GAO-16-448, Mortgage Servicing: Community Lenders Remain Active under New Rules, but CFPB Needs More Complete Plans for Reviewing Rules

MORTGAGE

SERVICING

Community Lenders

Remain Active under

New Rules, but CFPB

Needs More

Complete Plans for

Reviewing Rules

Report to the Chairman, Committee on

Financial Services, House of

Representatives

June 2016

GAO-16-448

United States Government Accountability Office

Highlights of GAO-16-448, a report to the

Chairman, Committee on Financial Services,

House of Representatives

June 2016

MORTGAGE SERVICING

Community Lenders

Remain Active under New Rules

but

CFPB Needs More Complete Plans for Reviewing

Rules

Why GAO Did This Study

As of September 30, 2015, community

lenders held about $3.1 billion in MSRs

on their balance sheets. Servicing is a

part of holding all mortgage loans, but

an MSR generally becomes a distinct

asset when the loan is sold or

securitized. In response to the 2007–

2009 financial crisis, regulators have

implemented new rules related to

mortgage servicing and regulatory

capital to protect consumers and

strengthen the financial services

industry. GAO was asked to review the

effect of these rule changes on U.S.

banks and credit unions, particularly

community lenders. This report

examines (1) community lenders’

participation in the mortgage servicing

market and potential effects of CFPB’s

mortgage servicing rules on them, (2)

potential effects of the treatment of

MSRs in capital rules on community

lenders’ decisions about holding or

selling MSRs, and (3) the process

regulators used to consider impacts of

these new rules on mortgage servicing

and the capital treatment of MSRs.

GAO analyzed financial data, reviewed

relevant laws and documents from

regulatory agencies, and interviewed

16 community lenders selected based

on size and volume of mortgage

servicing activities, as well as industry,

consumer groups, and federal officials.

What GAO Recommends

CFPB should complete a plan to

measure the effects of its new

regulations that includes specific

metrics, baselines, and analytical

methods to be used. CFPB agreed to

take steps to complete its plan for

conducting a retrospective review of

the mortgage servicing rules and refine

the review’s scope and focus.

What GAO Found

Community banks and credit unions (community lenders) remained active in

servicing mortgage loans under the Consumer Financial Protection Bureau’s

(CFPB) new mortgage-servicing rules. Among other things, these rules are

intended to provide more information to consumers about their loan obligations.

The share of mortgages serviced by community lenders in 2015—about 13

percent—remained small compared to larger lenders, although their share

doubled between 2008 and 2015. Large banks continue to service more than half

of residential mortgages. Many lenders GAO interviewed said changes in

mortgage-related requirements resulted in increased costs, such as hiring staff

and updating systems. However, many also stated that servicing mortgages

remained important to them for the revenue it can generate and their customer-

focused business model.

Banking and credit union regulators’ new capital rules changed how mortgage

servicing rights (MSR) are treated in calculations of required capital amounts, but

GAO found that these new rules appear unlikely to affect most community

lenders’ decisions to retain or sell MSRs. For example, GAO found that in the

third quarter of 2015, about 1 percent of community banks had to limit the

amount of MSRs that counted in their capital calculations due to the amount of

these assets they held. This may result in some institutions choosing to raise

additional capital or sell MSRs to meet required minimum capital amounts,

depending on banks’ holdings of other types of assets. A few banks with large

concentrations of MSRs that GAO spoke with said they were considering selling

MSRs or other changes to their capital but market participants told us that the

MSR capital treatment was only one of several factors influencing their decisions.

Separate capital rules for credit unions also are unlikely to affect most credit

unions. For example, credit unions told GAO they did not expect to make

changes to their MSR holdings and one credit union explained that it is because

MSRs represented a small percentage of their overall capital.

Banking regulators and CFPB estimated the potential impacts of their new rules

prior to issuing them by, for example, estimating potential costs of compliance.

Banking regulators included the capital rules in a retrospective review of all their

rules required by statute, although this review is to be completed before the MSR

requirements are fully implemented by the end of 2018. Banking regulators also

said they often conduct other informal reviews as needed to evaluate their rules’

effectiveness. CFPB also has a statutory retrospective review requirement, but

its plans for retrospectively reviewing its mortgage-servicing rules are incomplete.

CFPB has not yet finalized a retrospective review plan or identified specific

metrics, baselines, and analytical methods, as encouraged in Office of

Management and Budget guidance. In addition, GAO found that agencies are

better prepared to perform effective reviews if they identify potential data sources

and the measures needed to assess rules’ effectiveness. CFPB officials said it

was too soon to identify relevant data and that they wanted flexibility to design an

effective methodology. However, without a completed plan, CFPB risks not

having time to perform an effective review before January 2019—the date by

which CFPB must publish a report of its assessment.

View GAO-16-448. For more information,

contact

Mathew J. Sciré at (202) 512-8678 or

sciremj@gao.gov

Page i GAO-16-448 Mortgage Servicing

Letter 1

Background 4

Community Lenders Continue to Service Mortgages as Regulatory

Requirements Increase 14

New Capital Treatment of Mortgage Servicing Rights Likely Will

Not Affect Many Community Banks and Credit Unions 24

Regulators Estimated Impacts of New Rules Using Public Input

and Data Analysis, but CFPB’s Plans for Reviewing Rules Have

Limitations 33

Conclusions 44

Recommendation for Executive Action 44

Agency Comments and Our Evaluation 44

Appendix I Objectives, Scope, and Methodology 47

Appendix II Mortgage Servicing Rights Transfer Activity 56

Appendix III Comments from the Bureau of Consumer Financial Protection 58

Appendix IV Comments from the National Credit Union Administration 61

Appendix V GAO Contact and Staff Acknowledgments 62

Tables

Table 1: Federal Prudential Regulators and Their Basic Prudential

Functions, as of June 2016 6

Table 2: Median Risk-Based Capital Ratios for Community Banks

by Size and Required Minimum Capital Ratios, Third

Quarter 2015 28

Table 3: Number of Community Banks and Credit Unions

Interviewed, by Size 54

Contents

Page ii GAO-16-448 Mortgage Servicing

Table 4: Percentage of Residential Mortgage Transfers of

Servicing Approved by Freddie Mac or Ginnie Mae by

Institution Type, 2010 through 2015 56

Table 5: Net Transfers of Mortgage Servicing Rights Approved by

Freddie Mac or Ginnie Mae by Institution Type and Size

(in billions of dollars of unpaid principal balance), 2010

through 2015 57

Figures

Figure 1: Mortgage Servicing and Creation of a Mortgage

Servicing Right 8

Figure 2: Estimated Share of Residential Mortgages Serviced by

Community Banks, Credit Unions, and Nationwide,

Regional, and Other Banks, First Quarter 2008 through

Third Quarter 2015 (percentage) 14

Figure 3: Percentage of Community Banks, Credit Unions, and

Nationwide, Regional, and Other Banks with Residential

Mortgages on their Balance Sheets by Size, First Quarter

2001 through Third Quarter 2015 (percentage) 22

Figure 4: Mortgage Servicing Rights and Risk-Based Capital

Ratios 25

Figure 5: Community Banks’ Capital Treatment of Mortgage

Servicing Rights, 2015Q3 26

Figure 6: Sample Public Comments from the “Regulation Room”

on CFPB’s Proposed Mortgage Servicing Rule 35

Page iii

GAO-16-448 Mortgage Servicing

Abbreviations

ABA American Bankers Association

CET1 common equity tier 1

CFPB Consumer Financial Protection Bureau

CUNA Credit Union National Association

Dodd-Frank Act Dodd-Frank Wall Street Reform and Consumer

Protection Act

EGRPRA

Economic Growth and Regulatory Paperwork

Reduction Act of 1996

FDIC Federal Deposit Insurance Corporation

Federal Reserve Board of Governors of the Federal Reserve System

FFIEC Federal Financial Institutions Examination Council

MBS mortgage-backed securities

MSR mortgage servicing rights

NCUA National Credit Union Administration

OMB Office of Management and Budget

PRA Paperwork Reduction Act of 1995

RFA Regulatory Flexibility Act

SBREFA Small Business Regulatory Enforcement Fairness

Act of 1996

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-16-448 Mortgage Servicing

441 G St. N.W.

Washington, DC 20548

June 23, 2016

The Honorable Jeb Hensarling

Chairman

Committee on Financial Services

House of Representatives

Dear Mr. Chairman:

Many community banks and credit unions (community lenders) view

servicing mortgages as important to maintaining their business and

satisfying their customers. As of September 30, 2015, community lenders

held about $3.1 billion in mortgage servicing rights on their balance

sheets.

Recently, these relatively small financial institutions as well as

larger banks have become subject to regulatory changes developed in

response to the 2007‒2009 financial crisis.

These changes are designed, in part, to strengthen the financial services

industry, and some are specific to mortgage servicing. The 2010 Dodd-

Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank Act)

directed or gave authority to federal agencies to issue mortgage servicing

regulations. In 2013, the Bureau of Consumer Financial Protection

(commonly known as the Consumer Financial Protection Bureau or

CFPB) issued regulations that require, among other things, prompt

crediting of mortgage payment and early intervention for delinquent

borrowers.

Further, also in 2013, federal banking regulators adopted new

requirements for risk-based capital that are based on international

standards (the Basel III framework) developed by the Basel Committee

on Banking Supervision, a global standard-setter for prudential bank

Although no commonly accepted definition of a community bank exists, the term often is

associated with smaller banks (e.g., under $1 billion in assets) that provide relationship

banking services to the local community and have management and board members who

reside in the local community. In this report, we use the term “community lenders” to mean

community banks and credit unions. Credit union membership is based on a common

bond, such as residing in a specific geographic area or working in the same profession.

Mortgage Servicing Rules Under the Real Estate Settlement Procedures Act (Regulation

X), 78 Fed. Reg. 10696 (Feb. 14, 2013); Mortgage Servicing Rules Under the Truth in

Lending Act (Regulation Z), 78 Fed. Reg. 10902 (Feb. 14, 2013).

Letter

Page 2 GAO-16-448 Mortgage Servicing

regulation.

These requirements define how much capital banks must

hold for a variety of activities—including for holding mortgage servicing

rights (MSR), which are distinct assets that generally are created when a

mortgage loan is sold or securitized.

Both the mortgage servicing regulations and the regulatory capital

changes for MSRs are relatively new and the effect of these new

requirements has not yet been formally evaluated. Industry and trade

associations have raised concerns about the potential effect of mortgage

servicing regulations on U.S. banks and, in particular, on community

lenders. You asked us to examine the effect of mortgage servicing and

risk-based capital regulations on community lenders and their

customers.

This report examines (1) community lenders’ participation in

the mortgage servicing market and potential effects of CFPB’s mortgage

Regulatory Capital Rules: Regulatory Capital, Implementation of Basel III, Capital

Adequacy, Transition Provisions, Prompt Corrective Action, Standardized Approach for

Risk-weighted Assets, Market Discipline and Disclosure Requirements, Advanced

Approaches Risk-Based Capital Rule, and Market Risk Capital Rule, 78 Fed. Reg. 62018

(Oct. 11, 2013) (Federal Reserve and OCC) and 78 Fed. Reg. 55340 (Sept. 10, 2013)

(FDIC Interim Final Rule). With minor changes, the September 2013 FDIC interim final

rule became a final rule in April 2014. See 79 Fed. Reg. 20754 (Apr. 14, 2014). The Basel

III framework has no legal force but was issued by the agreement of the Basel Committee

members with the expectation that individual national authorities would implement the

standards.

In this report, we use the term “mortgage servicing rights” to mean mortgage servicing

assets that are recognized on a servicer’s balance sheet and are subject to capital

deductions and risk-weighting under the federal prudential regulators’ risk-based capital

rules. Depending upon the facts and circumstances of a given loan sale or securitization

transaction, mortgage servicing rights may or may not actually be recognized on a

servicer’s balance sheet, and when recognized, could actually constitute either a

mortgage servicing asset or a mortgage servicing liability. Also, under U.S. Generally

Accepted Accounting Principles, certain accounting criteria must be met for the sale to

qualify as an accounting sale of servicing assets. For example, the transferor must

surrender control of the financial assets to the transferee.

We use the Federal Deposit Insurance Corporation’s practical definition of a community

bank, which incorporates an asset size threshold—generally including banks with less

than $1 billion in assets—and other characteristics, including if it is part of a banking

organization that has loans or core deposits, has limited amounts of foreign assets, has

limited amounts of assets in specialty banks like credit card banks or trust companies, and

is either relatively small or has large amounts of loans and core deposits and a limited

geographic scope. The asset size threshold is not a strict requirement. Larger banks may

be considered community banks if they meet other criteria such as having a loan-to-asset

ratio greater than 33 percent. See Federal Deposit Insurance Corporation, FDIC

Community Banking Study, December 2012.

Page 3 GAO-16-448 Mortgage Servicing

servicing rules on them, (2) potential effects of the risk-based capital

treatment of MSRs on decisions about holding or selling MSRs, and (3)

the process regulators used to estimate the impact of regulations

addressing mortgage servicing requirements and the risk-based capital

treatment of MSRs.

To address these objectives, we analyzed quarterly data on banks

obtained from the Federal Deposit Insurance Corporation (FDIC) and the

Federal Financial Institutions Examination Council (FFIEC) for the period

from the first quarter of 2001 through the third quarter of 2015, quarterly

data on credit unions obtained from the National Credit Union

Administration (NCUA) for the period from the second quarter of 2002

through the third quarter of 2015, and quarterly data on outstanding

residential mortgages obtained from the Board of Governors of the

Federal Reserve System (Federal Reserve) for the period from the first

quarter of 2008 through the third quarter of 2015. We used these data to

estimate the shares of residential mortgages serviced by banks and credit

unions of different sizes, to assess the extent to which banks and credit

unions participate in residential mortgage lending, and to analyze the

potential effect on banks of the capital treatment of MSRs under risk-

based capital rules. We grouped banks into five equal-sized groups, or

quintiles, based on their size as measured by total assets, with the first

quintile containing the smallest banks and the fifth quintile containing the

largest banks. We did the same for credit unions. We assessed the

reliability of the data from FDIC, FFIEC, the Federal Reserve, and NCUA

for the purposes described above by reviewing relevant documentation

and electronically testing the data for missing values, outliers, and invalid

values and found the data to be sufficiently reliable for these purposes.

We also analyzed data on transfers of mortgage servicing rights obtained

from Fannie Mae, Freddie Mac, and Ginnie Mae for the period from 2010

to 2015. We analyzed the amount of MSRs associated with mortgage

pools that were sold via bulk sales by banks, credit unions, and nonbank

entities. We assessed the reliability of these data for this purpose by

electronically testing the variables for missing values, invalid values, and

outliers. We found these data to be sufficiently reliable for this purpose.

Finally, we reviewed relevant laws and regulations, as well as past GAO

reports on the financial crisis and the implementation of the Dodd-Frank

Act. We also interviewed officials from a variety of organizations,

including community banks, credit unions, regulators, industry

organizations, credit union service organizations, consumer groups, and

academics and other industry participants such as mortgage brokers. Our

Page 4 GAO-16-448 Mortgage Servicing

interviews with a small sample of community lenders and an additional

regional bank provided further insights on their participation in mortgage

servicing and effects of regulations. The responses are not generalizable

to the population of community lenders. Appendix I provides a more

detailed description of our scope and methodology.

We conducted this performance audit from April 2015 to June 2016 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.

Various institutions, including banks, credit unions, nonbank entities, and

subservicers, service mortgage loans.

These institutions are defined as

follows:

• Banks. Institutions of various types that may be chartered under

federal or state law.

One type of bank, community banks, is often

associated with smaller banks (e.g., under $1 billion in assets) that

provide relationship banking services to the local community and have

For more information about nonbank mortgage servicers, see GAO, Nonbank Mortgage

Servicers: Existing Regulatory Oversight Could Be Strengthened, GAO-16-278

(Washington D.C.: Mar. 10, 2016).

For purposes of this report, banks include bank holding companies, financial holding

companies, savings and loan holding companies, and insured depository institutions,

including any subsidiaries or affiliates of these institutions.

Background

Types of Mortgage

Servicers and Federal

Regulators

Page 5 GAO-16-448 Mortgage Servicing

management and board members who reside in the local community.

In addition to mortgage servicing, banks offer a variety of financial

products to consumers, including deposit products, loan products,

such as mortgage and auto loans, and credit card products.

• Credit unions. Member-owned cooperatives run by member-elected

boards with an historical emphasis on serving people of modest

means. Like banks, credit unions offer a variety of financial, deposit,

and loan products to consumers.

• Nonbank entities. Entities that are not financial institutions and may be

involved in a variety of mortgage activities, including servicing and

originating loans. Nonbank entities generally do not offer deposit or

credit card products to consumers.

• Subservicers. Third-party servicers that have no investment in the

loans they service. Banks, credit unions, and nonbanks may

outsource loan servicing activities to a subservicer that performs the

same administrative functions the bank, credit union, or nonbank

would to service the mortgage loan.

All U.S. depository institutions that have federal deposit insurance have a

federal prudential regulator that generally may issue regulations and take

enforcement actions against institutions within its jurisdiction. The federal

prudential regulators, which are the Office of the Comptroller of the

Currency (OCC), FDIC, and Federal Reserve, along with the credit-union-

regulating NCUA, oversee depository institutions for safety and

soundness purposes and for compliance with other laws and regulations

We use the FDIC’s definition of a community bank, which incorporates an asset size

threshold—generally including banks with less than $1 billion in assets—and other

characteristics, including if it is part of a banking organization that has loans or core

deposits, has limited amounts of foreign assets, has limited amounts of assets in specialty

banks like credit card banks or trust companies, and is either relatively small or has large

amounts of loans and core deposits and a limited geographic scope. The asset size

threshold is not a strict requirement. Larger banks may be considered community banks if

they meet other criteria such as having a loan-to-asset ratio greater than 33 percent. See

Federal Deposit Insurance Corporation, FDIC’s Community Banking Study.

Page 6 GAO-16-448 Mortgage Servicing

that fall within the scope of the relevant prudential regulator’s authority

(see table 1).

Table 1: Federal Prudential Regulators and Their Basic Prudential Functions, as of June 2016

Agency

Basic function

Office of the Comptroller of the Currency

Charters and supervises national banks, federal savings associations (also known as

federal thrifts), and federal branches and agencies of foreign banks, supervises

subsidiaries of national banks and federal thrifts.

Board of Governors of the Federal

Reserve System

Supervises state-chartered banks that opt to be members of the Federal Reserve System,

depository institution holding companies (bank holding companies and savings and loan

holding companies), and the nonbanking subsidiaries of those entities.

Federal Deposit Insurance Corporation

Supervises state-chartered banks that are not members of the Federal Reserve System,

as well as state savings banks and thrifts and state chartered branches of foreign banks;

insures the deposits of all banks and thrifts that are approved for federal deposit

insurance;

has the authority to conduct insurance or backup examinations for any insured

institutions; resolves all failed insured banks and thrifts; and has the authority to resolve

certain large bank holding companies and nonbank financial companies, if appointed

receiver by the Secretary of the Treasury after the statutory process.

National Credit Union Administration

Charters and supervises federally chartered credit unions and insures savings in federal

and most state-chartered credit unions.

Source: GAO. | GAO-16-448

Note: The Federal Deposit Insurance Corporation (FDIC) insures deposits in insured banks and thrifts

for at least $250,000. FDIC promotes the safety and soundness of these institutions by identifying,

monitoring, and addressing risks to the deposit insurance funds and limiting the effect on the financial

system when a bank or thrift fails.

Additionally, the Dodd-Frank Act transferred consumer financial

protection regulation and some other authorities regarding certain federal

consumer financial laws from other federal banking regulators to CFPB,

to help foster consistent enforcement of federal consumer financial

laws.

CFPB has supervision and primary enforcement authority for most

federal consumer financial laws for insured depository institutions with

more than $10 billion in assets and their affiliates as well as certain

nonbank entities. The prudential regulators—the Federal Reserve, Office

For a more detailed discussion of the regulatory framework for bank holding companies

and savings and loan holding companies, see GAO, Bank Holding Company Act:

Characteristics and Regulation of Exempt Institutions and the Implications of Removing

the Exemptions, GAO-12-160 (Washington, D.C.: Jan. 19, 2012). Nonbank servicers are

subject to different safety and soundness regulation and different capital rules. See

GAO-16-278.

These authorities were transferred on July 21, 2011.

Page 7 GAO-16-448 Mortgage Servicing

of the Comptroller of Currency, Federal Deposit Insurance Corporation,

and NCUA—have primary supervision and exclusive enforcement

authority for federal consumer financial laws for institutions that have $10

billion or less in assets. CFPB also has rulemaking authority to implement

provisions of federal consumer financial law. The Dodd-Frank Act

authorized CFPB to exercise its authorities for a number of purposes,

including ensuring that consumers are provided with timely and

understandable information that will help them make responsible

decisions about financial transactions; protecting consumers from unfair,

deceptive, or abusive acts and practices and from discrimination; and

identifying and addressing outdated, unnecessary, or unduly burdensome

regulations.

In the primary market, lenders make, or originate, mortgage loans that are

secured by property or real estate.

Originators can choose to hold

mortgages in their own portfolios or sell them into the secondary market.

Servicing is a part of holding mortgage loans in portfolio, but the right to

service a mortgage generally becomes a distinct asset—an MSR—when

contractually separated from the loan if the loan is sold or securitized. In

the secondary market, the government-sponsored enterprises Fannie

Mae and Freddie Mac purchase mortgages that meet their underwriting

criteria and either hold them in their own portfolios or pool them into

mortgage backed securities (MBS) and sell them to investors. Ginnie Mae

guarantees the timely principal and interest payments to investors in

securities issued by approved institutions through its MBS program. Once

the loan origination process is complete, the loan must be serviced until it

is paid in full or foreclosure occurs (see fig. 1).

Servicers perform various loan management functions, including

collecting payments from the borrower, sending monthly account

See 12 U.S.C. § 5511(b). CFPB’s rules implementing federal consumer financial laws

apply to all entities within the scope of a particular rule, including smaller banks. See 12

U.S.C. § 5512(b)(4).

For a more complete discussion of the primary and secondary mortgage markets, see

GAO, Housing Finance System: A Framework for Assessing Potential Changes,

GAO-15-131 (Washington, D.C.: Oct. 7, 2014) and Sean M. Hoskins, Katie Jones, and N.

Eric Weiss, Congressional Research Service, An Overview of the Housing Finances

System in the United States, R42995 (Washington, D.C.: Feb. 19, 2015).

Mortgage Servicing

Market

Page 8 GAO-16-448 Mortgage Servicing

statements and tax documents, responding to customer service inquiries,

maintaining escrow accounts for property taxes and hazard insurance,

and forwarding monthly mortgage payments to the mortgage owners. In

the event that borrowers become delinquent on their loan payments,

servicers may offer borrowers the loss mitigation options made available

by the owners of the loan, which may include a workout or a loan

modification that permits the borrower to stay in the home or other

options, such as a short sale. In some cases, the servicer is the same

institution that originated the loan. However, servicers may change over

the life of a mortgage as MSRs are sold or transferred to other

institutions.

Figure 1: Mortgage Servicing and Creation of a Mortgage Servicing Right

Note: In addition to banks, credit unions and nonbanks, independent mortgage servicers and

subservicers may be involved in these arrangements. For instance, subservicers—which are typically

nonbanks but can also be banks—may perform some or all servicing functions for the servicer of

record but they do not own the mortgage servicing right.

Page 9 GAO-16-448 Mortgage Servicing

Mortgage Origination Rules. CFPB issued new rules related to

underwriting of mortgage loans that became effective in January 2014.

Generally, lenders making mortgage loans must make a reasonable,

good faith determination of the borrower’s ability to repay the loan. This

ability-to-repay determination requires lenders to meet minimum

underwriting standards, including consideration and verification of a

borrower’s income or assets, debt, and credit history. Lenders are

presumed to comply with the ability-to-repay (determination) requirement

when they make a qualified mortgage, which is a loan that meets specific

product feature and underwriting criteria.

Mortgage Servicing Rules. CFPB issued mortgage-servicing-related

rules covering nine major topics that became effective January 10,

2014.

According to CFPB, the goals of the servicing rules are to provide

better disclosure to consumers regarding their mortgage loan obligations,

and to inform and assist them with options that may be available if they

Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act

(Regulation Z), 78 Fed. Reg. 6408 (Jan. 30, 2013). Under CFPB’s ability-to-repay and

qualified mortgage rule, which, as amended, became effective on January 10, 2014, the

agency identified eight underwriting factors a lender must consider in relation to making

the required good faith determination of a borrower’s ability to repay, including a

borrower’s income, assets, employment, credit history, and monthly expenses. In general,

the borrower also must have a total monthly debt-to-income (DTI) ratio, including

mortgage payments of 43 percent or less for the loan to have qualified mortgage status.

12 C.F.R. § 1026.43(e)(2)(vi). The ratio represents the percentage of a borrower’s income

that goes toward all recurring debt payments, including the mortgage payment. Lenders

use the DTI ratio as a key indicator of a borrower’s capacity to repay a loan. A higher ratio

is generally associated with a higher risk that the borrower will have cash flow problems

and may miss mortgage payments.

According to CFPB, a qualified mortgage is a category of loans that have certain, more

stable features that help make it more likely that a borrower is able to afford the loan. A

lender must make a good-faith effort to determine that borrowers have the ability to repay

a mortgage loan before it is made. This is known as the “ability-to-repay” rule. A creditor

is presumed to have complied with the ability-to-repay requirements if the creditor makes

a qualified mortgage loan.

Mortgage Servicing Rules Under the Real Estate Settlement Procedures Act (Regulation

X), 78 Fed. Reg. 10696 (Feb. 14, 2013); Mortgage Servicing Rules Under the Truth in

Lending Act (Regulation Z), 78 Fed. Reg. 10902 (Feb. 14, 2013). The nine topics are:

periodic billing statements; interest rate adjustment notices; prompt payment crediting and

payoff statements; force-placed insurance; error resolution and information requests;

general servicing policies, procedures, and requirements; early intervention with

delinquent borrowers; continuity of contact with delinquent borrowers; and loss mitigation

procedures.

Recent Changes to

Selected Mortgage

Regulations and

Regulatory Capital

Requirements for MSRs

Page 10 GAO-16-448 Mortgage Servicing

have difficulty making their mortgage payments. The servicing rules also

aim to ensure that borrowers are protected from harm in connection with

the process of evaluating a borrower for a loss mitigation option or

proceeding to foreclosure.

The rules also address critical servicer

practices relating to, among other things, correcting errors, imposing

charges for force-placed insurance, crediting mortgage loan payments,

and providing payoff statements.

CFPB included a small servicer exemption from certain parts of the

mortgage servicing rules.

Generally, entities that service 5,000 or fewer

mortgage loans, all of which they own or originated, are exempt, for

example, from providing periodic statements. In general, entities that

service one or more loans they neither originated nor own do not qualify

as small servicers, even if they service 5,000 or fewer loans overall.

Regulatory Capital Requirements for MSRs.

The Basel III framework

addressed MSRs as part of its effort to improve the banking sector’s

ability to absorb shocks arising from financial and economic stress,

whatever the source; improve risk management and governance; and

For complete loss mitigation applications received more than 37 days before a

foreclosure sale, servicers must evaluate borrowers for all available loss mitigation options

and provide notice of decision; borrowers may appeal a denial of a loan modification so

long as the borrower’s complete loss mitigation application is received 90 days or more

before a scheduled foreclosure sale. Servicers are restricted from dual-tracking,

simultaneously evaluating a borrower for a loss mitigation option while preparing to

foreclose on the property.

Force-placed insurance is hazard insurance obtained by a servicer on behalf of the

owner or assignee of a mortgage loan that insures the property securing the loan.

A small servicer is a servicer that (1) services 5,000 or fewer mortgage loans for all of

which the servicer (or an affiliate) is the creditor or assignee; (2) is a Housing Finance

Agency (as defined in 24 C.F.R. § 266.5); or (3) is a nonprofit entity that services 5,000 or

fewer mortgage loans for all of which the servicer or an associated nonprofit entity is the

creditor. 12 C.F.R. § 1026.41(e)(4)(ii).

For the purposes of this report, regulatory capital requirements for U.S. banking

organizations related to Basel III capital standards establish more restrictive capital

definitions, higher risk-weighted assets, additional capital buffers, and higher requirements

for minimum capital ratios.

Page 11 GAO-16-448 Mortgage Servicing

strengthen banks’ transparency and disclosures.

In 2013, the U.S.

federal banking regulators adopted revised capital rules to implement

many aspects of the Basel III capital framework and the Dodd-Frank Act

that apply to banks, savings associations, and top-tier U.S. bank and

savings and loan holding companies. The revised capital rules

significantly changed the risk-based capital requirements for banks and

bank holding companies, applied capital requirements to certain savings

and loan companies, and introduced new leverage standards. These

requirements include provisions related to MSRs that will be fully phased

in by 2018, including:

• An amount equal to MSRs in excess of 10 percent of a bank’s

common equity tier 1 (CET1) capital is deducted from CET1 capital.

• An amount equal to the sum of MSRs, certain deferred tax assets

arising from temporary differences, and significant investment in the

capital of unconsolidated financial institutions in the form of common

stock in excess of 15 percent of CET1 capital is also deducted from

tier 1 capital.

• MSRs that have not been deducted from CET1 capital are added to

risk-weighted assets with a 100 percent risk-weight during the

transition period and will be subject to a 250 percent risk-weight once

the revised regulatory capital rule is fully phased-in.

In October 2015, NCUA also issued risk-based capital regulations.

However, unlike the banking regulators, which applied the MSR

provisions to all supervised banks, NCUA exempted credit unions with

See Bank For International Settlements, Basel III: A Global Regulatory Framework; and

Basel Committee on Banking Supervision, Basel III: International Framework for Liquidity

Risk Measurement, Standards and Monitoring (Basel, Switzerland: December 2010,

revised June 2011). The Basel framework was adopted in 2010 and revised in 2011 and

2013.

Common equity tier 1 capital includes in part common shares and retained earnings.

Tier 1 capital, in part is the sum of common equity tier 1 capital and additional tier 1 (which

can include non-cumulative perpetual preferred shares).

Prior to the new regulations taking effect, MSRs were included in tier 1 capital up to 100

percent of their remaining unamortized book value (net of any related valuation

allowances) reported on an institution’s balance sheet or 90 percent of their fair value,

whichever was lower. See 78 Fed. Reg. at 62069.

Risk-Based Capital, 80 Fed. Reg. 66626 (Oct. 29, 2015).

Page 12 GAO-16-448 Mortgage Servicing

$100 million or less in assets from its risk-based capital regulations.

Also, all MSRs have a 250 percent risk-weight under the NCUA rule.

According to NCUA’s final rule, the intent is to reduce the likelihood that a

relatively small number of high-risk credit unions will exhaust their capital

to cover their financial obligations and cause systemic losses under the

Federal Credit Union Act, as amended. Under the act, all federally

insured credit unions would have to pay through the National Credit Union

Share Insurance Fund.

NCUA’s risk-based capital rules will take effect

on January 1, 2019.

The CFPB and federal prudential regulators face several statutory

requirements in the rule development process, including the Regulatory

Flexibility Act (RFA), as amended by the Small Business Regulatory

Enforcement Fairness Act of 1996 (SBREFA), and the Paperwork

Reduction Act of 1995 (PRA). The RFA requires that federal agencies

consider the impact on small entities of certain regulations they issue and,

in some cases, alternatives to lessen the regulatory burden on small

entities.

The PRA requires agencies to minimize the paperwork burden

of their information collections and evaluate whether a proposed

information collection is necessary for the proper performance of the

functions of the agency.

In addition, when promulgating any rule that

would have a significant economic impact on a substantial number of

small entities, CFPB must convene a review panel to collect the advice

and recommendations of small-entity representatives about the potential

NCUA defined small institutions in its October 2015 regulatory capital rules as those

having $100 million or less in assets. See 80 Fed. Reg. 66626.

The National Credit Union Share Insurance Fund is the federal fund created by

Congress in 1970 to insure members’ deposits in federally insured credit unions. Federally

insured credit unions must maintain 1 percent of their deposits in the Share Insurance

Fund. The purpose of the Share Insurance Fund’s capitalization deposit is to cover losses

in the credit union system.

Regulatory Flexibility Act, Pub. L. No. 96-354, 94 Stat. 1164 (1980) (codified as

amended at 5 U.S.C. §§ 601-612). Under RFA, agencies, including financial regulators,

generally must prepare a regulatory flexibility analysis in connection with certain proposed

and final rules, unless the head of the issuing agency certifies that the proposed rule

would not have a significant economic impact on a substantial number of small entities.

Paperwork Reduction Act of 1995, Pub. L. No. 104-13, 109 Stat. 163 (codified as

amended at 44 U.S.C. §§ 3501-3520).

Rule Development and

Retrospective Review

Processes

Page 13 GAO-16-448 Mortgage Servicing

impacts of the proposed rule prior to publishing the required initial

regulatory flexibility analysis.

Both CFPB and the banking regulators also have requirements to

retrospectively review rules after rules are in effect. The Dodd-Frank Act

requires CFPB to review its significant rules within 5 years of such rules

taking effect.

CFPB is required to assess the effectiveness of the rules

in meeting the purposes and objectives of Title X of the Dodd-Frank Act

and the specific goals stated by the agency, which include ensuring that

consumers receive timely and understandable information to make

responsible decisions about financial transactions and that markets for

consumer financial products and services are fair, transparent, and

competitive. In addition, the federal banking regulators are required by the

Economic Growth and Regulatory Paperwork Reduction Act of 1996

(EGRPRA) to review their regulations at least every 10 years to identify

outdated, unnecessary, or unduly burdensome regulations, consider how

to reduce regulatory burden on insured depository institutions, and

eliminate unnecessary regulations as appropriate.

The report from the

first EGRPRA review was submitted to Congress in 2007 and the second

review is underway.

The banking regulators have publicly stated they

anticipate completing the current EGRPRA review by the end of 2016.

NCUA conducts a voluntary review of its regulations on the same cycle

and in a manner consistent with the EGRPRA review. Additionally, per

NCUA’s internal policies, NCUA conducts a review of all of its regulations

every 3 years and produces a non-binding memorandum for its board

with suggestions on rules that should be revised or streamlined.

See 5 U.S.C. § 609(b). We have work underway looking at CFPB’s Small Business

Review Panel process addressing the extent that CFPB considered small business inputs

into its rulemaking and plan to issue a report later in 2016.

12 U.S.C. § 5512(d). Within 5 years of the effective date of each significant rule, CFPB

must conduct an assessment of the rule and publish a report of its assessment.

Pub. L. No. 104-208, § 2222, 110 Stat. 3009, 3009-414 (codified at 12 U.S.C. § 3311).

Joint Report to Congress, July 31, 2007; Economic Growth and Regulatory Paperwork

Reduction Act, 72 Fed. Reg. 62036 (Nov. 1, 2007).

Page 14 GAO-16-448 Mortgage Servicing

Many of the representatives from 16 community lenders—9 community

banks and 7 credit unions—that we interviewed noted that they have

maintained their customer-focused business models and continued to

service mortgages over the past 7 years. The total share of all mortgages

serviced by community banks and credit unions has increased since

2008. Some representatives of community banks and credit unions told

us that to manage their increased compliance costs of CFPB’s mortgage-

related rules required under the Dodd-Frank Act, they made adjustments

to certain business practices.

Based on our analysis, the total share of all U.S. residential mortgages

serviced by community lenders increased between 2008 and 2015.

Specifically, between the first quarter of 2008 and the third quarter of

2015, the share of mortgages serviced by community banks increased

from about 3.4 percent to about 6.8 percent, and the share serviced by

credit unions rose from about 3.1 percent to about 5.7 percent (see fig. 2).

The largest community banks and credit unions accounted for most of the

growth in the share of servicing by community banks and credit unions.

Over this period, the amount of residential mortgages outstanding fell

from about $11.3 trillion to about $10 trillion.

Figure 2: Estimated Share of Residential Mortgages Serviced by Community Banks, Credit Unions, and Nationwide, Regional,

and Other Banks, First Quarter 2008 through Third Quarter 2015 (percentage)

Community Lenders

Continue to Service

Mortgages as

Regulatory

Requirements

Increase

The Share of Mortgages

Serviced by Community

Lenders Has Increased

Since 2008

Page 15 GAO-16-448 Mortgage Servicing

Note: We used data on banks and credit unions that filed Call Reports for the period from the first

quarter of 2008 through the third quarter of 2015. We identified community banks using Federal

Deposit Insurance Corporation’s Historical Community Banking Reference Data. Banks that are not

community banks include nationwide and regional banks, internationally active banks, and specialty

banks. We estimated the share of outstanding residential mortgages a bank services by adding the

unpaid principal balance of residential mortgages held for investment, sale, or trading to the unpaid

principal balance of residential mortgages serviced for others and dividing the result by total

outstanding residential mortgages. We estimated the share of outstanding residential mortgages a

credit union services by adding the amount of residential mortgages on its balance sheet to the

amount of mortgages serviced for others and dividing the result by total outstanding residential

mortgages. These estimates may overstate the fraction of residential mortgages a bank or credit

union services because banks and credit unions may not service all of the residential mortgages on

their balance sheet.

Nationwide, regional, and other banks continue to service more than half

of the market. However, nonbank servicers—servicers that are not banks

or credit unions and also are not affiliates of banks or credit unions—have

increased their market presence. We previously estimated that the share

of U.S. residential mortgages serviced by nonbank servicers increased

from approximately 6.8 percent in the first quarter of 2012 to

approximately 24.2 percent in the second quarter of 2015.

At the same

time, the share serviced by the largest nationwide, regional, and other

banks decreased from about 75.4 percent to about 58.6 percent.

Many of the 16 community lenders we interviewed, which included

representatives from 9 community banks and 7 credit unions, and several

industry associations we spoke with told us that community banks and

credit unions serviced mortgages held in portfolio or held MSRs because

these activities generated income and allowed them to maintain strong

relationships with their customers. Some of these community lenders and

industry associations noted that holding mortgage loans in portfolio and

servicing these mortgage loans helped with overall profitability. For

example, the servicing revenue can offset a reduction in income from

originating loans when interest rates rise. Conversely, when interest rates

decline, borrowers are more likely to prepay or refinance their mortgage

loans, and servicing revenue may decline, while income from new

mortgage loan originations might increase. Also, representatives at these

institutions and two industry associations noted that servicing mortgages

allowed them to offer customers other revenue-producing products and

services. For example, representatives at one credit union told us that

See GAO-16-278.

Page 16 GAO-16-448 Mortgage Servicing

servicing mortgages provides it with the opportunity to develop borrowers

into full members with checking and savings accounts and car loans.

In addition to revenue, many community lenders noted that they and their

customers benefit from the close relationship maintained when these

institutions service mortgages. Representatives at several institutions we

interviewed emphasized that they were well positioned to work directly

with customers experiencing hardship to mitigate losses. For example,

representatives at one community bank told us that a customer who could

not make a mortgage payment could meet directly with a bank

representative to develop a payment plan. Customers of community

lenders whose mortgages are serviced by these institutions may

potentially also benefit by not being at risk of errors occurring during a

transfer of servicing, a process that has resulted in violations of consumer

protection laws and other regulations. As we noted in our March 2016

report on nonbank servicers, transfer errors can be especially harmful for

borrowers in delinquency or in the middle of loss mitigation proceedings.

Representatives at several industry associations and community lenders

that we interviewed for this report told us that community banks and credit

unions preferred to retain MSRs even if they sold the mortgages in the

secondary market because they were able to maintain close customer

contact should issues arise. A representative at one credit union told us

that it had sometimes needed to step in on behalf of customers to help

resolve issues, such as escrow errors, on mortgage loans they had sold

without retaining the MSRs.

See GAO-16-278.

Page 17 GAO-16-448 Mortgage Servicing

Many community lenders that we interviewed noted that they continued to

service mortgages in their portfolio or to hold MSRs on loans sold to the

secondary market in spite of increased compliance costs of mortgage-

related requirements resulting from new rules instituted pursuant to the

Dodd-Frank Act.

These rules cover both origination and servicing of

mortgage loans. They include new requirements, such as minimum

underwriting standards for mortgage loan originators, disclosures to

consumers about their mortgage loan obligations, and loss mitigation

procedures. Some community lenders that we spoke with noted that they

had increased staff, updated their data systems, or hired third parties to

assist with compliance activities to meet CFPB’s servicing rules. Similarly,

in our December 2015 report on the effect of Dodd-Frank Act regulations

on community banks and credit unions, we noted that representatives at

community lenders we interviewed and CFPB stated that the compliance

costs incurred by community lenders to implement new disclosures

included costs of having to work with third-party vendors to update their

loan origination and documentation system software.

In an industry

survey of a nonprobability sample of banks in which the majority of

respondents had less than $1 billion in assets, over 80 percent of

respondents noted that increased personnel costs and staff time allocated

to compliance issues were the primary drivers of increased compliance

See Escrow Requirements Under the Truth in Lending Act (Regulation Z), 78 Fed. Reg.

4726 (Jan. 22, 2013); Ability-to-Repay and Qualified Mortgage Standards Under the Truth

in Lending Act (Regulation Z), 78 Fed. Reg. 6408 (Jan. 30, 2013); Mortgage Servicing

Rules Under the Real Estate Settlement Procedures Act (Regulation X), 78 Fed. Reg.

10696 (Feb. 14, 2013); Mortgage Servicing Rules Under the Truth in Lending Act

(Regulation Z), 78 Fed. Reg. 10902 (Feb. 14, 2013).

GAO, Dodd-Frank Regulations: Impacts on Community Banks, Credit Unions and

Systemically Important Institutions, GAO-16-169 (Washington, D.C.: Dec. 30, 2015).

Many Community Lenders

Reported Spending More

in Response to Regulatory

Requirements and Some

Adjusted Business

Activities, Potentially

Affecting Customers

Page 18 GAO-16-448 Mortgage Servicing

costs. In addition, nearly 70 percent cited increased costs for third-party

vendor services.

Some representatives at community banks and credit unions we spoke

with commented that CFPB’s exemptions for small servicers and creditors

had been helpful to their businesses and customers. Several community

lenders noted that CFPB’s small servicer exemption, which excludes from

certain parts of CFPB’s mortgage servicing rules entities that service

5,000 or fewer mortgages, had been helpful in reducing some of their

compliance requirements.

For example, representatives at one

community bank noted that it was nearing the threshold and would have

to create additional processes and install software, but would not

intentionally limit its growth to qualify for the exemption. The community

bank representatives also noted that the bank was preparing for the

additional regulatory requirements by forming a mutual holding company

with another bank to achieve greater economies of scale.

Representatives at another bank noted that CFPB’s small creditor

exemption has allowed the bank to make loans that are considered

qualified mortgages based on the bank’s evaluation of a customer’s debt

American Bankers Association, 22nd Annual ABA Real Estate Survey Report

(Washington D.C.: 2015). The 22nd Real Estate Lending Survey had the participation of

182 banks. The web survey was sent out to over 3,000 banks and elicited response from

182 banks for a response rate of 6 percent overall. The data were collected from March 4,

2015, to April 17, 2015, and in most cases report calendar year or year-end results. In

other cases, data reflect current activities and expectations at the time of data collection.

Of the survey participants, 68 percent of respondents were commercial banks and 32

percent were savings institutions. About 77 percent of the participating institutions had

assets of less than $1 billion. Because the survey relies on a nonprobability sample, the

results cannot be used to make generalizations about all commercial banks and thrifts.

According to CFPB documentation, this definition covers substantially all of the

community banks and credit unions that are involved in servicing mortgages. Although the

rules exempt small servicers from certain provisions, they require, for example, all

servicers to respond to written notices of errors received from borrowers, and with respect

to loss mitigation, a small servicer is required to comply with two requirements: (1) a small

servicer may not make the first notice or filing required for a foreclosure process unless a

borrower is more than 120 days delinquent, and (2) a small servicer may not proceed to

foreclosure judgment or order of sale, or conduct a foreclosure sale, if a borrower is

performing pursuant to the terms of a loss mitigation agreement.

Page 19 GAO-16-448 Mortgage Servicing

and income even though the customer did not meet the specific debt-to-

income ratio required under CFPB’s rule.

Some community lenders noted that to manage the increased compliance

costs, they made adjustments to both loan origination and servicing

business practices that could affect their customers’ costs or choices.

These adjustments included raising fees and interest rates and changing

product offerings. Representatives at one credit union noted that while it

had absorbed additional regulatory compliance costs into its general

overhead expenses, it had to raise additional revenue to cover these

additional expenses, such as increasing underwriting fees for mortgage

applications. Several institutions noted that they no longer offered

customers certain products because offering them would require

additional compliance testing to meet regulatory requirements. For

example, a representative at one community bank said that it no longer

offers home equity lines of credit to its customers due to costs associated

with complying with CFPB’s rules related to increased disclosures to the

customer.

Representatives at another community bank said it no longer

offered bridge loans—short-term loans typically used when a consumer is

Under CFPB’s Ability-to-Repay and Qualified Mortgage Standards Under the Truth in

Lending Act (Regulation Z), 78 Fed. Reg. 6408 (Jan. 30, 2013) (ATR/QM rule) rule, which,

as amended, became effective on January 10, 2014, generally, lenders making qualified

mortgages must make a good-faith effort to determine that the borrower has the ability to

repay the mortgage loan by documenting a borrower’s income, assets, employment, credit

history, and monthly expenses. The ATR/QM rule sets out several categories of qualified

mortgage: general, temporary, and small creditor. Although small creditors must consider

and verify the borrower’s debt-to-income ratio, these loans are not subject to a specific

debt-to-income ratio, such as the 43 percent or less determination under the general

category. These loans must be made by small creditors and generally held in portfolio for

at least 3 years. Small creditors are generally creditors that together with their affiliates

have less than $2 billion in assets (adjusted annually for inflation) and originated no more

than 2,000 first-lien mortgage loans in the preceding year. Ability-to-Repay and Qualified

Mortgage Standards Under the Truth in Lending Act (Regulation Z), 78 Fed. Reg. 35430,

35503 (June 12, 2013) and 80 Fed. Reg. 59944, 59968 (Oct. 2, 2015).

A home equity line of credit is a form of revolving credit in which a borrower’s home

serves as collateral. Many lenders set the credit limit on a home equity line by taking a

percentage of the home’s appraised value and subtracting from that the balance owed on

the existing mortgage.

Page 20 GAO-16-448 Mortgage Servicing

buying a new home before selling the consumer’s existing home—to its

customers.

Other community lenders told us that despite increased mortgage-related

regulatory requirements under the Dodd-Frank Act, they ensured their

customers’ access to credit and maintained close customer contact.

According to an FDIC report, community banks are often considered to be

“relationship” bankers and tend to base credit decisions on local

knowledge and long-term relationships with customers.

Representatives

at some community banks and credit unions we spoke with noted that

because of their familiarity with customers, they were retaining mortgage

loans in their portfolios that no longer met tighter credit restrictions in the

secondary market under the Dodd-Frank Act.

They explained that

holding these mortgage loans allowed them more flexibility in their

underwriting of these mortgage loans, which facilitated greater access to

credit for their customers. Another community lender told us that although

regulatory compliance costs necessitated having to outsource mortgage

servicing to a third-party servicer, a credit union service organization, for

its mortgage servicing activities, the credit union still speaks directly with

Usually secured by the existing home, a bridge loan provides financing for the new

home (often in the form of the down payment) or mortgage payment assistance until the

consumer can sell the existing home and secure permanent financing. Bridge loans

normally carry higher interest rates, points, and fees than conventional mortgages,

regardless of the consumer’s creditworthiness.

Community banks tend to focus on providing essential banking services in their local

communities, and are often considered to be “relationship” bankers. This means that they

have specialized knowledge of their local community and their customers. Because of this

expertise, community banks tend to base credit decisions on local knowledge and

nonstandard data obtained through long-term relationships and are less likely to rely on

the models-based underwriting used by larger banks. See Benjamin R. Backup and

Richard A. Brown, “Community Banks Remain Resilient Amid Industry Consolidation,”

FDIC Quarterly, vol. 8, no. 2, 2014 accessed on March 29, 2016 at

https://www.fdic.gov/bank/analytical/quarterly/2014_vol8_2/article.pdf.

The enterprises generally purchase conforming loans, which are mortgage loans that

meet certain criteria for size, features, and underwriting standards. The enterprises’

underwriting includes assessments of measures of the credit risk of purchasing

mortgages, such as borrower credit scores and debt-to-income ratios. For example, the

enterprises have a debt-to-income ceiling of 45 percent.

Page 21 GAO-16-448 Mortgage Servicing

borrowers who typically prefer to communicate with the mortgage loan

officer because they have a pre-existing relationship.

Although new regulations related to mortgage lending and servicing may

increase compliance costs for community banks, our analysis suggests

that these lenders generally appear to be participating in residential

mortgage lending much as they have in the past. We found that for every

quarter from the first quarter of 2001 through the third quarter of 2015,

over 97 percent of community banks of all sizes had residential

mortgages on their balance sheets (see fig. 3).

For most community

banks with residential mortgages, these mortgages continued to average

at least 10 percent of assets in their portfolio. Over the period from the

first quarter of 2001 through the third quarter of 2015, median residential

mortgages were 11 percent to 15 percent of assets for the smallest

community banks and 16 percent to 20 percent of assets for larger

community banks.

In addition, median residential mortgages as a

percentage of assets have generally increased in the past couple of years

for community banks of all sizes. Thus, community banks generally do not

appear to be shifting their portfolios away from mortgage lending.

Credit union service organizations are entities that are owned by federally chartered or

federally insured state-chartered credit unions and that are engaged primarily in providing

products or services to credit unions or credit union members. See 12 C.F.R. § 712.1(d).

We used total assets to measure size and we divided all banks—community banks and

nationwide, regional, and other banks—into five equal-sized groups, or quintiles, based on

their size each quarter. The first quintile contained the smallest banks, the second quintile

contained the next largest banks, and so on through the fifth quintile, which contained the

largest banks. In the third quarter of 2015, the largest bank in the first quintile had assets

of about $73.9 million, the largest bank in the second quintile had assets of about $138.5

million, the largest bank in the third quintile had assets of about $250.6 million, the largest

bank in the fourth quintile had assets of about $545.1 million, and the largest bank in the

fifth quintile had assets of about $2 trillion. There were 1,187, 1,248, 1,243, 1,234, and

899 community banks in the first, second, third, fourth, and fifth quintiles, respectively.

We calculated residential mortgages as a percentage of assets for every bank and then

calculated the median value of residential mortgages as a percentage of assets for

community banks in each size group, where the median value is the middle or 50

percentile value when the values are ordered from smallest to largest.

Community Lenders Are

Continuing Residential

Mortgage Lending and

Have Not Changed Their

Customer-Focused

Business Model

Page 22 GAO-16-448 Mortgage Servicing

Figure 3: Percentage of Community Banks, Credit Unions, and Nationwide, Regional, and Other Banks with Residential

Mortgages on their Balance Sheets by Size, First Quarter 2001 through Third Quarter 2015 (percentage)

Note: We used data on banks that filed Call Reports for the period from the first quarter of 2001

through the third quarter of 2015 and on credit unions that filed Call Reports from the second quarter

of 2002 through the third quarter of 2015. We assigned banks to groups each quarter using quintiles

based on the distribution of their total assets, where the first quintile contains the smallest 20 percent

of banks, the second group contains the next largest 20 percent of banks, and so on through the fifth

quintile, which contains the largest 20 percent of banks. We identified community banks using

Federal Deposit Insurance Corporation’s Historical Community Banking Reference Data. Banks that

are not community banks include nationwide and regional banks, internationally active banks, and

specialty banks. We also assigned credit unions to groups each quarter using quintiles based on the

distribution of their total assets. Banks that hold residential mortgages are those that hold residential

mortgages for investment, sale, or trading. Credit unions that hold residential mortgages are those

that hold first lien residential mortgages or any other real estate loan or line of credit, which typically

includes second mortgages, and home equity lines of credit, but may also include some member

business loans secured by subordinate real estate liens. Thus, our calculations may overstate the

fraction of credit unions with residential mortgages.

Similarly, the largest nationwide, regional, and other banks (those in the

fifth quintile) generally do not appear to be changing the extent to which

they participate in mortgage lending. Over the period from the first quarter

of 2001 through the third quarter of 2015, over 89 percent of the largest

nationwide, regional, and other banks had residential mortgages on their

Page 23 GAO-16-448 Mortgage Servicing

balance sheets.

Among those institutions that had residential

mortgages, median residential mortgages were between 13 percent and

16 percent of their assets. In contrast, the percentage of smaller

nationwide, regional, and other banks (those in the first, second, third,

and fourth quintiles) with residential mortgages on their balance sheet has

generally decreased over the period.

Finally, our analysis suggests that credit unions are generally participating

in residential mortgage lending at least as much as they have in the past.

Throughout the period from the first quarter of 2002 through the third

quarter of 2015, larger credit unions were more likely to have residential

mortgages than smaller credit unions.

However, for credit unions of all

sizes, the percentage with residential mortgages increased. Among credit

unions with residential mortgages, larger credit unions typically had more

residential mortgages as a percentage of assets than small credit unions.

Median residential mortgages ranged from 6 percent to 10 percent of

assets for the smallest credit unions, and up to 24 percent to 35 percent

of assets for the largest credit unions. While median residential

mortgages as a percentage of assets have decreased for the smallest

credit unions in recent quarters, they have remained constant or

increased in recent quarters for larger credit unions. Like community

banks, credit unions generally do not appear to be shifting their portfolios

away from mortgage lending, with the possible exception of the smallest

institutions.

There were 77, 16, 20, 30, and 364 nationwide, regional, and other banks in the first,

second, third, fourth, and fifth quintiles, respectively.

As we did with banks, we used total assets to measure size and we divided credit unions

into quintiles based on their size each quarter. In the third quarter of 2015, the largest

credit union in the first quintile had assets of about $5.1 million, the largest credit union in

the second quintile had assets of about $16.3 million, the largest credit union in the third

quintile had assets of about $42.3 million, the largest credit union in the fourth quintile had

assets of about $139.3 million, and the largest credit union in the fifth quintile had assets

of about $72 billion. There were 1,244 credit unions in the first quintile and 1,243 credit

unions in the second, third, fourth, and fifth quintiles.

Page 24 GAO-16-448 Mortgage Servicing

Our analysis suggests that the capital treatment of MSRs would not likely

have a material effect on most community banks because they generally

did not have large concentrations of MSRs. Most community banks we

interviewed confirmed that they did not need to make changes to their

capital because of these rules, but those with large concentrations of

MSRs were considering changes to their capital. We also do not expect

the capital treatment of MSRs to affect most applicable credit unions

based on our analysis and discussions with NCUA and credit unions.

Regardless of type or size, banks with large concentrations of MSRs may

need to raise capital as a result of the new risk-based capital treatment of

MSRs, but our analysis and most community banks we spoke with

confirmed that these new rules were currently not an issue for them.

stated earlier, new banking requirements being phased in by January 1,

2018, include requiring any amount of MSRs above 10 percent of a firm’s

common equity tier 1 (CET1) capital to be deducted from CET1 capital. In

addition, MSRs, adjusted for amounts deducted from CET1 capital, are

currently assigned a 100 percent risk weighting, which means that every

$1 in these MSRs adds $1 to risk-weighted assets. When the new

requirements are fully phased in, any MSRs not deducted from CET1

capital will be assigned a 250 percent risk weighting, which means every

$1 in these MSRs will add $2.50 to risk-weighted assets (see fig. 4).

Some banks may have to increase CET1 capital to ensure that their ratios

of CET1 capital to risk-weighted assets remain above the required

Prudential regulators’ capital rules implementing Basel III include provisions related to

MSRs that are to be fully phased in by 2018 and that will affect banks’ regulatory capital

as well as their risk-weighted assets. For a more complete discussion of Basel III, see

GAO, Bank Capital Reforms: Initial Effects of Basel III on Capital, Credit, and International

Competitiveness, GAO-15-67 (Washington, D.C.: Nov. 20, 2014).

Any amount of MSRs, certain deferred tax assets arising from temporary differences,

and significant investments in the capital of unconsolidated financial institutions in the form

of common stock (collectively, “threshold items”) above 15 percent of a firm’s CET1 capital

must be deducted from CET1 capital. Starting January 1, 2018, any amount of the

threshold items that is not deducted from CET1 capital will be risk weighted at 250

percent.

New Capital

Treatment of

Mortgage Servicing

Rights Likely Will Not

Affect Many

Community Banks

and Credit Unions

Most Community Banks

and Credit Unions Hold

Limited MSRs and Have

Sufficient Capital to Cover

Them

Page 25 GAO-16-448 Mortgage Servicing

minimum.

Those banks that need to raise capital could face increased

funding costs that they could, in turn, pass on to consumers through

increased cost or reduced availability of credit. However, most community

banks subject to the capital requirements we spoke with said that they did

not expect to need to make changes to their capital because they did not

hold large concentrations of MSRs and had sufficient capital.

Figure 4: Mortgage Servicing Rights and Risk-Based Capital Ratios

Note: Mortgage servicing rights (MSRs) up to 10 percent of common equity tier 1 (CET1) capital will

receive a 250 percent risk weight when the new risk-based capital requirements are fully phased in.

The remaining MSRs are not included in risk-weighted assets.

Based on our analysis, the capital treatment of MSRs likely would not

have a material effect on most community banks because they generally

did not have large concentrations of MSRs. Data on banks’ holdings of

MSRs for the period from the first quarter of 2001 through the third

quarter of 2015 showed that larger community banks were more likely to

have MSRs than smaller community banks and that more community

banks of all sizes were holding MSRs in 2015 than in 2001. However,

these assets were typically a small fraction of total assets. Specifically,

the median value of MSRs as a percentage of total assets has remained

at less than 1 percent in each quarter since the first quarter of 2001.

Additionally, our analysis showed that about 19 percent of community

banks held MSRs and about 1 percent made MSR-related deductions

from capital due to the amount of these assets they held as of the third

quarter of 2015 (see fig. 5).

Two banking regulators noted that a bank’s capital is a function of its overall risk profile

and not just the MSR amount.

Page 26 GAO-16-448 Mortgage Servicing

Figure 5: Community Banks’ Capital Treatment of Mortgage Servicing Rights,

2015Q3

Note: Estimates of the percentage of community banks holding mortgage servicing rights (MSR) and

the percentage making deductions from their capital because of their MSR holdings are based on Call

Report data from the third quarter of 2015.

Most community banks we interviewed confirmed that they did not need

to make changes to their capital because of these rules, but those with

large concentrations of MSRs were considering changes to their capital.

Institutions with large concentrations of MSRs may choose to raise

additional capital or sell MSRs to meet required minimum capital

amounts, depending on banks’ holdings of other types of assets. Most

representatives of community banks said that regulatory changes to the

capital treatment of MSRs did not require them to sell MSRs or raise

additional capital. For example, two banks stated that they were not close

to the 10 percent threshold and one of the banks noted that their total

MSR values were small relative to their total capital levels.

However, two other banks we interviewed with large concentrations of

MSRs (one of which was a midsized regional bank) stated that they would

likely be affected by the MSR provisions in the risk-based capital rules

and were considering options to raise capital levels by selling MSRs or

other ways. For example, one community bank with less than $1 billion in

assets told us that MSRs equaled 47 percent of total capital and that

under the new risk-based capital rules the bank would need to make a

$12 million deduction from regulatory capital. The bank officials stated

that the rules would prevent the bank from growing as much as it would

Page 27 GAO-16-448 Mortgage Servicing

like. Another midsized regional bank that services mortgages it originates

and purchases MSRs from others said that it was evaluating how to

reduce its MSR holdings, which currently equaled about 22 percent of

total capital, without affecting revenue. Bank officials said that the 10

percent threshold and the 250 percent risk weight were hindrances. One

industry survey on capital treatment of MSRs found that about 5 percent

of respondents had sold MSRs in the past year, and 14 percent were

contemplating selling MSRs due to new regulatory requirements or capital

treatment of MSRs compared to 11 percent in 2013.

Further, our analysis of data on the ratios of CET1, tier 1, and total capital

to risk-weighted assets for the first three quarters of 2015 showed that

almost all community banks in every size group met or exceeded

regulatory minimums in each of the first three quarters of 2015. Over 99

percent of community banks of all sizes had CET1 capital ratios greater

than or equal to the minimum required ratio of 4.5 percent plus a 2.5

percent buffer in all three quarters. Similarly, over 98 percent had tier 1

capital ratios greater than or equal to the minimum of 6.0 percent plus a

2.5 percent buffer, and over 97 percent had a total capital ratio greater

than or equal to the minimum of 8.0 percent plus a 2.5 percent buffer.

Finally, we found that while most community banks of all sizes had capital

ratios well in excess of the regulatory minimums, small community banks

typically had higher capital ratios than larger community banks (see table

2).

The American Bankers Association’s The 22nd Real Estate Lending Survey had the

participation of 182 banks. The data were collected from March 4, 2015, to April 17, 2015,

and in most cases reports calendar year or year-end results. In other cases, data reflect

current activities and expectations at the time of data collection. Of the survey participants,

68 percent of respondents were commercial banks and 32 percent were savings

institutions. About 77 percent of the participating institutions had assets of less than $1

billion.

Page 28 GAO-16-448 Mortgage Servicing

Table 2: Median Risk-Based Capital Ratios for Community Banks by Size and Required Minimum Capital Ratios, Third Quarter

2015

Median for first

quintile

(smallest)

Median for

second

quintile

Median for

third

quintile

Median for

fourth

quintile

Median for

fifth quintile

(largest)

Required

minimum plus

2.5 percent

buffer

CET1 capital ratio

18.3

16.4

15.2

14.5

13.1

7.0

Tier 1 capital ratio

18.3

16.4

15.2

14.5

13.2

8.5

Total capital ratio

19.5

17.6

16.4

15.6

14.3

10.5

Source: GAO analysis of data from the Federal Deposit Insurance Corporation and Federal Financial International Examinations Council. | GAO-16-448

Note: We used data on common equity tier 1 (CET1), tier 1, and total risk-based capital ratios

reported by banks that filed Call Reports for the third quarter of 2015. We assigned banks to groups

based on the distribution of their total assets. We divided banks into quintiles, where the first quintile

contains the smallest 20 percent of banks, the second group contains the next largest 20 percent of

banks, and so on through the fifth quintile, which contains the largest 20 percent of banks. We

identified community banks using the Federal Deposit Insurance Corporation’s Historical Community

Banking Reference Data.

We estimated how much banks’ CET1 capital ratios might increase if they

replaced their MSRs with an asset such as U.S. Treasury securities that

could be held in any amount without reducing capital and that was not

included in risk-weighted assets. Analysis of Call Report data on capital

and risk-weighted assets for 2015 suggests that replacing MSRs with

U.S. Treasury securities would have little effect on the capital ratios of

most community banks with MSRs. If a community bank replaced its

MSRs with U.S. Treasury securities, which do not require a deduction in

CET1 capital and which have a zero percent risk-weight, then its CET1

capital would increase and its risk-weighted assets would fall, all else

being equal. However, our estimates suggest that for most community

banks with MSRs, the increase in the CET1 capital ratio would be less

than 1 percentage point.

NCUA finalized its risk-based capital rule on October 29, 2015. The rule

applies to credit unions regulated by NCUA with assets over $100 million

Our analysis has limitations and should be interpreted with caution. Specifically,

available data do not include all of the amounts required to calculate how the CET1 capital

ratio would change if banks with MSRs replaced their MSRs with U.S. Treasury securities,

so we made assumptions that likely caused us to overestimate that change. In addition,

the new risk-based capital requirements are not fully phased-in, and banks may not have

fully adjusted to them, so these results may not be indicative of the extent to which banks

will be affected by the new risk-based capital requirements in the future. See appendix I

for more details of our analysis.

Page 29 GAO-16-448 Mortgage Servicing

and assigns a 250 percent risk-weight to MSRs.

Based on discussions

with NCUA and market participants, we also do not expect the capital

treatment of MSRs to affect most applicable credit unions. NCUA officials

told us that it did not expect the risk-based capital standards to prevent

credit unions from holding MSRs. Some credit unions we spoke with also

stated that NCUA’s risk-based capital rule was not likely to affect them.

One credit union told us that it was well capitalized and that the risk-

based capital rule would likely not affect it. This credit union has an MSR

value of about $1.2 million, which equates to 3 percent of its capital.

Another credit union stated that it would not need to do anything

differently, and another did not see the risk-based capital threshold as an

issue. NCUA’s risk-based capital rules will take effect on January 1, 2019.

Changes in banks’ funding costs and costs to consumers as a result of

the regulatory capital rules are likely to be modest. Since most community

banks likely would not have to increase capital because of these rules,

costs to consumers similarly are not expected to increase. However,

banks that do have to increase capital may need to cover increased

funding costs, which could affect the cost and availability of credit to

consumers.

We found in a 2015 report on risk-based capital rules that

raising capital to cover any capital shortfalls associated with new

minimum capital requirements, including provisions related to MSRs,

would likely have a modest effect on the cost and availability of credit to

consumers.

First, in that report we estimated that the total amount of

capital that banks would need to raise to meet the new minimum capital

ratios would likely amount to less than 1 percent of total assets. Second,

our report noted that although funding costs for banks that increase equity

capital to meet new minimum capital requirements could increase, we

estimated that the change would likely be small, generally about 0.3

percentage points or less. Finally, we estimated in our 2015 report that

any increases in loan rates consumers experience due to increases in

funding costs also are likely to be small in part because banks may be

limited in the extent to which they can raise lending rates. In part, banks

respond to changes in their funding costs in several ways. Banks that

Risk-Based Capital, 80 Fed. Reg. 66626 (Oct. 29, 2015).

Equity funding is generally more expensive than debt funding, so increasing capital will

cause overall funding costs to increase, all else being equal.

See GAO-15-67.

Page 30 GAO-16-448 Mortgage Servicing

need to raise capital may cover their increased funding costs by raising

loan rates, shifting lending activity to lower-risk borrowers, and increasing

efficiency.

Our analyses of loan transfer data show that smaller banks and credit

unions maintained a relatively small level of participation in the MSR

market. Between 2010 and the first half of 2015, MSRs representing the

right to service totaled approximately $2.1 trillion worth of unpaid principal

balance were sold via bulk sales associated with Fannie Mae, Freddie

Mac, and Ginnie Mae mortgage pools.

Over this period, banks and

credit unions with assets of less than $10 billion represented about 13

percent of MSR sales and 4 percent of MSR purchases as a portion of

the unpaid principal balance within Freddie Mac and Ginnie Mae loan

pools.

Nonbanks accounted for most of the remaining purchases of

MSRs backed by mortgages in Freddie Mac and Ginnie Mae mortgage

pools over the period, and large banks and credit unions—those with

more than $10 billion in assets—and nonbanks had roughly equal

proportions of sales. See appendix II for more information about MSR

purchases and sales.

Market participants, regulators, and mortgage brokers commented that

recent trends in the market for MSRs have been influenced by a number

of factors, including volatility in the value of MSRs, compliance risk,

interest rates and prepayments, and regulatory capital requirements. For

example, volatility in the value of MSRs makes pricing this asset difficult,

which could affect banks’ decisions about retaining or selling MSRs. One

regulator explained that MSR values could be volatile because of

assumptions about their perceived value and noted that it advised

institutions with MSR concentrations to use a third-party valuator to test

these assumptions. One mortgage broker we spoke with explained that

the fair value of MSRs is difficult to determine using observable measures

The MSR associated with these loans often is measured as a function of the unpaid

principal balance. In addition to bulk sales, MSRs may be sold through co-issue

transactions in which the MSR is sold concurrently with the sale of the associated

mortgage into the secondary market. According to one government-sponsored enterprise,

these types of arrangements are more common among nonbanks than banks.

Data provided by Fannie Mae did not allow GAO to perform a comparable analysis to

Freddie Mac and Ginnie Mae.

Community Lenders Are

Likely to Remain Minor

Participants in MSR Sales,

Which Can Be Influenced

by Several Factors

Page 31 GAO-16-448 Mortgage Servicing

such as market prices or models. Also, fair value can be difficult to

determine due to the lack of comparable active trades. According to one

bank, it prefers not to manage the volatility associated with MSRs, given

the need to reflect MSR values on the balance sheet.

The bank sells its

loans along with the servicing. The bank official added that he does not

believe the bank’s ability to retain customers is negatively affected by its

decision not to perform mortgage servicing. However, some community

lenders we spoke with did not see volatility in MSR values as a concern

for their business operations. These lenders viewed MSRs as a part of

their broader business decisions (e.g., selling loans into the secondary

market and retaining servicing) rather than as assets to be monitored

closely.

Increased concerns about compliance risk, especially uncertainty

associated with new regulations, could potentially decrease the value of,

and thus demand for, MSRs. Failure to comply with CFPB’s new

consumer financial laws related to originating and servicing mortgages

can result in possible losses from litigation and enforcement actions. One

government-sponsored enterprise expressed concern that regulatory

costs and penalties might push competent servicers out of the market.

One mortgage broker stated that aside from operational risks, there has

been a cumulative effect of recent regulations on servicers, which

includes increased costs, as well as uncertainty about future costs. In

terms of compliance, a regional bank we spoke with stated that although

the regulations brought more clarity around requirements and consistency

in how bank staffs interact with customers, the resources needed to keep

up with the regulations had the biggest effect on the bank. For example,

the bank officials stated that in 2013 the bank prepared a project

management document to map out the implementation and systems

requirements needed to comply with all the new regulations. The bank set

up committees around a number of regulatory changes. FHA officials

shared this concern, stating that they had heard servicers complaining

about uncertainty associated with the enforcement of regulations and

penalties they might face. These officials believed that servicers needed

more time to adapt their systems to existing regulations and increase their

Under the fair value method of accounting, changes in the value of MSRs must be

reflected on the bank’s balance sheet immediately.

Page 32 GAO-16-448 Mortgage Servicing

capacity to implement changes before additional regulations were

developed.

Interest rates and prepayments may also affect the value of MSRs. Rising

interest rates can reduce homebuyers’ willingness or ability to finance a

real estate loan. Higher rates could negatively affect both the volume of

loan originations and profitability. Rising interest rates and lower

prepayments can result in increased value of MSR portfolios. In contrast,

when interest rates fall, prepayment speeds increase as borrowers

refinance their mortgages, causing a decline in MSRs’ value, because

MSRs associated with prepaid loans are eliminated.

Though MSR

values are expected to decrease when prepayments increase, MSR

holders may be able to recapture lost revenue if they have a mortgage

origination business. While servicing is seen as a business hedge that

offsets fluctuations in mortgage lending, changes in MSR values may

occur more rapidly than changes in the rates of mortgage originations.

In addition, a market development or policy change that encourages or

discourages borrowers to refinance their loans could impact prepayments

and thus the value of MSRs. For instance, in a January 2015 press

release, the Department of Housing and Urban Development announced

that the Federal Housing Administration would reduce annual insurance

premiums that new borrowers pay by half a percent. This reduction in

insurance premiums would lower the cost of owning a home and may

contribute to a decrease in mortgage prepayments. According to the

press release, the action is projected to save more than 2 million FHA

homeowners an average of $900 annually and spur 250,000 new

homebuyers to purchase their first homes over the next 3 years.

Finally, a few market participants we spoke with attributed some sales by

large banks to nonbanks in the MSR market to the anticipated changes to

the treatment of MSRs in risk-based capital requirements. For example,

officials from one government-sponsored entity explained that in some

One community bank with a relatively large concentration of MSRs related to

commercial real estate said that they had prepayment penalties built into the terms of their

loans that could protect investors from prepayments.

Servicing offsets fluctuations in mortgage lending. On the production side (i.e.,

originations), margins and volumes are highest when interest rates are low. Conversely,

on the servicing side, MSRs gain value when interest rates rise (i.e., number of

prepayments expected to decrease).

Page 33 GAO-16-448 Mortgage Servicing

cases banks with high levels of MSRs on their balance sheet may be

selling their MSRs to nonbank subsidiaries but would continue to service

the loan as a subservicer. Additionally, one mortgage broker told us that

regulatory capital requirements motivated some banks to sell MSRs, but

added that for a majority of banks, regulatory capital requirements are a

non-issue. As mentioned earlier, two banks with large concentrations of

MSRs stated that they expected to be impacted by mortgage servicing

regulatory capital rules in the future. These banks were considering

options to adjust their required capital levels, such as selling MSRs, but

they noted that selling MSRs could limit growth and decrease their overall

profitability.

CFPB used several methods to estimate the impact of its mortgage

servicing rules prior to finalizing them in 2013, and banking regulators

incorporated changes to the treatment of MSRs when they estimated the

overall impact of new regulatory capital rules in 2013. Both CFPB and

banking regulators have begun preparing to retrospectively review these

rules. However, CFPB’s plans are limited because the agency has not

finalized an approach for retrospectively reviewing its mortgage servicing

rules. Banking regulators have incorporated the regulatory capital rules as

part of a retrospective review of all their rules—which is required every 10

years—although the review will be complete before these rules are fully

phased in.

CFPB used several methods to estimate the impact of its mortgage

servicing rules prior to finalizing the rules in 2013. For example, prior to

issuing proposed rules, CFPB officials met with representatives of small

entities that would be subject to the rule to obtain input about CFPB’s

proposals on mortgage servicing requirements. As noted earlier, when

promulgating any rule that would have a significant economic impact on a

substantial number of small entities, CFPB must convene a review panel

to collect the advice and recommendations of small entity representatives

prior to issuing a proposed rule.

The panel received comments from

representatives of 16 small entities, including 5 banks and 5 credit unions.

These representatives’ comments played a significant role in persuading

See 5 U.S.C. § 609(b).

Regulators Estimated

Impacts of New Rules

Using Public Input

and Data Analysis,

but CFPB’s Plans for

Reviewing Rules

Have Limitations

CFPB Used Multiple

Methods to Obtain Public

Comments and Estimate

Impact Prior to Issuing

Final Mortgage Servicing

Rules

Page 34 GAO-16-448 Mortgage Servicing

CFPB to propose exempting smaller servicers from certain mortgage-

servicing requirements, agency officials said.

CFPB also sought comments on its proposed rule both by issuing a

formal request for public comment in the Federal Register and seeking

public input via an interactive website called “Regulation Room” operated

by Cornell University’s eRulemaking Initiative. Agency officials said they

hoped the Regulation Room would allow CFPB to obtain a broader range

of perspectives than those coming from a formal request for comment.

CFPB received 347 comments from the Regulation Room on questions

on topics such as options for avoiding foreclosure and getting errors fixed

(see fig. 6). CFPB officials said comments on the proposed rules

prompted the agency to scrutinize the definition of small servicers and

ultimately led to CFPB expanding eligibility for the small-servicer

exemption in the final rule. The final rule expanded the small-servicer

exemption to institutions with up to 5,000 mortgages, up from the

proposed rule’s 1,000-loan threshold. Specifically, public comments

prompted the agency to analyze state-level averages on loan balances

rather than the national averages used for the proposed rule.

This

change in the underlying data from national to state-specific information

led CFPB to determine that a 5,000-loan threshold would better meet the

agency’s goal of having the small-servicer exemption apply to nearly all

banks with $2 billion or less in assets, officials said.

The eRulemaking Initiative describes the Regulation Room website as a pilot project that

provides an online environment for people and groups to learn about, discuss, and react

to selected regulations proposed by federal agencies. It expands the types of public input

available to agencies in the rulemaking process, while serving as a teaching and research

platform.

Using state-specific data on average loan balance—rather than national data—changed

the results of the agency’s analysis of the proportion of banks that would qualify for the

exemption, agency officials said.

In the preamble to its final rule, CFPB said two bank trade associations and the Small

Business Administration recommended an exemption threshold of up to 10,000 loans.

CFPB determined such a threshold was too high because it would also exempt 50 percent

of banks and credit unions with more than $2 billion in assets, including 20 percent of

those institutions with more than $10 billion in assets. See Mortgage Servicing Rules

Under the Truth in Lending Act (Regulation Z), 78 Fed. Reg. 10902, 10981 (Feb. 14,

2013).

Page 35 GAO-16-448 Mortgage Servicing

Figure 6: Sample Public Comments from the “Regulation Room” on CFPB’s Proposed Mortgage Servicing Rule

CFPB identified the exemption for small servicers as an element in

mitigating compliance costs for small banks and credit unions. The

agency said estimates showed that 99 percent of banks and credit unions

with less than $1 billion in assets—and 98 percent of banks with $2 billion

or less in assets—would be eligible for the small servicer exemption that

would eliminate the requirements to comply with many components of the

mortgage-servicing rules. For example, small servicers are not required to

engage in certain types of early intervention with delinquent borrowers,

Page 36 GAO-16-448 Mortgage Servicing

such as providing a written description of loss-mitigation options within 45

days of a borrower’s delinquency.

In addition to its approaches to obtain public comment, CFPB conducted

several required analyses before issuing the final rule, including an

analysis of the potential impact on small banks and credit unions, as

required by the RFA.

CFPB also considered the potential costs and

benefits for consumers and all banks and credit unions, as required by

the Dodd-Frank Act.

For both analyses, CFPB described potential

impacts but did not give a specific financial estimate of expected impacts,

citing data limitations as a barrier to quantifying impacts. For example,

CFPB cited data gaps in servicers’ costs for vendor services, such as the

one-time and ongoing costs that vendors were likely to charge for

creating and sending periodic statements to mortgage holders and new

disclosures to customers for the reset of adjustable-rate mortgages.

CFPB officials said they were hindered in trying to estimate these costs

because many vendor contracts required banks to keep terms of their

agreements confidential.

CFPB also cited limitations in trying to quantify

the benefits for customers of additional disclosures and other provisions

of the mortgage servicing rules, noting that the rules were designed to

Examples of requirements that small servicers must follow include prohibitions on

making the first notice or filing required for a foreclosure process unless a borrower is

more than 120 days delinquent and proceeding to foreclosure judgment or order of sale,

or conducting a foreclosure sale, if a borrower is performing under the terms of a loss-

mitigation agreement.

CFPB defined small banks and credit unions using a definition prepared by the Small

Business Administration as those with $175 million or less in assets.

See Pub. L. No. 111-203, § 1022(b)(2)(A), 124 Stat. 1376, 1980 (2010) (codified at 12

U.S.C. § 5512(b)(2)(A)). We reviewed CFPB’s Dodd-Frank Act section 1022(b)(2) analysis

conducted in connection with the final rule, available at Mortgage Servicing Rules Under

the Truth in Lending Act (Regulation Z), 78 Fed. Reg. 10902, 10978 (Feb. 14, 2013), but

did not assess the quality of the analysis.

In the preamble to its final rule, CFPB discussed some elements that the agency

believed would likely mitigate the total costs for regulated entities, even if those costs

could not be quantified. For example, CFPB created sample disclosure forms that banks

could directly adopt and allows coupon books to be provided in lieu of monthly statements

for certain fixed-rate mortgages. See Mortgage Servicing Rules Under the Real Estate

Settlement Procedures Act (Regulation X), 78 Fed. Reg. 10696, 10697 (Feb. 14, 2013). In

addition, the one-time costs incurred by any single vendor for producing each new form

were likely to be spread among a large number of servicers, which would mitigate the cost

to each servicer.

Page 37 GAO-16-448 Mortgage Servicing

address a failure of the servicing market that exists because consumers

dissatisfied with their mortgage servicer cannot easily change servicers.

Benefits are especially hard to quantify when rule changes are intended

to address market failures, CFPB said in the preamble to its final rule.

The agency noted that none of the public commenters proposed methods

for addressing this measurement limitation.

CFPB also analyzed the expected information collection burden on

affected entities—an analysis required under the Paperwork Reduction

Act of 1995—and produced a specific financial estimate. CFPB’s

estimates show that the mortgage servicing regulations would lead

servicers regulated by CFPB to spend an average of about 91 hours

annually and an additional 4 hours in one-time efforts to comply with the

information requirements associated with the rules. In addition, servicers

would spend an average of about $900 each year for vendor services and

another $700 in one-time costs to comply.

Specifically, CFPB stated: “These potential benefits and costs, and these impacts,

however, are not generally susceptible to particularized or definitive calculation in

connection with this rule. The incidence and scope of such potential benefits and costs,

and such impacts, will be influenced very substantially by economic cycles, market

developments, and business and consumer choices, which are substantially independent

from adoption of the rule. No commenter has advanced data or methodology that it claims

would enable precise calculation of these benefits, costs, or impacts. Moreover, the

potential benefits of the rule on consumers and covered persons in creating market

changes that are anticipated to address market failures are especially hard to quantify.”

Mortgage Servicing Rules Under the Real Estate Settlement Procedures Act (Regulation

X), 78 Fed. Reg. 10696, 10844 (Feb. 14, 2013).

Costs likely would not be spread evenly, however. For example, banks that do not

service mortgages on adjustable-rate loans would not need to address the new

requirements for providing notice for adjustable-rate loans. CFPB officials said estimates

of vendor costs were based on information obtained from larger banks. The information

could not be extrapolated to its Regulatory Flexibility Analysis estimates pertaining to the

costs for small banks to hire vendors, CFPB officials said.

Page 38 GAO-16-448 Mortgage Servicing

Banking regulators included the planned changes to the treatment of

MSRs when estimating the impact of the regulatory capital rules issued in

2013. However, they did not specifically isolate the impact only from the

changes to the treatment of MSRs. Regulators’ impact estimates were

prepared to comply with the RFA, which requires that regulators describe

the impact of significant rules on small entities.

Although the RFA

requires that regulators describe the overall rule’s impact on small

businesses, it does not require regulators to isolate the impact of specific

parts of an overall rule.

Each of the three prudential banking regulators—the Federal Reserve,

FDIC, and OCC—included the 10 percent threshold on MSRs when

estimating the number of banks that would not conform with the new

regulatory capital rules and the amount of their capital shortfall. NCUA,

which issued a regulatory capital rule for credit unions in October 2015

that exempted credit unions with less than $100 million in assets, did not

conduct a Regulatory Flexibility Act analysis on small credit unions

because it concluded that its rule would not have a material effect on

those institutions.

Although each banking regulator included the 10

percent MSR deduction threshold, only one of the three regulators—

FDIC—also included the change in MSR risk weights in its estimation

model.

Officials from the Federal Reserve said that they did not

incorporate the impact of increased risk weights in their analyses because

the change in MSRs’ risk weights would not have a meaningful effect on

the regulatory capital rules’ impact on small banks. OCC officials told us

omitting the impact of the 250 percent risk weight was an oversight. When

The banking regulators defined small banks and credit unions using a definition

prepared by the Small Business Administration as those with $500 million or less in

assets.

See Risk-Based Capital, 80 Fed. Reg. 66626, 66704 (Oct. 29, 2015). Small credit

unions, under NCUA’s definition, see 12 C.F.R. § 702.103(b), were those with $50 million

or less in assets, and therefore NCUA’s regulatory capital rules would not impact small

credit unions, NCUA concluded. When an agency determines and certifies that a

regulation would not have a significant economic impact on a substantial number of small

entities, a Regulatory Flexibility Act analysis is not required. See 5 U.S.C. § 605(b).

As we previously noted, banking regulators changed risk weights for MSRs from 100

percent to 250 percent for MSRs that are included in the calculation of CET1 capital under

the risk-based capital framework, though these requirements are being phased in through

2018. FDIC did not isolate the impact of changing risk weights when it estimated the

overall rule’s impact on small businesses.

Banking Regulators

Included Changes in the

Treatment of MSRs in

Their Impact Estimates

Prior to Issuing New

Regulatory Capital Rules

Page 39 GAO-16-448 Mortgage Servicing

OCC included the 250 percent risk weight in a re-analysis of the same

bank data after our inquiry, OCC officials found that the number of banks

needing capital would not change. In their analyses, the three banking

regulators estimated that a combined 124 small banks would need to

raise about $408 million by the end of 2018, when the regulatory capital

rules are fully phased in.

Banking regulators estimated that the 124

banks would face costs of about $2.2 million per year collectively—or less

than $18,000 per bank, on average—to raise this $408 million by

converting debt to equity.

Banking regulators did not identify how much

of these costs would be attributable to MSRs specifically, but the $2.2

million per year represents an upper limit of the maximum possible impact

related to the MSR provisions.

In a separate analysis, the banking regulators estimated that small banks

as a group would spend about $242 million—about $43,000 per bank—to

comply with the rule. For some banks, this would represent a significant

impact to the bank, regulators found. For example, OCC estimated that

compliance costs would represent a significant impact for about 19

Regulators’ estimates were distributed as follows: The Federal Reserve estimated 9

small banks would need to raise a total of about $11 million in additional capital; OCC

found that 41 small banks would need to raise a total of about $164 million; and FDIC

projected that 74 small banks would need to raise a total of about $233 million. See 78

Fed. Reg. at 62153 (Federal Reserve) and 62154-55 (OCC); 79 Fed. Reg. at 20757

(FDIC). When the OCC re-estimated impact to include the 250 percent risk weight

provision, OCC officials determined that the amount of capital needing to be raised was

still about $164 million. Federal Reserve officials re-estimated impact to include the 250

percent risk weight provision using updated data from June 30, 2015, and found that 2

small, state member banks would not meet the minimum capital ratio. The total capital

shortfall for these 2 banks would be $8.6 million. (The Federal Reserve’s analysis that was

published with the rule used data on state member banks from March 31, 2013.)

The costs calculated were based on the loss of tax benefits associated with converting

debt to equity. Converting debt to equity would be one possible method for complying with

regulatory capital requirements.

It is not possible to determine from the banking regulators’ analyses whether the

projected capital shortfall was the result of changes in the capital treatment of MSRs or

from some other aspect of the regulatory capital rule. Banks could fall short of the new

regulatory capital standards even if MSRs represented far less than 10 percent of their

common equity tier 1 capital, or even if they had no MSRs at all. Conversely, banks could

meet the various regulatory requirements for common equity tier 1 capital, overall tier 1

capital, and total regulatory capital even if their MSR holdings exceeded 10 percent of

their common equity tier 1 capital if they had sufficient cushion from other types of

regulatory assets.

Page 40 GAO-16-448 Mortgage Servicing

percent of small banks it supervised.

In addition, banking regulators

estimated the expected burden associated with additional information

collection—as required by the Paperwork Reduction Act—and estimated

that banks would not face any added monetary costs associated with

information collection.

Banking regulators obtained public input about the MSR provisions of the

regulatory capital rules during the rulemaking but said they did not revise

their proposed changes to the treatment of MSRs as a result. In the

preamble to the final rule, they noted that some commenters advocated

different approaches both for the threshold for including MSRs in CET1

capital and for risk weights assigned to MSRs. Regulators noted in the

final rule that MSRs have long been fully or partially excluded from

regulatory capital because of the high level of uncertainty regarding the

ability of banking organizations to realize value from these assets.

Officials from the three banking regulators also stated that their

organizations, which had participated in the international Basel III

agreement, wanted to adhere to the principles described in the

agreement.

CFPB’s plans for retrospectively reviewing its mortgage servicing rules

are incomplete, as the agency has not finalized its planned approach.

This review is required under the Dodd-Frank Act within 5 years from the

effective date of significant rules, which would be January 2019 for the

mortgage servicing rules.

The Dodd-Frank Act requires CFPB to publish

a report assessing the effectiveness of significant rules in meeting the

goals described by CFPB and the purposes and objectives of Title X of

the Dodd-Frank Act. CFPB staff prepared a preliminary planning

document in May 2015, but as of April 2016, the agency’s plan is not final.

The Dodd-Frank Act also requires CFPB to invite public comments on

recommendations to modify significant rules before the report is

In its published rules, OCC and FDIC calculated that compliance costs would have a

significant impact when the compliance costs exceeded 2.5 percent of banks’ total

noninterest expense or 5 percent of banks’ annual salaries and employee benefits. The

Federal Reserve did not publish a calculation in its final rule for the number of banks that

would be significantly impacted by estimated compliance costs.

See Pub. L. No. 111-203, § 1022(d), 124 Stat. 1376, 1984-85 (2010) (codified at 12

U.S.C. § 5512(d)).

CFPB’s Plans for

Conducting Retrospective

Reviews of the Mortgage

Servicing Rules Are

Incomplete, and Banking

Regulators’ Reviews of

Capital Rules Are Ongoing

Page 41 GAO-16-448 Mortgage Servicing

published.

CFPB has not determined how the agency will obtain public

comment on any recommendations.

OMB has provided guidance to independent agencies about effective

approaches for retrospectively reviewing existing rules. While not required

for independent agencies such as CFPB, OMB guidance suggests that

plans for retrospective reviews of regulations should specify the outcomes

and methodologies—such as the specific metrics, baselines, and

analytical methods—that agencies plan to use in their reviews.

CFPB’s

plan has identified some potential methodologies for measuring

outcomes, but the agency has not determined which specific approaches

it will employ. OMB also has cited the benefits of providing members of

the public with an opportunity to comment on draft plans because they

may have useful information and perspectives for improving the quality of

agencies’ planned approaches for retrospective reviews.

In addition, we

found in 2014 that public input from informed stakeholders such as

regulated entities and policy advocacy groups could be useful in

evaluating regulatory reforms.

CFPB officials said that as of April 2016

the agency had not decided whether to incorporate an opportunity for

public input into its methodologies for analyzing the rules.

CFPB officials said a specific plan for reviewing the mortgage servicing

rules had not been completed because it was too soon to identify the

relevant data and because the agency wanted the flexibility to design the

See § 1022(d)(3), 124 Stat. at 1985 (codified at 12 U.S.C. § 5512(d)(3)).

OMB memorandums M-11-10 and M-11-28 provide guidance to independent

agencies—and OMB Circular A-4 provides guidance to all federal agencies—related to

creating an overall process for reviewing existing rules. However, we believe the same

guidance is relevant when agencies are required to review specific rules, as CFPB is

required to do under section 1022(d) of the Dodd-Frank Act. For example, we found in

2007 that if agencies fail to plan for how they will measure the performance of their

regulations, and what data they will need to do so, they may continue to be limited in their

ability to assess the effects of their regulations. See GAO, Reexamining Regulations:

Opportunities Exist to Improve Effectiveness and Transparency of Retrospective Reviews,

GAO-07-791 (Washington, D.C.: July 16, 2007).

See OMB Memorandum M-11-19 and the related Executive Order 13563.

GAO, Reexamining Regulations: Agencies Often Made Regulatory Changes, but Could

Strengthen Linkages to Performance Goals, GAO-14-268 (Washington, D.C.: Apr. 11,

2014).

Page 42 GAO-16-448 Mortgage Servicing

most effective method to analyze the rules. In addition, they noted there

were trade-offs associated with seeking public input at the planning stage

because it may add time and cost to the review. However, we found in a

2007 report on retrospective reviews that agencies are better prepared to

perform effective reviews if they identify potential sources of data and the

measures that would be needed to assess effectiveness of the rules.

For example, soliciting input during the planning process could better

prepare CFPB by clarifying the feasibility of certain methodologies that

the agency is considering, and potentially save time and agency

resources if public input suggests that certain potential methodologies

would not be feasible. Further delay in finalizing a plan may preclude an

effective review if the agency were to forgo complex or time-consuming

methodologies due to time constraints.

Banking regulators reported two separate ongoing efforts to assess the

potential impacts of the regulatory capital rule, including the provisions

related to MSRs. The first effort is part of a review process established

under EGRPRA, which requires that banking regulators review all of their

rules at least once every 10 years.

However, the current EGRPRA

review is to be completed later this year—before the regulatory capital

rules are fully in effect in 2019—and therefore banking regulators may not

be able to determine the full impact from the EGRPRA review. The

sources for public input as part of the EGRPRA process include six

See GAO-07-791. We recommended that OMB officials provide guidance to regulatory

agencies to consider, during the promulgation of certain new rules, how to measure the

performance of the regulation—including how and when they will collect, analyze, and

report the data needed to conduct a retrospective review. OMB implemented the

recommendation by providing such guidance in 2011.

Agency officials do not have direct experience with how long this review process will

take because the agency has not yet completed any of these required 5-year retrospective

reviews of significant rules. The agency was still developing its overall plan for conducting

these reviews in February 2016. The first such review is expected be completed in late

2018 for a rule addressing electronic remittance transactions, CFPB officials said.

The first EGRPRA review was completed in 2007. If banking regulators complete the

current review in 2016 as planned, the next review would be required to be completed by

2026. NCUA also is conducting a similar review of its regulations. Although NCUA is not

technically required to participate in the EGRPRA review process, it is conducting its own

review “in keeping with the spirit of the law,” according to NCUA. NCUA has participated

along with the banking regulators in the EGRPRA planning process but has developed its

own regulatory categories that are comparable with those developed by the banking

regulators.

Page 43 GAO-16-448 Mortgage Servicing

separate outreach sessions at various locations across the United States

that allow the public to provide input during panel discussions.

The

regulatory capital rules were not part of the original information request

for the current EGRPRA process because these rules had been recently

promulgated and have not been fully phased in, but banking regulators

added the regulatory capital rules to the list of those eligible for comment.

Regulators also included the regulatory capital rules in one of its requests

for written public comments as part of the EGRPRA process. The second

effort is a separate study of the capital requirements for MSRs that was

mandated by the Consolidated Appropriations Act, 2016. The law

requires federal banking regulators and NCUA to study such issues as

the impact of the MSR provisions on competition in the mortgage

servicing business and on services to mortgage customers, as well as the

risk to banking institutions in holding MSRs.

As of June 8, 2016, the

study had not been issued.

In addition to these planned reviews, banking regulators said that they

often conducted other informal reviews as needed to evaluate the

effectiveness of rules. For example, OCC officials said the agency

considers the need to revise regulations based on other factors such as

changes in the broader economy and feedback from financial institutions

during agency outreach and interactions with OCC supervisory staff.

Federal Reserve representatives said the agency periodically reviews its

existing rules to see if they need to be updated, though there is no formal

schedule for conducting such a review. FDIC officials also cited its bank

examination process and other outreach sessions with a community bank

advisory committee as other mechanisms for obtaining public input that

could lead to additional review of existing rules. In 2007, we reported that

agencies often found their discretionary reviews to be more productive

and likely to generate action than mandatory reviews.

However, we

found that the mandatory EGRPRA process was an exception, with the

Between December 2014 and December 2015, EGRPRA outreach meetings were held

in Los Angeles, Dallas, Boston, Kansas City, Chicago, and Washington, D.C.

Pub. L. No. 114-113, Div. E, tit. VI, § 634(b), 129 Stat. 2242, 2471 (2015). The report is

required to be submitted to the Senate Committee on Banking, Housing, and Urban Affairs

and the House Committee on Financial Services not later than 180 days after enactment

of the appropriations act, or June 2016.

See GAO-07-791.

Page 44 GAO-16-448 Mortgage Servicing

first EGRPRA cycle ending in 2006 generating at least four regulatory

changes and more than 180 legislative proposals for regulatory relief.

Mortgage servicing is a substantial business for many community banks

and credit unions, and the value of their MSRs exceeded $3 billion as of

September 30, 2015. The Dodd-Frank Act requires that CFPB review

significant rules to evaluate the rules’ effectiveness, seek public input on

recommendations for modifying significant rules, and report on those

review findings no later than 5 years after the rules take effect. For the

mortgage servicing rules, this review must be completed by January

2019. To allow CFPB to understand the rules’ impact, including for

consumers and the market for consumer financial products, these

retrospective reviews must be carefully planned to specify the metrics,

baselines, and analytical methods to be used. We and OMB have found

that these elements, as well as seeking public input before finalizing

plans, can benefit the overall quality of a review. However, as of April

2016, CFPB had not finalized a plan for reviewing the mortgage-servicing

rules, including what methodologies to use and when to seek public input.

Without a completed plan, CFPB risks not having enough time to perform

an effective review.

To enhance the effectiveness of preparations for conducting a

retrospective review of its mortgage servicing regulations, the Director of

the Consumer Financial Protection Bureau should complete a plan to

identify the outcomes CFPB will examine to measure the effects of the

regulations, including the specific metrics, baselines, and analytical

methods to be used. For example, in developing such a plan, CFPB could

seek public input for information and perspectives to improve the quality

of its review through feedback on available data or improvements on

proposed methodologies.

We provided a draft of this report for review and comment to CFPB; HUD,

including the Federal Housing Administration and Ginnie Mae; FDIC;

Federal Housing Finance Agency (FHFA), including Fannie Mae and

Freddie Mac; the Federal Reserve; NCUA; OCC; the Department of

Veterans Affairs (VA); and the U.S. Department of Agriculture (USDA).

CFPB and NCUA provided written comments, which we have reprinted in

appendixes III and IV, respectively. FDIC; FHFA, including Freddie Mac;

the Federal Reserve; OCC; and VA also provided technical comments

which we have incorporated, as appropriate. HUD, including the Federal

Conclusions

Recommendation for

Executive Action

Agency Comments

and Our Evaluation

Page 45 GAO-16-448 Mortgage Servicing

Housing Administration and Ginnie Mae; and USDA did not provide

comments.

In its written comments, CFPB agreed to take steps to complete its plan

for conducting a retrospective review of the mortgage servicing rules.

While CFPB has developed a preliminary planning document that

identifies potential methodologies for measuring outcomes, the agency

stated that it is continuing to work on its plan. Specifically, CFPB intends

to refine the scope and relative emphases of the assessment, the

outcomes the assessment will likely focus on, potential ways to measure

outcomes, the qualitative and quantitative information used, and the cost

to gather such information. CFPB also stated that it intends to finalize its

plan by incorporating each of these elements, while ensuring that

sufficient time is allotted to make modifications should potential

improvements or identified costs necessitate revisions.

In its written comments, NCUA stated that it agrees with the report’s

conclusions as they relate to NCUA and the credit union industry. In

particular, NCUA agrees with the conclusion that NCUA’s capital

treatment of MSRs is unlikely to affect most credit unions.

As agreed with your office, unless you publicly announce the contents of

this report earlier, we plan no further distribution until 30 days from the

report date. At that time, we will send copies of this report to CFPB,

HUD, FDIC, FHFA, the Federal Reserve, NCUA, OCC, VA, and USDA

and other interested parties. In addition, the report will be available at no

charge on the GAO website at http://www.gao.gov.

Page 46 GAO-16-448 Mortgage Servicing

If you or your staff have any questions about this report, please contact

me at (202) 512-8678 or sciremj@gao.gov. Contact points for our Offices

of Congressional Relations and Public Affairs may be found on the last

page of this report. Key contributors to this report are listed in appendix V.

Sincerely yours,

Mathew J. Sciré

Director, Financial Markets and Community Investment

Appendix I: Objectives, Scope, and

Methodology

Page 47 GAO-16-448 Mortgage Servicing

The objectives of our report were to examine (1) community lenders’

participation in the mortgage servicing market and potential effects of the

Bureau of Consumer Financial Protection (commonly known as the

Consumer Financial Protection Bureau or CFPB) mortgage servicing

rules on those lenders, (2) potential effects of the risk-based capital

treatment of mortgage servicing rights (MSR) on decisions about holding

or selling MSRs, and (3) the process regulators used to estimate the

impact of regulations addressing mortgage servicing requirements and

the risk-based capital treatment of MSRs.

To describe the extent to which banks and credit unions are engaged in

mortgage lending and servicing activities, we used data from the Board of

Governors of the Federal Reserve System (Federal Reserve), the Federal

Deposit Insurance Corporation (FDIC), the Federal Financial Institutions

Examination Council (FFIEC), and the National Credit Union

Administration (NCUA). For banks, we used quarterly data on banks that

filed Call Reports (forms FFIEC 031 and 041) for the period from the first

quarter of 2001 through the third quarter of 2015. We divided banks into

groups based on their size and type.

• Size. We used total assets to measure size and we divided banks into

five equal-sized groups, or quintiles, based on their size each quarter.

The first quintile contained the smallest banks, the second quintile

contained the next largest banks, and so on through the fifth quintile,

which contained the largest banks.

• Type. We assigned banks to groups based on whether or not they

were a community bank—that is, we classified a bank as a community

bank if it was identified as such in FDIC’s Historical Community Bank

Reference Data dataset. In this dataset, a bank is a community bank if

it is part of a banking organization that has loans or core deposits, a

foreign assets-to-total assets ratio less than 10 percent, and has less

than 50 percent of assets in certain specialty banks, including credit

card specialists, consumer nonbank banks, industrial loan companies,

trust companies, and bankers’ banks. (Consumer nonbank banks are

financial institutions with limited charters that can make commercial

loans or take deposits, but not both.) In addition, the banking

organization must either have total assets less than an indexed asset

size threshold or total assets greater than the indexed asset size

threshold, a loan-to-assets ratio greater than 33 percent, a core

deposits-to-assets ratio greater than 50 percent, more than one office

but no more than the indexed maximum number of offices, offices in

less than three large metropolitan statistical areas, offices in less than

four states, and no single office with deposits greater than the indexed

Appendix I: Objectives, Scope, and

Methodology

Appendix I: Objectives, Scope, and

Methodology

Page 48 GAO-16-448 Mortgage Servicing

maximum branch deposit size. (The asset size threshold is indexed to

equal $250 million in 1985 and $1 billion in 2010. The maximum

number of offices is indexed to equal 40 in 1985 and 75 in 2010. The

maximum branch deposit size is indexed to equal $1.25 billion in 1985

and $5 billion in 2010.)

In the third quarter of 2015, the largest bank in the first quintile had assets

of about $73.9 million, the largest bank in the second quintile had assets

of about $138.5 million, the largest bank in the third quintile had assets of

about $250.6 million, the largest bank in the fourth quintile had assets of

about $545.1 million, and the largest bank in the fifth quintile had assets

of about $2 trillion. There were 1,187, 1,248, 1,243, 1,234 and 899

community banks in the first, second, third, fourth, and fifth quintiles,

respectively. There were 77, 16, 20, 30, and 364 nationwide, regional,

and other banks in the first, second, third, fourth, and fifth quintiles,

respectively.

To describe the extent to which banks of different sizes and types in our

sample were engaged in mortgage lending and servicing activities, we

determined whether each bank held residential mortgages for investment,

sale, or trading and had MSRs.

In addition, we calculated residential

mortgages held for investment, sale, or trading as a percentage of total

assets and MSRs as a percentage of total assets. Finally, we estimated

the fraction of all outstanding residential mortgages that each bank

serviced by adding the unpaid principal balance of residential mortgages

held for investment, sale, or trading to the unpaid principal balance of

residential mortgages serviced for others and divided the result by total

outstanding residential mortgages. This calculation may overstate the

fraction of residential mortgages banks service because the institutions

may not service all mortgages held for investment, sale, or trading. We

summarized the values of these variables for each quarter for each group

of banks. Our analysis of banks has limitations and should be interpreted

with caution. In particular, some new regulations related to mortgage

lending and servicing have only recently become effective or are not yet

fully implemented, and banks may not have fully adjusted to them. As a

result, current trends may not be indicative of the extent to which banks

participate in residential mortgage lending and servicing in the future. We

MSRs may include mortgage servicing assets that are commercial mortgage servicing

assets or other nonresidential mortgage servicing assets.

Appendix I: Objectives, Scope, and

Methodology

Page 49 GAO-16-448 Mortgage Servicing

assessed the data we used for this analysis by reviewing relevant

documentation and by electronically testing the variables for missing

values, invalid values, and outliers. We found them to be sufficiently

reliable for the purpose of conducting our analysis.

For credit unions, we used quarterly data on those that filed Call Reports

(form NCUA 5300) for the period from the second quarter of 2002 through

the third quarter of 2015. We divided the credit unions in our sample into

groups based on their size. As we did with banks, we used total assets to

measure size and we divided credit unions into quintiles each quarter

based on their size. Again, the first quintile contained the smallest credit

unions, the second quintile contained the next largest credit unions, and

so on through the fifth quintile, which contained the largest credit unions.

In the third quarter of 2015, the largest credit union in the first quintile had

assets of about $5.1 million, the largest credit union in the second quintile

had assets of about $16.3 million, the largest credit union in the third

quintile had assets of about $42.3 million, the largest credit union in the

fourth quintile had assets of about $139.3 million, and the largest credit

union in the fifth quintile had assets of about $72 billion. There were 1,244

credit unions in the first quintile and 1,243 credit unions in the second,

third, fourth, and fifth quintiles.

To describe the extent to which credit unions of different sizes were

engaged in mortgage lending and servicing activities, we determined

whether a credit union held residential mortgages on its balance sheet,

and had MSRs. In addition, we calculated residential mortgages as a

percentage of total assets and MSRs as a percentage of total assets.

Finally, we estimated the fraction of all outstanding residential mortgages

that each credit union serviced by adding residential mortgages on the

balance sheet to mortgages serviced for others and dividing the result by

total outstanding residential mortgages. As with community banks, this

calculation may overstate the fraction of residential mortgages serviced

because the institutions may not service all of the residential mortgages

on their balance sheet and because mortgages serviced for others may

include commercial mortgages. We summarized the values of these

variables for each quarter for each group of credit unions. As with our

analysis of banks, our analysis of credit unions has limitations. In

particular, current trends may not be indicative of the extent to which

credit unions participate in residential mortgage lending and servicing in

the future. We assessed the data we used for this analysis by reviewing

relevant documentation and electronically testing the variables for missing

values, invalid values, and outliers. We found the data to be sufficiently

reliable for the purpose of conducting our analysis.

Appendix I: Objectives, Scope, and

Methodology

Page 50 GAO-16-448 Mortgage Servicing

To understand the regulatory framework for mortgage servicers, we

reviewed applicable laws and regulations and relevant literature on

community banks’ and credit unions’ participation in the mortgage

servicing market. As part of this review, we selected academic studies

and research by industry organizations, federal agencies, GAO, and

others since the 2007–2009 financial crisis on the mortgage servicing

market that focused on the role of community banks and credit unions. In

particular, to examine the potential impacts of compliance with mortgage-

related regulatory requirements we reviewed the 2010 Dodd-Frank Wall

Street Reform and Consumer Protection Act (Dodd-Frank Act) and the

Consumer Financial Protection Bureau (CFPB) mortgage servicing rules

established in accordance with the act. We focused our analysis on the

final rules issued pursuant to the Dodd-Frank Act that became effective

since 2013. For each rule, we reviewed the Federal Register release of

the final rule document.

To assess the potential effects on banks of the capital treatment of MSRs

under the prudential regulators’ capital rules implementing Basel III, we

used data from FDIC and FFIEC. We used quarterly data on banks that

filed Call Reports (forms FFIEC 031 and 041) for the first three quarters of

2015. We divided banks into groups based on their size and type, as

described previously. We collected data on banks’ common equity tier 1

(CET1), tier 1, and total capital ratios and we determined whether the

ratios met or exceeded regulatory minimum amounts, with and without a

capital conservation buffer. We also counted the number of banks with

MSRs that did and did not make deductions from CET1 capital based on

the amount of MSRs on their balance sheets. Finally, for banks with

MSRs, we estimated what the CET1 capital ratio would be if banks

replaced MSRs with U.S. Treasury securities or some other asset with a

zero risk weight that is not deducted from CET1 (hereafter, the

counterfactual CET1 capital ratio). The CET1 capital ratio is the ratio of

CET1 capital to risk-weighted assets. Under the prudential regulators’

capital rules implementing Basel III, replacing MSRs with U.S. Treasury

securities would affect both CET1 capital (the numerator) and risk-

weighted assets (the denominator). We took the following steps to

estimate the numerator and the denominator of our counterfactual CET1

ratio:

• To estimate the numerator, we assumed that if banks replaced their

MSRs with U.S. Treasury securities, then they would make no MSR-

related deductions from CET1 capital. Thus, we estimated the

numerator by adding a bank’s MSR-related deductions back to its

CET1 capital. Our assumption leads us to overestimate counterfactual

Appendix I: Objectives, Scope, and

Methodology

Page 51 GAO-16-448 Mortgage Servicing

CET1 capital because replacing MSRs with U.S. Treasury securities

may not cause all MSR-related deductions to fall to zero. However,

we make this assumption because we cannot calculate exactly how

all of the MSR-related deductions would change with available data.

• To estimate the denominator, we assumed that if banks replaced their

MSRs with U.S. Treasury securities, their risk-weighted assets would

decrease by an amount equal to 10 percent of their CET1 capital, risk-

weighted at 100 percent during the transition period and at 250

percent fully phased in. Under the prudential regulators’ capital rules

implementing Basel III, banks include in their risk-weighted assets the

amount of MSRs adjusted for deductions from CET1. At most this

amount is equal to 10 percent of CET1. We note that this assumption

leads us to underestimate counterfactual risk-weighted assets

because the amount of MSRs adjusted for deductions from CET1 and

thus added to risk-weighted assets may be equal to less than 10

percent of CET1. We made this assumption because we cannot

calculate exactly how risk-weighted assets would change with

available data.

Our analysis has limitations and should be interpreted with caution. In

particular, the new risk-based capital requirements are not fully phased-

in, and banks may not have fully adjusted to them. As a result, current

trends may not be indicative of the extent to which banks will be affected

by the new risk-based capital requirements in the future. We assessed

the data we used for this analysis by reviewing relevant documentation

and by electronically testing the variables for missing values, invalid

values, and outliers. We determined that the data were sufficiently reliable

for the purpose of estimating the potential effects on banks of the capital

treatment of MSRs under the prudential regulators’ capital rules

implementing Basel III.

To determine community banks’ and credit unions’ participation in sales

or purchases of MSRs in the secondary market, we analyzed data on

transfers of mortgage servicing rights obtained from Fannie Mae, Freddie

Mac, and Ginnie Mae for the period from 2010 through the second

quarter of 2015. We used these data to determine the amount of MSRs

(unpaid principal balance) associated with mortgage pools which were

sold via bulk sales by banks, credit unions, and nonbank entities. We also

determined the percentage of transfers between entity types such as

bank to bank and bank to nonbank transactions, and dollar volume of

these transfers. Additionally, we calculated the number of loans

associated with transfers by year. We assessed the reliability of these

data for this purpose by electronically testing the variables for missing

Appendix I: Objectives, Scope, and

Methodology

Page 52 GAO-16-448 Mortgage Servicing

values, invalid values, and outliers. We determined the data to be

sufficiently reliable for the purpose of determining community banks’ and

credit unions’ participation in sales or purchases of MSRs.

To determine the process regulators used to estimate the impact of

regulations addressing mortgage servicing requirements and the risk-

based capital treatment of MSRs, we reviewed the final and proposed

rules to understand the results of regulators’ estimates of the rules’

impacts. We also reviewed other agency documentation, including

CFPB’s report summarizing the results of a small business review panel.

We also interviewed CFPB officials and banking regulators to understand

their approaches for estimating the impact of the regulations before the

final rules were issued.

To examine the extent to which CFPB and banking regulators have plans

in place to monitor and assess the effects of new rules related to

mortgage servicing or MSRs, we identified and reviewed requirements

and guidance relating to agencies’ efforts to monitor and assess

regulations (criteria). We also identified and reviewed Office of

Management and Budget memorandums and related Executive Orders

related to agencies’ efforts to conduct retrospective reviews. To evaluate

CFPB’s approach toward these retrospective reviews, we also reviewed

provisions of the Dodd-Frank Act that require CFPB to assess its

significant rules and CFPB’s planning documents for conducting these

assessments. In addition, to evaluate banking regulators’ approach

toward the assessment of the regulatory capital rules that include

changes to MSRs, we reviewed requirements contained in the Economic

Growth and Regulatory Paperwork Reduction Act of 1996 (EGRPRA) for

reviewing existing rules, as well as requirements contained in the

Consolidated Appropriations Act, 2016 for banking regulators to study the

impact of changes to the capital treatment of MSRs. We also interviewed

officials from CFPB and banking regulators about their plans for

retrospective reviews. Additionally, we reviewed prior GAO work that

addresses important characteristics for conducting high-quality reviews of

regulations’ impacts.

To address all the objectives, we conducted semistructured interviews

with federal regulators, including officials from the Board of Governors of

the Federal Reserve System (Federal Reserve), CFPB, Federal Deposit

Insurance Corporation (FDIC), Office of the Comptroller of the Currency,

National Credit Union Administration (NCUA), and officials from industry

associations selected for their affiliation with community banks or credit

unions, including American Bankers Association (ABA), Independent

Appendix I: Objectives, Scope, and

Methodology

Page 53 GAO-16-448 Mortgage Servicing

Community Bankers of America (ICBA), Credit Union National

Association (CUNA), National Association of Federal Credit Unions, and

academics and other industry participants such as mortgage brokers and

a rating agency selected based on our literature review. We also

interviewed representatives from three credit union service organizations

and two consumer groups—the Center for Responsible Lending and the

National Consumer Law Center.

We also interviewed representatives

from Fannie Mae and Freddie Mac; Ginnie Mae; and the Federal Housing

Administration and the Department of Veterans Affairs, which insures or

guarantees, respectively, loans in Ginnie Mae-guaranteed mortgage-

backed securities (MBS).

We interviewed community lenders of a range of sizes and different levels

of experience with respect to mortgage servicing and MSRs. Our

selection factors were based on available data in the Consolidated

Reports of Condition and Income (commonly referred to as Call Reports)

and the Credit Union 5300 Call Reports. For example, for credit unions

we used the number of mortgage loans as an indicator of size because

CFPB’s mortgage servicing rules include an exemption from some rules

for small servicers. The term “small servicer” generally applies to

institutions that service 5,000 mortgages or fewer. For community banks,

we used asset size as our measure of size for the community banks

rather than loan count because bank Call Reports do not track the

number of loans.

For our core sample of community lenders, we interviewed

representatives from a total of eight community banks and six credit

unions. However, we also interviewed a few additional community lenders

for additional perspective for a total of nine community banks and seven

credit unions. Although no commonly accepted definition of a community

Credit union service organizations are entities that are owned by federally chartered or

federally insured state-chartered credit unions and that are engaged primarily in providing

products or services to credit unions or credit union members. See 12 C.F.R. § 712.1(d).

We used the Federal Deposit Insurance Corporation’s definition of a community bank,

which incorporates an asset size threshold—generally including banks with less than $1

billion in assets—and other characteristics. The asset size threshold is not a strict

requirement. Larger banks may be considered community banks if they meet other criteria

such as having a loan-to-asset ratio greater than 33 percent. See Federal Deposit

Insurance Corporation, FDIC’s Community Banking Study.

Appendix I: Objectives, Scope, and

Methodology

Page 54 GAO-16-448 Mortgage Servicing

bank exists, the term often is associated with smaller banks that provide

relationship banking services to the local community and have

management and board members who reside in the local community.

While most credit unions are relatively small, we focused our interview

selections on larger credit unions as additional servicing requirements

apply to institutions with more than 5,000 loans. We also sought to gain

perspectives from a credit union that may be approaching the 5,000 loan

threshold. In addition to size, we included as a selection factor whether

institutions had MSRs, including some institutions that had MSRs and

some that did not (see table 1). Finally, separate from our core sample of

interviews, we interviewed one community bank and one credit union as

we developed the methodology and discussion questions for our interview

sample. We selected these banks based on referrals from trade

associations. For additional context, we also interviewed one regional

bank with a large concentration of MSRs. Based on our small judgmental

sample of community lenders, the responses are not generalizable to the

population of community lenders.

Table 3: Number of Community Banks and Credit Unions Interviewed, by Size

Number of

community banks

Description

of size

Number of credit

unions

Description

of size

Small

Less than $100 million in assets

11 to 500 mortgage loans

Medium

$100 million to $1 billion in assets

501-500 mortgage loans

Large

More than $1 billion in assets

More than 5000 mortgage

loans

Total

Source: GAO analysis of Call Report information. | GAO-16-448

We also completed a literature search and reviewed recent reports and

articles related to the mortgage servicing industry, including industry and

academic reports. We identified some literature recommended to us by

the institutions we interviewed. In particular, to assess the impact of

mortgage-related requirements and mortgage servicing rules issued

pursuant to the Dodd-Frank Act, we reviewed an industry survey of a

nonprobability sample of banks on the regulatory impact of Dodd-Frank

Appendix I: Objectives, Scope, and

Methodology

Page 55 GAO-16-448 Mortgage Servicing

Act regulations on banks in 2015.

Since the survey relied on a

nonprobability sample, the results cannot be used to make

generalizations to the population of community lenders or commercial

banks and thrifts. We reviewed the methodologies used in the study and

determined that they were reasonable for analyzing the issues raised but

note that the study has limitations and is not necessarily definitive.

We conducted this performance audit from April 2015 to June 2016 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 finding and

conclusions based on our audit objectives.

The American Banking Association’s 22nd Real Estate Lending Survey had the

participation of 182 banks. The data were collected from March 4, 2015, to April 17, 2015,

and in most cases reports calendar year or year-end results. In other cases, data reflect

current activities and expectations at the time of data collection. Of the survey participants,

68 percent of respondents were commercial banks and 32 percent were savings

institutions. About 77 percent of the participating institutions had assets of less than $1

billion. The survey was sent out as a web survey to over 3,000 banks including both

American Bankers Association member and nonmember banks (commercial banks and

thrifts) and elicited responses from 182 banks for a response rate of 6 percent overall.

Since the survey relies on a non-probability sample, the results cannot be used to make

inferences about all commercial banks and thrifts.

Appendix II: Mortgage Servicing Rights

Transfer Activity

Page 56 GAO-16-448 Mortgage Servicing

This appendix provides additional information about mortgage servicing

rights (MSR) transfers. We analyzed data on MSR transfers from Freddie

Mac and Ginnie Mae to determine the volume of transfers by type and

size of institutions.

Table 1 shows the volume of MSR transfers by

different types of institutions.

Table 4: Percentage of Residential Mortgage Transfers of Servicing Approved by

Freddie Mac or Ginnie Mae by Institution Type, 2010 through 2015

Institution type (seller-to-

buyer)

Percentage of MSRs

transferred based on

unpaid principal balance,

2010–2012

Percentage of MSRs

transferred based on

unpaid

principal balance,

2013–2015

Bank-to-bank

26.8

9.0

Bank-to-nonbank

33.5

47.7

Nonbank-to-bank

13.4

7.1

Nonbank-to-nonbank

25.0

36.4

Total

100

Source: GAO analysis of Freddie Mac and Ginnie Mae data. | GAO-16-448

Note: Totals may not sum to 100 percent due to rounding.

Bank refers to any insured depository institution (e.g., bank, credit union, thrift). Nonbanks could

potentially be owned by a bank even if they are not an insured depository themselves.

Period is from January 2013 to June 2015, which was the most recent data available.

Table 2 shows additional information about the net MSR transfers by

banks of different sizes and nonbanks. In the 2010 to 2012 period,

nonbanks and multiple categories of small banks were net purchasers of

MSRs approved by Freddie Mac and Ginnie Mae, while the largest banks

sold more MSRs than they purchased. In the 2013 to 2015 period, banks

in each of the four size categories were net sellers of MSRs, while

nonbanks were net purchasers.

Data provided by Fannie Mae did not allow GAO to perform a comparable analysis to

Freddie Mac and Ginnie Mae.

Appendix II: Mortgage Servicing Rights

Transfer Activity

Appendix II: Mortgage Servicing Rights

Transfer Activity

Page 57 GAO-16-448 Mortgage Servicing

Table 5: Net Transfers of Mortgage Servicing Rights Approved by Freddie Mac or

Ginnie Mae by Institution Type and Size (in billions of dollars of unpaid principal

balance), 2010 through 2015

Institution type and size

Net transfers

2010–2012

Net transfers

2013–2015

Banks with less than $1 billion in assets

-2

Banks with $1 billion and up to $10

billion in assets

-77

Banks with more than $10 billion and up

to $50 billion in assets

-4

-10

Banks with more than $50 billion in

assets

-35

-146

All nonbanks

236

Source: GAO analysis of Freddie Mac and Ginnie Mae data. | GAO-16-448

Note: The negative amounts indicate that the unpaid principal balance on outstanding loans for which

mortgage servicing rights (MSR) were transferred exceeded the unpaid principal balance on

outstanding loans for which MSRs were purchased. Data provided by Fannie Mae did not allow GAO

to perform a comparable analysis to Freddie Mac and Ginne Mae data. Net transfers for all institution

types do not equal zero because institution size could not be determined for some buyers or sellers,

and these transactions are not included in the table.

Bank refers to any insured depository institution (e.g., bank, credit union, thrift). We did not

categorize nonbanks by size. Nonbanks could potentially be owned by a bank even if they are not an

insured depository themselves.

Period is from January 2013 to June 2015, which were the most recent data available.

Appendix III: Comments from the Bureau of

Consumer Financial Protection

Page 58 GAO-16-448 Mortgage Servicing

Appendix III: Comments from the Bureau of

Consumer Financial Protection

Appendix III: Comments from the Bureau of

Consumer Financial Protection

Page 59 GAO-16-448 Mortgage Servicing

Appendix III: Comments from the Bureau of

Consumer Financial Protection

Page 60 GAO-16-448 Mortgage Servicing

Appendix IV: Comments from the National

Credit Union Administration

Page 61 GAO-16-448 Mortgage Servicing

Appendix IV: Comments from the National

Credit Union Administration

Appendix V: GAO Contact and Staff

Acknowledgments

Page 62 GAO-16-448 Mortgage Servicing

Mathew J. Sciré (202) 512-8678 or scir[email protected]

In addition to the contact named above, Jill M. Naamane (Assistant

Director), Janet Fong (Analyst-in-Charge), Farah B. Angersola, Bethany

M. Benitez, Emily R. Chalmers, Pamela R. Davidson, Janet C. Eackloff,

Kendra R. Froshman, Justin P. Gordinas, Cynthia L. Grant, Courtney L.

LaFountain, Marc W. Molino, Steve Robblee, Jennifer W. Schwartz and

James D. Vitarello made key contributions to this report.

Appendix V: GAO Contact and Staff

Acknowledgments

GAO Contact

Staff

Acknowledgments

(250818)

The Government Accountability Office, the audit, evaluation, and

investigative arm of Congress, exists to support Congress in meeting its

constitutional responsibilities and to help improve the performance and

accountability of the federal government for the American people. GAO

examines the use of public funds; evaluates federal programs and

policies; and provides analyses, recommendations, and other assistance

to help Congress make informed oversight, policy, and funding decisions.

GAO’s commitment to good government is reflected in its core values of

accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no

cost is through GAO’s website (http://www.gao.gov). Each weekday

afternoon, GAO posts on its website newly released reports, testimony,

and correspondence. To have GAO e-mail you a list of newly posted

products, go to http://www.gao.gov and select “E-mail Updates.”

The price of each GAO publication reflects GAO’s actual cost of

production and distribution and depends on the number of pages in the

publication and whether the publication is printed in color or black and

white. Pricing and ordering information is posted on GAO’s website,

http://www.gao.gov/ordering.htm.

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or

TDD (202) 512-2537.

Orders may be paid for using American Express, Discover Card,

MasterCard, Visa, check, or money order. Call for additional information.

Connect with GAO on Facebook, Flickr, Twitter, and YouTube.

Subscribe to our RSS Feeds or E-mail Updates.

Listen to our Podcasts and read The Watchblog.

Visit GAO on the web at www.gao.gov.

Contact:

Website: http://www.gao.gov/fraudnet/fraudnet.htm

E-mail: fraudnet@gao.gov

Automated answering system: (800) 424-5454 or (202) 512-7470

Katherine Siggerud, Managing Director, [email protected], (202) 512-

4400, U.S. Government Accountability Office, 441 G Street NW, Room

7125, Washington, DC 20548

Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800

U.S. Government Accountability Office, 441 G Street NW, Room 7149

Washington, DC 20548

GAO’s Mission

Obtaining Copies of

GAO Reports and

Testimony

Order by Phone

Connect with GAO

To Report Fraud,

Waste, and Abuse in

Federal Programs

Congressional

Relations

Public Affairs

Please Print on Recycled Paper.