JANUARY | 2022
Ginnie Mae and the
Securitization of Federally
Guaranteed Mortgages
© Yeti studio/Shutterstock.com
At a Glance
e Government National Mortgage Association (Ginnie Mae), a part of the Department of Housing
and Urban Development, works to attract capital to the market for federally insured mortgages. It
does so by guaranteeing the timely payment of principal and interest on mortgage-backed securities
(MBSs) that private nancial institutions create from mortgages that are insured or guaranteed by
other federal programs.
Ginnie Maes guarantee operations have increased signicantly in recent years. In addition, the types
of mortgages included in the MBSs that Ginnie Mae guarantees—and the types of lenders that issue
those MBSs—have shifted in ways that could increase Ginnie Maes exposure to the risk of losses on
its guarantees.
In this report, the Congressional Budget Oce provides an overview of its baseline budget projec-
tions for Ginnie Mae and analyzes a scenario in which Ginnie Mae could be exposed to losses in a
period of severe economic stress.
In its July 2021 baseline, CBO estimates that new guarantees issued by Ginnie Mae in 2022 will
produce a budgetary savings of $2.2 billion in that year (using the measures for the cost of federal
credit programs specied in the Federal Credit Reform Act).
In the stress scenario—which represents the worst 1 percent of potential economic outcomes
over the coming decade—nancial institutions that issue Ginnie Mae–backed MBSs would fail
at higher-than-expected rates. Losses on the mortgages underlying those securities would also
be larger than expected. Under that scenario, Ginnie Maes new guarantees in 2022 would not
produce budgetary savings but instead would increase the decit by $3.0 billion in that year,
CBOestimates.
www.cbo.gov/publication/57176
Boxes
1. Eects of the Coronavirus Pandemic on Ginnie Mae 7
2. Possible Changes to Ginnie Mae’s Role From a Restructuring of the HousingFinanceSystem 9
Contents
Summary 1
Overview of Ginnie Mae and Its Exposure to Future Losses 2
The Role of Ginnie Mae in the SecondaryMortgageMarket 2
Ginnie Mae’s MBS Programs 3
Ginnie Mae’s Exposure to Losses 5
Volume and Loan Characteristics of GinnieMaesGuarantees 5
Ginnie Mae’s Exposure to Future Losses FromChanges in Its Guarantees 8
Estimating Federal Subsidy Costs for Ginnie Mae in CBO’s Baseline and UnderaStress Scenario 8
Estimating the Volume of Guarantees 9
Estimating the Subsidy Rate on GinnieMae’sGuarantees 10
Estimating the FCRA Subsidy Rate UnderaStressScenario 12
Estimates of Ginnie Mae’s Budgetary Costs 15
List of Tables and Figures 16
About This Document 17
Notes
Unless this report indicates otherwise, all years referred to are federal scal years, which run from
October 1 to September 30 and are designated by the calendar year in which they end.
Numbers in the text and tables may not add up to totals because of rounding.
References to the Congressional Budget Oces baseline budget projections refer to the July 2021
baseline.
Ginnie Mae and the Securitization of
Federally Guaranteed Mortgages
Summary
e Government National Mortgage Association,
known as Ginnie Mae, is a government-owned corpo-
ration within the Department of Housing and Urban
Development. Since its establishment in 1968, Ginnie
Mae has aimed to attract capital to the market for
federally insured mortgages, and thus reduce costs to
mortgage borrowers, while minimizing risk to taxpayers.
Ginnie Mae carries out that mission by guaranteeing the
timely payment of principal and interest on mortgage-
backed securities (MBSs) that private nancial institu-
tions create from home loans that are insured or guar-
anteed by other federal programs, such as those of the
Federal Housing Administration (FHA), the Department
of Veterans Aairs (VA), and the Department of
Agricultures Rural Housing Service (RHS).
Ginnie Maes MBS guarantees are similar to the ones
provided by Fannie Mae and Freddie Mac, except that
those two government-sponsored enterprises focus on
conventional mortgages that are not federally insured.
1
In addition, eective federal support for Fannie Maes
and Freddie Mac’s guarantees is limited, whereas Ginnie
Maes guarantees are explicitly backed by the full faith
and credit of the federal government.
2
1. Fannie Mae and Freddie Mac are government-sponsored
enterprises that were established by federal law to provide a
stable ow of funding for home loans. ey buy mortgages
that are not insured or guaranteed by a federal agency, pool
them to create MBSs, and sell the securities to investors with a
guarantee against most losses from defaults on the underlying
loans. Although Fannie Mae and Freddie Mac were originally
independent entities, they have been under the government’s
control (in federal conservatorships) since the nancial crisis of
2008. For more details, see Congressional Budget Oce, Fannie
Mae, Freddie Mac, and the Federal Role in the Secondary Mortgage
Market (December 2010), www.cbo.gov/publication/21992.
2. For a description of the limits on federal support for Fannie Mae
and Freddie Mac, see Congressional Budget Oce, e Eects of
Increasing Fannie Maes and Freddie Mac’s Capital (October 2016),
www.cbo.gov/publication/52089.
e volume of Ginnie Maes business has increased
sharply in the past 15 years, reecting the growth in
the volume of federally insured mortgages over that
period. e composition of Ginnie Maes business
has also changed. For example, a growing share of the
MBSs guaranteed by Ginnie Mae contain loans issued
by nonbank nancial institutions. Such institutions are
generally subject to a dierent regulatory regimen and a
more uncertain process in the event of failure than tradi-
tional banks are. In addition, a growing share of Ginnie
Mae–guaranteed MBSs contain loans backed by federal
mortgage guarantee programs, such as VAs, that provide
a partial rather than a full guarantee against losses when
a borrower defaults. ose changes may increase Ginnie
Maes risk to taxpayers, although the extent of the addi-
tional risk is uncertain.
In its baseline budget projections, the Congressional
Budget Oce estimates that Ginnie Maes new guaran-
tees will have a negative subsidy rate throughout the next
10 years, according to the accounting approach pre-
scribed by the Federal Credit Reform Act of 1990. e
subsidy rate measures the budgetary cost to the federal
government per dollar of new guarantees. A negative
subsidy rate represents a budgetary savings.
Although Ginnie Maes activities are projected to reduce
the budget decit under normal, or even moderately
stressful, economic conditions, the corporation could be
exposed to losses in a period of severe economic stress.
To explore the potential for such losses, CBO analyzed a
scenario with higher-than-expected failure rates among
issuers of Ginnie Mae–guaranteed securities and larger-
than-expected losses on the mortgages underlying those
securities.
3
at stress scenario—which represents the
3. Ginnie Maes exposure to risk ultimately comes from the
institutions that service the mortgages in a Ginnie Mae–
guaranteed MBS. (Servicers collect the payments that borrowers
owe on those loans.) e servicer is not always the same institution
that originated the loan for the borrower or that issued the MBS.
In this report, CBO uses “issuer” to describe an institution that
plays all three roles—originator, issuer, and servicer.
2 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES JANUARY 2022
worst 1 percent of economic outcomes over the coming
decade, CBO estimates—reects some of the increased
uncertainty for Ginnie Mae stemming from the recent
increases in the number of nonbank issuers and in the
volume of VA-guaranteed loans.
e conditions of the stress scenario would signicantly
alter Ginnie Maes eect on the federal budget. For
example, CBO projects that in 2022, Ginnie Maes new
guarantees would increase the decit by $3.0 billion
under the stress scenario, as opposed to reducing the
decit by $2.2 billion in CBO’s baseline estimate.
Overview of Ginnie Mae and Its
Exposure to Future Losses
e federal government operates various programs to
encourage lenders to provide mortgages to people who
might otherwise have trouble getting a home loan, such
as rst-time homebuyers, veterans, people with relatively
low income, or people who live in places where access
to credit is limited. Ginnie Mae was created to make it
easier for lenders that make loans under those programs
to sell the loans in the secondary (resale) market and
thus replenish their funds to make additional mortgages.
Ginnie Mae pursues that goal by guaranteeing that
investors who buy mortgage-backed securities based on
such loans will continue to receive scheduled principal
and interest payments on the loans if the MBS issu-
ers or mortgage borrowers fail to meet their nancial
obligations.
In recent years, Ginnie Maes guarantee operations
have grown substantially. At the same time, the types of
mortgages included in the MBSs that Ginnie Mae guar-
antees—and the types of lenders issuing those MBSs—
have changed in ways that could increase Ginnie Maes
exposure to the risk of losses on its guarantees.
The Role of Ginnie Mae in the
SecondaryMortgageMarket
Ginnie Mae oversees the process in which private issuers
create MBSs backed by the full faith and credit of the
federal government by pooling mortgages that are
guaranteed or insured by various federal agencies. ose
agencies include the Federal Housing Administration
and the Oce of Public and Indian Housing (PIH),
both part of the Department of Housing and Urban
Development; the Department of Veterans Aairs; and
the Rural Housing Service, part of the Department of
Agriculture. Some of those agencies’ programs, such as
FHAs, provide mortgage insurance that protects lend-
ers against losses if a borrower defaults on a mortgage.
When that happens, FHA pays a claim to the lender
for the unpaid principal balance of the defaulted mort-
gage. Other programs, such as VAs, guarantee to repay a
certain percentage of a mortgage in default. ose federal
backstops typically mean that lenders provide mortgages
to qualifying borrowers on better terms (such as with
lower interest rates, lower closing costs, or smaller down
payments) than they might otherwise.
Ginnie Maes role in turning such loans into
government-guaranteed mortgage-backed securities,
which issuers can sell to investors, involves steps at every
stage of the securitization process:
Approving new issuers of MBSs;
Managing the issuers and the infrastructure (such
as technology, program rules, legal documentation,
and personnel) for pooling loans to serve as collateral
forMBSs;
Maintaining the technology, operations, legal
disclosures, and personnel necessary to guarantee the
timely payment of principal and interest to investors
that buy the MBSs; and
Managing outstanding MBSs that Ginnie Mae
takes over from issuers that have defaulted on their
obligations (see Figure 1).
e fact that Ginnie Maes guarantee is explicitly backed
by the full faith and credit of the federal government
makes the underlying mortgages more liquid, helping to
reduce costs for borrowers and increasing the availability
of nancing for lenders.
4
(In a liquid market, investors
can quickly buy or sell large quantities of an asset with-
out aecting its price.) Without Ginnie Maes guarantee,
there would be less demand for MBSs created from
federally insured mortgages because investors would be
exposed to the risk of losses if MBS issuers failed, partic-
ularly during times of economic distress and increased
delinquencies by borrowers.
Unlike with traditional mortgages, which lenders
typically sell to securitizers (such as Fannie Mae and
Freddie Mac), issuers of Ginnie Mae–guaranteed MBSs
do not sell the underlying mortgages. Instead, the issuer,
4. e explicit federal backing for Ginnie Maes guarantee diers
from the eective federal support for the guarantees of Fannie
Mae and Freddie Mac. For a description of the federal support
oered to Fannie Mae and Freddie Mac, see Congressional
Budget Oce, e Eects of Increasing Fannie Mae’s and Freddie
Mac’s Capital (October 2016), www.cbo.gov/publication/52089,
and Eects of Recapitalizing Fannie Mae and Freddie Mac
rough Administrative Actions (August 2020), www.cbo.gov/
publication/56496.
3JANUARY 2022 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES
or a designated party, continues to receive payments
from borrowers and forwards part of those payments
to the MBS investors (see Figure 2). If borrowers fail
to make their payments, the issuer is required to send
the expected principal and interest payments to the
investors, as long as the underlying loans remain in the
MBSs. Once the issuer determines that a borrower is
unable or unwilling to resume regular payments, it will
remove that borrower’s loan from the MBS, repay the
investor the remaining principal balance, and try to
recover the amount owed from the borrower. At the end
of that recovery process, the issuer may work to settle any
potential claim with the primary government guarantor,
typically FHA or VA. As long as issuers perform their
duties, Ginnie Maes only involvement in those cash
ows is providing the infrastructure necessary for issuers
to remit payments to MBS investors and collecting the
guarantee fee it charges issuers.
Ginnie Maes role and exposure to risk change signi-
cantly if an issuer fails to perform its duties. In that
case, Ginnie Mae steps into the role of issuer—servicing
and administering the MBSs (by promptly forwarding
principal and interest payments on the securities and
providing disclosures to investors), handling defaults and
foreclosures on the underlying loans, and working with
the primary government guarantors to settle claims.
5
Ginnie Mae’s MBS Programs
Ginnie Maes largest securitization program—and the
focus of this report—involves MBSs composed of mort-
gages on single-family homes. However, Ginnie Mae also
oers securitization programs for other types of loans,
including multifamily mortgages (for properties with ve
or more units), mortgages for manufactured housing,
and reverse mortgages (in which households with at least
one member age 62 or older can borrow money by using
the equity in their home as collateral).
6
5. Rather than perform the role of issuer itself, Ginnie Mae could
transfer those responsibilities to another issuer or hire a third
party (known as a subservicer) to handle them. In either case,
Ginnie Maes guarantee, and thus its exposure to risk, would
remain in eect.
6. For more information about multifamily mortgages and reverse
mortgages, see Congressional Budget Oce, e Federal Role in
the Financing of Multifamily Rental Properties (December 2015),
www.cbo.gov/publication/51006, and e Role of the Federal
Housing Administration in the Reverse-Mortgage Market (May
2019), www.cbo.gov/publication/55247.
Figure 1 .
The Role of Ginnie Mae in the Market for Mortgage-Backed Securities
During Repayment of
Loans Underlying MBSs
Before Issuance of
MBSs
After MBS
Issuers Fail
Approves Financial
Institutions to Issue
MBSs Guaranteed by
Ginnie Mae
Guarantees Principal
and Interest
Payments to Investors
That Buy MBSs
Monitors the Financial and Operational
Capacity of Issuers
Manages the MBS
Programs’
Infrastructure
a
Manages
the MBSs of Failed
Issuers
Data source: Congressional Budget Oce.
MBSs = mortgage-backed securities.
a. Program infrastructure consists of the technology, legal documentation, personnel, and other things needed to pool mortgages and issue MBSs.
4 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES JANUARY 2022
In addition to dierent types of mortgages, Ginnie Mae
oversees the creation of dierent types of MBSs. e two
main types are single-class securities called Ginnie MaeI
and Ginnie Mae II.
7
A Ginnie Mae I security is based
on mortgages from the same issuer that total at least
$1million and have the same interest rate. A Ginnie
Mae II security, by contrast, can include mortgages from
dierent issuers and with dierent interest rates (which
can range from 0.25 percentage points to 0.75 percent-
age points above the rate on the MBS). e multi-issuer
feature of Ginnie Mae II MBSs is particularly important
for smaller issuers, which may have trouble acquiring
7. Ginnie Mae also guarantees multiclass securities, which are based
on a combination of existing MBSs. ose securities are not
covered in this report.
enough loans to meet the size requirement for a Ginnie
Mae I MBS.
e multi-issuer pool of loans underlying a Ginnie
MaeII MBS functions in much the same way as the pool
of loans from a single issuer. Each issuer receives a share
of the Ginnie Mae II security that is based on the unpaid
principal balance of the loans it contributed. After the
MBS is issued, a central agent collects payments from
all of the issuers and makes a single monthly payment to
each investor holding that security.
8
8. See Ted Tozer, A Primer and Perspective on Ginnie Mae (Milken
Institute, October 2019), https://tinyurl.com/54kw29fm.
Figure 2 .
Ginnie Mae’s Cash Flows, by Financial Condition of MBS Issuer
Solvent Issuer
Failed Issuer
Mortgage
Borrower
FHA, VA, and
Other Federal
Mortgage
Programs
FHA, VA, and
Other Federal
Mortgage
Programs
MBS Investor
MBS
Issuer
Ginnie
Mae
Ginnie
Mae
Mortgage
Borrower
MBS Investor
Payments
From Borrowers
and Payments
on Delinquent
Mortgages
Payments
From Borrowers
and Payments
on Delinquent
Mortgages
Mortgage
Payments
Payments to Settle
Claims on Insured
Mortgages
Guarantee Fee
Payments
Payments to Settle
Claims on Insured
Mortgages
Mortgage
Payments
Data source: Congressional Budget Oce.
FHA = Federal Housing Administration; MBS = mortgage-backed security; VA = Department of Veterans Aairs.
5JANUARY 2022 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES
e volume of Ginnie Mae II securities far exceeds
that of Ginnie Mae I securities. At the end of 2020,
the outstanding balance of single-family Ginnie Mae II
MBSs totaled about $1.5 trillion, compared with about
$0.1trillion of single-family Ginnie Mae I MBSs.
9
Ginnie Mae’s Exposure to Losses
Because Ginnie Maes guarantee is limited to providing
timely principal and interest payments to investors in its
MBSs, its exposure to losses on its guarantees is bu-
ered by other parties in a mortgage transaction. e rst
level of protection comes from the homeowner’s down
payment and resources for repaying the mortgage. Any
equity contributed by the homeowner (dened as the
dierence between the value of the home and the balance
of the mortgage) reduces both the likelihood and the
severity of default.
If a borrower defaults and the default leads to losses,
the primary government guarantee from FHA, VA, or
another federal mortgage program covers a signicant
portion of losses. FHAs mortgage insurance provides a
full credit guarantee, which covers all reasonable losses
from a default. Guarantees by VA and the Rural Housing
Service are partial—from 25 percent to 50 percent of
the unpaid mortgage balance for VA and 90 percent for
RHS. In many cases, those amounts are large enough to
cover much of the expected losses from a default because
some money is eventually recovered from the borrower.
Despite down payments and primary government guar-
antees, MBS issuers frequently bear costs on mortgages
for borrowers who fail to make their monthly payments.
For a delinquent mortgage—in which the borrower
is behind on payments but is not yet declared to be
in default—the issuer is required to make scheduled
principal and interest payments to MBS investors. e
issuer generally recoups those costs once the borrower
starts making scheduled monthly payments again, or
once a claim is settled with the government guarantor.
Nevertheless, such an obligation presents a liquidity risk
for the issuer. For example, an issuer might have to raise
the money to cover a shortfall resulting from an increase
in late payments by borrowers during an economic
slowdown, when the issuer might have more trouble
borrowing money than at other times. at liquidity risk
is particularly acute for nonbank nancial institutions,
which do not have access to the same types of funding
9. See Ginnie Mae, “Unpaid Principal Balance (UPB) Summary
(September 2020), https://tinyurl.com/fxfk2db.
that banks do (such as deposits from customers and
short-term loans from the Federal Reserves discount
window).
10
An MBS issuer can also face solvency con-
cerns if too many costs associated with defaults are not
repaid as part of the claim process—whether because of
the partial guarantee from VA and RHS or because some
foreclosure costs are not reimbursable even under FHAs
full guarantee.
Ginnie Maes guarantee that investors in qualifying
MBSs will receive timely payments of principal and
interest requires it to assume the obligations for an issu-
er’sentire portfolio of outstanding MBSs when the issuer
fails. As a result, Ginnie Mae faces the same concerns
that issuers face, including having to cover uncollected
mortgage payments for MBS investors and settle claims
with primary government guarantors for defaulted
mortgages. Ginnie Maes liquidity concerns are alleviated
by the fact that it can borrow funds from the Treasury.
But that borrowing exposes taxpayers to the risk that the
funds will not be recovered from the mortgage borrowers
or primary government guarantors.
To compensate for that risk, Ginnie Mae charges issuers
a guarantee fee of 6 basis points (0.06 percent) of the
outstanding balance of an MBS. at fee is set by law
as a part of the National Housing Act, which mandates
that Ginnie Mae charge no more than 6 basis points for
MBSs made up of single-family loans.
11
Ginnie Mae can
reduce that fee for certain securities, but it has limited
ability to adjust its fee to account for the expected cost
of its guarantee. Ginnie Maes fee is in addition to the
up-front and ongoing guarantee fees that the primary
guarantor charges lenders.
Volume and Loan Characteristics of
GinnieMae’sGuarantees
e dollar amount of new MBSs that Ginnie Mae
guarantees in a year is directly related to the amount of
new mortgages guaranteed by the primary government
guarantors. As a result, the volume and loan character-
istics of Ginnie Maes guarantees closely track those of
10. See Karan Kaul and Ted Tozer, e Need for a Federal Liquidity
Facility for Government Loan Servicing (Urban Institute, July
2020), https://tinyurl.com/3c6b2exm; and Karan Kaul and
Laurie Goodman, Should Nonbank Mortgage Companies Be
Permitted to Become Federal Home Loan Bank Members? (Urban
Institute, June 2020), https://tinyurl.com/vsuck6bf.
11. Sec. 306(g)(3)(A) of the National Housing Act, 12 U.S.C.
§1721(g)(3)(A).
6 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES JANUARY 2022
FHA and VA. From a low of $77 billion in calendar
year 2006, the volume of new Ginnie Mae guarantees
grew to more than $775 billion in 2020 (see Figure 3).
12
e volume of new guarantees was especially high in
2020, partly because low interest rates, which the Federal
Reserve put in place to try to lessen the economic eects
of the coronavirus pandemic, led to a wave of mortgage
renancing. (For more details about how the pandemic
aected Ginnie Maes operations in 2020, see Box 1.)
e total value of outstanding Ginnie Mae–guaranteed
MBSs also increased over the 2006–2020 period, from
about $0.4 trillion at the end of calendar year 2006 to
more than $2.1 trillion at the end of 2020.
13
12. See Ginnie Mae, Global Markets Analysis Report (October 2021),
https://tinyurl.com/ppdhuwkx.
13. See Ginnie Mae, “Unpaid Principal Balance (UPB) Summary
(December 2020), https://tinyurl.com/fxfk2db, and “Issuance
Summary” (December 2020), https://tinyurl.com/2p857z3r.
e total dollar value of outstanding MBSs grows each year by
net issuance, which is the dierence between the amount of
new MBSs issued in a year and the amount of previously issued
MBSs repaid during that year. e gap between annual new
issuance and net issuance depends mainly on the renancing
of outstanding mortgages. at gap is small in years with little
Ginnie Maes share of new agency MBSs—dened as
the total MBSs issued by Fannie Mae, Freddie Mac, and
Ginnie Mae—has also increased in recent years. Ginnie
Mae was responsible for about 8 percent of the agency
MBSs issued in calendar year 2006. Its share rose steadily
after that, peaking at 34 percent from 2015 through
2018. Ginnie Maes share of new agency MBSs declined
to 24 percent in 2020, despite record growth in its newly
guaranteed securities in that year.
In addition to the growth in the volume of Ginnie Maes
MBS guarantees since the nancial crisis of 2008, the
characteristics of the mortgages included in those MBSs
have changed in several ways. One signicant shift is
the increase in the volume of mortgages originated or
serviced by nonbank institutions—nancial rms that
renancing and large in years with a great deal of renancing.
For example, with the surge in renancing in calendar year 2020
that resulted from low interest rates, new issuance of Ginnie
Mae–guaranteed MBSs exceeded $775 billion, but net issuance
declined by about $14 billion. e result was a small decrease
in the total value of outstanding Ginnie Mae–guaranteed
MBSsbetween the end of December 2019 and the end of
December 2020.
Figure 3 .
Amount of New Mortgage-Backed Securities Guaranteed by Ginnie Mae, CalendarYears
2000 to 2020, by the Mortgages’ Primary Guarantor
Billions of Dollars
0
100
200
300
400
500
600
700
800
900
2000 2005 2010 2015 2020
FHA
VA
Other
a
Data source: Ginnie Mae, Global Markets Analysis Report (October 2021), https://tinyurl.com/ppdhuwkx. See www.cbo.gov/publication/57176#data.
FHA = Federal Housing Administration; VA = Department of Veterans Aairs.
a. Includes the Oce of Public and Indian Housing (part of the Department of Housing and Urban Development) and the Rural Housing Service (part of the
Department of Agriculture).
7JANUARY 2022 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES
do not have a banking license and do not accept depos-
its. At the end of calendar year 2013, loans by nonbank
nancial institutions made up about 50 percent of the
mortgages in new Ginnie Mae–guaranteed MBSs. By the
end of 2020, that share had risen to more than 90 per-
cent. e share of mortgages from nonbank institutions
included in Fannie Mae’s and Freddie Macs new MBSs
also increased during that period, but by less than the
growth for Ginnie Mae–guaranteed MBSs.
14
14. Several factors have been cited as reasons for the increased role
of nonbank nancial institutions in the market for Ginnie
Mae–guaranteed MBSs. ose factors include the processes
that the federal government has used since the 2008 nancial
crisis to pursue recoveries on FHA-insured mortgages in default,
tax changes that have aected the value of mortgage-servicing
rights,and technological advances by nonbank institutions. See
Another signicant shift is the increasing prevalence of
loans guaranteed by VA in the securities that Ginnie
Mae guarantees. From calendar year 2000 through 2009,
VA loans accounted for 21 percent of the mortgages in
Ginnie Mae–guaranteed MBSs, on average. Since 2015,
that share has averaged 44 percent, reaching a peak of
55percent in 2020. e total dollar amount of home
loans guaranteed by VA has also grown, for a number of
possible reasons, including the stricter eligibility require-
ments that mortgage lenders adopted for non-VA orig-
inations as a result of the nancial crisis; the greater use
by borrowers of renancing options available through
You Suk Kim and others, “Liquidity Crises in the Mortgage
Market,Brookings Papers on Economic Activity (Spring 2018),
https://tinyurl.com/uu53b5a4.
Box 1 .
Eects of the Coronavirus Pandemic on Ginnie Mae
The coronavirus pandemic that began in 2020 and the
resulting recession in the United States have had a negative
eect on the financial condition of many borrowers, including
those with mortgages contained in Ginnie Mae–guaranteed
mortgage-backed securities (MBSs). In response to the
economic slowdown, the Federal Housing Administration
(FHA), the Department of Veterans Aairs (VA), and the Rural
Housing Service all announced that borrowers with loans
backed by their guarantee programs could reduce or suspend
mortgage payments.
1
Such loans are referred to as being in
forbearance.
As a result of those actions, the percentage of borrowers with
mortgages included in Ginnie Mae–guaranteed MBSs who
were behind on their monthly payments increased significantly.
For example, the share of people with FHA-insured mortgages
who were at least 90 days past due (including loans in forbear-
ance) or who were in foreclosure rose from 3.5 percent at the
end of calendar year 2019 to 10.8 percent at the end of 2020.
For VA-guaranteed mortgages, the share of borrowers who
were at least 90 days past due (including loans in forbearance)
1. See Department of Housing and Urban Development, “COVID-19 Resources
for Homeowners” (June 25, 2021), www.hud.gov/coronavirus/homeowners;
Department of Veterans Aairs, “VA Home Loans: Information for VA Home
Loan Borrowers During COVID-19” (October 20, 2021), https://tinyurl.
com/3xcs9cyb; and Department of Agriculture, “Rural Development
COVID-19Response” (June 24, 2021), www.rd.usda.gov/coronavirus.
or who were in foreclosure increased from 1.9 percent to
5.8percent in 2020.
2
A sharp increase in delinquent borrowers can create stress
for issuers of Ginnie Mae–guaranteed MBSs because they are
required to forward the missed payments to MBS investors. In
an attempt to alleviate that stress, Ginnie Mae announced a
series of actions early in the pandemic, including the Pass-
Through Assistance Program (PTAP).
3
Under that program, a
Ginnie Mae–approved issuer with a shortfall of available funds
can ask Ginnie Mae to make advance payments to investors
without being declared a failed issuer (whose assets are taken
over by Ginnie Mae). In exchange, the issuer is required to
repay the advance, with interest, in a specified period.
PTAP was designed to be a program of last resort for issuers,
requiring them to exhaust all other funding options before
applying for an advance from Ginnie Mae. As a result, the
volume of PTAP advances has been fairly low. From April 2020
through September 2021, Ginnie Mae issued only 23 advances,
totaling $13.1 million.
4
By the end of September 2021, the
balance of all advances had been repaid.
2. See Ginnie Mae, Global Markets Analysis Report (prepared by State Street
Global Advisors and the Urban Institute’s Housing Finance Policy Center,
March 2020 and February 2021), https://tinyurl.com/ppdhuwkx.
3. See Ginnie Mae, “All Participant Memorandum 20-03: Availability of Pass-
Through Assistance Program for Participants in Ginnie Mae’s Single-Family
MBS Program” (April 10, 2020), https://tinyurl.com/5bx3nua2.
4. See Ginnie Mae, “Issuers: Ginnie Mae PTAP Assistance” (accessed
November 9, 2021), https://tinyurl.com/y82xjvba.
8 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES JANUARY 2022
VA; and, since January 1, 2020, the removal of the cap
on the size of a mortgage eligible for a VA guarantee.
15
Ginnie Mae’s Exposure to Future Losses
FromChanges in Its Guarantees
e protected position of Ginnie Maes guarantees limits
expected losses, meaning that Ginnie Mae is not likely
to suer large losses from its guarantees under normal
economic conditions. But Ginnie Mae would be exposed
to the risk of large losses in a period of severe economic
stress that included high default rates on mortgages
and widespread insolvency among issuers of its guaran-
teed MBSs. e two recent shifts in those MBSs—the
increasing prevalence of loans from nonbank institutions
and of VA-guaranteed loans—could increase Ginnie
Maes losses during times of stress.
Nonbank nancial institutions are subject to state-level
oversight, and Ginnie Mae reviews them as MBS issuers.
Nevertheless, nonbank nancial institutions are typically
subject to less regulation of their safety and soundness
than banks are. ey also have less access to the sources
of liquidity available to banks, such as consumer depos-
its and Federal Reserve lending programs.
16
As a result,
the possibility of insolvency by issuers of Ginnie Mae–
guaranteed MBSs may increase along with the share of
mortgages from nonbank institutions included in those
MBSs. (Investors that lend to nonbank nancial institu-
tions may exert more market discipline on those insti-
tutions than they do on banks because their payments
are not protected by a government guarantee, such as
federal deposit insurance. Such market discipline would
oset some of the eect that less regulation could have
on nonbank institutions’ risk of insolvency.) Nonbank
nancial institutions are also subject to a more uncertain
resolution process in the event of failure than banks are,
which could expose Ginnie Mae to additional nancial
and operational risk.
Ginnie Mae has attempted to address those issues by
tightening standards for its MBS issuers, such as require-
ments for their liquidity, net worth, and ratio of capital
15. For more about the lifting of that cap, see Veterans Benets
Administration, “Blue Water Navy Vietnam Veterans Act of
2019” (February 5, 2021), https://tinyurl.com/2p9f33vb.
16. See You Suk Kim and others, “Liquidity Crises in the Mortgage
Market,Brookings Papers on Economic Activity (Spring 2018),
https://tinyurl.com/uu53b5a4.
to assets. Ginnie Mae has also increased its ongoing
monitoring of issuers’ nancial condition.
17
All mortgages included in Ginnie Mae–guaranteed secu-
rities have a primary federal guarantee, but those federal
guarantees are not identical. In particular, VAs guaran-
tee covers only part of the losses on a defaulted loan,
with MBS issuers responsible for the rest. As a result,
the possibility of insolvency by issuers of Ginnie Mae–
guaranteed MBSs may increase along with the share
of VA-guaranteed loans included in those securities.
(Ginnie Maes risks could change if the corporations role
was modied as part of broader changes to the housing
nance system. For details, see Box 2.)
Estimating Federal Subsidy Costs for
Ginnie Mae in CBO’s Baseline and
UnderaStress Scenario
In the federal budget, the subsidy outlays recorded for
Ginnie Mae in a given year equal the dollar amount
of new mortgage-backed securities that Ginnie Mae
guarantees in that year multiplied by the budgetary cost
per dollar of new guarantees, also known as the sub-
sidy rate.
18
To project subsidy outlays for Ginnie Mae,
CBO needs estimates of both the expected volume of
Ginnie Maes guarantees and Ginnie Maes subsidy rate.
17. See Ginnie Mae, An Era of Transformation (September 2014),
https://tinyurl.com/3fkn9ty6 (PDF, 464 KB), and Ginnie Mae
2020: Roadmap for Sustaining Low-Cost Homeownership (June
2018), https://tinyurl.com/taz85ma7 (PDF, 6 MB). e capital
requirement for nonbank nancial institutions is a total adjusted
net worth (based on Ginnie Maes denition) equal to at least
6 percent of total assets. In comparison, banks, thrifts, bank
holding companies, and savings and loan holding companies
must meet one of the following targets: tier 1 capital (including
retained earnings and stock) equal to at least 5 percent of total
assets or 6 percent of risk-based assets, or total capital equal to at
least 10 percent of risk-based assets. See Ginnie Mae, Ginnie Mae
MBS Guide, 5500.3, Rev. 1 (accessed March 9, 2021), Chapter 2,
pp. 2-9 and 2-10, https://tinyurl.com/d58c7twd.
18. e subsidy cost associated with the current years guarantees
(which is net of the guarantee fees that Ginnie Mae collects from
MBS issuers) is the largest component of Ginnie Maes eect
on the federal budget. But the budget reects other aspects of
Ginnie Maes operations as well. Ginnie Mae earns interest on
its investments in Treasury securities, which it makes primarily
with cash from previously collected fees not used to cover
expenses. Itis also credited with osetting collections from fees
other than its guarantee fee. Furthermore, Ginnie Mae receives
Congressional appropriations to pay for salaries and other
operating expenses. In addition, the budget reects credit subsidy
reestimates, which are revisions to estimates of the subsidy costs
of guarantees made in previous years.
9JANUARY 2022 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES
e subsidy rate that CBO uses for its baseline budget
projections is calculated using the measures for the cost
of federal credit programs specied in the Federal Credit
Reform Act (FCRA).
For this analysis, CBO also estimated Ginnie Maes
FCRA subsidy rate under a scenario of severe economic
stress. at scenario is meant to illustrate the risks that
Ginnie Mae could be exposed to in a major recession.
ose risks include the additional uncertainty stem-
ming from the recent changes to the types of mortgages
included in Ginnie Mae–guaranteed MBSs. In CBO’s
baseline, which represents average economic conditions,
Ginnie Maes new guarantees in 2022 are projected to
produce budgetary savings of $2.2 billion in that year
(on a FCRA basis). But under the stress scenario, those
guarantees would have a budgetary cost of $3.0 billion
in2022, CBO estimates.
Estimating the Volume of Guarantees
For each of the 10 years of its baseline budget projec-
tions, CBO uses its macroeconomic forecast to estimate
the total value of mortgages expected to be originated in
a given year. On the basis of that projection for the entire
single-family mortgage market—including mortgages
that meet the criteria for a government guarantee and
Box 2 .
Possible Changes to Ginnie Mae’s Role From a Restructuring of the
HousingFinanceSystem
Policymakers’ interest in Ginnie Mae extends beyond the
rapid growth of its guarantees on mortgage-backed securi-
ties (MBSs) and the shifts in the types of loans and financial
institutions covered by those guarantees. One particular area
of focus has been the role that Ginnie Mae might play in the
mortgage markets as a part of a broader restructuring of the
U.S. housing finance system. Members of Congress and market
participants have discussed such restructuring since 2008,
when the federal government assumed control of Fannie Mae
and Freddie Mac, government-sponsored enterprises (GSEs)
that are central to the housing finance system. Not all propos-
als for restructuring include a modified role for Ginnie Mae, but
most recognize that changes to the GSEs might also require
the Federal Housing Administration (FHA), the Department
of Veterans Aairs (VA), the Rural Housing Service (RHS), and
Ginnie Mae to adapt to those changes.
A number of proposals for altering the housing finance system
would extend Ginnie Mae’s guarantees to MBSs that contain
mortgages currently guaranteed by the GSEs—that is, mort-
gages without a primary government guarantee.
1
Under such
proposals, Ginnie Mae would continue to guarantee the timely
payment of principal and interest to MBS investors, much as
1. See Michael Bright and Ed DeMarco, Toward a New Secondary Mortgage
Market (Milken Institute, September 2016), https://tinyurl.com/3ekfzs48; and
statements by witnesses at a hearing of the House Committee on Financial
Services titled “A Legislative Proposal to Provide for a Sustainable Housing
Finance System: The Bipartisan Housing Finance Reform Act of 2018”
(December 21, 2018), https://tinyurl.com/ywjk328w.
it does today. The main dierence would be that private firms
would provide the capital necessary to protect Ginnie Mae
against the sorts of losses that FHA, VA, and RHS cover under
their guarantee programs. That private capital could take many
forms, similar to the approach that the GSEs use in their exist-
ing credit-risk-transfer transactions.
2
With such a change, Ginnie Mae would need to take several
steps to address the additional risks involved in its expanded
guarantees. First, it would need to acquire the data and
expertise to assess and manage the riskiness of the private
firms whose capital would substitute for a primary government
guarantee. Second, it would need to determine whether exist-
ing systems and personnel could handle the added volume
associated with a larger group of issuers and mortgages.
3
Third, it might need to reassess its current guarantee fees and
servicing requirements to ensure that they were sucient to
safeguard taxpayers from the new risks. Ginnie Mae might be
able to use the expertise of the GSEs’ regulator, the Federal
Housing Finance Agency, to help it manage its expanded role
under such proposals.
2. For a description of the credit-risk-transfer activities of Fannie Mae and
Freddie Mac, see Congressional Budget Oce, Transferring Credit Risk on
Mortgages Guaranteed by Fannie Mae or Freddie Mac (December 2017),
www.cbo.gov/publication/53380.
3. For a discussion of Ginnie Mae’s current stang levels and risk
management procedures, see Government Accountability Oce, Ginnie
Mae: Risk Management and Stang-Related Challenges Need to Be
Addressed, GAO-19-191 (April 3, 2019), www.gao.gov/products/gao-19-191.
10 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES JANUARY 2022
those that do not—CBO allocates shares to the guaran-
tee programs of Fannie Mae, Freddie Mac, FHA, VA,
and RHS.
19
e allocations are based on those programs
past shares of the mortgage market and any implemented
or scheduled changes in law or policy that might aect
future shares, such as changes in a programs guarantee
fees.
CBO then uses its projections of the volume of guar-
antees by FHA, VA, and RHS to project the volume of
guarantees by Ginnie Mae. at estimate is based on
the share of mortgages from each primary government
guarantor that have been part of Ginnie Mae–guaranteed
MBSs in recent years and any announced changes to
Ginnie Maes securitization programs.
20
On the basis of
that analysis, CBO estimates that in 2022, Ginnie Mae
will guarantee more than 90 percent of FHA-backed
loans and 98 percent of VA- and RHS-backed loans.
(Lenders are expected to hold the rest of those loans in
their portfolios rather than pooling the mortgages into
MBSs.) In CBO’s July 2021 baseline, those percentages
are projected to result in a total of $577 billion in new
Ginnie Mae MBS guarantees in 2022.
Estimating the Subsidy Rate on
GinnieMae’sGuarantees
e Federal Credit Reform Act requires that the impact
of Ginnie Maes new MBS guarantees each year—Ginnie
Maes subsidy cost—be recorded in the federal budget
on a present-value basis. (A present value is a single
number that expresses a ow of income or payments in
terms of an equivalent lump sum received or paid at a
particular point in time.) at subsidy cost is calculated
as the dierence between the present value of the losses
that Ginnie Mae is expected to incur on a given years
cohort of newly guaranteed MBSs over their lifetime and
the present value of the guarantee fees that Ginnie Mae
expects to collect on those guarantees over their lifetime.
19. See Congressional Budget Oce, “Details About Baseline
Projections for Selected Programs: Federal Programs
at Guarantee Mortgages” (July 2021), https://tinyurl.
com/5x9w63j2.
20. e annual appropriation act that provides funding for Ginnie
Mae sets a limit on the dollar volume of Ginnie Maes guarantees.
In recent years, the limit has been set high enough to not
constrain Ginnie Maes ability to meet its MBS issuers’ demand
for guarantees. at limit does not aect CBO’s projections
because, in recent years, it has been higher than CBO’s projection
of the volume of new guarantees.
For Fannie Mae, Freddie Mac, FHA, and other large
federal credit programs, CBO uses its own models to
estimate subsidy rates on a FCRA basis for its base-
line projections. For Ginnie Mae, however, CBO uses
the estimated FCRA subsidy rate published in the
Administrations Federal Credit Supplement.
21
CBO does
not model Ginnie Maes guarantee programs, because
of a lack of data about the past performance of MBS
issuers and the relatively small budgetary cost of those
programs.
In the Federal Credit Supplement for scal year 2021,
Ginnie Mae is projected to have a FCRA subsidy rate of
−0.38 percent in 2022—meaning that the present value
of projected losses on new guarantees made in 2022 is
smaller than the present value of the fees that Ginnie
Mae is projected to collect in exchange for providing
those guarantees. Programs with negative subsidy rates
produce savings for the federal budget.
e Administrations estimate of Ginnie Maes FCRA
subsidy rate is based on four sets of cash ows: defaults,
recoveries, fees, and miscellaneous items.
Defaults. e losses that Ginnie Mae incurs from
mortgage defaults are a function of the number
of MBS issuers that fail to meet their obligations,
the number of mortgages in those failed issuers
MBSs that go into delinquency or default, and the
losses experienced on those defaults (see Figure 4).
Delinquent and defaulted loans in failed issuers
MBSs result in cash outows from Ginnie Mae for
several reasons. First, Ginnie Mae must forward the
scheduled monthly principal and interest payments
to MBS investors for failed issuers’ delinquent
mortgages that have not yet been removed from
(bought out of) an MBS.
22
Second, Ginnie Mae
must repay the remaining principal to MBS
investors for failed issuers’ defaulted mortgages that
have been bought out of an MBS. ird, it must
cover maintenance costs for properties associated
with defaulted mortgages until those properties
21. See Oce of Management and Budget, Budget of the U.S.
Government, Fiscal Year 2022: Federal Credit Supplement (May
2021), www.whitehouse.gov/omb/supplemental-materials.
22. If, instead of servicing the loans of failed issuers itself, Ginnie
Mae transferred the servicing duties to a solvent issuer, it
would have little or no cash outow. Alternatively, if Ginnie
Mae retained the servicing obligation and hired a third-party
subservicer to carry out those duties, it would have cash outows
to pay the subservicer.
11JANUARY 2022 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES
can be foreclosed on or transferred to another
party. In the Administrations 2022 FCRA subsidy
estimate—which reects the average eects of each
of Ginnie Maes potential resolution options for
addressing a failed issuer—Ginnie Maes outows
for delinquencies and defaults are projected to equal
0.19percent of the original principal balance of all
MBSs newly guaranteed by Ginnie Mae in 2022.
Recoveries. After Ginnie Mae steps in to assume
the obligations of failed issuers, it receives cash
inows from several sources: repayment of forwarded
principal and interest payments for delinquent
loans; recoveries after a loan is modied or a home
is foreclosed on or transferred; claim payments from
FHA, VA, and other primary government guarantors
for defaulted mortgages in failed issuers’ MBSs; and
miscellaneous reimbursements from those guarantors.
In the Administrations 2022 FCRA subsidy estimate,
Ginnie Mae is expected to recover 96 percent of its
outows for delinquencies and defaults. On the basis
of those default and recovery estimates, the present
value of default costs, net of recoveries, is projected to
equal 0.01 percent of the original principal balance of
all MBSs newly guaranteed in 2022.
Fees. Ginnie Mae also receives cash inows from
the guarantee fee it charges solvent issuers (set at
0.06percent of the outstanding balance of an MBS).
In the Administrations subsidy estimate, those fee
collections are projected to have a present value in
2022 equal to −0.37 percent of the original principal
balance of all MBSs newly guaranteed in that year.
Miscellaneous Cash Flows. Ginnie Maes activities
produce a number of other cash ows that are not
directly related to defaults, recoveries, or fees. ose
cash ows—which are associated with guarantees on
multifamily mortgages, reverse mortgages, and VA
mortgages to service members—represent a small part
of Ginnie Maes subsidy cost. In the Administrations
2022 FCRA estimate, those cash ows are
projected to have a present value of −0.02percent
of the original principal balance of all MBSs newly
guaranteed in that year.
Figure 4 .
Representation of Ginnie Mae’s Losses as a Share of Its Annual Guarantees
Guarantees on
Defaulted Mortgages
From Failed Issuers
Guarantees on
Mortgages From
Failed Issuers
MBS Guarantees
on Mortgages
From All Issuers
Losses on Defaulted
Mortgages From
Failed Issuers
Data source: Congressional Budget Oce.
MBS = mortgage-backed security.
12 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES JANUARY 2022
Although FCRA estimates are used in the federal budget
for most credit programs, including Ginnie Maes, CBO
also frequently prepares fair-value estimates for credit
programs to provide a more comprehensive picture of
the programs’ long-term costs.
23
e fair-value approach
recognizes that, in the private sector, uncertain cash
ows that grow or shrink along with the economy are
less valuable than cash ows that are stable regardless of
economic conditions. us, fair-value estimates account
for market risk, which is the element of nancial risk
that is correlated with overall economic conditions (and
that therefore cannot be eliminated by diversifying a
portfolio of investments). For assets, such as loan guaran-
tees, that are more likely to go into default when eco-
nomic conditions are poor, fair-value estimates discount
the value of future cash ows at a higher rate than the
interest rates on Treasury securities, which are consid-
ered risk-free. ose fair-value estimates show lower
savings than present-value estimates made using the
method prescribed for federal loan programs by FCRA,
which involves discounting future cash ows at Treasury
interestrates.
CBO publishes a fair-value subsidy rate for Ginnie Mae
as part of its fair-value estimates for various federal credit
programs. CBO projects that in 2022, the fair-value
subsidy rate for Ginnie Mae will eectively be zero—
meaning that Ginnie Maes guarantee programs will not
produce any budgetary costs or savings when measured
on a fair-value basis.
24
Estimating the FCRA Subsidy Rate
UnderaStressScenario
e cash ows that the Administration uses to estimate
Ginnie Maes FCRA subsidy rate are based on a pro-
jection of how the MBSs that Ginnie Mae is expected
to guarantee in 2022 will perform on average. at
projection is based on data about the past performance
of Ginnie Mae–guaranteed MBSs and their issuers, as
well as the Administrations forecast of future economic
conditions. e Administrations subsidy rate estimate
represents an average of expected defaults, recoveries,
fees, and other cash ows across a range of scenarios and
economic outlooks.
23. See Congressional Budget Oce, Estimates of the Cost of
Federal Credit Programs in 2022 (October 2021), www.cbo.gov/
publication/57412.
24. Ibid.
For this analysis, CBO adjusted the Administrations esti-
mates of average cash ows to calculate a FCRA subsidy
rate for Ginnie Mae under a scenario of economic stress.
at scenario, which reects a severe recession, represents
the worst 1 percent of economic outcomes that CBO
models over the coming decade.
Although CBO does not have detailed performance data
about the MBSs underlying Ginnie Maes cash ows, it
can consider the factors that would inuence the issu-
ers and mortgages in Ginnie Maes guarantee programs
under a stress scenario. For example, CBO can use its
models of FHAs and VAs loan guarantee programs to
project how mortgage cash ows might vary under dier-
ent future economic conditions.
25
CBO used those mod-
els for this analysis to develop projections of defaults,
prepayments, and recoveries on the mortgages that make
up Ginnie Mae–guaranteed MBSs (see Table 1). In addi-
tion, CBO analyzed failure rates for thrift institutions
insured by the Federal Deposit Insurance Corporation
(FDIC) to estimate how much the failure rate for MBS
issuers might rise under a stress scenario.
How CBO Developed the Stress Scenario. e default,
prepayment, and recovery rates in the stress scenario are
based on results from CBO’s FHA model. at model
uses 1,000 combinations of macroeconomic variables
(such as interest rates, home prices, and unemployment
rates) to simulate future economic conditions. On the
basis of those conditions and the portfolio of mortgages
that FHA is expected to guarantee, the model generates a
series of cash ows—representing mortgage repayments,
mortgage defaults, and recoveries on those defaults—for
each period from a loans origination to its repayment or
default for each of the 1,000 paths for economic con-
ditions. To calculate the FCRA subsidy rate for FHA,
CBO converts those cash ows to present values by dis-
counting them back to the origination date of each loan
using the appropriate FCRA discount rate.
26
25. See Francesca Castelli and others, Modeling the Budgetary Costs of
FHA’s Single Family Mortgage Insurance, Working Paper 2014-05
(Congressional Budget Oce, September 2014), www.cbo.gov/
publication/45711; and Congressional Budget Oce, e Role
of the Department of Veterans Aairs in the Single-Family Mortgage
Market (September 2021), www.cbo.gov/publication/57024.
26. ose discount rates are the interest rates on Treasury securities
of comparable maturity. For example, the projected yield on
Treasury securities maturing in two years is used to discount
cash ows two years from the loan origination date, a three-year
Treasury rate is used for cash ows three years from origination,
and so on.
13JANUARY 2022 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES
Although the FHA subsidy rate that CBO publishes
is the average rate across all 1,000 paths for economic
conditions, the model produces a subsidy rate for each
path. For this analysis, CBO ranked the 1,000 paths by
subsidy rate, from lowest to highest. e severe economic
stress scenario is dened as the set of average cash ows
from the 10 paths with the highest subsidy rates.
ose 10 highest subsidy rates result from a set of projec-
tions for interest rates, home prices, and unemployment
rates that would lead to very high default rates and low
recoveries on FHA-insured mortgages. For example,
in the 10 paths, nominal home prices are projected to
decline by a total of 21 percent, on average, over the
rst ve years of the projection and by 38 percent over
10years. In comparison, average home prices declined
by a total of 20 percent over the ve years from 2007
to 2012 (which included the nancial crisis), before
rebounding to show total growth of 4 percent over the
10-year period from 2007 to 2017.
27
27. ose gures for actual changes in home prices are based on the
Federal Housing Finance Agencys monthly, purchase-only, and
Projected Cash Flows in the Stress Scenario. Using
the results from the stress scenario, CBO adjusted the
Administrations estimates of average cash ows for
Ginnie Mae to develop a set of cash ows representing
a severe recessionary environment for Ginnie Mae. e
scale of the adjustments that CBO made depended on
the category of cash ows—defaults, recoveries, or fees.
(CBO did not adjust miscellaneous cash ows for the
stress scenario because of their negligible eect on Ginnie
Maes subsidy rate.)
Defaults. CBO used two multipliers to adjust the
Administrations estimate of Ginnie Maes annual cash
outows for delinquencies and defaults. e rst mul-
tiplier is based on defaults in the 10 paths with the
highest subsidy rates from CBO’s FHA model (the stress
scenario). e second multiplier is derived from CBO’s
analysis of FDIC data on failure rates for thrift institu-
tions. By looking at the share of thrifts that failed in each
year from 1934 to 2016, CBO developed an estimate
of the share of Ginnie Mae issuers that would fail in a
severe recession.
28
Together, those two multipliers—which were adjusted
to reect the correlation between mortgage defaults
and failures of nancial institutions—increase the
Administrations estimate of default-related cash ows
20-fold for the stress scenario. As a result, Ginnie Maes
estimated cash outows in 2022 for delinquent and
defaulted loans in failed issuers’ MBSs, as a share of
the original balance of all MBSs newly guaranteed by
Ginnie Mae in 2022, increase from 0.19 percent in
the Administrations subsidy estimate (which is used in
CBO’s baseline) to 3.76 percent under CBO’s severe
stress scenario (see Table 2).
Recoveries. e Administration calculates Ginnie Maes
cash ows from recoveries as a share of its cash out-
ows for delinquent and defaulted mortgages in failed
issuers’ MBSs. us, the rst step in CBO’s estimate
of recoveries in a severe recession was to increase the
Administrations estimate of average recovery cash ows
by the same 20-fold multiplier used to adjust defaults.
not seasonally adjusted house price index. See Federal Housing
Finance Agency, “House Price Index Datasets” (accessed October
26, 2021), https://tinyurl.com/cy6bjtby.
28. In CBO’s judgment, thrifts operating during the 1934–2016
period are a good proxy for the risks associated with nonbank
institutions’ increased participation in Ginnie Mae’s MBS
programs today.
Table 1 .
Data Sources That CBO Used to Estimate
Ginnie Mae’s Cash Flows Under a
StressScenario
Variable
Source of Projection
for the Stress Scenario
Default Rates for MBS
Issuers
Analysis of data from the Federal Deposit
Insurance Corporation and the Bureau of
Economic Analysis
Default Rates for
Mortgages
Simulations from CBO’s model of FHA’s
insurance program for single-family
mortgages
Prepayment Rates for
Mortgages
Simulations from CBO’s model of FHA’s
insurance program for single-family
mortgages
Recoveries on Defaulted
Mortgages
Simulations from CBO’s model of FHA’s
insurance program for single-family
mortgages
Analysis of losses on FHA-insured mortgages
that would exceed the guarantee percentage
on VA-guaranteed mortgages
Data source: Congressional Budget Oce.
FHA = Federal Housing Administration; MBS = mortgage-backed security;
VA = Department of Veterans Aairs.
14 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES JANUARY 2022
CBO further adjusted recovery cash ows to reect the
expectation that home prices would decline in a severe
recession, resulting in greater default losses. In particular,
mortgages guaranteed by VA would be expected to incur
losses in excess of VAs partial guarantee, resulting in
costs to Ginnie Mae (in its role as a replacement for the
failed issuer). CBO estimated the size of those additional
losses by determining the projected losses on FHA-
insured loans under the stress scenario that would exceed
the amount covered by a VA guarantee and then apply-
ing that estimate of excess losses to the expected share of
VA-guaranteed loans in Ginnie Mae–guaranteed securi-
ties. at estimate reects the lower expected default rate
on mortgages guaranteed by VA than on loans insured
byFHA.
In the stress scenario, those adjustments reduce the
Administrations estimate of Ginnie Maes recovery cash
ows by 28 percent. As a result, Ginnie Mae is projected
to recover 70 percent of its outows for delinquencies and
defaults in 2022 under CBO’s severe stress scenario, com-
pared with 96 percent in the baseline subsidy estimate.
With those default and recovery estimates, the present
value of default costs, net of recoveries, would equal
1.07percent of the original principal balance of all
MBSsnewly guaranteed in 2022 under CBO’s severe
stress scenario, compared with 0.01 percent in the
baseline subsidy estimate.
Table 2 .
Estimates of Ginnie Mae’s Cash Flows and Budgetary Cost in 2022
UnderCBO’sBaselineand a Stress Scenario
CBO’s Baseline Severe Economic Stress Scenario
a
Ginnie Mae’s Cash Flows From MBSs Newly Guaranteed in 2022
Lifetime default costs (Percentage of original principal
balance of MBSs) 0.19 3.76
Recoveries on defaulted mortgages (Percentage of
default costs) 96.17 69.72
Present value of net default costs (Percentage of
original principal balance of MBSs) 0.01 1.07
Present value of income from guarantee fees
(Percentage of original principal balance of MBSs) -0.37 -0.52
Present value of miscellaneous cash ows
(Percentage of original principal balance of MBSs) -0.02 -0.02
b
Ginnie Mae’s Subsidy Rate (Per dollar of new guarantees) -0.38 0.53
Ginnie Mae’s Subsidy Cost (Billions of dollars) -2.2 3.0
Data source: Congressional Budget Oce, using data from Oce of Management and Budget, Budget of the U.S. Government, Fiscal Year 2022: Federal Credit
Supplement (May 2021), www.whitehouse.gov/omb/supplemental-materials. See www.cbo.gov/publication/57176#data.
MBSs = mortgage-backed securities.
a. This scenario represents the worst 1 percent of economic outcomes that CBO models over the coming decade. It represents a severe recession that includes
higher-than-expected failure rates among issuers of Ginnie Mae–guaranteed MBSs and larger-than-expected losses on the mortgages underlying those
securities. CBO based the stress scenario on its model of the Federal Housing Administration’s (FHA’s) mortgage guarantee programs. That model uses
1,000 combinations of dierent values for major macroeconomic variables to simulate future economic conditions and projects FHA’s cash flows for loan
guarantees under those diering conditions. CBO defined the stress scenario as the 10 out of those 1,000 sets of economic conditions that produced the
highest estimated subsidy rates for FHA’s loan guarantees (and thus presumably for Ginnie Mae’s guarantees of MBSs composed of loans guaranteed by FHA
and other federal agencies).
b. CBO did not adjust miscellaneous cash flows for the stress scenario because of their negligible eect on Ginnie Mae’s subsidy rate.
15JANUARY 2022 GINNIE MAE AND THE SECURITIZATION OF FEDERALLY GUARANTEED MORTGAGES
Fees. CBO adjusted the Administrations estimate of
annual cash ows from Ginnie Maes guarantee fees for
the stress scenario to reect the expected dierence in
borrowers’ voluntary prepayments of mortgages. Under
the economic conditions implied by the stress scenario,
borrowers would voluntarily prepay their mortgages
more slowly—driven in part by declining home prices
and tighter lending standards for renancing loans—
leaving higher outstanding mortgage balances, which
would generate more fee income for Ginnie Mae. As
it did with recoveries, CBO used results from its FHA
model to estimate the size of that adjustment.
In the stress scenario, the adjustment for slower prepay-
ments increases the Administrations estimate of cash ows
from guarantee fees by 42 percent. As a result, Ginnie Maes
fee income in 2022 is projected to have a present value of
−0.52 percent of the original balance of all MBSs newly
guaranteed in that year under the stress scenario, compared
with −0.37 percent in the baseline subsidy estimate.
Estimates of Ginnie Mae’s Budgetary Costs
Using the Administrations estimates of cash ows for
Ginnie Mae, CBO projects in its July 2021 baseline
that Ginnie Maes outows in 2022 for delinquent
and defaulted loans in failed issuers’ MBSs will equal
0.19 percent of the original principal balance of the
new MBSs that Ginnie Mae will guarantee in 2022.
Recoveries on those defaults are projected to equal
96percent, leaving net losses to Ginnie Mae equal to
0.01 percent of guarantees. Fees collected on those
guarantees (equal to −0.37 percent) and miscellaneous
cash ows (equal to −0.02 percent) more than oset
losses, resulting in an estimated FCRA subsidy rate of
−0.38percent for 2022 (see Table 2). Because CBO pro-
jects that Ginnie Mae will guarantee $577 billion in new
MBSs in 2022, the budgetary impact of those guarantees
is estimated to be a savings of $2.2 billion under the
baseline.
Ginnie Maes estimated subsidy rate is higher under the
scenario of severe economic stress that CBO constructed
for this analysis because of the changes to defaults, recov-
eries, and fee income described above. With defaulted
mortgages from failed issuers increased and recoveries
decreased, net losses to Ginnie Mae in 2022 under the
stress scenario would equal 1.07 percent of the original
principal balance of MBSs newly guaranteed in that year.
Although fee income would increase to −0.52 percent
of guarantees, that increase would not be sucient to
oset the larger losses, resulting in a FCRA subsidy rate
of 0.53 percent under the stress scenario. Given CBO’s
projection of $577 billion in new Ginnie Mae guarantees
in 2022, the budgetary impact of those guarantees would
be a cost of $3.0 billion under the stress scenario—an
increase of $5.2 billion from CBO’s baseline.
29
e estimates that Ginnie Maes net losses (on a FCRA
basis) would equal 0.01 percent of guarantees on average
and 1.07 percent in a stress scenario that has a 1 percent
chance of occurring over the coming decade suggest
that Ginnie Mae would eectively incur no losses in the
99percent of economic scenarios not covered by CBO’s
stress scenario.
30
at estimate is consistent with Ginnie
Maes past performance, including during and after the
nancial crisis of 2008.
31
CBO did not estimate losses in the stress scenario under
the fair-value approach, but the results would follow a
similar pattern: Ginnie Mae would be expected to incur
few or no losses in the majority of economic scenarios
and much higher losses in the stress scenario. e main
dierence from the FCRA estimates is that Ginnie Maes
net losses in the stress scenario would probably be much
higher under the fair-value approach.
29. e stress scenario described here would probably also increase
the costs associated with Ginnie Maes guarantees of MBSs issued
before or after 2022. us, the total cost increase resulting from
that scenario would exceed the $5.2 billion associated with new
Ginnie Mae guarantees in 2022.
30. Although CBO did not calculate Ginnie Maes net losses
in scenarios other than the one with a 1 percent chance of
occurring, the average for 99 paths with no net losses and 1 path
with net losses of 1.07 percent is net losses equal to 0.01 percent.
31. According to Ginnie Maes annual reports, since 2007 the
corporations posted revenues have exceeded its expenses every
year except 2019. (In that year, a large write-down in the value
of Ginnie Maes future fee income from its guarantees, driven
by market conditions and by modeling and accounting changes,
caused the corporation to post an operating loss.) Any revenues
it earns that exceed those costs are used to accumulate a capital
reserve, which is held as cash by the Treasury or invested in
Treasury securities. See Ginnie Mae, “Annual Reports” (accessed
September 28, 2021), https://tinyurl.com/3bhpmybv.
List of Tables and Figures
Figures
1. The Role of Ginnie Mae in the Market for Mortgage-Backed Securities 3
2. Ginnie Mae’s Cash Flows, by Financial Condition of MBS Issuer 4
3. Amount of New Mortgage-Backed Securities Guaranteed by Ginnie Mae,
CalendarYears 2000 to 2020, by the Mortgages’ Primary Guarantor 6
4. Representation of Ginnie Mae’s Losses as a Share of Its Annual Guarantees 11
Tables
1. Data Sources That CBO Used to Estimate Ginnie Mae’s Cash Flows Under a StressScenario 13
2. Estimates of Ginnie Mae’s Cash Flows and Budgetary Cost in 2022
UnderCBO’sBaselineand a Stress Scenario 14
About This Document
e Congressional Budget Oce prepared this report in fulllment of its eorts to be more
transparent. In keeping with CBO’s mandate to provide objective, impartial analysis, the report
makes no recommendations.
Mitchell Remy prepared the report with guidance from Sebastien Gay. Nabeel Alsalam,
MichaelFalkenheim, Paul B. A. Holland, Erik O’Donoghue, Jerey Perry, Robert Reese,
AuroraSwanson, David Torregrosa, and Susan Willie oered comments.
Michael Bright of the Structured Finance Association, Ed DeMarco and Angel Hernandez of the
Housing Policy Council, Laurie Goodman of the Urban Institute, Karen Pence of the Board of
Governors of the Federal Reserve System, and Mark Willis of New York Universitys Furman Center
commented on an earlier draft. e assistance of external reviewers implies no responsibility for the
nal product; that responsibility rests solely with CBO.
Mark Doms, Jerey Kling, and Robert Sunshine reviewed the report. Christian Howlett edited it,
and Jorge Salazar created the graphics and prepared the text for publication. e report is available at
www.cbo.gov/publication/57176.
CBO seeks feedback to make its work as useful as possible. Please send any comments to
Phillip L. Swagel
Director
January 2022