB), Calendar Years 2015–2018

Calendar year

2015

2016

2017

2018

Total

Number of complaints

628

741

594

509

2,472

Source: GAO analysis of CFPB Consumer Complaint Database. | GAO-19-702

• By state. The states with the most complaints were California

(accounting for 16 percent of reverse mortgage complaints), Florida

(11 percent), New York (8 percent), and Texas (7 percent). These

states are among the most populous, and three of them (California,

Florida, and Texas) also had the greatest numbers of HECMs.

• By submission method. A majority of reverse mortgage complaints

(56 percent) were submitted through CFPB’s website. The remaining

complaints were submitted through referrals to CFPB from other

agencies (22 percent), by phone (12 percent), by postal mail (7

percent), and by fax (3 percent). Compared to the percentage of all

types of mortgage complaints filed during the 4-year period, the

percentage of reverse mortgage complaints filed through the website

(56 percent) was lower than the corresponding percentage for

complaints about all types of mortgages (67 percent). Representatives

from legal aid organizations representing HECM borrowers said that

reverse mortgage consumers may be less likely to file a complaint

through a website because of limitations sometimes related to aging—

for example, lack of internet access or computer skills. Additionally,

representatives from three of the five organizations said seniors may

suffer from health or capacity issues, such as hearing, vision, or

memory loss, that may make it difficult for them to file or follow up on

a complaint. For these reasons, seniors may not be submitting

complaints through CFPB’s website and seniors’ complaints about

reverse mortgages may be underreported in general.

Page 48 GAO-19-702 Reverse Mortgages

• By company. Companies that were the subject of reverse mortgage

complaints included both lenders and servicers.

103

From 2015 through

2018, five companies were the subject of more than 100 complaints

each, ranging from a low of 116 to a high of 506. Together, these five

companies accounted for 61 percent (1,509) of the reverse mortgage

complaints CFPB received. Additionally, one company received the

most complaints in 4 out of the 5 years reviewed.

104

We also conducted a more detailed analysis of a random, generalizable

sample of 100 consumer complaint narratives from among the 2,472 total

reverse mortgage complaints CFPB received in calendar years 2015

through 2018. The purpose of this analysis was to identify patterns in

consumer-described issues about reverse mortgages. We created issue

categories by reading the consumer narratives. Figure 6 shows the

estimated percentage of reverse mortgage complaints received by CFPB

over the 4-year period by consumer-described issue categories, based on

our sample of 100 complaint narratives.

103

As previously noted, some companies originate reverse mortgages and then continue

to own and service them. However, in many instances, the reverse mortgage may be

originated by one company, owned by another, and serviced by yet another. This makes

analysis by company difficult because not all companies are either originating or servicing

reverse mortgages.

104

Making finer comparisons across companies would require a standardized measure

such as a complaint rate—that is, the number of complaints divided by a uniform indicator

of reverse mortgage business volume (number of loans originated or serviced by a

company). However, differences in the focus of the companies (for example, originating,

servicing, or owning reverse mortgages) and annual shifts in the size of the companies’

reverse mortgage portfolios (for example, due to business decisions to exit the market or

increase market share) complicate development of such a measure.

Page 49 GAO-19-702 Reverse Mortgages

Figure 6: Types of Reverse Mortgage Complaints Received by the Consumer Financial Protection Bureau, Calendar Years

2015–2018

Note: GAO created the complaint categories by reading a random generalizable sample of 100

consumer complaint narratives. Percentages add to more than 100 percent because some consumer

complaints included multiple issues and, as a result, were included in more than one complaint

category. Confidence intervals are rounded to the nearest whole number.

Among the largest consumer-described issue categories were

foreclosures; poor communication from lenders or servicers; problems at

loan origination; estate management; and unfair interest rates, fees, or

costs.

• Being at risk of foreclosure or in foreclosure. The largest

consumer-described issue category (47 percent) involved consumers

(or someone complaining on behalf of the consumer) who said they

were at risk of foreclosure or in the foreclosure process.

105

For

example, some consumers said they or the borrower they represent

had received a notice of default, were in due and payable status, or

105

Our analysis of consumer complaint narratives found that a number of complaints were

submitted on behalf of the borrower by another party, such as family members, potentially

because of limitations from aging previously discussed.

Page 50 GAO-19-702 Reverse Mortgages

were at risk of foreclosure. Some consumers sought help in

preventing foreclosure or felt they were wrongly being foreclosed on.

In 16 of the 47 complaints about being at risk of or in foreclosure,

consumers also cited concerns about property taxes, insurance, or

other property charges.

• Poor communication on a servicing or lending issue. The second

largest consumer-described issue category (42 percent) involved

complaints about poor communication on a reverse mortgage

servicing or lending issue. These complaints included concerns about

a lack of communication or communications that were unclear or

unresponsive to the consumer’s needs. Complaints in this category

often overlapped with those about being at risk of or in foreclosure.

For example, some of these complaints included consumers’

concerns that they had not received information about the status of or

reason for a possible foreclosure from their servicer or did not get

responses to their inquiries.

• Loan origination issues. The third largest complaint category

involved problems occurring at loan origination (29 percent). These

complaints included consumers’ concerns that the amount of funds

available from their reverse mortgage was less than expected or that

interest rates or fees were not disclosed or explained to them. The

complaints also included cases where the adult children of borrowers

said they felt the lender took advantage of their parents.

• Estate-management issues. Twenty-seven percent of consumer

complaints were about estate management issues. Complaints

involving estate-management were often submitted by deceased

borrowers’ families or heirs. In some cases, heirs said that they were

unable to get information about the status of the reverse mortgage. In

other cases, the heirs said that because of the reverse mortgage, they

were at risk of losing the home, which was also their place of

residence.

• Unfair interest rates, fees, or costs. Twenty-seven percent of

consumer complaints were about being charged higher-than-expected

costs, fees, or interest. For example, in a few complaints, consumers

said that their servicers required them to pay for insurance products

(for example, flood insurance) that they felt were not needed.

According to CFPB officials, the bureau (1) refers consumer complaints

about financial products and services to the companies the complaints

are about or other federal regulators with supervisory jurisdiction over

those companies or (2) makes complaint information available to other

federal agencies with jurisdiction over the relevant product or service.

Page 51 GAO-19-702 Reverse Mortgages

CFPB officials said the bureau does not currently refer reverse mortgage

complaints to FHA; however, they told us reverse mortgage complaints

are available to FHA through CFPB’s public website and through a secure

portal FHA can access that has more data available than on the public

website.

FHA collects and records inquiries and complaints about HECMs and, as

previously mentioned, has access to CFPB data on reverse mortgage

complaints. However, FHA does not use its inquiry and complaint data to

help inform HECM program policies and oversight, and the way data are

collected does not produce quality information for these purposes.

Additionally, FHA has not leveraged CFPB’s complaint data for HECM

program oversight. Federal internal control standards state that agencies

should use quality information to achieve the entity’s objectives, including

using relevant data from reliable internal and external sources.

106

Additionally, in prior work we identified practices to enhance collaboration

across agencies, including leveraging agency resources.

107

According to agency officials, FHA’s two main methods for collecting

customer inquiries and complaints are hotlines operated by the agency’s

National Servicing Center and the FHA Resource Center.

108

Historically,

the National Servicing Center was FHA’s primary method for collecting

inquiries and complaints about the HECM program.

109

From calendar

years 2015 through 2018, the National Servicing Center received about

105,000 HECM-related calls. During this same period, the FHA Resource

Center received 147 HECM-related calls. In April 2019, the FHA

Resource Center became the primary entity for collecting, recording, and

responding to all HECM-related calls. FHA officials told us they

106

GAO-14-704G.

107

GAO-12-1022.

108

The National Servicing Center is a customer assistance center that works with FHA

homeowners and their lenders or servicers to avoid foreclosure. Customers can submit

their inquiries and complaints via telephone, email, postal mail, or fax. In addition to its two

main methods, FHA receives complaints and inquiries through congressional and White

House correspondence. FHA officials said complaints received through these channels

were less frequent than complaints received through other methods and sometimes

involved prospective borrowers who did not meet HECM eligibility requirements.

109

We use the term “calls” to refer to any inquiry or complaint submitted to FHA and

logged through its two main collection methods.

FHA Does Not Analyze

Data on Consumer

Complaints to Help Inform

HECM Program Policies

Page 52 GAO-19-702 Reverse Mortgages

transferred these responsibilities from the National Servicing Center to

the FHA Resource Center to help improve call management.

While this change could help improve customer service, it would not fully

resolve limitations we found in FHA’s approach to collecting and

recording HECM inquiries and complaints that diminish the usefulness of

the information for program oversight. These limitations include the

following:

• Information is not suitable for thematic analysis. Both the National

Servicing Center and the FHA Resource Center do not collect call

information in a way that would allow FHA to readily analyze the data

for themes. For example, both centers do not reliably differentiate

between inquiries and complaints—a potentially important distinction

for determining appropriate agency-level responses (for example,

creating informational materials to address frequently asked questions

from borrowers or investigating problematic servicing practices after

repeated complaints). Additionally, while both the centers collect data

on the reason for calls, neither did so in a systematic way that would

allow FHA to readily determine how frequently issues are being

raised. For example, neither centers’ data systems contained

standardized categories or menus with options for recording reasons

for calls. As a result, the FHA Resource Center’s data from 2015

through 2018 contained more than 100 separate reasons for 147

HECM-related calls. Some of the reasons the center recorded were

too specific (for example, a property address or a case number) to be

useful for identifying themes, while others were so similar that they do

not provide meaningful distinctions (but could be combined into fewer,

potentially more useful categories).

110

We noted similar limitations in

the National Servicing Center’s data, which included ambiguous call

reasons such as “history” and “documents,” and categories that could

be collapsed, which hinders thematic analysis.

111

• Customer type is not recorded. The National Servicing Center,

which received the large majority of HECM-related calls to FHA, did

not record information on the type of customer that made the call.

National Servicing Center guidance for staff says customers include

110

For example, among the call reasons were 36 that included the name of the contractor

that services loans assigned to FHA, three of which consisted of the contractor’s name

and the word “feedback,” “issue,” or “problem.”

111

Because of these limitations, we did not analyze FHA customer calls in a manner

similar to our analysis of CFPB consumer complaint data and narratives.

Page 53 GAO-19-702 Reverse Mortgages

borrowers, nonprofit organizations, government entities, real estate

brokers and agents, title companies, attorneys, lenders and servicers,

and HUD employees, but its data system does not include these

categories.

112

Information on customer type could be useful in

identifying issues facing different populations of callers and could help

FHA tailor strategies for addressing their concerns. In contrast, the

FHA Resource Center’s data system does include categories for

customer type for the smaller number of HECM-related inquiries and

complaints it received. Because the FHA Resource Center’s system is

now FHA’s primary repository for new HECM-related calls, information

on customer type should be available for future inquiries and

complaints. However, this information is not available for the bulk of

HECM-related calls FHA received in prior years.

FHA officials said the agency uses customer complaint and inquiry data

to improve customer service. For example, FHA officials said the National

Servicing Center monitors calls on a daily basis to ensure that prompt

responses are provided. Similarly, FHA officials said they review call data

monthly to identify training needs of servicers or contractors and potential

process changes to improve customer experience with the call process.

However, FHA does not analyze data for other purposes that could

enhance program oversight, such as determining which HECM servicers

and lenders receive the most complaints, targeting entities for on-site

reviews, or identifying topics that may need additional borrower

education.

FHA also does not use CFPB’s consumer complaint data to inform

management and oversight of the HECM program, even though some of

the information could be useful to the agency. For example, according to

CFPB’s complaint data for 2015 through 2018, approximately 6 percent of

reverse mortgage complaints were about FHA’s servicing contractor. FHA

officials said they do not review CFPB’s complaint data because they

believe the data are too limited to be useful and because they have

concerns about CFPB’s controls over data integrity. However, as our

analysis shows, CFPB’s data can be used to identify consumer

concerns—such as difficulties avoiding or navigating foreclosure or

problems communicating with servicers—that may merit additional

attention by FHA. Additionally, CFPB’s Office of Inspector General

recently reviewed CFPB’s management controls for the Consumer

112

Department of Housing and Urban Development, Expanded Customer Service National

Servicing Center Hotline Desk Guide (Oklahoma City, Okla.: 2013).

Page 54 GAO-19-702 Reverse Mortgages

Complaint Database and did not identify major data integrity issues that

would preclude use of the data for general oversight purposes.

113

Periodically analyzing CFPB consumer complaint data and internally

collected consumer complaint data could help FHA to detect and respond

to consumer protection issues regarding HECMs.

Since 2000, the take-up rate—the ratio of HECM originations to eligible

senior homeowners—has been limited (see fig. 7).

114

This rate, which

provides an indication of how popular HECMs are among the population

of senior homeowners, has not reached 1 percent and has fallen in recent

years. In addition, since calendar year 2010, the volume of HECM

originations has declined and is about half of what originations had been

at their peak. For example, in calendar years 2007–2009, more than

100,000 new HECMs were originated each year, compared with roughly

42,000 in calendar year 2018.

113

Consumer Financial Protection Bureau, Office of Inspector General, Opportunities Exist

to Enhance Management Controls Over the CFPB’s Consumer Complaint Database,

2015-FMIC-C-016, (Washington, D.C.: Sept.10, 2015); Bureau Efforts to Share Consumer

Complaint Data Internally Are Generally Effective; Improvements Can Be Made to

Enhance Training and Strengthen Access Approval, 2019-FMIC-C-008, (Washington,

D.C.: June 3, 2019); and Semiannual Report to Congress April 1, 2018–September 30,

2018, (Washington, D.C.: Oct. 31, 2018).

114

For more information on how we calculated HECM take-up rates and the sources of

data used, see app. II.

Housing Market

Conditions, Exit of

Large Bank Lenders,

and Policy Changes

Help Explain the

Decline in Reverse

Mortgages since

2010

Page 55 GAO-19-702 Reverse Mortgages

Figure 7: Home Equity Conversion Mortgage (HECM) Originations and Take-Up Rates, Calendar Years 1989–2018

Note: We were unable to calculate a take-up rate for calendar year 2018 because Census Bureau

data were not yet available.

The relatively high homeownership rate and low retirement savings of

U.S. seniors suggest that reverse mortgages could be a way for many

older Americans to tap their home equity and supplement retirement

income.

115

However, the popularity of reverse mortgages has declined in

recent years for a number of possible reasons. We developed an

econometric model to examine the relationship between HECM take-up

115

According to the Census Bureau’s 2016 American Community Survey data, the

homeownership rate for householders aged 65 and older is 78 percent, compared to 63

percent for all households. According to the Board of Governors of the Federal Reserve

System’s 2016 Survey of Consumer Finances, 50 percent of American families with a

head of household age 65 to 74 have no retirement accounts. Additionally, we found that

the percentage of households headed by someone aged 55 or over that had no retirement

savings was about 48 percent in 2016. See GAO, Retirement Security: Most Households

Approaching Retirement Have Low Savings, an Update, GAO-19-442R (Washington,

D.C.: Mar. 26, 2019).

Page 56 GAO-19-702 Reverse Mortgages

rates and a number of explanatory variables.

116

For additional information

and detailed results from our econometric model of factors associated

with HECM take-up rates, see appendix II. Among other factors, our

model results indicate that house price changes, home equity, and prior

use of other home equity lending products were statistically significant (at

the 1 percent level) in explaining the decrease in HECM take-up rates

since 2010.

117

• Changes in house prices. The decline in take-up rates may reflect

lower house prices, which have limited the number of households with

sufficient home equity (as a percentage of home value) to benefit from

a HECM. Our model estimated that, controlling for other factors, take-

up rates were higher when house price growth was large and there

was a history of house price volatility compared to either relatively low

house price appreciation or stable house prices. This result is

consistent with senior homeowners using reverse mortgages to insure

against house price declines.

118

For example, researchers have noted

that in states where house prices are volatile and the current level is

above the long-term norm, seniors anticipate future reductions in

house prices and lock in their home equity gains by obtaining a

reverse mortgage.

• Home equity and prior home equity borrowing. Additionally, we

found that controlling for other factors, take-up rates were higher

where home equity (house value minus any mortgage debt) was high.

In these cases, senior homeowners tap into their high home equity to

help supplement income with proceeds from the HECM. Further, we

found that among seniors who had previously used other home equity

lending products, such as home equity loans, take-up rates were high.

116

We estimated a panel of state-year observations using fixed-effects estimation.

Because this estimation technique controls for both observable and unobservable factors

that vary across states but not over time, it estimates only the within-state variation in

take-up rates.

117

We were not able to include some factors that could affect HECM take-up rates,

including FHA program policy changes and behavioral and structural factors discussed

later in this report. However, the state fixed-effects and year fixed-effects that we used are

expected to control for some of these factors.

118

For similar results, see Donald Haurin, Chao Ma, Stephanie Moulton, Maximilian

Schmeiser, Jason Seligman, and Wei Shi, “Spatial Variation in Reverse Mortgages

Usage: House Price Dynamics and Consumer Selection,” Journal of Real Estate Finance

and Economics, vol. 53 (2016): pp. 392-417.

Page 57 GAO-19-702 Reverse Mortgages

This result is consistent with seniors using HECMs to pay off these

loans.

119

Academics and industry experts have also noted possible reasons why

the popularity of reverse mortgages is limited. For example, senior

homeowners can tap their home equity by other means, such as home

equity loans, home equity lines of credit, and cash-out refinancing.

120

Some of these options may be less expensive than reverse mortgages.

Seniors can also downsize––sell their current home and buy or rent a

less expensive one—and keep the difference to supplement retirement

savings. Seniors have other ways to supplement their retirement income

and age in place—for example, one academic noted that some seniors

rent out rooms in their homes, potentially using online marketplaces such

as Airbnb.

121

Additionally, our literature review and interviews with

academics identified other factors that have may have contributed to

limited interest in reverse mortgages, including the following:

• Exit of large bank lenders. As previously noted, banks, thrifts, and

credit unions were historically the primary lenders and servicers of

mortgage loans. Following the 2007–2009 financial crisis and

subsequent revisions to regulatory bank capital requirements, banks

reevaluated the benefits and costs of being in the mortgage lending

market, as well as retaining mortgages and the right to service them.

Today, the reverse mortgage market is dominated by a relatively

small number of nonbank entities. The exit of large, well-known

lenders, such as Bank of America and Wells Fargo, from the HECM

market created opportunities for smaller nonbank lenders to enter the

market.

122

According to an academic we spoke with, in addition to

119

See Donald L. Redfoot, Ken Scholen, and S. Kathi Brown, Reverse Mortgages: Niche

Product or Mainstream Solution? A report on the 2006 AARP National Survey of Reverse

Mortgage Shoppers (Washington, D.C.: December 2007).

120

A cash-out refinance is when a mortgage borrower refinances into a new mortgage that

exceeds the existing loan balance. The difference between the two mortgages is given to

the borrower in cash.

121

For example, according to an Airbnb publication, the fastest-growing host demographic

is seniors, with more than 200,000 senior hosts. See 2019 Airbnb Statistics: User and

Market Growth Data, April 2019, accessed May 3, 2019,

https://ipropertymanagement.com/airbnb-statistics.

122

One study found that the exit of Bank of America and Wells Fargo was associated with

an 11 percent reduction in HECM volume. See Stephanie Moulton, Samuel Dodini,

Donald R. Haurin, and Maximilian Schmeiser, “Seniors’ Home Equity Extraction: Credit

Constraints and Borrowing Channels,” available at: https://ssrn.com/abstract=2727204.

Page 58 GAO-19-702 Reverse Mortgages

new capital requirements, large banks may have exited the market

partly out of concern that they risked damage to their reputations if

they foreclosed on seniors who defaulted on their HECMs.

Additionally, a 2018 survey of lenders found a variety of reasons why

lenders have stopped originating HECMs, including potential

reputation risk and concerns about HECMs being a distraction from

their forward mortgage business.

123

Although the HECM market is

currently served by several nonbank lenders, their smaller scale,

limited access to capital, and limited name recognition may limit their

ability to reach more potential borrowers.

124

• FHA policy changes to the HECM program. FHA has made several

policy changes in recent years to help stabilize the financial

performance of the HECM portfolio and strengthen financial criteria for

HECM borrowers. Although many of the HECM policy changes

introduced since 2010 were intended to minimize program losses,

they also may have reduced take-up rates. For example, in 2010 FHA

reduced the amount of money a borrower can get from a HECM.

125

Some academics we interviewed said reductions in the loan amounts

that borrowers can receive likely reduced demand for HECMs. In

2015, FHA changed financial requirements for HECMs to include a

financial assessment of the prospective borrower prior to loan

approval. Some academics said these changes made other home

equity extraction options that already had similar requirements more

competitive with HECMs.

• Consumers’ misunderstanding and product complexity. A 2013

survey of U.S. homeowners aged 58 and older revealed a lack of

knowledge of reverse mortgages.

126

The survey found that awareness

123

Jim Cameron, “Moving Forward in Reverse,” STRATMOR Insights, vol. 3, no. 2

(February 2018).

124

Karan Kaul and Laurie Goodman, Seniors’ Access to Home Equity: Identifying Existing

Mechanisms and Impediments to Broader Adoption (Washington, D.C.: Urban Institute,

February 2017).

125

As of October 4, 2010, the amount of HECM loan proceeds available to borrowers was

reduced due the implementation of new principal limit factors. See Department of Housing

and Urban Development, Mortgagee Letter 2010-34 (Sept. 21, 2010). According to FHA

officials, the amount of HECM loan proceeds available to borrowers has been further

reduced since Mortgagee Letter 2010-34, and the most recent reduction was effective

October 2, 2017. See Department of Housing and Urban Development, Mortgagee Letter

2017-12 (Aug. 29, 2017).

126

Thomas Davidoff, Patrick Gerhard, and Thomas Post, Reverse Mortgages: What

Homeowners (Don’t) Know and How It Matters (Oct. 24, 2016). Available at

https://ssrn.com/abstract=2528944.

Page 59 GAO-19-702 Reverse Mortgages

of reverse mortgages is high, but knowledge of mortgage terms is

limited. Additionally, the survey found that respondents perceived

reverse mortgages to be fairly complex.

• Consumers’ perception of the product. Academics we spoke with

told us that consumers’ negative perception of reverse mortgages

likely has a negative influence on take-up rates. For example, three

academics elaborated that consumers build their perception of the

product based on the industry’s marketing and advertising, which

includes television commercials with celebrity spokespeople that may

appeal to individuals facing economic hardship. Additionally, a 2016

survey of Americans aged 55 to 75 found that many respondents had

reservations about reverse mortgages, including that they are often

considered a financial tool of last resort.

127

For example, only 27

percent of survey respondents stated that, in general, it was better to

use a reverse mortgage earlier in retirement as opposed to using it as

a last resort.

• Relatively high origination costs and fees. HECMs also may be

unpopular with borrowers because they can be more expensive than

other home equity lending products, such as home equity lines of

credit.

128

For example, HECM borrowers are charged various fees,

such as the up-front insurance premiums that FHA charges as

compensation for its insurance guarantee and origination fees lenders

charge. The up-front insurance premium is 2 percent of the

mortgage’s maximum claim amount. Also, for origination fees, lenders

can charge the greater of $2,500 or 2 percent of the first $200,000 of

the mortgage’s maximum claim amount plus 1 percent of the

maximum claim amount over $200,000. However, origination fees are

currently capped at $6,000. Further, because borrowers do not make

127

Hopkins, Jamie, “The Effect of Low Reverse Mortgage Literacy on Usage of Home

Equity and Retirement Income Plans,” Journal of Financial Planning (May 2017): pp. 44–

52.

128

Deborah Lucas, “Hacking Reverse Mortgages” (Oct. 26, 2015), working paper

accessed on June 14, 2019,

http://mitsloan.mit.edu/shared/ods/documents/?DocumentID=4596. See also Hopkins.

Page 60 GAO-19-702 Reverse Mortgages

monthly payments on the loans, the interest will accumulate over time,

and compounding the interest, the loan balance can rise quickly.

129

• Seniors’ attitudes toward debt and desire to leave a bequest.

Some academics have noted that seniors tend to be financially

conservative and avoid debt in old age––behavior driven by their

desire to leave a bequest or save for emergency expenses or long-

term care costs.

130

For example, academics have noted that some

impediments to home equity extraction are behavioral and have to do

with seniors’ long-held values, beliefs, and attitudes, such as to

maximize wealth transfer to heirs by leaving a bequest. As a result,

they may be reluctant to take out a HECM, even if it could help pay for

some future expenses.

HECMs allow seniors to tap a portion of their home equity to supplement

their retirement income, but these loans can present risks to borrowers

and their spouses. The growing number of borrowers who have defaulted

on their HECMs and faced foreclosure in recent years highlights the

importance of monitoring loan outcomes and overseeing loan servicing

policies and practices in the HECM program. FHA has taken some steps

to enhance the data it receives from servicers and has created

foreclosure prevention options for distressed borrowers. However, FHA

could significantly improve its monitoring of loan outcomes and oversight

of servicing in the HECM program in the following areas:

• FHA’s lack of comprehensive termination data limits understanding of

the reasons why HECMs end, how the debt is satisfied, and how well

the program is helping seniors age in place. By, for example, updating

and providing more guidance to servicers on how to record

termination reasons, FHA could improve the completeness and

accuracy of HECM termination data.

• FHA has not effectively assessed the performance of the HECM

program. By establishing performance indicators and periodically

assessing them, FHA could better oversee the program and

communicate information on program performance to Congress.

129

The interest rate on a HECM is agreed upon between the lender and the borrower and

can be either fixed for the life of the loan or adjustable, with adjustments occurring either

monthly or yearly. For such loans, the interest rate is based on an index such as the

LIBOR index (London Interbank Offer Rate) and a fixed premium charged by, and at the

discretion of, the lender (known as the lender’s spread or margin).

130

Karan Kaul and Laurie Goodman, Seniors’ Access to Home Equity.

Conclusions

Page 61 GAO-19-702 Reverse Mortgages

Further, FHA could use the performance data to help make informed

decisions about any needed program changes in the future.

• FHA’s internal monitoring and reporting on loan outcomes has been

limited. Adopting analytic tools could better position FHA to evaluate

loan outcomes and help ensure senior officials have information

needed to make key decisions.

• FHA has not fully analyzed the implications of how it prioritizes

foreclosures for assigned HECMs. FHA’s current process generally

results in no foreclosures on assigned loans with property charge

defaults. Analyzing the implications of this process could help FHA

optimize how it services assigned loans.

• Because FHA does not currently perform on-site reviews of HECM

servicers, it lacks assurance that servicers are complying with rules

and program requirements. While FHA plans to begin reviewing

HECM servicers in fiscal year 2020, its plan does not include

development of a risk-rating system to prioritize reviews and identify

servicers that should be reviewed more frequently.

• CFPB does not share the results of its examinations of HECM

servicers with FHA, in part because the two agencies have not

completed a formal information-sharing agreement. Sharing these

results could aid FHA’s oversight of HECM servicers by providing

additional information about the servicers’ performance and

operations.

• FHA’s collection and use of consumer complaint data could be

improved. More organized collection of complaints and better

monitoring of internal and external complaint data could help FHA

detect and respond to emerging consumer protection issues regarding

HECMs.

By addressing these issues, FHA could help ensure that the HECM

program achieves program goals, effectively oversees servicers, and

provides appropriate borrower protections.

We are making a total of nine recommendations, eight to FHA and one to

CFPB:

The FHA Commissioner should take steps to improve the quality and

accuracy of HECM termination data. These steps may include updating

the termination reasons in the HERMIT system or updating the HERMIT

Recommendations for

Executive Action

Page 62 GAO-19-702 Reverse Mortgages

User Guide to more clearly instruct servicers how to record termination

reasons. (Recommendation 1)

The FHA Commissioner should establish, periodically review, and report

on performance indicators for the HECM program—such as the

percentage of terminations due to borrower defaults, the proportion of

active HECMs with delinquent property charges, the amount of servicer

advances, and the percentage of distressed borrowers who have

received foreclosure prevention options—and examine the impact of

foreclosure prevention options in the forthcoming HECM program

evaluation. (Recommendation 2)

The FHA Commissioner should develop analytic tools, such as

dashboards or watch lists, to better monitor outcomes for the HECM

portfolio, such as reasons for HECM terminations, defaults, use of

foreclosure prevention options, or advances paid by servicers on behalf of

HECM borrowers. (Recommendation 3)

The FHA Commissioner should evaluate FHA’s foreclosure prioritization

process for FHA-assigned loans. Such an analysis should include the

implications that the process may have for HECM borrowers,

neighborhoods, and FHA’s insurance fund. (Recommendation 4)

The FHA Commissioner should develop and implement procedures for

conducting on-site reviews of HECM servicers, including a risk-rating

system for prioritizing and determining the frequency of reviews.

(Recommendation 5)

The FHA Commissioner should work with CFPB to complete an

agreement for sharing the results of CFPB examinations of HECM

servicers with FHA. (Recommendation 6)

The CFPB Director should work with FHA to complete an agreement for

sharing the results of CFPB examinations of HECM servicers with FHA.

(Recommendation 7)

The FHA Commissioner should collect and record consumer inquiries and

complaints in a manner that facilitates analysis of the type and frequency

of the issues raised. (Recommendation 8)

The FHA Commissioner should periodically analyze available internal and

external consumer complaint data about reverse mortgages to help

Page 63 GAO-19-702 Reverse Mortgages

inform management and oversight of the HECM program.

(Recommendation 9)

We provided HUD and CFPB with a draft of this report for review and

comment. HUD provided written comments, which are reproduced in

appendix V, that communicate FHA’s response to the report. CFPB’s

written comments are reproduced in appendix VI.

CFPB said that it did not object to our recommendation to complete an

agreement for sharing the results of CFPB examinations of HECM

servicers with FHA (recommendation 7) and that it would work to

complete such an agreement with FHA.

FHA agreed with six of our eight recommendations and neither agreed

nor disagreed with the remaining two.

• Recommendation 1. FHA agreed with our recommendation to

improve HECM termination data and said it would convene a working

group to update the HERMIT system and User Guide and develop

clear directions for HECM servicers to record termination reasons in

HERMIT.

• Recommendation 2. Regarding our recommendation on HECM

performance indicators and program evaluation, FHA agreed that

periodic review and reporting of HECM performance indicators is

critically important and said it would work to expand its reporting to

include the level of foreclosure prevention activity. However, FHA

added that there were no HECM metrics for early default or

delinquency rates, as those measures are linked to the amortizing

nature of forward mortgages.

131

We agree that early default and

delinquency rates are not suitable metrics for HECMs, and our draft

report did not suggest that they are. Our report focuses on metrics

that would be pertinent to HECMs and that would provide additional

insight into HECM loan performance. These include the percentage of

HECM terminations due to borrower defaults, the proportion of active

HECMs with delinquent property charges, and the amount of funds

servicers have advanced on behalf of borrowers. We revised the

recommendation in our final report to more specifically describe the

types of performance indicators that FHA should establish and report

131

Early defaults or delinquencies occur when a borrower quickly falls behind on mortgage

payments—for example, within the first six mortgage payments.

Agency Comments

and Our Evaluation

Page 64 GAO-19-702 Reverse Mortgages

on. In addition, FHA disagreed with a statement in our draft report that

its evaluation of the HECM program has been limited. FHA said it

engages in robust HECM program evaluation and cited an example

that led to recent changes in FHA’s appraisal practices for HECMs.

While our draft report described the change in FHA’s appraisal

practices, we updated our final report to include reference to the FHA

study that prompted the appraisal change. However, we maintain that

FHA’s evaluation of the HECM program has been limited because the

last comprehensive program evaluation was completed 19 years ago

and FHA has not assessed the impact of HECM foreclosure

prevention options.

• Recommendations 3 and 4. FHA agreed with our recommendations

to develop analytic tools for monitoring HECM loan outcomes and to

evaluate its foreclosure prioritization process for FHA-assigned loans.

Regarding the latter, FHA said that it is evaluating alternative

disposition options to reduce the number of loans that must go

through foreclosure and that it would take steps to evaluate the impact

of its prioritization process to assist in future decision-making.

• Recommendation 5. FHA agreed with our recommendation to

develop and implement procedures for conducting on-site reviews of

HECM servicers, including a risk-rating system for prioritizing and

determining the frequency of reviews. As noted in our draft report,

FHA said it is in the process of updating procedures for on-site

reviews and plans to implement them in fiscal year 2020. FHA

disagreed with a statement in our draft report that FHA’s oversight of

HECM servicers is limited. FHA said the HECM servicing community

is small, which allows the agency to maintain regular communication

with HECM servicers, including through training sessions and industry

working group meetings. Our draft report acknowledged FHA’s

communications with servicers, but these activities are not a

substitute for in-depth compliance reviews of servicers’ operations. As

our draft report stated, FHA has not conducted on-site HECM servicer

reviews since fiscal year 2013. Given the 5-year lapse in FHA’s use of

this key oversight tool, we maintain that FHA’s oversight of HECM

servicers has been limited.

• Recommendation 9. FHA agreed with our recommendation to

periodically analyze internal and external consumer complaint data

about reverse mortgages. FHA said it is expanding its data and

reporting capabilities as part of an information technology

modernization initiative. FHA also said that routing consumer inquiries

through the FHA Resource Center should improve data collection and

analysis.

Page 65 GAO-19-702 Reverse Mortgages

FHA did not explicitly agree or disagree with our recommendations to

work with CFPB to complete an agreement for sharing examination

results and to collect and record consumer inquiries and complaints in a

manner that facilitates analysis (recommendations 6 and 8, respectively).

FHA said it would explore opportunities to coordinate with CFPB where

appropriate. FHA also said that routing inquiries through the FHA

Resource Center would help identify common issues, track servicer

performance, and inform policy decisions. Fully implementing our

recommendations will help ensure that FHA has the information it needs

to effectively oversee the HECM program.

We are sending copies of this report to the Secretary of the Department

of Housing and Urban Development, the Director of the Consumer

Financial Protection Bureau, appropriate congressional committees, and

other interested parties. In addition, the report is available at no charge on

the GAO website at http://www.gao.gov.

If you or your staff have any questions about this report, please contact

me at (202) 512-8678 or CackleyA@gao.gov. Contact points for our

Offices of Congressional Relations and Public Affairs may be found on

the last page of this report. GAO staff who made key contributions to this

report are listed in appendix VII.

Alicia Puente Cackley

Director, Financial Markets and Community Investment

Appendix I: Objectives, Scope, and

Methodology

Page 66 GAO-19-702 Reverse Mortgages

This report examines issues related to reverse mortgages made under

the Home Equity Conversion Mortgage (HECM) program administered by

the Department of Housing and Urban Development’s (HUD) Federal

Housing Administration (FHA). The Consumer Financial Protection

Bureau (CFPB) also plays a role in overseeing reverse mortgages,

including HECMs. Our objectives were to examine (1) what FHA data

show about HECM terminations, servicer advances, and the use of

foreclosure prevention options; (2) FHA’s assessment and monitoring of

HECM portfolio performance, servicer advances, and foreclosure

prevention options; (3) FHA’s and CFPB’s oversight of HECM servicers;

(4) how FHA and CFPB collect, analyze, and respond to consumer

complaints about HECMs; and (5) how and why the market for HECMs

has changed in recent years.

To address all of our objectives, we reviewed relevant laws, regulations,

and requirements, such as HECM authorizing legislation, the Reverse

Mortgage Stabilization Act of 2013, FHA regulations, and mortgagee

letters governing the HECM program.

We also interviewed FHA and

CFPB officials and staff from other relevant HUD offices such as the

Office of Policy Development and Research and the Office of General

Counsel. We reviewed FHA’s annual reports to Congress on the financial

status of the Mutual Mortgage Insurance Fund, actuarial reports on the

HECM portfolio, and FHA’s annual management reports. We also

reviewed our prior reports and reports by HUD’s Office of Inspector

General about the HECM program.

Congress authorized the HECM program in 1988 by adding Section 255 to Title II of the

National Housing Act. See Housing and Community Development Act of 1987, Pub. L. No.

100-242, § 417 (1988) (codified as amended at 12 U.S.C. § 1715z-20). The Reverse

Mortgage Stabilization Act of 2013, Pub L. No. 113-29, 127 Stat. 509 (2013) (codified at

12 U.S.C. § 1715z-20(h)(3)), allows FHA to make changes to HECM program

requirements by notice or mortgagee letter in addition to regulation. FHA’s regulations for

the HECM program can be found in 24 C.F.R. Part 206. A mortgagee letter is a written

communication to lenders and servicers about changes in FHA operations, policies, and

procedures.

For example, see GAO, Federal Housing Administration: Capital Requirements and

Stress Testing Practices Need Strengthening, GAO-18-92 (Washington, D.C.: Nov. 9,

2017); GAO, Reverse Mortgages: Product Complexity and Consumer Protection Issues

Underscore Need for Improved Controls over Counseling for Borrowers, GAO-09-606

(Washington, D.C.: June 29, 2009); and Department of Housing and Urban Development

Office of Inspector General, HUD Was Not Tracking Almost 13,000 Defaulted HECM

Loans With Maximum Claim Amounts of Potentially More Than $2.5 Billion, 2010-FW-

0003 (Aug. 25, 2010).

Appendix I: Objectives, Scope, and

Methodology

Appendix I: Objectives, Scope, and

Methodology

Page 67 GAO-19-702 Reverse Mortgages

Additionally, we identified the largest HECM servicers using FHA data on

the number of loans serviced as of the end of fiscal year 2018. We found

that five companies serviced more than 99 percent of the HECM portfolio

(excluding loans assigned to FHA, which are serviced by an FHA

contractor) as of the end of fiscal year 2018. We developed a

questionnaire to solicit information applicable to our objectives from these

five servicers. We took steps to verify the information gathered in the

questionnaire. We reviewed responses for completeness and held

teleconferences with each HECM servicer to discuss, clarify, and amend

responses. Where possible, we corroborated servicers’ responses with

information or analysis from other sources, such as our analysis of FHA

loan-level data or FHA documents. We use summary statements and

illustrative examples from these questionnaires and our interviews with

the five servicers throughout the report.

We also interviewed representatives from five legal aid organizations

representing HECM borrowers in the states of California, Florida, New

York, Texas, and Washington. We selected these states because they

had the highest number of HECM originations in the past decade and

because they provided some geographic diversity; the five states span

the West (California), Northwest (Washington), Northeast (New York),

Southeast (Florida), and South (Texas). We selected the specific legal aid

organizations within those states because they represented a large

number of HECM borrowers, according to organization representatives.

We conducted semistructured interviews with organization

representatives that included questions on the top consumer protection

issues facing HECM borrowers, how recent HECM program changes may

have helped borrowers delay or avoid foreclosure, and characteristics of

HECM borrowers that may affect their ability to file consumer complaints.

We use summary statements and illustrative examples from these

interviews throughout the report.

To address the first objective, we analyzed FHA data to determine the

number of and reasons for HECM terminations, the amounts of servicer

advances, and the number of borrowers approved for selected

foreclosure prevention options (for example, repayment plans). We used

data from the Home Equity Reverse Mortgage Information Technology

(HERMIT) system, which FHA adopted in fiscal year 2013. FHA provided

us a HERMIT case detail table from its Single Family Data Warehouse

that contained loan-level information as of the end of fiscal year 2018. We

separately obtained several ad hoc HERMIT reports from FHA’s HERMIT

system contractor, as described below. For some of our analyses, we

HECM Terminations,

Servicer Advances, and

Foreclosure Prevention

Options

Appendix I: Objectives, Scope, and

Methodology

Page 68 GAO-19-702 Reverse Mortgages

merged data from the ad hoc reports with data from the case detail table

using the unique FHA case number for each HECM. Unless otherwise

noted, we analyzed data for the 5-year period spanning fiscal years

2014–2018.

We assessed the reliability of data from the HERMIT system by reviewing

FHA documentation about the data system and data elements. For

example, we reviewed the HERMIT User Guide and notes on HERMIT

system updates.

Additionally, we interviewed FHA and contractor staff

knowledgeable about the HERMIT system and data to discuss

interpretations of data fields and trends we observed in our analyses. We

also conducted electronic testing, including checks for duplicate loans,

outliers, missing data fields, and erroneous values. Where appropriate,

we removed from our analyses any loans missing an endorsement

(insurance approval) date as well as cases with erroneous values. When

possible, we corroborated our analyses with external reports such as

FHA’s annual reports to Congress, management reports, and production

reports. Based on these steps, we determined the data we used from the

HERMIT system were sufficiently reliable for summarizing trends and

generating descriptive statistics for HECM terminations, servicer

advances, and selected foreclosure prevention options over the 5-year

period.

We analyzed FHA loan-level data from the HERMIT system to determine

the total number of terminated HECMs and reasons for terminations by

fiscal year. We first identified terminations occurring in fiscal years 2014–

2018 using data fields for case status and termination date (see table 5).

Table 5: Reported Home Equity Conversion Mortgage Terminations, Fiscal Years

2014–2018

Fiscal year

2014

2015

2016

2017

2018

Total

Number of

loans

24,272

39,075

82,157

66,479

60,172

272,155

Source: GAO analysis of Federal Housing Administration data. | GAO-19-702

Home Equity Conversion Mortgage Servicer Provider HERMIT User Guide, Servicing

Module, version 2.17 (September 2018).

Termination Analysis

Appendix I: Objectives, Scope, and

Methodology

Page 69 GAO-19-702 Reverse Mortgages

We then removed any terminated loans that had previously been

assigned to FHA (16,008) using the data field that records the date FHA

accepted assignment of the loan. We removed these loans because FHA

officials told us the agency generally does not foreclose on FHA-assigned

HECMs that default and keeping them in the analysis would have resulted

in understating the proportion of terminations stemming from defaults.

Accordingly, the denominator for our terminations analysis was 256,147

loans (272,155 total terminations minus the 16,008 loans previously

assigned to FHA).

We then identified the reported termination reasons for the 256,147 loans.

We analyzed loan-level data from the HERMIT system to identify the

number of loans that fell into various termination reason categories. To

identify terminations stemming from a HECM becoming due and payable,

we used data from two reports that we obtained from FHA’s HERMIT

system contractor: the Default Key Dates Report and the Due and

Payable Delinquency Report. From these reports, we identified the

number of terminations due to a borrower’s death, conveyance of title,

default due to unpaid property charges, default due to failure to meet

occupancy or residency requirements, and default due to failure to keep

the home in good repair. To identify terminations stemming from

repayment, refinancing, moving, or other (undetermined) reasons, we

used information on case substatus from the HERMIT case detail table

from the Single Family Data Warehouse.

Our undetermined reasons

category included loans for which the case substatus either was labeled

“terminate-other” or showed how the debt was satisfied (such as through

a deed-in-lieu of foreclosure, foreclosure, or short sale) rather than

providing a termination reason. For the full results of our terminations

analysis, see appendix III.

We analyzed servicer advances to HECM borrowers using data from an

ad hoc HERMIT report we requested from FHA’s HERMIT system

contractor. We analyzed the data to determine the amounts and types of

servicer advances in fiscal years 2014 through 2018 for terminated

HECMs. We distinguished between servicer advances for unpaid property

Using this data table, we identified additional HECMs that had terminated because of a

borrower’s death. We added these cases to the death cases from the Default Key Dates

report and present summary information for terminations due to borrower death. We also

identified in this data table additional cases of terminations due to borrowers moving. We

added these cases to the borrower conveyed title cases we had identified in the Default

Key Dates report and present summary information for terminations due to the borrower

conveying title or moving.

Servicer Advances Analysis

Appendix I: Objectives, Scope, and

Methodology

Page 70 GAO-19-702 Reverse Mortgages

charges and servicer advances for other costs. Examples of the latter are

attorney, trustee, and appraisal fees typically incurred during the

foreclosure process. For each year and for the 5-year period as a whole,

we calculated total servicer advances and the amount and percentage of

advances for property charges and for other costs.

Additionally, we distinguished between servicer advances for unpaid

property charges before and after a HECM borrower’s death using the

date of death of the last surviving borrower in HERMIT.

This allowed us

to determine the amount of servicer advances for unpaid property

charges on behalf of living borrowers. We calculated the total amount of

these advances over the 5-year period as well as the mean, median, and

25th and 75th percentile values. We also calculated the number and

percentage of loans for which property charge advances on behalf of

living borrowers were less than $2,000 (the threshold for one of FHA’s

foreclosure prevention options).

We analyzed data from HERMIT on the use of selected foreclosure

prevention options—repayment plans and at-risk extensions—for

borrowers who defaulted because of unpaid property charges. We

analyzed data from April 2015 (the effective date of FHA’s current

repayment plan and at-risk extension policies) through fiscal year 2018.

To conduct the analysis of repayment plans, we used the HERMIT Due

and Payable Delinquency Report noted previously, which includes data

fields for loan default status and the dates borrowers were approved for a

repayment plan. We calculated the percentage of borrowers with property

charge defaults who were approved for repayment plans during the

period examined. To conduct the analysis of at-risk extensions, we

requested an ad hoc report from FHA’s HERMIT system contractor

showing whether and when borrowers had been approved for at-risk

extensions and appended it to the default status within the Due and

Payable Delinquency Report using FHA case numbers. We calculated the

percentage of borrowers with property charge defaults who were

approved for at-risk extensions during the period examined.

We also reviewed and summarized information that FHA provided us

from HERMIT on nonborrowing spouses who applied for mortgagee

optional election assignments from June 2015 (the effective date of FHA’s

A servicer may make an advance for property charges after a borrower dies if, for

example, a property tax bill comes due before the HECM is paid off and the property is

sold to another party.

Foreclosure Prevention

Options Analysis

Appendix I: Objectives, Scope, and

Methodology

Page 71 GAO-19-702 Reverse Mortgages

mortgagee optional election assignment policy) through fiscal year 2018.

FHA provided information on the number of requested, approved, and

denied mortgagee optional election assignments during that period. We

also reviewed documentation from FHA and interviewed agency officials

about the mortgagee optional election assignment process and reasons

for denials. For the denied mortgagee optional election assignments, we

reviewed information that FHA provided us from HERMIT on the current

status of the associated loans as of May 31, 2019. For example, for the

denied mortgagee optional election assignments, FHA determined

whether the loan had been terminated as of that date. For those that had

terminated, we summarized whether the debt was paid off or whether the

debt was satisfied because of a foreclosure, deed-in-lieu of foreclosure,

or short sale.

To address the second objective, we reviewed agency reports and

interviewed agency officials to determine how the agency assesses the

performance of the HECM program, including the use of any performance

indicators or program evaluations. For example, we reviewed HUD’s

strategic plan for fiscal years 2018–2022 and its most recent annual

performance report to identify any goals and performance indicators

related to the HECM program.

Additionally, we reviewed program

evaluations completed for the HECM program.

We also interviewed FHA

and HUD Office of Policy Development and Research officials about

previous program evaluations and HUD’s plans for forthcoming

evaluations of the HECM program. We compared FHA’s practices against

leading practices identified in our previous work on assessing program

performance and against Office of Management and Budget (OMB)