GAO, Federal Student Loans: Education Could Improve Direct Loan Program Cost Estimates (GAO-22-105365, 2022-07)
Summary
GAO-22-105365, a GAO report to congressional requesters dated July 28, 2022, titled Student Loans: Education Has Increased Federal Cost Estimates of Direct Loans by Billions due to Programmatic and Other Changes. It examines how and why the Department of Education's Direct Loan cost estimates changed over time for loans made in fiscal years 1997 through 2021. The report states that Education originally estimated these loans to generate $114 billion in income, while as of fiscal year 2021 they are estimated to cost the federal government $197 billion. It attributes the largest estimated cost increases, $102 billion in total, to emergency relief under the CARES Act and related administrative actions. GAO states that it is not making recommendations.
Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used
Cite this document
“GAO, Federal Student Loans: Education Could Improve Direct Loan Program Cost Estimates (GAO-22-105365, 2022-07).”
Original at www.gao.gov. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/gao-federal-student-loans-education-could-improve-direct-loan-program-d3c27836f4/
, SHA-256 859f2683634f6a324cb09e578376801cae275c444b08c50f3226781e0d962ccd.Full text
United States Government Accountability Office
Report to Congressional Requesters
STUDENT LOANS
July 2022
Education Has
Increased Federal
Cost Estimates of
Direct Loans by
Billions due to
Programmatic and
Other Changes
GAO-22-105365
July 2022
STUDENT LOANS
Education Has Increased Federal Cost Estimates of
Direct Loans by Billions due to Programmatic and
Highlights of GAO-22-105365, a report to
Other Changes
congressional requesters
Why GAO Did This Study What GAO Found
Over the last three decades, the Direct Although the Department of Education originally estimated federal Direct Loans
Loan program has grown in size and made in the last 25 years would generate billions in income for the federal
complexity, with almost $1.4 trillion in government, its current estimates show these loans will cost the government
outstanding federal student loans. The billions. Education originally estimated these loans to generate $114 billion in
Direct Loan program provides financial income for the government. Although actual costs cannot be known until the end
assistance to students and their of the loan terms, as of fiscal year 2021 these loans are estimated to cost the
parents to help pay for postsecondary federal government $197 billion. This swing of $311 billion was driven both by
education. GAO was asked to review programmatic changes and by reestimates using revised assumptions (e.g.,
changes in Education’s cost estimates
economic factors and loan performance) as additional data became available
and factors contributing to them.
(see figure).
This report examines how and why
Education’s Direct Loan cost estimates Original and Current (Fiscal Year 2021) Estimated Cost or Income for Direct Loans Made in
Fiscal Years 1997–2021
have changed over time. GAO
reviewed budget documents and data
covering Direct Loans made from fiscal
years 1997 through 2021. GAO also
conducted a model-based analysis on
a hypothetical group of borrowers
beginning repayment to demonstrate
how changing economic assumptions
can affect both repayment plan
selection and estimated loan
payments. Additionally, GAO
interviewed Education budget officials
about their process for estimating
student loan costs and how these
estimates are calculated and
documented.
A forthcoming report will examine The largest estimated cost increases—$102 billion in total—stemmed from
government and private sector emergency relief provided to most federal student loan borrowers under the
estimation methods and Education’s CARES Act and related administrative actions in response to the COVID-19
approach to estimating Direct Loan pandemic. This relief included suspending (1) all payments due, (2) interest
costs. accrual, and (3) involuntary collections for loans in default. The suspensions,
which are programmatic changes dating back to March 13, 2020, are currently
Education provided written comments set to expire on August 31, 2022. Reestimates based on updated data and
with additional context about some assumptions about borrowers in Income-Driven Repayment plans also
factors that contribute to its revised
substantially increased estimated costs.
estimates. GAO is not making
recommendations. Among the factors that make estimating the cost of Direct Loans difficult are the
lack of historical data when new programmatic changes are introduced, and
assumptions Education must make about borrower behavior over the life of the
loan. For example, the monthly payment amount for borrowers in Income-Driven
Repayment plans can change based on their economic situation. Using a
hypothetical group of borrowers, GAO found that borrowers’ income growth and
inflation, which are difficult to predict, affect borrowers’ payments. For example,
View GAO-22-105365. For more information, GAO found that when income grows at a slower rate, borrowers’ payments to the
contact Melissa Emrey-Arras at (617) 788-
0534 or emreyarrasm@gao.gov, Cheryl E.
government decrease, which increase government costs.
Clark at (202) 512-9377 or clarkce@gao.gov,
or Lawrance L. Evans, Jr. at (202) 512-4802 or
evansl@gao.gov.
United States Government Accountability Office
Contents
Letter 1
Background 3
Programmatic Changes and Updated Assumptions Contributed to
Substantial Increases in Cost Estimates 10
Agency Comments 18
Appendix I Objectives, Scope, and Methodology 20
Appendix II Additional Data on Direct Loan Estimated Costs 25
Appendix III Comments from the Department of Education 26
Appendix IV GAO Contacts and Staff Acknowledgments 28
Figures
Figure 1: Direct Loan Dollars in Income-Driven and Other
Repayment Plans, Third Quarter Fiscal Year 2013
through First Quarter Fiscal Year 2022 5
Figure 2: Calculation of Direct Loan Original Subsidy Costs 7
Figure 3: Original and Current (Fiscal Year 2021) Estimated Cost
or Income by Direct Student Loan Cohort, Fiscal Years
1997–2021 11
Figure 4: Increases in Budget Cost Estimates due to Direct
Student Loan Programmatic Changes (modifications),
Fiscal Years 2005–2021 13
Figure 5: Timeline and Estimated Direct Student Loan Costs of
COVID-19 Emergency Relief, Fiscal Years 2020–2022 14
Figure 6: Original and Current (fiscal year 2021) Estimated Direct
Student Loan Cost or Income per $100 by Loan Cohort,
Fiscal Years 1997–2021 25
Page i GAO-22-105365 Student Loan Cost Estimates
Abbreviations
CSC Credit Subsidy Calculator
Direct Loan William D. Ford Federal Direct Loan
FCRA Federal Credit Reform Act of 1990
IDR Income-Driven Repayment
PSLF Public Service Loan Forgiveness
TEPSLF Temporary Expanded Public Service Loan Forgiveness
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 ii GAO-22-105365 Student Loan Cost Estimates
Letter
441 G St. N.W.
Washington, DC 20548
July 28, 2022
The Honorable Richard Burr
Ranking Member
Committee on Health, Education, Labor and Pensions
United States Senate
The Honorable Virginia Foxx
Republican Leader
Committee on Education and Labor
House of Representatives
The Honorable Mike Braun
United States Senate
The Honorable Greg Murphy
House of Representatives
Over the last three decades, the William D. Ford Federal Direct Loan
(Direct Loan) program has grown in size and complexity with almost $1.4
trillion in outstanding federal student loans as of December 2021. Every
year the Department of Education estimates the lifetime costs of the
program for the President’s budget. 1 These estimates include costs for
new loans made and updates, or reestimates, for outstanding loans due
to changes in economic factors and actual loan performance, such as
how many borrowers defaulted. These cost estimates are also revised to
account for programmatic changes stemming from legislative or
administrative actions. 2
You asked us to examine these changes and the factors contributing to
them. This report examines how and why Education’s Direct Loan cost
1As required by the Federal Credit Reform Act of 1990 (FCRA), Direct Loan cost
estimates are included annually in the President’s budget. Cost estimates are calculated
based on the net present value of lifetime estimated cash flows to and from the
government associated with these loans. Direct Loan cash flows from the government
include loan disbursements to borrowers, while cash flows to the government include
repayments of loan principal, interest and fee payments, and recoveries on defaulted
loans.
2Programmatic changes as a result of federal government actions are referred to as
modifications.
Page 1 GAO-22-105365 Student Loan Cost Estimates
estimates have changed over time. A forthcoming report will examine
government and private sector estimation methods and Education’s
approach to estimating Direct Loan costs.
To answer how and why Education’s cost estimates have changed over
time, we examined Education’s cost estimates for Direct Loans made
during the most recent 25-year period (fiscal years 1997–2021) based on
available data and budget documents. 3 This includes identifying major
cost fluctuations and comparing those fluctuations to the timing of
programmatic changes and Education’s reestimates due to updated data
and assumptions. We reviewed relevant federal laws and regulations and
interviewed Education budget officials about their process for estimating
student loan costs and how these estimates are calculated and
documented.
We also conducted a model-based analysis on a hypothetical group of
undergraduate and graduate borrowers beginning repayment to
demonstrate how changing borrower characteristics and economic
assumptions such as income growth can both affect repayment plan
selection and estimated loan payments. The loan and income
characteristics of the hypothetical group of borrowers were based on
loans entering repayment in fiscal year 2017. We determined that the
budget data and the data used in our model-based analysis were
sufficiently reliable for our purposes. For more information see appendix I.
We conducted this performance audit from August 2021 through July
2022, 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.
3We reviewed Education’s underlying data for its annual submissions to the President’s
budget for each of the loan cohorts included in our review for fiscal years 1997 through
2021. We excluded data for the fiscal year 2022 cohort from our review because the
cohort is still disbursing and a cost reestimate will not be available until a later budget
cycle. A loan cohort includes all loans made in a given fiscal year, even if disbursements
occur in subsequent years.
Page 2 GAO-22-105365 Student Loan Cost Estimates
Background
Direct Loan Program and The Direct Loan program provides financial assistance to students and
Repayment Plans their parents to help pay for higher education. Under the Direct Loan
program, Education issues several types of student loans (see sidebar).
William D. Ford Federal Direct Loan Types Borrowers are responsible for repaying the loan, generally upon leaving
Subsidized Stafford Loans: Available to school. 4 A variety of repayment plans are available to eligible Direct Loan
undergraduate students with financial need borrowers, including Standard, Graduated, Extended, and several
(generally the difference between their cost of
attendance and a measure of their ability to Income-Driven Repayment (IDR) plans; borrowers may switch among
pay, known as expected family contribution). plans throughout the repayment term, depending on eligibility. Borrowers
Borrowers are not responsible for paying
interest on these loans while in school and
are automatically enrolled in the Standard plan if they do not choose
during certain periods of deferment, an option another option and generally make fixed monthly payments over a period
that allows eligible borrowers to temporarily of up to 10 years (or 10 to 30 years for consolidation loans). IDR is an
postpone loan payments.
umbrella term that describes five repayment plans available to Direct
Unsubsidized Stafford Loans: Available to
undergraduate and graduate school students Loan borrowers who meet specific eligibility requirements. Monthly
irrespective of financial need. Borrowers must payments for IDR plans are based on a borrower’s income and family
pay all interest on these loans.
size and extend the repayment period up to 20 or 25 years, depending on
PLUS Loans: Available to graduate student
borrowers and parents of dependent the plan. IDR plans also offer forgiveness of the loan’s balance at the end
undergraduates. Borrowers must pay all of the repayment period.
interest on these loans.
Consolidation Loans: Available to eligible
borrowers wanting to combine multiple federal The Direct Loan program’s terms and conditions have changed over time
student loans (including those listed above) with the addition of new repayment plans and forgiveness options, such
into one loan. Repayment periods are
extended up to a maximum of 30 years,
thereby lowering monthly payments.
Note: Interest rates for all loan types are set
by federal law and currently tied to the U.S.
Treasury’s 10-year note rate and can vary by
loan type. In addition, there are limits on the
annual and aggregate amounts that can be
borrowed for certain loan types.
Source: GAO summary of Department of Education
documents. | GAO-22-105365
4Borrowers are not required to make loan payments on Stafford loans and PLUS loans
made to graduate student borrowers when they are enrolled in school at least half-time or
during the grace period. The grace period usually ends 6 months after a borrower leaves
school or drops below half-time enrollment. Different repayment terms apply to PLUS
loans made to parent borrowers and consolidated loans.
Page 3 GAO-22-105365 Student Loan Cost Estimates
as Public Service Loan Forgiveness (PSLF). 5 Additionally, legislative and
administrative actions in response to the COVID-19 pandemic have
temporarily suspended loan payments for borrowers with Direct Loans
and other student loans owned by Education. For example, the CARES
Act suspended payments due, accrual of interest, and involuntary
collections on defaulted loans through September 30, 2020, for most
federal student loans. 6 This COVID-19 emergency relief for student loans
has been extended several times through administrative actions, most
recently through August 31, 2022. Although borrowers are not required to
make payments during this time, the months covered under the COVID-
19 emergency relief period count towards loan forgiveness and loan
rehabilitation. 7
The size of the Direct Loan program has also grown substantially over the
last decade, and about half of the loan volume is now being repaid
through IDR plans (see fig. 1). According to a February 2020 report by the
5The Public Service Loan Forgiveness (PSLF) program, established by statute in 2007, is
intended to encourage individuals to enter and continue in public service by forgiving the
remaining balances of Direct Loan borrowers who have made at least 10 years of loan
payments while working in qualifying public service jobs and meeting other requirements.
The Consolidated Appropriations Act, 2018 provided limited, additional conditions under
which borrowers may become eligible for loan forgiveness if some or all of the payments
made on Direct Loans were under a non-qualifying repayment plan, which is referred to as
the Temporary Expanded Public Service Loan Forgiveness (TEPSLF) opportunity. In
October 2021, Education announced a time-limited waiver so that borrowers can count
payments from all federal loan programs or repayment plans toward forgiveness, known
as the PSLF waiver.
6The CARES Act was enacted on March 27, 2020. See Pub. L. No. 116-136, § 3513, 134
Stat. 281, 404-05 (2020). Education implemented this COVID-19 emergency relief for
federal student loans retroactive to March 13, 2020, the date COVID-19 was declared a
national emergency. Involuntary collections may include wage garnishments and offsets of
tax refunds or federal benefit payments.
7Under the CARES Act and related administrative actions, most federal student loans
were placed in forbearance beginning March 13, 2020, with a current planned end date of
August 31, 2022. Unlike other periods of forbearance, interest does not accrue on the
loans during this period and the months without payments count as qualifying payments
towards IDR and PSLF forgiveness. Loan rehabilitation is an option for resolving defaulted
federal student loans, which allows borrowers who make nine on-time monthly payments
within 10 months to have the default removed from their credit reports. Borrowers may
rehabilitate a loan only once. During the COVID-19 emergency relief period, borrowers
who have entered into a rehabilitation agreement can get credit toward rehabilitating their
loans (for each month after the start of the rehabilitation agreement) even without making
a payment.
Page 4 GAO-22-105365 Student Loan Cost Estimates
Congressional Budget Office, borrowers who enroll in IDR plans tend to
borrow more and earn less than borrowers in fixed-payment plans. 8
Figure 1: Direct Loan Dollars in Income-Driven and Other Repayment Plans, Third
Quarter Fiscal Year 2013 through First Quarter Fiscal Year 2022
Note: Includes loan dollars in repayment, forbearance, and deferment. Does not include loans in
school, the grace period, and default.
8Congressional Budget Office, Income-Driven Repayment Plans for Student Loans:
Budgetary Costs and Policy Options (Washington, D.C.: February 2020).
Page 5 GAO-22-105365 Student Loan Cost Estimates
Cost Estimates and As required by the Federal Credit Reform Act of 1990 (FCRA), each year
Reestimates Education estimates the lifetime costs of a new cohort of Direct Loans for
inclusion in the President’s budget. For the Direct Loan program, these
Direct Loan Cost Estimation Key Terms costs—sometimes referred to as subsidy costs—represent the estimated
Assumptions: Estimates of the future cost to the government of extending credit over the life of the loan. These
operating and functional characteristics of a
loan or group of loans, such as estimates of
costs exclude administrative costs, such as expenses related to
loan maturity, default rate, and the impact of processing loan applications and servicing existing loans. Original
changes in economic factors, that are used to subsidy costs are calculated based on the net present value of estimated
forecast cash flows over time.
cash inflows and outflows (see side bar). Credit programs can result in
Cash flows: Payments to or from the
government over the life of a loan or group of costs or income for the government, depending on whether the present
loans. For direct loans these may include: value of estimated cash inflows (e.g., collections) exceeds the present
loan disbursements, repayments of principal,
payments of interest, and any other payments value of estimated cash outflows (e.g., loan disbursements). Figure 2
such as prepayments, fees, penalties, and illustrates the types of cash flows that affect the Direct Loan original
recoveries on defaulted loans. subsidy cost.
Discount rate: An interest rate that is used in
present value calculations to equate amounts
that will be received or paid in the future to
their value in today’s dollars.
Loan cohort: All loans of a loan program
made in a given fiscal year, even if
disbursements occur in subsequent years.
Modification: A programmatic change due to
federal government action, including new
legislation or administrative actions, that
directly or indirectly alters the estimated
subsidy cost of outstanding loans.
Net present value: The value today of a
future stream of cash flows, discounted using
an appropriate discount rate (generally the
average annual interest rate for marketable
zero-coupon U.S. Treasury securities with the
same maturity from the date of disbursements
as the cash flow being discounted).
Reestimate: Revision of estimated lifetime
costs for each loan cohort, incorporating
updated information on actual loan
performance and revised assumptions about
economic factors and future loan
performance. Reestimates are calculated
each year for the President’s budget and the
agency’s financial report. The sum of a
cohort’s original estimated subsidy cost and
subsequent annual reestimates is the cohort’s
lifetime cost.
Subsidy cost: The estimated long-term cost
to the government of extending credit,
calculated on a net present value basis and
excluding administrative costs.
Source: GAO summary of the 2023 Federal Credit
Supplement and Federal Accounting Standards Advisory
Board standards. | GAO-22-105365
Page 6 GAO-22-105365 Student Loan Cost Estimates
Figure 2: Calculation of Direct Loan Original Subsidy Costs
Education also annually updates, or reestimates, the cost of loans made
in prior years. 9 Reestimates take into account actual loan performance as
well as revised assumptions about economic factors and future loan
performance, such as how many borrowers are expected to default or
how many are expected to participate in different repayment plans.
Reestimates may result in increases or decreases in subsidy cost
estimates. No Direct Loan cohort has finished repaying all of its loans
9Education performs reestimates twice per year: once to be recorded in Education’s
financial statements, and once for presentation in the President’s budget. Unless
otherwise specified, this report presents the results of analysis of data supporting
reestimates for the President’s budget.
Page 7 GAO-22-105365 Student Loan Cost Estimates
since the start of the program in 1994. 10 Therefore, Education continues
to update estimates for all cohorts each year for the President’s budget
and actual costs will not be known until the end of the loan terms.
Education also updates its cost estimates to reflect any programmatic
changes that alter the estimated costs of outstanding loans as they occur,
referred to as modifications. 11
Factors that Can Change To estimate Direct Loan costs, Education developed a student loan cash
Direct Loan Cost flow model (the student loan model) that incorporates a variety of
assumptions about loan performance for the entire life of the loans, up to
Estimates
40 years in the future. These assumptions relate both to the program’s
terms and to various aspects of loan performance, such as how many
borrowers will prepay their loans and how many borrowers will default.
Education uses a supplementary model to estimate repayment patterns
for loans in IDR plans. Multiple factors make predicting actual loan
performance difficult, leading to changes from the original estimates:
• Programmatic changes. Changing terms and conditions of the Direct
Loan program due to legislative, regulatory, and policy changes make
costs difficult to estimate given the lack of data and uncertainty on
how these changes will affect loan performance. These programmatic
changes are captured as modifications.
• Data availability and limitations. Education continually updates data
to estimate future loan performance. However, limitations in available
data can introduce uncertainty into Education’s estimates. For
example, borrower income is a key variable for estimating the cost of
loans in IDR plans. For Education’s original cost estimates, Education
must estimate future borrower incomes for up to 40 years in the future
to determine borrower eligibility for the plan and repayment amount
based on limited historical income data. 12
• Loan characteristics and borrower behavior. Education makes
assumptions about borrowers’ loan characteristics and future
10Borrowers may not begin repaying their loan for several years after the loans are made,
and repayment periods can be up to 30 years, depending on the plan a borrower selects.
11While reestimates are funded with permanent indefinite budget authority, modifications
based on programmatic changes require new appropriations.
12We previously reported on the challenges associated with estimating borrowers’ future
income and the costs of IDR plans. GAO, Federal Student Loans: Education Needs to
Improve Its Income-Driven Repayment Plan Budget Estimates, GAO-17-22 (Washington,
D.C.: November 15, 2016).
Page 8 GAO-22-105365 Student Loan Cost Estimates
behavior, such as which repayment plan they will select and how
many and to what extent borrowers will pay early or default on their
loans and at what point in time.
• Timing. For the President’s budget, Education calculates the original
estimate before any loans are made. This requires Education to
estimate the types of loans borrowers will obtain, how much they will
borrow, and the riskiness of borrowers who will enter the program. 13
In addition, loans in a cohort may not enter repayment for several
years.
• Macroeconomic changes. Changes in macroeconomic conditions,
including interest rates, the unemployment rate, or wage growth, also
affect cost estimates. For example, the interest rates that borrowers
must pay on their loans are established in advance of the upcoming
school year. However, the interest rate Education must pay the
Treasury to fund those loans is not set until after the loans are 90
percent disbursed, which can be up to 18 months later. The difference
between the estimated and actual rates paid to Treasury can result in
additional subsidy costs or income depending on whether the actual
interest rate paid to Treasury is higher or lower than the estimated
rate.
Education is in the process of replacing its cohort-based student loan
model with a borrower-based microsimulation model to incorporate some
of the factors outlined above into its cost estimates. 14 According to
Education, the microsimulation model will allow it to provide more detailed
analysis of the factors affecting student loan costs, such as how long it
will take borrowers to pay off their loans and how many loans will enter
deferment and forbearance. Education said it expects to use this model
for the estimates in the President’s fiscal year 2026 budget.
13According to Education officials, the department has little to no influence over the
volume of higher-risk borrowers who enter the portfolio during certain parts of each
macroeconomic cycle.
14While Education’s current student loan model groups loan dollars into different
categories for which different assumptions are prepared, the microsimulation model will
make estimates based on assumptions about individual borrowers. According to
Education, this will allow for more flexibility to better address the complexity of the
program.
Page 9 GAO-22-105365 Student Loan Cost Estimates
Education’s Direct Loan budget cost estimates for loans made in the last
Programmatic 25 years are now $311 billion more than originally estimated due to
Changes and programmatic changes and reestimates based on updated assumptions
as additional loan performance data became available. Of this increase,
Updated Assumptions $122 billion (39 percent) is based on programmatic changes including the
Contributed to suspension of loan payments and interest due to the COVID-19
pandemic. The remaining $189 billion (61 percent) increase is due to
Substantial Increases reestimates based on actual data on how loans have performed, including
in Cost Estimates updated income data for borrowers in IDR plans.
Education Currently Although Education originally estimated federal Direct Loans made in the
Estimates That Direct last 25 years would generate income for the federal government, as of
fiscal year 2021, Education estimates these loans will cost the
Loans Made in the Last 25
government $197 billion. This estimate is the aggregated cost to the
Years Will Have Higher government of providing the approximately $1.8 trillion in loans disbursed
Costs to the Government from fiscal year 1997 to 2021. Education originally estimated that these
than Originally Estimated loans would generate $114 billion in income for the government, making
Education’s current cost estimates $311 billion higher than originally
estimated (see fig. 3). 15 According to Education officials, Direct Loan
cohorts from fiscal years 2009 through 2019 were originally expected to
generate income primarily because borrowers were charged higher
interest rates than Education’s cost of borrowing during that period. 16
15Estimated costs are in nominal dollars.
16In 2014, we found the borrower interest rate was greater than the government’s cost of
borrowing between fiscal years 2007 and 2012. In instances where the borrower rate is
greater than the government borrowing costs, Education would be expected to receive
more in interest payments from borrowers than what it pays in interest to Treasury,
increasing the likelihood that revenues will exceed costs for the loan. See GAO, Federal
Student Loans: Borrower Interest Rates Cannot Be Set in Advance to Precisely and
Consistently Balance Federal Revenues and Costs, GAO-14-234 (Washington, D.C.:
January 31, 2014).
Page 10 GAO-22-105365 Student Loan Cost Estimates
Figure 3: Original and Current (Fiscal Year 2021) Estimated Cost or Income by Direct Student Loan Cohort, Fiscal Years 1997–
2021
Note: A loan cohort represents all Direct Loans made in a given fiscal year. Estimated costs are in
nominal dollars.
The estimated cost for each cohort is comprised of two elements that can
increase or decrease total cost estimates: (1) the cost (or income)
Page 11 GAO-22-105365 Student Loan Cost Estimates
expected for every $100 lent based on the terms of the loan and
predicted loan performance (which allows for comparison irrespective of
the amount lent); and (2) the volume of loans made for each cohort.
Changes to both of these elements have contributed to increases in the
overall estimated cost. Specifically, in total, loans made from the fiscal
year 1997 through 2021 cohorts were originally estimated to generate $6
in income per every $100 disbursed. As of fiscal year 2021, those cohorts
are expected to cost the government almost $9 for every $100 disbursed
(see appendix II for additional details on the cost per $100 for each of the
last 25 loan cohorts). 17 The amount disbursed to each Direct Loan cohort
increased after the expansion of the Direct Loan program in 2010,
amplifying the effect of the cost increase for each $100 of loans. 18
Estimated Costs Have Cost increases since Education’s original estimates are driven by
Increased Primarily due to programmatic changes, including the suspension of loan payments and
interest due to the COVID-19 pandemic (which are reflected in
COVID-19 Emergency
modifications), and reestimates based on revised assumptions about
Relief and Revised borrowers in IDR plans. The specific programmatic changes and
Assumptions Related to reestimates are described in more detail below.
Income-Driven Repayment
Plans
Programmatic Changes Programmatic changes leading to Direct Loan cost estimate increases,
Resulted in Cost Increases including legislative and administrative actions, have occurred in 7 of the
last 25 fiscal years. Education reported that these changes contributed
$122 billion of the estimated $311 billion in total cost increases, ranging
from approximately $7 million to over $68 billion per change, at the time
they were implemented. Changes in cost estimates as a result of
programmatic changes are referred to as modifications. Modifications
resulting from COVID-19 emergency relief provided to borrowers under
the CARES Act and related administrative actions made up the largest
share of Direct Loan modifications. Modifications resulting from new IDR
plans, such as the Income-Based Repayment plan created in 2007 and
17The cost per $100 for all loans made from fiscal years 1997 through 2021 is weighted
based on each loan cohort’s total disbursements.
18The volume of loans issued under the Direct Loan program expanded dramatically after
the SAFRA Act terminated the authority to make or insure new Federal Family Education
Loans on June 30, 2010. Pub. L. No. 111-152, tit. II, § 2201, 124 Stat. 1029, 1074 (2010).
For example, the volume of loans more than doubled from the fiscal year 2009 to the fiscal
year 2010 cohort, from about $41 billion to almost $98 billion.
Page 12 GAO-22-105365 Student Loan Cost Estimates
the Revised Pay As You Earn plan in 2015, also resulted in increases to
program cost estimates (see fig. 4).
Figure 4: Increases in Budget Cost Estimates due to Direct Student Loan
Programmatic Changes (modifications), Fiscal Years 2005–2021
Note: Modifications are one-time federal government actions, including new legislation or
administrative actions that directly or indirectly alter the estimated subsidy cost of outstanding loans.
The amounts in this figure represent Education’s cost estimates associated with each programmatic
change for existing loan cohorts starting with the fiscal year 1997 cohort at the time each change was
implemented. Accordingly, they do not represent the full cost of the action and actual costs for a given
Page 13 GAO-22-105365 Student Loan Cost Estimates
cohort will not be known until the end of that cohort’s loan terms. For example, the $4 billion
programmatic change in 2008, which included creating the PSLF program, represents the estimated
cost for the fiscal year 1997–2008 cohorts at the time this change was implemented. The years in this
figure reflect the fiscal year in which each modification was recorded in the budget, and may not
reflect the year of the programmatic change. This figure includes modifications associated with
COVID-19 emergency relief for federal student loans as of the fiscal year 2021 Agency Financial
Report, which covered relief provided from March 13, 2020 to January 31, 2022.
As previously discussed, COVID-19 emergency relief for federal student
loans has been extended several times through administrative actions,
most recently through August 31, 2022. The cost of these extensions has
spanned multiple budget years, and the modification included in the
budget for fiscal year 2022 covered additional relief through May 1, 2022,
when the relief was previously set to expire. According to our review of
Education data, COVID-19 emergency relief provided from March 13,
2020 through April 30, 2022 was estimated to increase Direct Loan costs
to the federal government by about $102 billion (see fig. 5). 19
Figure 5: Timeline and Estimated Direct Student Loan Costs of COVID-19 Emergency Relief, Fiscal Years 2020–2022
Note: In response to the COVID-19 pandemic, on March 27, 2020, the CARES Act was enacted,
which suspended federal student loan payments due, interest accrual, and involuntary collection for
most federal student loans through September 30, 2020. The Department of Education implemented
this COVID-19 emergency relief for student loans retroactively to March 13, 2020, the date COVID-19
was declared a national emergency. The relief has been extended several times through
administrative actions, and most recently has been extended until August 31, 2022. These amounts
represent estimated cost increases for all existing loan cohorts (since fiscal year 1994). Education
makes predictions about the estimated costs associated with programmatic changes affecting these
19As of May 2022, Education had not developed cost estimates related to the most recent
extension of COVID-19 emergency relief set to expire on August 31, 2022. Education
provided data on COVID-19 modifications for the entire Direct Loan portfolio. As a result,
the COVID-19 modification amounts include updated costs for the earlier loan cohorts
(fiscal years 1994–1996). These earlier cohorts represent $15 billion in loan volume, less
than one percent of the $1.8 trillion total Direct Loans disbursed.
Page 14 GAO-22-105365 Student Loan Cost Estimates
loans as these actions occur. Accordingly, they do not represent the full cost of the action and actual
costs for a given cohort will not be known until the end of that cohort’s loan terms.
Education estimates the impact of programmatic changes on Direct Loan
cash flows and costs at the time these changes are implemented. As
previously discussed, these costs are inherently difficult to estimate
because there is often no existing data on which Education can base its
estimates. Therefore, as Education updates its assumptions based on
actual loan performance in subsequent years, continuing effects of these
programmatic changes are reflected in reestimates. For example,
Education originally estimated in 2008 that programmatic changes, which
included creating the PSLF program, would increase Direct Loan subsidy
costs by $4 billion for cohorts that existed at that time. 20 In subsequent
years, actual data on how many borrowers have had their loans forgiven
through PSLF has been incorporated into more recent cost estimates.
Although the total cost of PSLF will not be known until all eligible
borrowers receive forgiveness, $9.5 billion in loan debt was forgiven
through PSLF, Temporary Expanded PSLF (TEPSLF), and the recent
PSLF waiver as of May 2022. 21
Reestimates due to Updated Education is required to reestimate its costs annually for the budget. As
Data and Assumptions previously discussed, multiple factors make estimating actual loan
Contributed to Cost Increases performance difficult, such as data limitations and the need to estimate
future borrower behavior. Reestimates based on updates to the
assumptions underlying Education’s student loan model that account for
actual loan performance, changes in expected future performance, and
20This amount includes the estimated cost increase associated with creating the PSLF
program and does not include costs associated with Temporary Expanded PSLF
(TEPSLF) or the PSLF waiver, which did not exist at the time. In addition, this amount
represents the estimated cost increase for loan cohorts starting in fiscal year 1997 and
does not include estimated increases for the fiscal year 1994–1996 cohorts.
21This amount represents the total amount forgiven as of May 2022 across all loan cohorts
since fiscal year 1994.
Page 15 GAO-22-105365 Student Loan Cost Estimates
economic changes—have increased Direct Loan costs by about $189
billion for loans made since fiscal year 1997. 22
According to data provided by Education, the three categories that
accounted for the largest increases in cost estimates since fiscal year
2013 were (1) updated assumptions about borrowers’ repayment plan
selection, (2) changes related to the underlying data and estimated
income growth for borrowers in IDR plans, and (3) updated assumptions
about the probability that borrowers will default on their loans. 23 For
example, IDR-specific changes have been the largest component of
Education’s reestimates in the last two years, largely driven by
Education’s efforts to incorporate updated income data into its estimates.
These updates decreased borrowers’ projected future income in
Education’s model to better match actual data, and lowered projected
payments borrowers would make in IDR plans. In contrast, assumption
changes related to discount rates and loan maturity have decreased cost
estimates over that time. 24
Student loan costs are difficult to predict in part because the monthly
payment amount for borrowers enrolled in an IDR plan can change based
on their economic situation. Borrowers’ expectations about the future may
also affect participation in IDR plans. 25 Specifically, our analysis shows
expected income growth may affect borrowers’ repayment plan selection
22Of the $189 billion, $34.4 billion represents interest related to the reestimates. Interest
related to the reestimates is the amount of interest that would have been earned or paid
on the subsidy reestimate, if the information had been included as part of the original
subsidy estimate. Education also noted changes in the portfolio mix, such as the types of
institutions and borrowers, can affect costs. For example, according to Education officials,
the department could not have predicted the Great Recession of 2007-2009, which
temporarily drew many more high-risk borrowers into the loan program.
23We examined available budget information going back to fiscal year 2013 that outlined
the specific assumption updates that contributed to budget reestimates since fiscal year
2013. Prior to fiscal year 2013, Education recorded these data in different formats that are
not comparable. The three categories accounted for the following estimated cost
increases since fiscal year 2013: predictions about repayment plan selection (about $70
billion); changes related to the underlying data for borrowers in IDR plans (about $68
billion); and assumptions about the probability of borrowers defaulting on their loans
(about $23 billion).
24Since fiscal year 2013 assumptions about discount rates decreased estimated costs by
about $63 billion and assumptions about maturity rates, or the average time it takes
borrowers to pay off loans, decreased costs by about $36 billion.
25Education officials also noted that borrowers who expect to qualify for PSLF, which
provides forgiveness of remaining balances after 10 years of qualifying payments, may be
more likely to choose an IDR plan.
Page 16 GAO-22-105365 Student Loan Cost Estimates
and income growth and inflation affect borrowers’ payments (see text
box).
How might income growth affect Income Driven Repayment (IDR) enrollment and program costs?
Under IDR plans, because a borrower’s payments are directly tied to their income, income growth over time determines a
borrower’s total payments over the term of the loan. Therefore, borrowers’ expectations about how their incomes might grow
in the future and whether they prioritize lower total payments over lower monthly payments might also affect their plan
choice.
Using a hypothetical group of undergraduate and graduate IDR participants, we found that if borrowers expected higher
incomes in the future and prioritized lower total payments, borrowers might be more likely to choose the standard repayment
plan. Specifically, among undergraduate borrowers, increasing the expected rate of income growth from 3 percent to 7
percent would change the percentage of borrowers for whom the standard plan is the lowest cost option from 15 percent to
24 percent. Among graduate borrowers, the percentage would change from 20 percent to 37 percent. This may indicate that
changes in expectations of income could affect a borrower’s choice of repayment plan. Importantly, though, the actual extent
to which borrowers would choose different repayment plans would be determined by how individual borrowers make
decisions (as well as borrowers’ individual circumstances), which were not incorporated into this analysis.
In addition, IDR costs are sensitive to macroeconomic changes, such as borrowers’ income growth and inflation, which are
difficult to predict. Using a hypothetical group of borrowers beginning repayment in Standard and IDR plans, we found that
decreases in the rate of income growth tended to decrease these borrowers’ payments to the government, and increase the
cost to the government. For example, for graduate borrowers, decreasing the rate of income growth from 5 percent to 3
percent increased the government’s costs from $10 to $19 per $100 of the borrowers’ loan balance. Similarly, increasing the
rate of inflation but holding the income growth constant at 5 percent increased costs from $10 to $14 per $100 for graduate
loans (see below).
Sensitivity of Estimated Costs to Changes in Assumptions about Borrowers’ Income Growth and Inflation
Source: GAO analysis of Department of Education data. | GAO-22-105365
Page 17 GAO-22-105365 Student Loan Cost Estimates
We provided a draft of the report to Education for review and comment. In
Agency Comments its comments, reproduced in appendix III, Education stated that
consistent with federal requirements, the department regularly estimates,
modifies, and reestimates the costs of the Direct Loan program.
Education provided context about some factors that contribute to revised
estimates discussed in the report, including programmatic changes and
changes to available data and the department’s methodology for
estimating costs. For example, Education said that it has made
improvements to its income modeling for IDR plans based in part on
income data that borrowers in these plans provide. Education also stated
there is some inherent uncertainty in its student loan cost estimates due
to factors, such as changing interest rates and economic conditions.
Education also provided additional context about the Direct Loan
program, stating that the program has provided millions of students the
opportunity to enroll in education beyond high school and earn a
credential, including many low-income students. In addition, Education
stated its commitment to ensuring borrowers have access to fair and
affordable repayment plans, and highlighted the role of IDR plans and the
pandemic payment pause in helping borrowers avoid default and
delinquency. Education also provided technical comments, which we
incorporated as appropriate.
As agreed with your offices, 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 the
appropriate congressional committees, the Secretary of Education, and
other interested parties. In addition, the report will be available at no
charge on GAO’s website at http://www.gao.gov.
Page 18 GAO-22-105365 Student Loan Cost Estimates
If you or your staff have any questions about this report, please contact
Melissa Emrey-Arras at (617) 788-0534 or emreyarrasm@gao.gov,
Cheryl E. Clark at (202) 512-9377 or clarkce@gao.gov, or Lawrance L.
Evans, Jr. at (202) 512-4802 or evansl@gao.gov. Contact points for our
Offices of Congressional Relations and Public Affairs may be found on
the last page of this report. GAO staff who made key contributions to this
report are listed in appendix IV.
Melissa Emrey-Arras
Director, Education, Workforce, and Income Security Issues
Cheryl E. Clark
Director, Financial Management and Assurance
Lawrance L. Evans, Jr.
Managing Director, Applied Research and Methods
Page 19 GAO-22-105365 Student Loan Cost Estimates
Appendix I: Objectives, Scope, and
Appendix I: Objectives, Scope, and
Methodology
Methodology
This appendix discusses in detail our methodology for addressing how
and why the Department of Education’s William D. Ford Federal Direct
Loan (Direct Loan) cost estimates have changed over time.
To analyze how Education’s cost estimates for Direct Loans made during
Cost Trends the most recent 25-year period (fiscal years 1997–2021) have changed
over time (original and current estimates), we reviewed Education’s
output files from the Credit Subsidy Calculator (CSC) submitted for the
President’s 2023 budget. 1 We also reviewed Education’s annual Budget
Justifications from fiscal years 1997 through 2023 for qualitative
information on the assumption changes that made up its upward and
downward reestimates during this period. Additionally, we reviewed
Education’s Agency Financial Reports from fiscal years 1997 through
2021 for information about the total estimated cost and the specific
programmatic changes that made up its Direct Loan cost modifications
during that period.
We also reviewed supplemental unpublished data from Education on the
total reestimated subsidy rates for each Direct Loan cohort from fiscal
years 1997 through 2021 from CSC output files for the President’s budget
reestimates, and on the individual components that made up its
reestimates (both upwards and downwards) for all Direct Loan cohorts
from fiscal years 2013 through 2021. We compared the supplemental
unpublished data to published data from the President’s budget credit
supplement for fiscal years 2022 and 2023 that included loans made
through fiscal year 2021 and interviewed Education officials to clarify
reasons for minor discrepancies. We determined the data was reliable for
our purposes. We also interviewed Education budget officials to discuss
their process for estimating student loan costs and how these estimates
are calculated and documented.
We also conducted a model-based analysis on a hypothetical group of
Model-Based 1,000 undergraduate borrowers and 1,000 graduate borrowers to
Analysis demonstrate how changing borrower characteristics and economic
assumptions such as income growth can both affect repayment plan
1The CSC is an Office of Management and Budget tool agencies use to calculate the net
present value of estimated cash flows in order to produce the official credit subsidy rate for
the President’s budget. According to Office of Management and Budget Circular A-11, all
federal agencies must use the CSC and associated discount rates to ensure government-
wide comparability and uniformity of discounting.
Page 20 GAO-22-105365 Student Loan Cost Estimates
Appendix I: Objectives, Scope, and
Methodology
selection and estimated loan payments. 2 For each borrower, we
simulated loan payments under borrower related assumptions, such as
different expectations of income growth. We determined how the mix of
repayment plans would change if borrowers chose plans to minimize loan
payments. In addition, given the current mix of repayment plans, we
determined how loan payments would change based on economic factors
like changes in income growth or inflation. The loan and income
characteristics of the borrowers were based on loans entering repayment
in fiscal year 2017.
Development of the Model In order to test the sensitivity of program costs to assumptions about
income growth and other factors, we developed a hypothetical group of
student loans for the borrowers in our model. Then, we altered
assumptions about borrower characteristics, such as growth in income
and inflation. From this, we estimated changes in resulting loan
repayments based on the terms of the repayment plans. In order to
present the results in terms similar to a subsidy cost, we expressed them
as the difference between the original balance and the expected present
value of future payments, divided by the original balance. We referred to
this as the “subsidy cost per $100 dollars in loans.”
We developed repayment estimates by simulating total payments for
borrowers under the terms of the Income-Based Repayment, Pay As You
Earn, Revised Pay As You Earn, and 10-year Standard plans. In addition,
we made the following assumptions in the model-based analysis:
• Consistent with prior GAO reports, each borrower is single with no
dependents. As a result, our analysis did not take into account the
effect of family formation on income or the relevant poverty threshold
for that borrower.
• The poverty threshold will increase annually at the rate of inflation.
The economic assumptions underlying the fiscal year 2023
President’s budget were used to obtain expectations of future
inflation.
• A discount rate of 2 percent, based on the interest rate associated
with 10-year Treasury Notes in calendar year 2022, rounded to the
nearest digit.
2We performed separate analyses on the hypothetical graduate and undergraduate
borrowers. While we performed these analyses on an equal number of graduate and
undergraduate borrowers, they do not reflect the actual proportion of these groups in the
Direct Loan portfolio.
Page 21 GAO-22-105365 Student Loan Cost Estimates
Appendix I: Objectives, Scope, and
Methodology
• In our baseline model, income growth is assumed to be 5 percent.
• All Direct Loans are unsubsidized. This assumption means we might
understate total loan costs for Income-Based Repayment and Pay As
You Earn borrowers with subsidized loans whose payments do not
fully cover interest. In prior reports, Education officials have told us
this assumption would have only a slight effect on total loan costs. 3
• The terms of Income-Based Repayment plan would be based on the
terms set for income-eligible borrowers who received Direct Loans on
or after July 1, 2014. 4 We assumed that for borrowers in the Standard
repayment plan, loans would have 10-year terms.
• Besides the potential for forgiveness on Income-Driven Repayment
(IDR) plans, borrowers could not benefit from other forms of
forgiveness, like Public Service Loan Forgiveness.
• We did not incorporate default or whether borrowers might prepay a
loan into the analysis.
Changing the assumptions described above would change the total loan
costs for borrowers. These assumptions incorporate experiences that
could affect individual borrowers’ eligibility for IDR plans or their payment
amounts. For example, individual borrowers could experience periodic
unemployment or job promotions, get married, or form families. These
and other experiences could change income levels or household size,
which would affect loan payments and Direct Loan costs. In addition, we
did not consider the likelihood of changes in income or inflation, as they
were hypothetical scenarios. Moreover, changes in inflation would likely
affect income growth, as workers demand higher wages in response to
higher prices. This was not incorporated into our estimate of the
sensitivity of costs to changes in inflation.
3For example, see Federal Student Loans: Education Could Do More to Help Ensure
Borrowers are Aware of Repayment and Forgiveness Options, GAO-15-663 (Washington,
D.C.: August 25, 2015).
4There are two versions of the Income-Based Repayment plan. Because the data we
obtained from Education did not differentiate which version of the plan borrowers were in,
we assumed borrowers were in the updated version of the plan for borrowers who
received Direct Loans on or after July 1, 2014. The updated version of the plan limits
payments to 10 percent of discretionary income and offers loan forgiveness after 20 years
of repayment. In contrast, the original version of the Income-Based Repayment plan limits
payments to 15 percent of discretionary income and offers loan forgiveness after 25 years.
Page 22 GAO-22-105365 Student Loan Cost Estimates
Appendix I: Objectives, Scope, and
Methodology
Development of the To ensure our analysis would be informative, we attempted to calibrate
Hypothetical Group of our hypothetical group of borrowers to distributions of loan size, income,
and interest rates of an actual recent cohort of borrowers entering
Borrowers
repayment. Specifically, we developed a group of 1,000 graduate and
1,000 undergraduate borrowers with loan size, initial income, and interest
rates of loans. To develop the hypothetical group, we obtained data on
borrowers’ loan size, income, and interest rates from Education.
Education provided information on a recent cohort of borrowers entering
repayment based on the 2020 four percent sample from the National
Student Loan Data System, Education’s central database for federal
student aid information. Specifically, Education provided information on
the joint distribution of loan size and income that represented
approximately 170,000 undergraduate and 40,000 graduate borrowers
entering repayment in 2017. 5 According to Education, 2017 is the latest
year that these data could be provided. Moreover, income information in
the National Student Loan Data System is only available for individuals
participating in IDR plans. To maximize the coverage of the income
variable, Education included income reported in 2017 through 2018.
To assess the reliability and representativeness of the information
provided by Education, we estimated the extent to which the income data
was missing for specific IDR plans (Income-Based Repayment, Revised
Pay As You Earn, and Pay As You Earn). In addition, we tested the
degree to which missing income data were associated with the size of the
loans. 6 We found that income data was missing in 5 percent of cases for
both undergraduate borrowers and graduate borrowers.
We determined that the data were sufficiently reliable for our purposes of
developing a hypothetical group of income and loan size for borrowers in
IDR plans. First, the data was drawn from the National Student Loan Data
System, Education’s primary source for information on student loans.
Second, although there were some missing values, there was only a
weak correlation between loan size and income. Therefore, the missing
incomes would not cause our loan size estimates to be biased.
We conducted this performance audit from August 2021 through July
2022, in accordance with generally accepted government auditing
5Income was based on the “total income” variable in the National Student Loan Data
System.
6Loan size was not missing in the data.
Page 23 GAO-22-105365 Student Loan Cost Estimates
Appendix I: Objectives, Scope, and
Methodology
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.
Page 24 GAO-22-105365 Student Loan Cost Estimates
Appendix II: Additional Data on Direct Loan
Appendix II: Additional Data on Direct Loan
Estimated Costs
Estimated Costs
Figure 6: Original and Current (fiscal year 2021) Estimated Direct Student Loan Cost or Income per $100 by Loan Cohort,
Fiscal Years 1997–2021
Note: Estimated costs are in nominal dollars.
Page 25 GAO-22-105365 Student Loan Cost Estimates
Appendix III: Comments from the
Appendix III: Comments from the Department
of Education
Department of Education
Page 26 GAO-22-105365 Student Loan Cost Estimates
Appendix III: Comments from the Department
of Education
Page 27 GAO-22-105365 Student Loan Cost Estimates
Appendix IV: GAO Contacts and Staff
Appendix IV: GAO Contacts and Staff
Acknowledgments
Acknowledgments
Melissa Emrey-Arras, (617) 788-0534 or emreyarrasm@gao.gov
GAO Contacts
Cheryl E. Clark, (202) 512-9377 or clarkce@gao.gov
Lawrance L. Evans, Jr., (202) 512-4802 or evansl@gao.gov
In addition to the contacts named above, Benjamin Bolitzer, Marcia
Staff Carlsen, Debra Prescott (Assistant Directors), Rachel Beers (Analyst in
Acknowledgments Charge), Robert Dacey, Benjamin Netto DeYoung, Daniel Flavin, Marissa
Jones Friedman, Gina Hoover, Christopher Klemmer, and Briana Shen
made key contributions to this report. Additional assistance was provided
by James Bennett, Elizabeth Calderon, Holly Dye, Susan J. Irving, Brittni
Milam, Frank Todisco, and Adam Wendel.
(105365)
Page 28 GAO-22-105365 Student Loan Cost Estimates
The Government Accountability Office, the audit, evaluation, and investigative
GAO’s Mission 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
Obtaining Copies of through our website. Each weekday afternoon, GAO posts on its website newly
GAO Reports and released reports, testimony, and correspondence. You can also subscribe to
GAO’s email updates to receive notification of newly posted products.
Testimony
Order by Phone 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, https://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.
Connect with GAO Subscribe to our RSS Feeds or Email Updates. Listen to our Podcasts.
Visit GAO on the web at https://www.gao.gov.
Contact FraudNet:
To Report Fraud,
Website: https://www.gao.gov/about/what-gao-does/fraudnet
Waste, and Abuse in
Automated answering system: (800) 424-5454 or (202) 512-7700
Federal Programs
A. Nicole Clowers, Managing Director, ClowersA@gao.gov, (202) 512-4400, U.S.
Congressional Government Accountability Office, 441 G Street NW, Room 7125, Washington,
Relations DC 20548
Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800
Public Affairs U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, DC 20548
Stephen J. Sanford, Managing Director, spel@gao.gov, (202) 512-4707
Strategic Planning and U.S. Government Accountability Office, 441 G Street NW, Room 7814,
External Liaison Washington, DC 20548
Please Print on Recycled Paper.
File and source
- File
- gao-22-105365.pdf
- Size
- 1,693,241 bytes
- SHA-256
- 859f2683634f6a324cb09e578376801cae275c444b08c50f3226781e0d962ccd
- Our copy
- gao-22-105365.pdf
- Original
- www.gao.gov