Student Aid And Postsecondary Tax Preferences: Limited Research  
Exists on Effectiveness of Tools to Assist Students and Families 
through Title IV Student Aid and Tax Preferences (29-JUL-05,	 
GAO-05-684).							 
                                                                 
Federal assistance helps students and families pay for		 
postsecondary education through several policy tools--grant and  
loan programs authorized by title IV of the Higher Education Act 
of 1965 and more recently enacted tax preferences. In fiscal year
2004, about $14 billion in grants and $56 billion in loans were  
made under title IV while estimated outlay equivalents for	 
postsecondary tax preferences amounted to $10 billion. In light  
of the relative newness and financial significance of tax	 
preferences, this report examines (1) how title IV assistance	 
compares to that provided through the tax code, (2) the extent to
which tax filers effectively use postsecondary tax preferences,  
and (3) what is known about the effectiveness of federal	 
assistance.							 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-05-684 					        
    ACCNO:   A31561						        
  TITLE:     Student Aid And Postsecondary Tax Preferences: Limited   
Research Exists on Effectiveness of Tools to Assist Students and 
Families through Title IV Student Aid and Tax Preferences	 
     DATE:   07/29/2005 
  SUBJECT:   Aid for education					 
	     Federal aid programs				 
	     Federal grants					 
	     Federal taxes					 
	     Higher education					 
	     Income taxes					 
	     Student financial aid				 
	     Student loans					 
	     Tax administration 				 
	     Tax credit 					 
	     Federal Family Education Loan Program		 
	     National Postsecondary Student Aid Study		 
	     Pell Grant 					 
	     William D. Ford Federal Direct Loan		 
	     Program						 
                                                                 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-05-684

     

     * Report to the Committee on Finance, U.S. Senate
          * July 2005
     * STUDENT AID AND POSTSECONDARY TAX PREFERENCES
          * Limited Research Exists on Effectiveness of Tools to Assist
            Students and Families through Title IV Student Aid and Tax
            Preferences
     * Contents
          * Results in Brief
          * Background
               * Federal Grant and Loan Assistance to Postsecondary Students
               * Tax Preferences
               * Tax Preferences Are More Recent Than Most Title IV Programs
          * Tax Preferences Differ from Title IV Assistance in Timing,
            Distribution, and Students' and Families' Responsibility for
            Obtaining Benefits
               * Tax and Title IV Programs Differ in Benefit Timing
               * Beneficiaries of Title IV Programs and Tax Preferences
                 Differ
                    * Title IV Assistance
                    * Tax Preferences
               * Students and Families Have More Responsibility for Obtaining
                 Benefits of Tax Preferences in Comparison to Title IV Aid
          * Some Tax Filers May Not Effectively Use Postsecondary Tax
            Preferences, Possibly Due to Complexity
               * Some Tax Filers Appear to Make Suboptimal Choices
               * The Suboptimal Use of Postsecondary Tax Preferences May
                 Result from Their Complexity
                    * Multiple Tax Preferences
                    * Varying Qualified Expense Definitions
               * Tax Rules Coordinating the Use of Multiple Preferences
                    * Interaction between Tax Preferences and Student
                      Financial Aid
          * Research on Effectiveness of Federal Postsecondary Assistance Is
            Incomplete
               * Little Is Known about the Effectiveness of Federal Title IV
                 Postsecondary Programs and Tax Preferences
               * Limited Research about Federal Assistance Reaches Varying
                 Conclusions about Some Aspects of Effectiveness
               * Data and Methodological Challenges Hinder Research
               * Steps to Support Effectiveness Research
          * Concluding Observations
          * Agency Comments and Our Evaluation
     * Objectives, Scope, and Methodology
     * Confidence Intervals
     * Postsecondary-Education-Related Tax Preferences
     * Comments from the Department of Education
     * Bibliography
     * cover1errata.pdf
          * Report to the Committee on Finance, U.S. Senate
               * July 2005
          * STUDENT AID AND POSTSECONDARY TAX PREFERENCES
               * Limited Research Exists on Effectiveness of Tools to Assist
                 Students and Families through Title IV Student Aid and Tax
                 Preferences
     * cover1errata.pdf
          * STUDENT AID AND POSTSECONDARY TAX PREFERENCES
               * Limited Research Exists on Effectiveness of Tools to Assist
                 Students and Families through Title IV Student Aid and Tax
                 Preferences

                 United States Government Accountability Office

Report to the Committee on Finance, U.S. Senate

July 2005

STUDENT AID AND POSTSECONDARY TAX PREFERENCES

    Limited Research Exists on Effectiveness of Tools to Assist Students and
           Families through Title IV Student Aid and Tax Preferences

This report was amended on November 17, 2005 to remove estimated median
income data for households with Section 529 Qualified Tuition Program
accounts or Coverdell Education Savings Accounts, due to a sample
selection bias in the private sector research report that was our source.
Median income estimates were removed from Table 2 and the text on page 18.

                                       A

GAO-05-684

STUDENT AID AND POSTSECONDARY TAX PREFERENCES

Limited Research Exists on Effectiveness of Tools to Assist Students and
Families through Title IV Student Aid and Tax Preferences

  What GAO Found

Title IV student aid and tax preferences provide assistance to a wide
range of students and families in different ways. While both help students
meet current expenses, tax preferences also assist students and families
with saving for and repaying postsecondary costs. While both serve
students and families with a range of incomes, some forms of title IV
aid-grant aid, in particular-provide assistance to those whose incomes are
lower, on average, than is the case with tax preferences. While both
require students and families to fill out forms, tax preferences require
more responsibility on the part of students and families because they must
identify applicable tax preferences, understand complex rules concerning
their use, and correctly calculate and claim credits or deductions. While
the tax preferences are a newer policy tool, the number of tax filers
using them has grown quickly, surpassing the number of students aided
under title IV in 2002.

Recipients of Title IV Assistance and Tax Filers Claiming an Education Tax
Credit or Tuition Deduction, 1997-2002

Some tax filers do not appear to make optimal education-related tax
decisions. For example, among the limited number of tax returns available
for our analysis, 27 percent of eligible tax filers did not claim either
the tuition deduction or a tax credit. In so doing, these tax filers
failed to reduce their tax liability by $169, on average, and 10 percent
of these filers could have reduced their tax liability by over $500. One
explanation for these taxpayers' choices may be the complexity of
postsecondary tax provisions, which experts have commonly identified as
difficult for tax filers to use.

Little is known about the effectiveness of title IV aid or tax preferences
in promoting, for example, postsecondary attendance or choice, in part
because of research data and methodological challenges. As a result,
policymakers do not have information that would allow them to make the
most efficient use of limited federal resources to help students and
families.

                 United States Government Accountability Office

Contents

  Letter 1

Results in Brief 3 Background 5 Tax Preferences Differ from Title IV
Assistance in Timing,

Distribution, and Students' and Families' Responsibility for

Obtaining Benefits 14 Some Tax Filers May Not Effectively Use
Postsecondary Tax

Preferences, Possibly Due to Complexity 20 Research on Effectiveness of
Federal Postsecondary Assistance Is

Incomplete 27 Concluding Observations 34 Agency Comments and Our
Evaluation 34

  Appendixes

    Appendix I: Objectives, Scope, and Methodology 37 Appendix II: Confidence
Intervals 41 Appendix III: Postsecondary-Education-Related Tax Preferences
                 51 Appendix IV: Comments from the Department of Education 54

  Bibliography

Table 1: Description of Federal Student Aid Programs Authorized

  Tables

under Title IV of the Higher Education Act 7 Table 2: Selected
Postsecondary Education Tax Preferences 10 Table 3: Comparison of
Assistance by Timing of Benefit for

Selected Programs and Tax Preferences 15 Table 4: Percentage of Aid
Recipients and Dollars of Aid by Income

Category for Dependent Students Served by Selected Title

IV Programs 16 Table 5: Percentage of Aid Recipients and Dollars of Aid by
Income

Category for Independent Students Served by Selected

Title IV Programs 17 Table 6: Percentage of Tax Filers Claiming Hope and
Lifetime

Learning Credits and Tuition Deduction and Tax

Preference Dollars by Income Category, Tax Year 2002 18 Table 7:
Description of Federal Student Aid Programs Authorized

under Title IV of the Higher Education Act 41 Table 8: Selected
Postsecondary Education Tax Preferences 42

                               Contents Contents

Table 9:            Tax Filers Claiming an Education Tax Credit or Tuition
                       Deduction                                           42 
Table 10: Percentage of Aid Recipients and Dollars of Aid by   
Income                                                         
         Category for Dependent Students Served by Selected Title 
IV Programs                                                             42 
Table 11: Percentage of Aid Recipients and Dollars of Aid by   
Income                                                         
Category for Independent Students Served by Selected           
Title IV Programs                                                       43 
Table 12: Percentage of Tax Filers Claiming Hope and Lifetime  
           Learning Credits and Tuition Deduction and Tax         
Preference Dollars by Income Category, Tax Year 2002                    43 
Table 13: Percentage of Form 1098-Ts with Postsecondary        
Expense                                                        
Information in 2002: Point Estimates                                    44 
Table 14: Percentage of Form 1098-Ts with Postsecondary        
Expense                                                        
Information in 2002: Confidence Intervals                               44 
Table 15: Percentage of Taxpayers Apparently Eligible to Claim 
an                                                             
         Education Tax Credit or Tuition Deduction in 2002: Point 
Estimates                                                               44 
Table 16: Percentage of Taxpayers Apparently Eligible to Claim 
an                                                             
         Education Tax Credit or Tuition Deduction in 2002:       
Confidence Intervals                                                    44 
Table 17: Percentage of Apparently Eligible Taxpayers to Claim 
an                                                             
         Education Tax Credit or Tuition Deduction That Failed to 
Do So in 2002: Point Estimates                                          45 
Table 18: Percentage of Apparently Eligible Taxpayers to Claim 
an                                                             
         Education Tax Credit or Tuition Deduction That Failed to 
Do So in 2002: Confidence Intervals                                     45 
Table 19: Amounts by Which Apparently Eligible Taxpayers       
Failed to                                                      
            Reduce Their Tax Liability: Point Estimates                    45 
Table 20: Amounts by Which Apparently Eligible Taxpayers       
Failed to                                                      
          Reduce Their Tax Liability: Confidence Intervals                 46 
Table 21: Percentage of Apparently Eligible Taxpayers That     
Claimed                                                        
the Tuition Deduction but Would Have Been Better off           
        Claiming the Lifetime Learning Credit in 2002: Point      
Estimates                                                               46 
Table 22: Percentage of Apparently Eligible Taxpayers That     
Claimed                                                        
the Tuition Deduction but Would Have Been Better off           
        Claiming the Lifetime Learning Credit in 2002: Confidence 
Intervals                                                               46 
Table 23: Amounts by Which Apparently Eligible Taxpayers Could 
          Have Reduced Their Tax Liability in 2002: Point         
Estimates                                                               47 

               Table 24: Amounts by Which Apparently Eligible Taxpayers Could
                         Have Reduced Their Tax Liability in 2002: Confidence
                                  Intervals                                47 
        Table 25: Percentage of Apparently Eligible Taxpayers That Claimed 
                   the Lifetime Learning Credit but Would Have Been Better 
                         off Claiming the Tuition Deduction in 2002: Point 
                                  Estimates                                48 
        Table 26: Percentage of Apparently Eligible Taxpayers That Claimed 
                   the Lifetime Learning Credit but Would Have Been Better 
                    off Claiming the Tuition Deduction in 2002: Confidence 
                                  Intervals                                48 
            Table 27: Amounts by Which Apparently Eligible Taxpayers Could 
                           Have Reduced Their Tax Liability in 2002: Point 
                                  Estimates                                48 
            Table 28: Amounts by Which Apparently Eligible Taxpayers Could 
                      Have Reduced Their Tax Liability in 2002: Confidence 
                                  Intervals                                49 
        Table 29: Percentage of Apparently Eligible Taxpayers That Claimed 
                   a Hope Credit but Would Have Been Better off Claiming a 
                         Lifetime Learning Credit in 2002: Point Estimates 49 
        Table 30: Percentage of Apparently Eligible Taxpayers That Claimed 
                   a Hope Credit but Would Have Been Better off Claiming a 
                    Lifetime Learning Credit in 2002: Confidence Intervals 50 
               Table 31: Percentage of Suboptimal Choices Made by Paid Tax 
                         Preparers: Point Estimates                        50 
               Table 32: Percentage of Suboptimal Choices Made by Paid Tax 
                    Preparers: Confidence Intervals                        50 
Figure Figure 1: Recipients of Title IV Assistance and Tax Filers       
                    Claiming an Education Tax Credit or Tuition Deduction, 
                    1997-2002                                              13 

Contents

                                 Abbreviations

AQEE                adjusted qualified education expenses                  
EFC                 expected family contribution                           
ESA                 Educational Savings Accounts                           
FAFSA               Free Application for Federal Student Aid               
FDL                 Federal Direct Loan                                    
FFEL                Federal Family Educational Loan                        
FWS                 Federal Work-Study                                     
GAO                 Government Accountability Office                       
GED                 General Educational Development                        
IRS                 Internal Revenue Service                               
NPSAS               National Postsecondary Student Aid Study               
PART                Program Assessment Rating Tool                         
PLUS                Parent Loans for Undergraduate Students                
SEOG                Supplemental Educational Opportunity Grant             
SOI                 Statistics of Income                                   
TIN                 taxpayer identification number                         

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

A

United States Government Accountability Office Washington, D.C. 20548

July 29, 2005

The Honorable Charles E. Grassley Chairman The Honorable Max Baucus
Ranking Minority Member Committee on Finance United States Senate

In recent decades, Congress has provided billions of dollars in assistance
each year to help the nation's students and families meet the costs of
postsecondary education. In the past, this assistance has primarily been
provided through student grant and loan programs authorized under title IV
of the Higher Education Act of 1965, as amended. Much of this aid, called
need-based aid, has been provided on the basis of the difference between a
student's cost of attendance and an estimate of the ability of the student
and the student's family to pay these costs, called the expected family
contribution. More recently, however, postsecondary assistance has also
been provided through a range of tax preferences,1 including postsecondary
tax credits, tax deductions, and tax-exempt savings programs. For example,
the Taxpayer Relief Act of 1997 allows eligible tax filers to reduce their
tax liability by receiving, for tax year 2005, up to a $1,500 Hope tax
credit and/or a $2,000 Lifetime Learning tax credit for tuition and
course-related fees paid. Furthermore, the 2001 Economic Growth and Tax
Relief Reconciliation Act created a new tax deduction for tuition
expenses, which eligible tax filers may use to deduct, in tax year 2005,
up to $4,000 from their taxable income for tuition and course-related fees
paid.2

Providing federal financial assistance in these varied ways presents
students and their families with an array of tools that may help them pay
postsecondary education expenses. There is concern, however, that the
postsecondary tax preferences create opportunities that are difficult for
families to understand and use correctly. Additionally, Congress has been

1 Tax preferences-also known as tax expenditures-are reductions in tax
liabilities that result from preferential provisions in the tax code, such
as exemptions and exclusions from taxation, deductions, credits,
deferrals, and preferential tax rates.

2 Tax filers may claim only one of the tax preferences above for a single
student. Families with more than one student with eligible expenses may
claim more than one of these tax preferences.

provided with little evidence concerning the effectiveness of assistance
provided under title IV or through tax preferences, such as whether
student assistance increases the rate at which students enroll in
postsecondary education or whether assistance increases the likelihood
that students will either earn a degree or continue their education (often
referred to as persistence).

You asked us to explain how the federal government assists students and
families in meeting the costs of postsecondary education, and to examine
the opportunities and challenges associated with these policies. To
address your interests, we answered the following questions: (1) How does
title IV grant and loan assistance compare with that provided through the
tax code? (2) To what extent are tax filers effectively using the
opportunities presented by postsecondary tax preferences? (3) What is
known about the effectiveness of federal assistance in promoting college
attendance, providing students with a wider range of choices among
postsecondary institutions, or encouraging students to persist in their
studies?

To compare title IV federal grant and loan assistance with that provided
through the tax code, we used Department of Education (Education) and
Internal Revenue Service (IRS) data, as well as agency documents and
statutory provisions describing federal student financial assistance
programs and tax preferences related to student financial assistance for
postsecondary education. For the purpose of this review, we focused on
federal grant and loan programs or tax preferences that served 500,000 or
more students and families. We also used two data sets, the 2003-2004
National Postsecondary Student Aid Study (NPSAS) from Education and IRS's
2002 Statistics of Income (SOI) individual tax file.3 We analyzed SOI data
to determine, among other things, whether tax filers made effective use of
the Hope and Lifetime Learning tax credits and the tuition deduction.
However, information required to analyze whether tax filers made effective
use of the tax credits and deduction was not consistently available for
tax filers included in the SOI data. Educational institutions must provide
IRS and students with Forms 1098-T, which document students' enrollment
status at the institution. Educational institutions may also, but are not
required to, report on students' qualified tuition and

3 Both data sets report 2002 adjusted gross income. In this report,
dependent student income equals 2002 parental adjusted gross income, while
independent student income equals the student's 2002 adjusted gross income
(and, if married, the spouse's adjusted gross income).

    Page 2 GAO-05-684 Federal Student Aid Programs

related expenses, scholarships, and grants. Consequently, our analysis of
SOI data for this purpose was limited to the approximately 1.8 million tax
filers for whom Forms 1098-T reported such information. We were unable to
determine whether these tax filers were representative of the
approximately 12.4 million tax filers for whom such information was
unavailable. To address issues of complexity as students and families use
postsecondary tax preferences, we reviewed relevant statutes, regulations,
and IRS documents, government reports, and academic research. We assessed
the reliability of NPSAS and SOI data sets in light of our data
reliability standards and found them to be useful for the purpose of this
review. Details of our data reliability assessment and other aspects of
our scope and methodology can be found in appendix I. To learn what is
known about the effectiveness of federal grant and loan programs and tax
preferences, we reviewed the academic literature on financial aid and
education-related tax preferences. We conducted our review from May 2004
through June 2005 in accordance with generally accepted government
auditing standards.

In general, assistance is provided to a wide range of students and
families

  Results in Brief

under both title IV programs and through several tax preferences, but
three key differences exist when comparing the two sources of aid. First,
while title IV aid and tax preferences focus primarily on helping students
meet current expenses, tax preferences also assist students and families
with saving for and repaying postsecondary costs. Second, while student
aid programs and tax preferences serve students and families across a wide
range of income groups, some title IV programs-particularly the Pell Grant
program--provide much of their financial assistance to students and
families whose incomes are lower, on average, than students and families
who receive student loans, tax credits and deductions, or who make use of
tax-exempt saving vehicles. Last, while both title IV aid and tax
preferences require students and families to fill out forms, tax
preferences require more responsibility on the part of students and
families for obtaining benefits because they must identify applicable tax
preferences, understand complex rules concerning their use, and correctly
calculate and claim credits or deductions on their returns.

Some tax filers do not appear to be making the most effective use of
certain postsecondary tax preferences and, as a result, fail to minimize
their federal income tax liabilities. Among tax filers included in the
2002 SOI data set for whom information indicating eligibility for an
education tax credit or tuition deduction was available, 27 percent failed
to claim either (1) a Hope or Lifetime Learning tax credit or (2) the
tuition deduction. By not claiming an education tax credit or tuition
deduction, we estimate that these tax filers failed to reduce their tax
liabilities by $169, on average. Furthermore, we estimate that 10 percent
of these apparently eligible tax filers failed to reduce their tax
liabilities by more than $500. We also found that 21 percent of tax filers
who claimed the tuition deduction would have reduced their tax liabilities
by an additional $83, on average, had they chosen to claim the Lifetime
Learning tax credit rather than the tuition deduction. While tax filers'
suboptimal choices may arise due to many factors, one reason may be the
complexity of these tax provisions. Experts commonly identify
postsecondary tax preferences as complex for tax filers to use because
many of these preferences have similar objectives, dissimilar definitions,
and rules that require extensive record keeping. Moreover, tax preferences
can interact with title IV aid in a manner that affects the net amount of
federal assistance received, adding to the complexity facing students and
families. Students and families may make costly mistakes to the extent
that they find using tax preferences to be difficult.

Little is known about the effectiveness of federal grant and loan programs
and education-related tax preferences in promoting attendance, choice, and
persistence. Many programs and tax preferences have not been studied, and
even among those that have, important aspects of their effectiveness
remain unexamined. When research exists, it suggests that federal programs
and tax preferences have a range of results, from no measurable effects to
modestly positive impacts on college attendance and persistence. Data and
methodological challenges limit the certainty with which the effects of
title IV programs and tax preferences, especially the latter, can be
identified, and result in widespread gaps in knowledge of their
effectiveness. Without this information, federal policy makers cannot
weigh the relative effectiveness of postsecondary assistance provided
through title IV and tax preferences. In 2002 we recommended, among other
things, that Education sponsor research into the impact of title IV
programs on postsecondary attendance and choice, completion, and college
costs. Little progress has been made by Education; however, according to
the department, it is in the process of establishing a postsecondary
research center that will sponsor such research.

In commenting on a draft of this report, Education agreed overall that
more research should be done on various federal programs that assist
students enrolled in postsecondary education, but disagreed with our
finding on the extent that title IV programs have been studied. Education
also said that we should have included its research in our description of
available program effectiveness research, including a particular study of
the Pell Grant program. In general, while Education's research
efforts-both the data collections conducted regularly by the agency and
studies based upon those data-provide useful descriptive information, the
data collections and related studies do not include information on those
who do not receive postsecondary education and are therefore of limited
use in establishing the effectiveness of title IV programs. Education's
letter is reprinted in appendix IV. Also, IRS provided technical comments
which we incorporated where appropriate.

The federal government helps students and families save, pay for, and
repay the costs of postsecondary education through grant and loan programs
authorized under title IV of the Higher Education Act of 1965, and through
tax preferences-reductions in federal tax liabilities that result from
preferential provisions in the tax code, such as exemptions and exclusions
from taxation, deductions, credits, deferrals, and preferential tax rates.
In fiscal year 2004, Education made approximately $14 billion in grants
and provided another $56 billion in loan assistance (face value) through
the title IV programs. The fiscal year 2004 outlay equivalent cost of the
postsecondary tax preferences reviewed in this study was estimated to be
$10 billion.

                                   Background

    Federal Grant and Loan Assistance to Postsecondary Students

Assistance provided under title IV programs include Pell Grants for
low-income students, parent loans known as PLUS loans, and Stafford
loans.4 While Pell Grants reduce the price paid by the student, student
loans help to finance the remaining costs and are to be repaid according
to varying terms. Stafford loans may be either subsidized or unsubsidized.
The federal government pays the interest cost on subsidized loans while
the

4 Consolidation loans are also authorized under title IV. These loans
allow borrowers to combine multiple student loans, possibly from different
lenders and from different loan programs, into a single new loan with
extended repayment periods. Because consolidation loans do not generally
result in an increase in loan principal, consolidation loans are not
addressed in this review. However, the federal government can incur
significant costs in providing borrowers with these loans. See GAO,
Student Loan Programs: As Federal Costs of Loan Consolidation Rise, Other
Options Should Be Examined, GAO-04-101 (Washington, D.C.: Oct. 31, 2003)
and Student Loan Programs: Lower Interest Rates and Higher Loan Volume
Have Increased Federal Consolidation Loan Costs, GAO-04-568T (Washington,
D.C.: Mar. 17, 2004).

student is in school, and during a 6-month period known as the grace
period, after the student leaves school. For unsubsidized loans, students
are responsible for all interest costs.5 Stafford and PLUS loans are
provided to students through both the Federal Family Education Loan
Program (FFELP) and the William D. Ford Direct Loan Program (FDLP). The
federal government's role in financing and administering these two loan
programs differs significantly. Under FFELP, private lenders, such as
banks, provide loan capital and make loans, and the federal government
guarantees FFELP lenders a minimum yield on the loans they make and
repayment if borrowers default. Under FDLP, federal funds are used as loan
capital and loans are provided through participating schools. The
Department of Education and its private-sector contractors jointly
administer the program. Title IV also authorizes programs funded by the
federal government and administered by participating higher education
institutions, including the Supplemental Educational Opportunity Grant
(SEOG), Perkins loans, and federal work-study aid, collectively known as
campus-based aid.

To receive title IV aid, students (along with parents, in the case of
dependent students) must complete a Free Application for Federal Student
Aid (FAFSA) form. Information from the FAFSA, particularly income and
asset information, is used to determine the amount of money-called the
expected family contribution (EFC)-that the student and/or family is
expected to contribute to the student's education. Statutory definitions
establish the criteria that students must meet to be considered
independent of their parents for the purpose of financial aid, and
statutory formulas establish the share of income and assets that are
expected to be available for the student's education.6 Once the EFC is
established, it is compared with the cost of attendance at the institution
chosen by the student. The cost of attendance comprises tuition and fees;
room and board; books and supplies; transportation; miscellaneous personal
expenses; and, for some

5 While called "unsubsidized," the federal government can still incur
costs on such loans, including the costs associated with borrowers who
default on their loans and, under the Federal Family Education Loan
Program, the costs of making payments to lenders to ensure them a minimum
federally guaranteed yield.

6 To be classified as an independent student for the purpose of receiving
title IV financial aid, students must meet one of the following criteria:
(1) be a veteran of the armed services; (2) be age 24 years or older by
December 31st of the award year; (3) be married; (4) be enrolled in a
graduate or professional education program; (5) have legal dependents
other than a spouse; or (6) be an orphan or ward of the court. Financial
aid administrators may also classify students as independent through the
exercise of their professional judgment.

students, additional expenses.7 If the EFC is greater than the cost of
attendance, the student is not considered to have financial need,
according to the federal aid methodology. If the cost of attendance is
greater than the EFC, then the student is considered to have financial
need. Title IV assistance that is made on the basis of the calculated need
of aid applicants is called need-based aid. Key characteristics of title
IV programs are summarized in table 1, below.

 Table 1: Description of Federal Student Aid Programs Authorized under Title IV
                          of the Higher Education Act

Title IV                                             Number and            
student aid                         Annual award     characteristics of    
program       Program details       amounts          beneficiaries         
Pell Grant    Grants are made on    $400 to $4,050   Dependent students:   
                 the basis of          for school year  About 2.1 million     
                 difference between    2004-2005.       grants were awarded   
                 the EFC and the                        in school year        
                 maximum Pell award or                  2003-2004, totaling   
                 the student's cost of                  $5.3 billion. The     
                 attendance, whichever                  average grant award   
                 is less. Grants are                    was $2,573; the       
                 not available for                      median income of      
                 postgraduate study.                    recipients was        
                                                        $24,576. Independent  
                                                        students: About 3     
                                                        million grants were   
                                                        awarded in school     
                                                        year 2003-2004,       
                                                        totaling $7.4         
                                                        billion. The average  
                                                        grant award was       
                                                        $2,436; the median    
                                                        income of recipients  
                                                        was $12,925.          
Supplemental Educational Opportunity Grant Schools administer grant funds,
which are made to undergraduates with exceptional financial need; priority
is given to Pell Grant recipients. Institutions must match a portion (at
least 25%) of the federal funds allocated. $100 to $4,000. Dependent
students: About 554,000 grants were awarded in school year 2003-2004,
totaling $494.2 million. The average grant award was $892; the median
income of recipients was $22,827. Independent students: About 715,000
grants were awarded in school year 2003-2004, totaling $391.9 million. The
average grant award was $548; the median income of recipients was $11,040.
Federal Work-Study Schools administer funds, which are used to provide
part-time jobs for undergraduate and graduate students with financial
need. Participating schools or nonprofit employers generally contribute at
least 25% of student's earnings (50% in the case of for-profit employers).
No annual minimum or maximum award amounts. Dependent students: About 1.1
million awards were awarded in school year 2003-2004, totaling $2 billion.
The average award was $1,901; the median income of recipients was $46,441.
Independent students: About 438,000 awards were awarded in school year
2003-2004, totaling $1 billion. The average award was $2,303; the median
income of recipients was $10,561.

7 These may include child care expenses for parents of young dependent
children or supportive services for disabled students.

Page 7 GAO-05-684 Federal Student Aid Programs

(Continued From Previous Page)

Title IV student aid program Program details Annual award amounts Number
and characteristics of beneficiaries

Federal Perkins Loan

Schools administer funds, comprised of federal capital contributions and
school matching funds (at least 1/3 of federal contributions), to make
low-interest (5 percent) loans for both undergraduate and graduate
students with exceptional financial need. Borrower repayments are owed to
the school.

$4,000 maximum for undergraduate students and $6,000 for graduate
students; no minimum award amount. (Aggregate limits: $8,000 for
undergraduates who have not completed 2 academic years; $20,000 for
undergraduates who have completed 2 years; and, $40,000 for graduate
students, including loans borrowed as an undergraduate.)

Dependent students: About 495,000 loans were made in school year
2003-2004, totaling $956 million. The average loan amount was $1,932; the
median income of recipients was $39,175.

Independent students: About 329,000 loans were made in school year
2003-2004, totaling $905.3 million. The average loan amount was $2,752;
the median income of recipients was $10,277.

Subsidized FFEL or Direct Stafford Loan

Loans made on the basis of financial need to undergraduate and graduate
students who are enrolled at least half-time. The federal government pays
the interest costs on subsidized loans while the student is in school, for
the first 6 months after the student leaves school, and during a period of
deferment.

$2,625 to $8,500 depending upon year of schooling. Aggregate limits are
$23,000 for undergraduates and $65,500 for graduate students.

Dependent students: About 2.6 million loans were made in school year
2003-2004, totaling $8.1 billion. The average loan amount was $3,188; the
median income of recipients was $44,678.

Independent students: About 3.8 million loans were made in school year
2003-2004, totaling $16.3 billion. The average loan amount was $4,340; the
median income of recipients was $19,430.

Unsubsidized FFEL or Direct Stafford Loan Loans made to undergraduate and
graduate students who are enrolled at least half-time. Unlike subsidized
loans, the federal government does not pay the interest costs on
subsidized loans while the student is in school, for the first 6 months
after the student leaves school, and during a period of deferment.
Otherwise, the terms and conditions of unsubsidized loans are the same as
those for subsidized loans.

$2,625 to $18,500 depending on year of schooling (including any subsidized
loan amounts received for the same period). Aggregate limits are $23,000
for dependent undergraduates, $46,000 for independent undergraduates, and
$138,500 for graduate students.

Dependent students: About 1.6 million loans were made in school year
2003-2004, totaling $5.3 billion. The average loan amount was $3,293; the
median income of recipients was $75,835.

Independent students: About 3.3 million loans were made in school year
2003-2004, totaling $18.5 billion. The average loan amount was $5,671; the
median income of recipients was $22,108.

FFEL or Direct PLUS Loan Loans made to parents on behalf of dependent
undergraduate students enrolled at least half-time. Borrowers are subject
to a credit check for adverse credit history and may be denied a loan.

Maximum loan amounts are limited to cost of attendance less other federal,
state, private, and institutional aid received for the period of
enrollment.

About 634,000 loans were made in school year 2003-2004, totaling $5.7
billion. The average loan amount was $9,019; the median income of
recipients was $71,397.

Source: GAO analysis of school year 2003-2004 NPSAS data.

    Tax Preferences

Prior to the 1990s, virtually all major federal initiatives to assist
students with the costs of postsecondary education were provided through
grant and loan programs authorized under title IV of the Higher Education
Act. Since the 1990s, however, federal initiatives to assist families and
students in paying for postsecondary education have largely been
implemented through the federal tax code. The federal tax code now
contains a range of tax preferences that may be used to assist students
and families in saving for, paying, or repaying the costs of postsecondary
education. These tax preferences include credits and deductions, both of
which allow tax filers to use qualified higher education expenses to
reduce their federal income tax liability. The tax credits reduce the tax
filers' income tax liability on a dollar-for-dollar basis but are not
refundable. Tax deductions permit qualified higher education expenses to
be subtracted from income that would otherwise be taxable. To benefit from
a higher education tax credit or tuition deduction, a tax filer must use
tax form 1040 or 1040A, have an adjusted gross income below the
provisions' statutorily specified income limits, and have a positive tax
liability after other deductions and credits are calculated, among other
requirements.

Tax preferences also include tax-exempt savings vehicles. Section 529 of
the tax code makes tax free the investment income from qualified tuition
programs. There are two types of qualified tuition programs: savings
programs established by states, and prepaid tuition programs established
either by states or by one or more eligible educational institutions.8
Another tax-exempt savings vehicle is the Coverdell Education Savings
Account. Tax penalties apply to both 529 programs and Coverdell savings
accounts if the funds are not used for allowable education expenses. Key
features of these and other education-related tax preferences are
described below, in table 2.

8 Certain aspects of the tax-favored treatment of section 529 programs
that were enacted in the Economic Growth and Tax Relief Act of 2001 are
subject to expire on December 31, 2010, if not extended.

Page 9 GAO-05-684 Federal Student Aid Programs

      Table 2: Selected Postsecondary Education Tax Preferences

Preference details
                         Income ranges                         Number and     
Tax                     for phasing                        characteristics 
                                   out                                     of 
preference Eligibility    benefitsa Eligible   Tax         beneficiaries   
                                       expenses   benefit   

Hope Credit Maximum credit: $1,500 per student. Credit rate is 100 percent
on first $1,000 of qualified higher education expenses, 50 percent on next
$1,000.

Tax filer on behalf of self,   Single filer:     Tuition and fees at 
spouse, or dependent           $42,000-        institutions eligible 
who is working toward a        $52,000       to participate in title 
degree or certificate at       Joint return: IV programs.            
least half-time in the first 2 $85,000-      
years of postsecondary         $105,000.b    
enrollment.                                  

Nonrefundable: if filer has no tax liability due to offsetting deductions,
exemptions, or competing tax credits, filer cannot receive credit.

In tax year 2002, 3.3 million tax filers claimed $3.2 billion in Hope
credits; the average credit claimed was $991, and the median income of
filers claiming the credit was $39,203.

Lifetime Learning Credit Tax filer on behalf of self, spouse, or dependent
who is enrolled in undergraduate or graduate courses, or any course that
aids in learning new or improving existing job skills, for as many years
as the student is enrolled.

Single filer: Tuition and fees at $42,000-institutions eligible $52,000 to
participate in title Joint return: IV programs. $85,000-$105,000.b

Maximum credit: $2,000 per tax filer. (20 percent of qualified higher
education expenses up to $10,000).

Nonrefundable: if filer has no tax liability due to offsetting deductions,
exemptions, or competing tax credits, filer cannot receive credit.

In tax year 2002, 3.5 million tax filers claimed $1.7 billion in Lifetime
Learning credits; the average credit claimed was $477, and the median
income of filers claiming the credit was $39,706.

Student Loan Interest Deduction Tax filer, on behalf of self, spouse, or
dependent, available even to those who do not itemize interest paid.
Student must have been enrolled at least half-time in a degree program.

Single filer: $50,000-$65,000 Joint return: $100,000-$130,000.

Eligible loans are those used to pay for tuition, fees, room and board,
and related expenses and include, for example, student loans provided
under title

IV.

Maximum deduction: $2,500 interest paid on eligible education loans is
deductible.

In tax year 2002, 6.6 million tax filers deducted $892.6 million of
student loan interest; the average deduction was $134, and the median
income of filers deducting student loan interest was $43,544.

(Continued From Previous Page)

      Preference details

Income ranges Number and Tax for phasing out characteristics of preference
Eligibility benefitsa Eligible expenses Tax benefit beneficiaries

Section 529 qualified tuition programs- prepaid tuition programs and
state-sponsored college savings programsc Specifics depend on No phaseout.
Tuition, fees, books, No tax is due on a About 7.2 million particular
program. supplies, and distribution from an account prepaid tuition and
Normally a prepaid equipment required unless the amount college savings
program is open for for attendance. distributed is greater than program
accounts contributions only on Room and board if the beneficiary's
adjusted had been established behalf of young children enrolled half time
or qualified education by December 31, and accounts must be more.
expenses. 2004, with a reported closed within some balance of $64.7 number
of years after the billion in both types of beneficiary reaches programs.
college age. Generally, savings programs do not have age restrictions.

Coverdell Education Savings Accounts Distributions can be used for
students enrolled on full-time, half- time, or less than half-time basis.

Account must be closed within 30 days after beneficiary reaches age

30.

For contributions, $95,000-$110,000 for single filers and
$190,000-$220,000 for joint returns.

Tuition, fees, books, supplies, and equipment required for attendance.
Room and board if enrolled half-time or more.

No tax is due on a distribution from an account unless the amount
distributed is greater than the beneficiary's adjusted qualified education
expenses.

Annual contribution limit is $2,000 per year per student (through age 17).

Approximately 1 million contributions were made to accounts in tax year
2002.

Tuition   Same as  Single filer:     Tuition and  Maximum        In tax    
Deduction Lifetime $65,000-80,000    fees at      deduction:     year      
(expires  Learning Joint Return:     institutions $4,000 per     2002, 3.4 
Dec. 31,  credit.  $130,000-160,000. eligible to  return for     million   
2005)                                participate  individual     tax       
                                        in title IV  filers whose   filers    
                                        programs.    modified       deducted  
                                                     adjusted gross $1.3      
                                                     income is less billion;  
                                                     than $65,000   the       
                                                                    average   
                                                                    deduction 
                                                                    was $377, 
                                                     ($130,000 for  and the   
                                                     joint filers); median    
                                                     $2,000 per     income of 
                                                     return for     filers    
                                                     individuals    using the 
                                                     whose modified deduction 
                                                                    was       
                                                     adjusted gross $54,326.  
                                                     income is more 
                                                     than $65,000   
                                                     ($130,000) but 
                                                     less than      
                                                     $80,000        
                                                     ($160,000).    

Sources: IRS and College Savings Plan Network; GAO analysis of IRS
Statistics of Income data for tax year 2002

a Modified adjusted gross income amounts are provided.

b Under the Taxpayer Relief Act of 1997, the income phaseout amounts are
indexed to inflation according to a formula specified in law for this
purpose, which may or may not result in a yearly increase.

c The Economic Growth and Tax Relief Reconciliation Act of 2001 provides
that from January 1, 2001 to December 31, 2011 eligible higher education
institutions, in addition to states, may offer tuition prepayment
programs. States remain the sole tax-exempt sponsors of college savings
programs.

    Tax Preferences Are More Recent Than Most Title IV Programs

Our review of tax preferences did not include exclusions from income,
which permit certain types of education-related income to be excluded from
the calculation of adjusted gross income on which taxes are based.9 We
also did not include special provisions in the tax code that also extend
existing tax preferences when tax filers support a postsecondary education
student.10 Appendix IV lists all tax preferences reported by IRS,
including ones not included in this review.

Title IV programs have been in place for decades, while most
education-related tax preferences were created much more recently. Between
the late 1950s and the early 1980s each major federal initiative to assist
students with the costs of postsecondary education was provided either
through grant or loan programs. From 1981 through 1995 no new federal
grant or loan-financing vehicles were adopted; however, new financing
options, such as loan consolidation, and new delivery systems, such as
direct student lending, were introduced for student loan programs. Since
1995, on the other hand, every federal initiative for the financing of
postsecondary education has been implemented through the federal tax code,
primarily through the Tax Relief Act of 1997 and the Economic Growth and
Tax Relief Reconciliation Act of 2001. By 2002 the number of tax filers
claiming a tax credit or tuition deduction was broadly comparable to the
number of students aided through title IV programs: about 8.4 million
students received a title IV grant and/or loan, and about 9.6 million tax

9 For example, scholarships covering tuition and fees and tuition
reductions for the children of employees of an educational institution are
not counted as income for income tax purposes. Similarly, student loans
forgiven when a graduate goes into certain professions are also not
subject to federal income taxes.

10 For example, tax filers may claim postsecondary education students as
dependents after age 18, even if the student has his or her own income
over the limit that would otherwise apply. Also, gift taxes do not apply
to funds used for certain postsecondary educational expenses, even for
amounts in excess of the usual $11,000 limit on gifts. In addition, funds
withdrawn early from an Individual Retirement Account are not subject to
the usual 10 percent penalty when used for either a tax filer's or his or
her dependent's postsecondary educational expenses.

filers filed returns claiming a Hope tax credit, Lifetime Learning tax
credit, or tuition deduction. (See fig. 1.)11

Figure 1: Recipients of Title IV Assistance and Tax Filers Claiming an
Education Tax Credit or Tuition Deduction, 1997-2002 Recipients and tax
filers (in millions)

8

6

4

2

0

      1997 1998 1999 2000 2001 2002 Year

Title IV aid recipients Tax returns claiming postsecondary tax credits
and/or tuition deduction Source: GAO analysis of Budget of the United
States Government, FY1994-2002 and Internal Revenue Service data
1994-2002.

11 The 8.4 million title IV aid recipient figure is an unduplicated count
of students, i.e., a student receiving both a grant and a loan is counted
only once. Each of the 9.6 million tax filers represents at least one
student and in some cases more than one student. A tax filer with more
than one dependent with qualified educational expenses, or with qualified
expenses of his or her own along with those of a dependent, may claim more
than one tax preference on his or her tax return as long as other
eligibility criteria are met.

Page 13 GAO-05-684 Federal Student Aid Programs

  Tax Preferences Differ from Title IV Assistance in Timing, Distribution, and
  Students' and Families' Responsibility for Obtaining Benefits

Postsecondary student financial assistance provided through programs
authorized under title IV of the Higher Education Act and the tax code
differ in three key ways. First, tax preferences and title IV programs
differ in the timing and the distribution of benefits they provide. Title
IV grant and loan programs primarily provide aid to students and families
while students are in college, whereas tax preferences help both during
the college years and before and after college by assisting with saving
for or repaying college costs. Additionally, although student aid programs
and tax preferences assist students and families across a wide range of
income levels, some title IV programs, such as the Pell Grant and
subsidized Stafford student loan programs, provide much of their financial
assistance to students and families whose incomes are lower, on average,
than students who receive unsubsidized Stafford loans, tax deductions, or
make use of tax-exempt saving vehicles. Last, students and families have
more responsibility for appropriately using and thereby obtaining the
benefits of tax preferences compared with title IV aid.

    Tax and Title IV Programs Differ in Benefit Timing

Title IV programs and education-related tax preferences differ
significantly as to when eligibility is established and in the timing of
the assistance they provide. Eligibility for title IV programs is
generally established at the time of enrollment and prior to each
subsequent school year thereafter, and title IV programs generally provide
benefits to students while they are in school. Education-related tax
preferences reach widely across the life span. They encourage saving for
college, especially among families with dependent children through
tax-exempt saving; assist enrolled students and their families in meeting
the current costs of postsecondary education through credits and a tuition
deduction; and assist students and families repaying the costs of past
postsecondary education by allowing tax filers to claim a deduction for
student loan interest paid. Thus, tax filers must determine their
eligibility for these preferences every year that contributions are made
to Coverdell Education Savings Accounts or every year that a former
student claims a student loan interest deduction. Table 3 shows a
comparison of the timing of title IV assistance and the assistance
provided through various tax preferences.

Table 3: Comparison of Assistance by Timing of Benefit for Selected Programs and
                                Tax Preferences

Type of assistance  Save for future  Pay current expenses  Repay expenses  
                          expenses                            
Grant programs                       Pell Grants           
                                        Supplemental          
                                        Educational           
                                        Opportunity Grants    
Loan programs                        Subsidized and        
                                        Unsubsidized Stafford 
                                        Loans Federal Perkins 
                                        Loans Federal PLUS    
                                        Loans                 
Tax preferences            Coverdell Hope Credit Lifetime  Student Loan    
                            Educational Learning Credit       Interest        
                       Savings Accounts Tuition Deduction     Deduction       
                            Section 529                       
                      Qualified Tuition                       
                               Programs                       
Work-Study program                   Federal Work Study    

                                  Source: GAO.

Beneficiaries of Title IV While title IV programs and tax preferences
assist many students and families, a variety of program and tax rules
affect students' and families'

    Programs and Tax

Preferences Differ eligibility for such assistance. These rules also
affect the distribution of title IV aid and the assistance provided
through tax preferences. As a result, the beneficiaries of title IV
programs and tax preferences differ, as discussed below.

Title IV Assistance Title IV programs have rules for calculating grant and
loan assistance that give different consideration to family income,
assets, and college costs in the award of financial aid.12 Pell Grant
awards are calculated by subtracting the student's expected family
contribution (EFC) from the maximum Pell Grant award ($4,050 in academic
year 2004-2005), or the student's cost of attendance, whichever is less.
Because the expected family contribution is closely linked to family
income and circumstances (such as the size of the family and the number of
dependents in school), and modest EFCs are required for Pell eligibility,
Pell awards are received primarily by families with modest incomes. The
maximum subsidized Stafford loan that a

12 Campus-based aid programs authorized under title IV differ from these
programs in funding and eligibility: institutions provide matching funding
for federal spending, and participating institutions distribute aid using
institution-specific criteria consistent with federal program
requirements. Because they have institution-specific criteria, the
relationship between program rules and the distribution of benefits is
more complex and excluded from our analysis.

Page 15 GAO-05-684 Federal Student Aid Programs

student may obtain is based upon his or her cost of attendance, minus the
expected family contribution and the estimated financial assistance that
the student will receive.13 For a given cost of attendance, therefore, the
amount of a subsidized Stafford loan increases as EFC decreases. In
contrast, the maximum unsubsidized Stafford loan amount is calculated
without direct consideration of financial need: students may borrow up to
their cost of attendance, minus the estimated financial assistance they
will receive.14

The different award rules for Pell Grants and subsidized and unsubsidized
Stafford loans result in different patterns of program participation among
students of different incomes, and different distributions of dollar
support among students. As table 4 shows, 92 percent of Pell financial
support in 2003-2004 was provided to dependent students whose family
incomes were $40,000 or below, and the 38 percent of Pell recipients in
the lowest income category ($20,000 or below) received a higher share (48
percent) of Pell financial support. With respect to subsidized Stafford
loans, 67 percent of recipients had family incomes of $60,000 or less and
received a proportional share of total subsidized loan volume. In
contrast, 65 percent of unsubsidized Stafford loan recipients had family
incomes above $60,000 and received 69 percent of total unsubsidized loan
volume.

Table 4: Percentage of Aid Recipients and Dollars of Aid by Income
Category for Dependent Students Served by Selected Title IV Programs

               Dependent           $21,001- $40,001- $60,001- $80,001-     More 
                                                                           than 
 Program      students   $0-20,000  40,000   60,000   80,000  100,000  $100,000 
 Pell Grant   Recipients        38       47       14        2        0        0 
              Dollars           48       44        8        1        0        0 
 Stafford     Recipients        16       28       23       17        9        7 
 Subsidized                                                            
 Loan         Dollars           16       28       24       17        9        6 
 Stafford     Recipients         7       14       14       19       18       28 
 Unsubsidized                                                          
 Loan         Dollars            7       12       12       18       19       32 

                 Source: GAO analysis of 2003-2004 NPSAS data.

13 Estimated financial assistance includes the Pell Grant and most other
sources of state, federal, private, and institutional aid.

14 Additionally, loan amounts for both subsidized and unsubsidized loans
are subject to statutory limits on annual and cumulative borrowing.

Page 16 GAO-05-684 Federal Student Aid Programs

Notes: See appendix II for confidence intervals associated with these
estimates. Numbers in rows may not add to 100 percent because of rounding.

Because independent students generally have lower incomes and accumulated
savings than dependent students and their families, patterns of program
participation and dollar distribution differ. Participation of independent
students in Pell, subsidized Stafford, and unsubsidized Stafford loan
programs is heavily concentrated among those with incomes of $40,000 or
less: from 74 percent (unsubsidized Stafford) to 95 percent (Pell) of
program participants have incomes below this level. As shown in table 5,
the distribution of award dollars follows a nearly identical pattern.

Table 5: Percentage of Aid Recipients and Dollars of Aid by Income
Category for Independent Students Served by Selected Title IV Programs

             Independent           $21,001- $40,001- $60,001- $80,001-     More 
                                                                           than 
 Program      students   $0-20,000  40,000   60,000   80,000  100,000  $100,000 
 Pell Grant   Recipients        67       28        5        0        0        0 
              Dollars           73       25        3        0        0        0 
 Stafford     Recipients        51       29       12        5        2        1 
 Subsidized                                                            
 Loan         Dollars           52       28       12        5        2        2 
 Stafford     Recipients        46       28       14        6        3        3 
 Unsubsidized                                                          
 Loan         Dollars           46       24       13        7        3        5 
                         Source: GAO analysis of 2003-2004 NPSAS data.
                         Notes: See appendix II for confidence intervals
                         associated with these estimates.
                         Numbers in rows may not add to 100 percent because of
                         rounding. 
 Tax                     Many education-related tax preferences have both de
 Preferences             facto lower limits
                         created by the need to have a positive tax liability
                         in order to obtain their
                         benefit and income ceilings on who may use them. For
                         example, the Hope
                         and Lifetime Learning tax credits require that tax
                         filers have a positive tax
                         liability to use them and income-related phase-out
                         provisions in 2004 that
                         begin at $42,000 and $85,000 for single and joint
                         filers, respectively. The
                         income-related phase-out provision for the tuition
                         deduction, in
                         comparison, begins in 2004 at $65,000 and $130,000 for
                         single and joint
                         filers, respectively. As a result, the majority of tax
                         filers claiming the Hope
                         and Lifetime Learning tax credits in 2002 had incomes
                         under $40,000.
                         Among those who claimed the tuition deduction, in
                         contrast, 38 percent of
                         tax filers had incomes in this range, while 62 percent
                         had incomes over
                         $40,000. Table 6 shows the income categories of tax
                         filers claiming the

three tax preferences available to current students and/or their families
along with the distribution of dollars through those preferences in 2002.

The reduction in tax liability associated with the use of the Hope and
Lifetime tax credits also differs from that associated with the use of the
tuition deduction. In 2002 tax filers claimed Hope credits worth about
$3.2 billion and Lifetime Learning credits totaling about $1.7 billion. As
shown in table 6, below, about 80 percent of the 2002 Hope and Lifetime
credits' reduction in tax liability went to tax filers with incomes
between $20,001 and $80,000. The distribution of benefits for the tuition
deduction shows a substantially different pattern: more than half (52
percent) of the approximately $1.3 billion reduction in tax liability
associated with the use of the deduction in 2002 went to families with
incomes of $80,001 and above.

Table 6: Percentage of Tax Filers Claiming Hope and Lifetime Learning
Credits and Tuition Deduction and Tax Preference Dollars by Income
Category, Tax Year 2002

                                $20,001- $40,001- $60,001- $80,001- More than 
Type of aid          $0-20,000 40,000  60,000   80,000  100,000   $100,000 
Hope Credit   Tax           18     33       20       18       11         0 
                filers                                              
               Dollars         10     33       23       24       10         0 
Lifetime      Tax                                                          
Learning     filers         16     34       25       16        8         0
Credit                                                           
               Dollars         12     34       25       21        7         0 
Tuition       Tax           20     18       16       13       16        17 
Deduction    filers                                              
               Dollars         11     10       13       14       23        29 

Source: GAO analysis of 2002 SOI data.

Notes: See appendix II for confidence intervals associated with these
estimates. Numbers in rows may not add to 100 percent because of rounding.

Although many families are eligible to participate in tax-exempt savings
programs, the programs are more advantageous to those with higher incomes
and tax liabilities. Families with higher than average incomes are more
likely to use tax-exempt savings opportunities for a range of reasons,
including, among others, that (1) these families hold greater assets to
invest in these vehicles; (2) these families have a higher marginal tax
rate, and thus benefit the most from the use of these vehicles; and (3)
higher-income

    Students and Families Have More Responsibility for Obtaining Benefits of Tax
    Preferences in Comparison to Title IV Aid

families may gain a reduction in tax liability even if withdrawals are not
used for postsecondary expenses.15

The federal government and postsecondary institutions have significant
responsibilities in assisting students and families in obtaining
assistance provided under title IV programs but only minor roles with
respect to tax filers' use of education-related tax preferences. To obtain
federal student aid, applicants must first complete the FAFSA form, which
in its 2004 paper version was over eight pages long and contained more
than 100 questions. While concerns have been raised that the FAFSA
application may deter potentially eligible students from participating in
title IV grant and loan programs,16 filling out the FAFSA and submitting
it to the Department of Education completes, by and large, students' and
families' responsibility in obtaining aid. To benefit from title IV
programs, students need not learn the rules of the federal student aid
methodology, eligibility rules for individual programs, or understand the
ways in which federal student aid programs interact with one another.
Rather, the Department of Education is responsible for calculating
students' and families' EFC on the basis of the FAFSA, and a student's
educational institution is responsible for determining aid eligibility and
the amounts and packaging of aid awards. In addition, title IV educational
institutions assist Education in verifying the information submitted on
the FAFSA form for a sample of aid applicants.

Higher education tax preferences, in contrast to federal grants and
student loans, require more responsibility on the part of students and
families. Although postsecondary institutions provide students and IRS
with information about higher education attendance, they have no other
responsibilities for higher education tax credits, deductions, or
tax-preferred savings. The federal government's primary role with respect
to higher education tax preferences is limited to the promulgation of
rules;

15 The earnings portion of a withdrawal from a Coverdell Education Savings
Account is taxed at the student's marginal rate, rather than the rate of
the parents. For parents with more than $100,000 in household income,
nonqualified withdrawals from such an account, even with a 10 percent tax
penalty, are taxed at a lower rate than withdrawals from a nonadvantaged
account. See Susan Dynarski, High Income Families Benefit Most from New
Education Savings Incentives, Tax Policy Issues and Options, No. 9
(February 2005).

16 Advisory Committee on Student Financial Assistance, The Student Aid
Gauntlet, Final Report of the Special Study of Simplification of Need
Analysis and Application for Title IV Aid, (January 2005). Kane (1995)
suggests that providing better information about financial aid and
streamlining the process of applying for aid could increase enrollment.

  Some Tax Filers May Not Effectively Use Postsecondary Tax Preferences,
  Possibly Due to Complexity

the provision of guidance to tax filers; and to the processing of tax
returns, including some checks on the accuracy of items reported on those
tax returns. In contrast, the primary responsibility for selecting among
and properly using tax preferences rests with tax filers: they must
understand the rules in light of their own situation, identify applicable
tax preferences, understand how these tax preferences interact with one
another and with federal student aid, keep records sufficient to support
their tax filing, and correctly claim the credit or deduction on their
return.

According to our analysis of IRS data on the use of Hope and Lifetime tax
credits and the tuition deduction, some tax filers appear to make
less-than-optimal choices among them. The apparently suboptimal use of
postsecondary tax preferences may arise, in part, from the complexity of
using these provisions; however, our analysis of tax data does not permit
us to identify why they are making these choices. Tax policy analysts
consistently identify postsecondary tax preferences as a set of tax
provisions that demand a particularly large investment of knowledge and
skill on the part of students and families or expert assistance purchased
by those with the means to do so.17 Additional complexity associated with
the use of postsecondary tax preferences also arises from the interaction
of tax preferences and title IV student aid.

               Some Tax Filers Appear to Make Suboptimal Choices

Making poor choices among tax preferences for postsecondary education may
be costly to tax filers. For example, families may strand assets in a
tax-exempt savings vehicle and incur tax penalties on their distribution
if their child chooses not to go to college. They may also fail to
minimize their federal income tax liability by claiming a tax credit or
deduction that yields less of a reduction in taxes than a different tax
preference or by failing to claim any of their available tax preferences.
For example, if a married couple filing jointly with one dependent in
his/her first 2 years of college had an adjusted gross income of $50,000,
qualified expenses of $10,000 in

17 See, for example, U.S. Congress, Joint Committee on Taxation, Study of
the Overall State of the Federal Tax System, vol. II (April 2001); U.S.
Congress, Joint Committee on Taxation, Present Law and Analysis Relating
to Tax Benefits for Higher Education (July 21, 2004); Nina E. Olson
(National Taxpayer Advocate), "Complexity, Compliance, and Communication:
Why Should Tax Filers Comply in a Complex and Changing Tax Environment?"
(presentation before the President's Advisory Panel on Federal Tax Reform,
Mar. 3, 2005).

2004, and tax liability greater than $2,000, their tax liability would be
reduced by $2,000 if they claimed the Lifetime Learning credit but only
$1,500 if they claimed the Hope credit.

To assess whether tax filers faced difficulty with choosing among these
preferences, we examined whether tax filers who were confronted with a
relatively common tax choice-whether to claim the Hope or Lifetime
Learning credit or the tuition deduction-chose the tax preference that
minimized their tax liability. We analyzed information that IRS was
provided with by educational institutions using Form 1098-T and tax return
information in IRS's 2002 Statistics of Income sample. Because 87 percent
of Form 1098-T returns did not contain educational expense information, we
were able to analyze only the remaining 13 percent of tax returns
(representing about 1.8 million returns) in the SOI sample that received a
Form 1098-T and contained information concerning students' educational
expenses. We were unable to determine if this 13 percent of returns is
representative of the entire population of Form 1098-Ts. (See appendix I
for details.) All estimates and their associated confidence intervals can
be found in appendix II.

We found that some people who appear to be eligible for tax credits and/or
the tuition deduction did not claim them. The filers of about 77 percent
of the tax returns that we were able to review were apparently eligible to
claim one or more of the three tax preferences: the tax filers appear to
have had a positive income tax liability, qualified educational expenses,
an adjusted gross income below statutory phase-out limits, and were
otherwise eligible.18 Among filers who were apparently eligible to claim
one of the three tax preferences, 27 percent, representing about 374,000
tax filers, failed to do so. The amount by which these tax filers failed
to reduce their tax averaged $169; 10 percent of this group could have
reduced their tax liabilities by over $500.

18 We examined whether tax filers had (1) tax liability after claiming
other tax credits, (2) net educational expenses after accounting for
scholarships and grants as reported on the Form 1098-T, and (3) taxable
income under program thresholds for tax year 2002. We also examined
whether tax filers were married filing separately or filed a Form 1040EZ
because this would prevent tax filers from being able to claim the
education tax credits or tuition deduction. We were unable to consider
other possible explanations, including whether tax filers did not meet
certain qualification requirements, such as, in the case of the Hope tax
credit, whether the student was in his or her first 2 years of
postsecondary education. Eligibility for more than one tax preference for
the same student does not mean that a tax filer may claim more than
one-the tax filer must choose just one of the three tax preferences we
discuss here per student.

    The Suboptimal Use of Postsecondary Tax Preferences May Result from Their
    Complexity

Some tax filers used a higher education tax credit or the tuition
deduction but chose one that yielded a smaller reduction in their tax
liability than they could have otherwise realized.19 Among those who
claimed the tuition deduction, we estimate that 21 percent (representing
about 51,000 tax filers) would have been better off claiming the Lifetime
Learning tax credit while 8 percent (representing about 22,000 tax filers)
of those claiming the Lifetime Learning credit would have reduced their
taxes by a greater amount if they had claimed the tuition deduction
instead. The average amount by which these tax filers failed to reduce
their taxes was $83 for tax filers claiming the tuition deduction and $138
for those claiming the Lifetime Learning credit. Some tax filers making
these decisions failed to realize larger reductions in their tax
liabilities-10 percent of suboptimal tuition deduction claimants could
have reduced their tax liabilities by about $158 or more and 10 percent of
suboptimal Lifetime Learning credit claimants could have reduced their tax
liabilities by about $237 or more. On the other hand, we found no cases
where tax filers claiming a Hope credit would have been better off by
claiming a Lifetime Learning credit instead.

Suboptimal choices were not limited to tax filers who prepared their own
tax returns. A possible indicator of the difficulty people face in
understanding education-related tax preferences is how often the
suboptimal choices we identified were found on tax returns prepared by
paid tax preparers. We estimate that about 50 percent of the returns we
found that appear to have failed to optimally reduce the tax filer's tax
liability were prepared by paid tax preparers. Generalized to the
population of tax returns we were able to review, returns prepared by paid
tax preparers represent about 223,000 of the approximately 447,000
suboptimal choices we found.

The apparently suboptimal use of postsecondary tax preferences may arise,
in part, because of the complexity of using these provisions. Tax policy
analysts have frequently identified postsecondary tax preferences as a set
of tax provisions that demand a particularly large investment of knowledge
and skill on the part of students and families or expert

19Our analysis considered the difference between (1) the 2002 maximum
allowable Lifetime Learning credit, calculated from 20 percent of each tax
filer's reported qualified educational expenses of up to $5,000 and (2)
the amount of possible deduction from income for qualified educational
expenses of up to $3,000 in combination with the tax filer's 2002 marginal
tax rate.

Page 22 GAO-05-684 Federal Student Aid Programs

                            Multiple Tax Preferences

assistance purchased by those with the means to do so. They suggest that
this complexity arises from multiple postsecondary tax preferences with
similar purposes, from key definitions that vary across these provisions,
and from rules that coordinate the use of multiple tax provisions.
Additional complexity associated with the use of these provisions may
arise from the interaction of tax preferences and title IV student aid.
Complexity may lead some not to claim a credit because they judge the
added costs of filing for the credit to outweigh its benefits.

Multiple tax preferences with similar purposes may place substantial
demands on the knowledge and skills of millions of students and families.
Twelve tax preferences are outlined in the IRS publication, Tax Benefits
for Education,20 including four different tax preferences for educational
saving.21 Three of these preferences-Coverdell Education Savings Accounts,
Qualified Tuition Programs, and U.S. education savings bonds- differ
across more than a dozen dimensions, including the tax penalty that occurs
when account balances are not used for qualified higher education
expenses, who may be an eligible beneficiary, annual contribution limits,
and other features.22 Attempting to learn about, compare, and choose from
among these tax-preferred higher education savings options may require
substantial knowledge and skill on the part of parents with young
dependents beyond that required of savings and investment decisions in
general.

Among the tax preferences we reviewed, three help students meet current
costs-the Hope credit, Lifetime Learning credit, and the tuition
deduction. These tax preferences also differ across many dimensions.23
Though similar in purpose, the three preferences have different
eligibility criteria, benefit levels, and income-related phase-outs. For
tax filers to obtain the maximum benefit from these preferences, they must
first ascertain which

20 Department of Treasury, IRS Publication 970, Tax Benefits for Education
(2004).

21 The 12 programs and 2 others not listed in the Publication are listed
in appendix III.

22 Albert J. Davis, Choice Complexity in Tax Benefits for Higher
Education, National Tax Journal, Vol. LV, No. 3 (September 2002).

23 See, for example, U.S. Congress, Joint Committee on Taxation, Present
Law and Analysis Relating to Tax Benefits for Higher Education (July 21,
2004); U.S. Congress, Joint Committee on Taxation, Study of the Overall
State of the Federal Tax System, vol. II (April 2001).

Varying Qualified Expense Definitions

preference they are eligible to use,24 and then correctly calculate which
preference minimizes their tax liability by making separate calculations
between the Hope and Lifetime Learning credits. If filers are within the
phase-out range, they must calculate whether the tuition deduction is
preferable to either credit. Tax filers with more than one student in
postsecondary education may be eligible to claim multiple preferences, and
may need to test different combinations of benefits to optimize tax
savings.

Additional demands on the skill and knowledge of students and families may
result from the fact that higher education tax preferences do not all use
the same definition of qualified higher education expenses. What tax
filers are allowed to claim as a qualified higher education expense
varies, for example, between tax-exempt savings vehicles and tax credits.
For example, while Coverdell Education Savings Accounts and section 529
Qualified Tuition Programs permit tax filers to include tuition, fees,
books, supplies, and equipment required for enrollment as part of their
qualified educational expenses, higher education tax credits permit only
tuition and fees required for enrollment to be counted as qualified higher
education

25

expenses. These dissimilar definitions require that tax filers keep track
of expenses separately, applying some expenses to some tax preferences,
but not others, and the Joint Committee on Taxation has suggested that
they may increase the likelihood of inadvertent errors and may also
increase taxpayer frustration.26

             Tax Rules Coordinating the Use of Multiple Preferences

In addition to learning about, comparing, and selecting tax preferences,
filers who wish to make optimal use of multiple tax preferences must
understand how the use of one tax preference affects the use of others.
The use of multiple education-related tax preferences is coordinated
through rules that prohibit the application of the same qualified higher
education expenses for the same student to more than one education-related
tax preference, sometimes referred to as "anti-double-dipping rules."
These rules are important because they prevent tax filers from
underreporting

24 Credit eligibility depends, in part, upon the academic year in which
the student is enrolled, the number of credits taken by the student, the
student's status with respect to a degree or certificate program, and the
adjusted gross income of the parents.

25 Fees and expenses are qualified only if they must be paid to the
institution as a condition of enrollment or attendance.

26 Study of the Overall State of the Federal Tax System, vol. II, 125-6.

Page 24 GAO-05-684 Federal Student Aid Programs

Interaction between Tax Preferences and Student Financial Aid

their tax liability. Nonetheless, anti-double-dipping rules are
potentially difficult for tax filers to understand and apply, and
misunderstanding them may have consequences for a filer's tax liability.

Consider, for example, how the use of college savings programs and the
tuition deduction is affected by these rules. To calculate whether a
distribution from a college savings program is taxable, a tax filer must
determine if the total distributions for the tax year are more or less
than the total qualified educational expenses reduced by any tax-free
educational assistance, i.e., their adjusted qualified education expenses
(AQEE). After subtracting tax-free assistance from qualified educational
expenses to arrive at the AQEE, tax filers multiply total distributed
earnings by the fraction (AQEE / total amount distributed during the
year). If parents of a dependent student paid $6,500 in qualified
education expenses from a $3,000 tax-free scholarship and a $3,600
distribution from a tuition savings program, they would have $3,500 in
AQEE. If $1,200 of the distribution consisted of earnings, then $1,200 x
($3,500 AQEE / $3,600 distribution) would result in $1,167 of the earnings
being tax-free, while $33 would be taxable. However, if the same tax filer
had also claimed a tuition deduction, anti-double-dipping rules would
require the tax filer to subtract the expenses taken into account in
figuring the tuition deduction from AQEE. If $2,000 in expenses had been
used toward the tuition deduction, then the taxable distribution from the
section 529 savings program would rise to $700.27 For families such as
these, anti-double-dipping rules increase the computational complexity
they face and may result in unanticipated tax liabilities associated with
the use of section 529 savings programs.

Because the use of federal higher education tax preferences may affect a
student's eligibility for title IV federal student assistance-and the
receipt of title IV federal student assistance may affect a student's
ability to use federal higher education tax preferences-many students and
families must develop knowledge and skill sufficient to understand the
relationship between the two. For example, the Internal Revenue Code
requires tax filers to reduce the qualified higher education expenses they
apply toward higher education tax credits by the amount of nontaxable aid
they receive, including federal aid such as a Pell Grant. As a result,
receiving a Pell Grant has the potential to reduce the amount of Hope or
Lifetime Learning tax credit for which a filer is eligible. More
generally, tax filers must take

27 The new nontaxable distribution figure is calculated $1,200 x
($1,500/$3,600) = $500. The taxable portion then becomes $1,200 - $500 =
$700.

Page 25 GAO-05-684 Federal Student Aid Programs

care to reduce their qualified higher education expenses by the amount of
all nontaxable assistance that they receive-federal and nonfederal-
applying only adjusted qualified higher education expenses toward credits,
deductions, and distributions from tax-exempt savings vehicles.

While some federal higher education tax preferences, such as tax credits,
have no effect on one's eligibility for title IV federal student
assistance,28 others do: like other family savings, assets held in
tax-exempt savings vehicles, such as a Coverdell Education Savings Account
or a section 529 savings account, are included in the calculation of the
expected family contribution. In those instances where students are on the
margin of eligibility to participate in need-based title IV federal aid
programs, using these accounts may reduce the aid for which a student is
eligible.

The federal financial aid methodology in place prior to January 2004
treated assets held in different savings vehicles in widely varying ways.
According to one set of calculations, each dollar of funds held in a 529
savings program resulted in a reduction of 15 cents in need-based aid,
while each dollar of funds held in a Coverdell Educational Savings Account
resulted in a $1.22 reduction in need-based aid.29 For families close to
the income limit for eligibility for need-based aid, a $1,000 pretax
investment in a Coverdell Educational Savings Account yielded a simulated
final return (net of income taxation and foregone student aid) of -$1,194,
leaving the family worse off than if they had not saved, or if they had
saved using a regular savings account (+$490) or a traditional individual
retirement account (+$844). In response to recommendations contained in
our 2002 report on student aid,30 the Department of Education modified its
guidance concerning federal financial aid methodology,31 announcing that
Coverdell assets would be treated as assets of the parent, rather than the
student-

28 The Higher Education Act stipulates that the Hope and Lifetime Learning
tax credits may not be considered either as estimated financial assistance
in the assessment of aid eligibility, or as income or assets in the
calculation of the expected family contribution.

29 See Dynarski (2004). These calculations include school need-based
grants and need-based federal aid (grant aid, work-study, and the Perkins
and Stafford subsidized federal loans).

30 GAO, Student Aid and Tax Benefits: Better Research and Guidance Will
Facilitate Comparison of Effectiveness and Student Use, GAO-02-751
(Washington, D.C.: Sept. 13, 2002).

31 Department of Education, "Treatment of Coverdell Accounts and 529
Tuition Plans," Dear Colleague Letter, DCL ID: GEN-04-02 (posted Jan. 22,
2004).

  Research on Effectiveness of Federal Postsecondary Assistance Is Incomplete

and, therefore, result in the same reduction in need-based aid as 529
savings program assets. Nonetheless, families that save in prepaid tuition
programs remain subject to a dollar-for-dollar reduction in federal aid
for each dollar distributed by the program.32

Little is known about the effectiveness of federal grant and loan programs
and education-related tax preferences in promoting attendance, choice, and
persistence. Many federal aid programs and tax preferences have not been
studied, and for those that have been studied, important aspects of their
effectiveness remain unexamined. When research exists, it reaches varying
conclusions about the effects of federal programs and tax preferences.
Some studies identify no measurable effects on college attendance and
persistence, while others find positive effects. (A bibliography listing
the studies we reviewed is included at the end of this report.) Data and
methodological challenges limit the certainty with which the effects of
title IV programs and tax preferences, especially the effects of the
latter, can be identified and result in widespread gaps in the knowledge
of their effectiveness.33 In 2002 we recommended that Education sponsor
research into key aspects of effectiveness of title IV programs and that
Education and the Department of the Treasury collaborate on such research
into the relative effectiveness of title IV programs and tax
preferences.34 Since our prior report, little has been done to implement
these recommendations, although Education is the process of establishing a
postsecondary research center.

32 Dr. Susan Dynarski, testimony on "The Role of Higher Education
Financing in Strengthening U.S. Competitiveness in a Global Economy"
before the U.S. Congress, Senate Committee on Finance (July 22, 2004).

33 We looked for studies addressing a program or tax preference's affect
on rates of postsecondary attendance, persistence, and choice because
these measures have been the focus of congressional concern as expressed
in committee reports, statutorily established study commissions, and
requests for our work from Congress.

34 GAO-02-751.

    Little Is Known about the Effectiveness of Federal Title IV Postsecondary
    Programs and Tax Preferences

We found no research on any aspect of effectiveness for several major
title IV federal postsecondary programs and tax preferences, including the
Federal Work-Study program, the tuition deduction, tax-exempt savings
programs, and others. For other federal programs, no research exists on
important aspects of program effectiveness. For example, no research has
examined the effects of federal postsecondary education tax credits on
students' persistence in their studies or on the type of postsecondary
institution they choose to attend.

    Limited Research about Federal Assistance Reaches Varying Conclusions about
    Some Aspects of Effectiveness

When research on the effectiveness of federal assistance does exist, it
reaches varying findings about title IV student aid and tax preferences.
Some studies find that title IV programs increase the rates of college
attendance and persistence, while others have identified no positive
effects. Research on Pell Grants shows that they generally have little or
no impact on attendance, with the exception of one study that found Pell
Grants to have increased attendance for students from 22 to 35 years of

35

age. The study attributed the program's impact to the fact that older
students generally attend less-expensive institutions where Pell Grants
represent a larger share of the cost of college than the same size grants
provided to students in general. The study also suggests that the limited
impact of Pell Grants on attendance in general may be due to the fact that
institutional aid awards may decrease at the same time as Pell Grant
awards increase, creating a substitution effect that could diminish the
impact of Pell Grants on attendance. According to the study, this occurs
less often for older students because they tend to go to institutions that
have less institutional aid for which Pell Grant awards might substitute.

Some research has also found that Pell Grants, like other grant programs,
appear to increase students' persistence toward completing their
studies.36

35 Hansen (1983), Kane (1995), and Kitmitto (2004) found that Pell Grants
had little or no impact on attendance, while Seftor and Turner (2002)
found they increased attendance for students of from 22 to 35 years of
age.

36 See Li (1999), Bettinger (2004), and Kitmitto (2004) for research on
the impact of Pell Grants on persistence. Angrist (1993) and Bound and
Turner (2002) found that the G.I. Bill and other veteran's benefits
increased the amount of college completed. Dynarski (2003) found that the
Social Security Student Benefit Program increased college completion as
well. Dynarski (2004) found that merit aid increased college completion.

Student loans have been found to modestly increase college attendance,
persistence, and choice, but there are various limitations to consider. Of
the three studies examining the effect of borrowing on college attendance,
only one focuses on lower-income students.37 The study estimates that a
$1,000 increase (in 1977 dollars) resulted in a 4.3 percentage point
increase in college enrollment among dependent students with family
incomes below $15,000. However, given long-term changes in lending
markets, returns to schooling, and other conditions, students' behavioral
responses to equivalent changes in loan amounts may be different today
from what they were in 1977.38 Findings about the impact of loans on
persistence and choice are each based on only one study, and focus only on
middle-income students.39

We found one study concerning how the Hope and Lifetime Learning tax
credits affect college attendance. The study found the credits to have no
effect on college attendance. This may be because the students who receive
these credits are likely to attend college anyway.40 The author
acknowledges several limitations of the study. For example, the study uses
income categories-as opposed to actual income-thereby introducing
measurement error and attenuating the estimated effects of the credits on
attendance. We also note that the study measured eligibility for the
credits, rather than the receipt of tax credits. Measuring eligibility
rather than the receipt of credits tends to underestimate the effect of
credits on attendance because many tax filers who appear to be eligible
for the credits do not

37 See Reyes (1995), Dynarski (2002), and Long (2004), which focuses on
low-income students.

38 Likewise, because the composition of financial aid has changed-the
frequency and levels of borrowing have increased-an equivalent (real)
change in loan amounts may elicit a different response. In particular,
some students, especially lower-income students, may already have large
loan levels and be less willing to increase their loan debt than they were
in 1977.

39 See Reyes (1995) for estimates on how loans affect persistence and
Dynarski (2002) for information on how loans affect college choice.

40 See Long (2003b), which does not separately examine the Hope and the
Lifetime Learning tax credits. A few additional papers have simulated the
effect of the Hope tax credit on college attendance, including Cronin
(1997) and Cameron and Heckman (1999). These estimates, however, are based
on the findings of how students respond to other financial aid policies,
which may be different from the response to these tax credits. In
addition, Cronin (1997) bases her analysis on a proposed version of the
tax credit as opposed to the enacted tax credit.

claim them.41 This study did not examine whether federal student tax
credits affect persistence or choice.

In addition to the research into the federal tax, grant, and loan programs
described above, a large body of research estimates the effects of both
tuition changes and non-title IV financial aid programs on postsecondary
attendance.42 These studies all found either an inverse relationship
between the cost of tuition and level of enrollment or that financial
assistance increased enrollment.43 One survey of this work concludes that
a $1,000 reduction in net costs (in 2001 dollars) would result in a 3 to 6
percent increase in college enrollment rates.44 Federal Pell Grants and
postsecondary tax credits do not show similar effects in the studies we
examined. Substitution effects caused by offsetting reductions in
nonfederal aid may diminish the enrollment effect of Pell awards. Also,
the impact of tax credits on enrollment may be limited by the fact that
tax filers receive higher education tax credits the year following tuition
outlays. Program design and information may also account for some of these
differences: tuition information and many nonfederal aid benefits are
known when a person is choosing whether or not to attend college; however,
students may not be aware of tax credits and Pell Grants or their effects
on price until after they have already decided to go to college.

While federal grants, student loans, and tax credits were created to
result in beneficial consequences, such as increasing college attendance,
they have

41 Bershadker and Cronin (2004).

42 There is also research on the effects of financial aid policy, in
general, on choice. Linsenmeier, Rosen, and Rouse (2002) and Van Der
Klaauw (2002) found that financial aid policy at specific schools
positively affected the likelihood of whether accepted students chose to
attend those schools. Avery and Hoxby (2003), focusing on high aptitude
students, found that larger amount of grants, loans, and work-study made
from all sources-federal, state, and institutions-were associated with
students who were more likely to attend that school. Although it is unlike
any existing federal program, Dynarski (2000) found that the Georgia Hope
program shifted some would-be 2-year school students to 4-year schools and
caused some students to choose to stay within state for college.

43 See Leslie and Brinkman (1988), McPherson and Shapiro (1991), Rouse
(1994), Kane (1995) and (1999), and Cameron and Heckman (1999) on the
effects of tuition changes. See studies of the Georgia Hope Scholarship by
Dynarski (2000) and Cornwell, Mustard, and Sridhar (2004) as well as those
conducted by Bound and Turner (2002) on the G.I. Bill, by Dynarski (2003)
on the Social Security Benefit Program, and by Kane (2003) on the Cal
Grant Program.

44 Kane (2002).

    Data and Methodological Challenges Hinder Research

also been found to result in the raising of tuition by institutions in
some circumstances.45 One potential explanation is that institutions raise
tuition because federal aid increases the amount students are able to pay.
Alternatively, federal aid may increase the demand for education, and with
no offsetting factors, this drives the price of tuition up.

Gaps in the research-based evidence of federal postsecondary program
effectiveness may be due, in part, to data and methodological challenges
that have proven difficult to overcome. The relative newness of most of
the tax preferences also presents challenges because relevant data are
just now becoming available. Additionally, to analyze the tax programs,
actual tax return data are preferred to indirect or processed data, but
researchers are often limited to publicly available data containing
self-reported information on income because of tax data confidentiality
protections.

Methodological challenges add to the difficulty of estimating the effects
of these programs. In general, researchers can reliably identify the
behavioral effects of a program when one factor changes while all others
remain unchanged. For example, if no other factors change, a researcher
could identify the impact of postsecondary tax credits on enrollment by
comparing enrollment rates before and after the tax credits are enacted.
Typically, however, many factors change simultaneously, either offsetting
or enhancing the effect of the policy intervention being studied. For
example, tuition rates may rise at the same time that postsecondary tax
credits are introduced, or other sources of postsecondary assistance-such
as federal, state, or institutional grants-may decrease. These changes
undermine researchers' ability to reliably isolate the behavioral effects
of the policy under study. While researchers in some fields address this
problem through the use of experiments, few opportunities exist for
experimentation in the study of postsecondary education finance.
Alternatively, researchers may address this problem through the use of
quasi-experimental research designs, which attempt to approximate the
random assignment of participants to treatment and control groups by
matching participants to nonparticipants having similar characteristics.
In practice, however, data limitations often make this difficult to
implement. To isolate the impact of tax credits on college attendance, for
example,

45 For studies of the effect on tuition, see Li (1999) for Pell Grants,
Long (2003b) for tax credits, and Acosta (2001) for federal grant and
federal loan aid. The Georgia Hope Scholarship has also been found to
increase college costs by Long (2003a).

Page 31 GAO-05-684 Federal Student Aid Programs

    Steps to Support Effectiveness Research

researchers may need to compare credit recipients with nonrecipients who
are similar in a range of ways-including academic preparation, family
income and wealth. National surveys often do not contain complete data on
all of these necessary factors both for those who attend and do not attend
college.

In 2002, we recommended that Education sponsor research into key aspects
of effectiveness of title IV programs, and that Education and the
Department of the Treasury collaborate on such research into the relative
effectiveness of title IV programs and tax preferences.46 In order to
provide Congress with information about the effectiveness of title IV
programs, we recommended in 2002 that Education sponsor research on the
impact of title IV programs on postsecondary attendance, choice,
completion, and costs. To provide information about the relative
effectiveness of Education's direct expenditure programs and Treasury's
postsecondary tax provisions, we recommended that the Secretaries of
Education and Treasury collaborate in studying the combined effects of tax
preferences and title IV aid. Few steps have been taken to implement these
recommendations. However, Education is in the process of establishing a
postsecondary research center that will, among other things, examine the
impact of title IV programs.

As we noted in our 2002 report, research into the effectiveness of
different forms of postsecondary education assistance is important.
Without such information federal policymakers cannot make fact-based
decisions about how to build on successful programs and make necessary
changes to improve less effective programs. As we stress in our recent
report

21st Century Challenges: Reexamining the Base of the Federal Government47,
the budget deficit and other major fiscal challenges facing the nation
necessitate rethinking the base of existing federal spending and tax
programs, policies, and activities by reviewing their results and testing
their continued relevance and relative priority for a changing society.

46 GAO, Student Aid and Tax Benefits: Better Research and Guidance Will
Facilitate Comparison of Effectiveness and Student Use, GAO-02-751
(Washington, D.C.: September 13, 2002).

47 GAO, 21st Century Challenges: Reexamining the Base of the Federal
Government, GAO- 05-325SP (Washington, D.C.: February 2005).

Page 32 GAO-05-684 Federal Student Aid Programs

  Concluding Observations

In our January 2004 report on OMB's Program Assessment Rating Tool
(PART),48 we recommended that OMB target PART assessments based on such
factors as the relative priorities, costs, and risks associated with
related clusters of programs and activities and that OMB select related or
similar programs for review in the same year to facilitate comparisons and
trade-offs.49 Furthermore, in our June 14, 2005 testimony before the
Senate Committee on Homeland Security and Governmental Affairs,
Subcommittee on Federal Financial Management, Government Information, and
International Security, we reported that it is often critical to
understand how programs fit with a portfolio of tools and strategies in
order to capture whether programs complement and support related programs,
are duplicative or redundant, or work at cross purposes to other
initiatives.50 The different tax preferences and title IV programs in
place to help students and families finance postsecondary education are a
good example of the sort of related clusters of programs and activities to
which we were referring.

Congress has adopted a range of tools to help students and families pay
for postsecondary education, including grants, loans, and tax preferences.
Many title IV financial aid programs have a long history, while most of
the tax preferences do not. The addition of tax preferences to the title
IV grant and loan programs has increased the number of options and given
students and families new choices about how to combine saving, borrowing,
and current income to meet the costs of postsecondary education.
Postsecondary tax preferences are widely thought to present challenges of
complexity to students and families. As we have shown, tax filers appear
to have some difficulty in making fully effective use of some
postsecondary

48 GAO, Performance Budgeting: Observations on the Use of OMB's Program
Assessment Rating Tool for the Fiscal Year 2004 Budget, GAO-04-174
(Washington, D.C.: January 30, 2004).

49 OMB describes the PART as a diagnostic tool meant to provide a
consistent approach to evaluating federal programs as part of the
executive budget formulation process. It applies 25 questions under four
broad topics: (1) program purpose and design, (2) strategic planning, (3)
program management, and (4) program results. It also uses additional
questions specific to the mechanism or approach used to deliver the
program, such as grants or credit programs (e.g. student loans). OMB has
not systematically applied the PART to tax preferences.

50 21st Century Challenges: Performance Budgeting Could Help Promote
Necessary Reexamination, GAO-05-709T (Washington D.C.: June 14, 2005).

  Agency Comments and Our Evaluation

tax preferences now in the tax code. Also, although millions of tax filers
are receiving billions of dollars in assistance with postsecondary costs
through both title IV programs and tax preferences, little is known about
the effectiveness of any of these forms of assistance in part because of
data and methodological issues. As we recommended in our 2002 report on
federal financial aid,51 it is important that information about the
effectiveness of both tax preferences and title IV federal grant and loan
programs be developed so that decision makers in Congress and in the
executive branch can make efficient use of limited federal resources and
reexamine, if necessary, the tools used to help students and families pay
for postsecondary education.

We provided a draft of this report to the Internal Revenue Service, the
Department of the Treasury, and the Department of Education for review and
comment. The Internal Revenue Service provided technical comments, which
we incorporated. The Department of the Treasury did not provide comments
on our report.

In its written comments, Education disagreed with our conclusion on the
extent that title IV programs have been studied. Education also said that
we did not properly acknowledge the role that its National Postsecondary
Student Aid Study (NPSAS) played in our analysis and that the report did
not cite many Departmental publications prepared on the basis of NPSAS and
other Departmental data collections, such as Persistence and

Attainment of Beginning Students with Pell Grants.52 In our report, we
describe NPSAS as a comprehensive study examining how students and their
families pay for postsecondary education and note that we relied upon
NPSAS data to conduct our analysis. Further, we explain in our report that
our findings about the effectiveness of federal postsecondary assistance
are based upon studies that meet professional standards of econometric
analysis and contain acceptably identified statistical estimates

51 GAO-02-751.

52 Wei, C.C., and Horn, L., Persistence and Attainment of Beginning
Students with Pell Grants, NCES 2002-169, U.S. Department of Education,
National Center for Education Statistics (Washington, D.C.: May 2002).

Page 34 GAO-05-684 Federal Student Aid Programs

of program effects. We do not include publications that fail to meet these
standards. While the data collections cited by the Department provide
useful descriptive information concerning title IV programs, they do not
provide data sufficient to determine key program effects. In particular,
because these data collections do not contain information about those who
do not attend postsecondary institutions, they are of limited use in
establishing the effectiveness of title IV programs, especially with
respect to postsecondary attendance-a challenge we note in our report.
Consequently, publications based upon these data share this limitation.

With respect to the study specifically cited by the Department-Persistence
and Attainment of Beginning Students with Pell Grants-it does not contain
acceptably identified estimates of Pell Grant effects on persistence. In
particular, the study does not implement a design and include variables
that can isolate the effect of Pell Grant receipt on persistence from
other factors that are associated both with persistence and with Pell
Grant receipt, including academic preparation, changes in family income,
and the net cost of education. Thus the study's estimate of the effect of
Pell Grants on persistence may reflect not only the influence of the Pell
Grant, but also the influence of these other factors. Consequently, its
results cannot be used to reliably assess the impact of Pell Grant receipt
on persistence.53

Although Education stated that our report presented an incomplete and
inaccurate assessment of its research, it nonetheless agreed that more
research should be done and that it is "committed to continuing to
increase its research associated with the effectiveness of the [its]
programs." The Department expressed a similar commitment in response to
our September 2002 report54 which found that Education had undertaken
little work identifying the impact of its grant and loan programs and
recommended that the Department sponsor research on the impact of title IV
programs on postsecondary education attendance and choice, completion, and
costs.

53 A similar conclusion is reached in Review of NCES Research on Financial
Aid and College Participation and Omitted Variables and Sample Selection
Issues in the NCES Research on Financial Aid and College Participation,
Reports Prepared for the Advisory Committee on Student Financial
Assistance by Donald E. Heller and William E. Becker (September 2003).

54 GAO, Student Aid and Tax Benefits: Better Research and Guidance Will
Facilitate Comparison of Effectiveness and Student Use, GAO-02-751
(Washington, D.C.: Sept. 13, 2002).

As agreed with your offices, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 30 days
after its date. At that time, we will send copies of this report to The
Commissioner of Internal Revenue, The Secretary of Education, The
Secretary of the Treasury, and other interested parties. This report is
available at no charge on GAO's web site at http://www.gao.gov.

If you or any of your staff have any questions, please contact Michael
Brostek at (202) 512-9110 or Cornelia Ashby at (202) 512-7215. You may
also reach us by e-mail at [email protected] and [email protected]. Contact
points for our Offices of Congressional Relations and Public Affairs may
be found on the last page of this report. Key contributors to this report
were David Lewis, Assistant Director, Jeff Appel, Assistant Director, Eric
Mader, Thomas Weko, John Mingus, Cynthia Decker, Jeffrey Weinstein, and
Katherine France.

Michael Brostek Director, Tax Issues Strategic Issues

Cornelia M. Ashby Director, Education Workforce and

Income Security Appendix I

                       Objectives, Scope, and Methodology

Our review focused on answering the following three questions: (1) How
does title IV grant and loan assistance compare with that provided through
the tax code? (2) To what extent are tax filers effectively using the
opportunities presented by postsecondary tax preferences? (3) What is
known about the effectiveness of federal assistance in promoting college
attendance, providing students with a wider range of choices among
postsecondary institutions, or encouraging students to persist in their
studies?

To compare title IV programs and tax preferences, we reviewed articles,
studies, and reports on federal assistance for postsecondary education
published by the Joint Committee on Taxation, the Internal Revenue Service
(IRS), the Department of Education's National Center for Education
Statistics and Office of Federal Student Aid, and other sources. Programs
or tax preferences that served more than 500,000 participants were judged
to be, for the purpose of our review, major. To obtain information on
participant numbers and other comparative information, we used (1) fiscal
year 2004 information from the President's fiscal year 2006 budget
request,1 (2) IRS's 2002 Statistics of Income (SOI) data set including IRS
Form 1098-T information for all tax filers in the 2002 SOI sample, (3)
2003-2004 school year data from the National Postsecondary Student Aid
Study (NPSAS),2 (4) data as of December 31, 2004, on the number of
accounts and amounts invested in section 529 Qualified Tuition Programs
from the College Savings Plans Network,3 and (5) tax year 2002 data on the
estimated number of contributions to Coverdell Education Savings Accounts
from IRS. We also reviewed studies conducted by Congressional Research
Service, the Congressional Budget Office, GAO, the Department of the
Treasury, the Urban Institute, and the College Board. We also interviewed
individuals from the Congressional Research Service, Education, IRS,
Treasury, and universities.

The 2002 SOI data and 2003-2004 NPSAS data were the most recent data
available. The SOI individual tax return file is a stratified probability

1 Office of Management and Budget, Appendix, Budget of the United States
Government, Fiscal Year 2006 (Washington, D.C.: Feb 7, 2005).

2 The NPSAS school year begins July 1 and ends on June 30 of the following
year.

3 The College Savings Plan Network collects information from the states
about the numbers of 529 accounts. The data are voluntarily provided by
the states. On the basis of our interview with College Savings Plan
Network staff, we determined that these data were sufficiently reliable
for our use in this study.

Appendix I Objectives, Scope, and Methodology

sample of income tax returns filed with IRS. The SOI sample of 175,000
returns represented the approximately 130 million tax returns filed for
2002. NPSAS is a comprehensive study that examines how students and their
families pay for postsecondary education. It includes nationally
representative samples of 79,852 undergraduates, 9,611 graduate students,
and 1,283 first-professional students enrolled during the 2003-2004
academic year. The NPSAS data are based on student interviews and
administrative records, and NPSAS includes survey results from both
students who received financial aid and those who did not. To assess the
reliability of the SOI and NPSAS sample data, we reviewed existing
information about the samples and performed electronic testing of the
required data elements to detect obvious problems in accuracy and
completeness. We determined that SOI and NPSAS data, as well as other data
used to provide certain specific pieces of information, were sufficiently
reliable for this report.

Because estimates from the SOI and NPSAS data are based on samples, they
are subject to sampling errors. These sampling errors measure the extent
to which the point estimates may vary from the actual values in the
population of tax filers. Each of our estimates is surrounded by a 95
percent confidence interval: an interval that 95 times out of 100 will
contain the true population value. The upper and lower bounds of the 95
percent confidence intervals for each estimate are presented in the tables
in appendix II.

To examine the extent to which tax filers are effectively using
postsecondary education tax preferences, we used much of the same
information we obtained for the description and comparison of the programs
and tax preferences. In addition we estimated the number of tax filers who
were eligible for an education tax credit or tuition and fees deduction
but either did not claim one at all or appeared to make a
less-than-optimal choice among these tax preferences and the amounts of
tax benefits lost as a result. To do this analysis we used IRS's SOI
sample of individual tax returns for tax year 2002 and all Form 1098-T
information returns for tax filers in the sample. Postsecondary
institutions participating in Education's student aid programs are
required to issue Form 1098-Ts to all enrolled students. Form 1098-Ts
include the student's name, address, and social security number, and the
school's taxpayer identification number (TIN). Form 1098-Ts also indicate
if the student was a graduate student and if he or she was enrolled at
least half-time. Postsecondary institutions had the option of providing
information concerning students' educational expenses, scholarships, and
grants but were not required to do

Appendix I Objectives, Scope, and Methodology

so. By combining information on the Form 1098-Ts with information on the
tax return, we were able to identify the postsecondary student population
in the SOI sample and the choices that tax filers made concerning
education-related tax preferences.

To conduct the analysis, we had to exclude two types of tax returns from
consideration. Because a person cannot claim the tuition deduction or the
Hope or Lifetime Learning credits if he or she is claimed as a dependent
on someone else's tax return, we excluded dependent tax returns in the SOI
sample from our analysis. We also excluded the tax returns of tax filers
that received a Form 1098-T with no information concerning students'
educational expenses because we could not analyze the tax returns without
these data. This included the tax returns of individuals who received an
education tax credit or tuition deduction but did not receive a Form
1098-T. These limitations excluded 87 percent of the returns in the
sample.

We considered different explanations for why those tax filers with
education expenses did not claim an education tax credit or tuition
deduction. For example, we examined whether tax filers had (1) income that
exceeded the program thresholds for tax year 2002, (2) no taxable income,
(3) no tax liability after claiming other tax credits, or (4) no net
educational expenses after accounting for scholarships and grants as
reported on the Form 1098-T. We also examined whether tax filers were
married filing separately or filed a Form 1040EZ because this would
prevent tax filers from being able to claim the education tax credits or
tuition deduction.

We calculated tax filers' optimal choice among the Hope and Lifetime
Learning credits, and the tuition deduction on the basis of program
eligibility criteria for tax year 2002. Tax filers are limited to claiming
either a tuition deduction or an education tax credit for the same
student. Eligibility is restricted by modified adjusted gross income and,
in the case of the Hope tax credit, whether or not students are enrolled
at least halftime.

We shared our methodology for this analysis with tax policy researchers
outside of GAO and incorporated their comments into our analysis.

To identify available academic research on the effectiveness of major
federal financial aid programs, we reviewed studies that examined whether
the programs or tax preferences affect college attendance, persistence,
and choice. We looked for these measures because they have been the focus
of

Appendix I Objectives, Scope, and Methodology

congressional concern as expressed in committee reports, statutorily
established study commissions, and requests for our work from Congress. We
examined studies we found through searches in EconLit, Digital
Dissertations from ProQuest, and the National Bureau of Economic Research
web site. These online sources are nationally recognized repositories of
research results. In addition, we also examined relevant studies cited in
studies found from our searches. Some of these studies were excluded from
further assessment because they did not undertake original data analysis
that could identify the effectiveness of federal financial aid programs.
We assessed studies that provided an original empirical analysis according
to professional standards of econometric analysis for their methodological
rigor. The results of the studies that we judged to contain acceptably
identified statistical estimates formed the basis for our findings about
the availability of information concerning the relative effectiveness of
major federal financial aid programs.

We conducted our review from May 2004 through June 2005 in accordance with
generally accepted government auditing standards.

Appendix II

Confidence Intervals

We used two data sets in this review: Education's 2003-2004 National
Postsecondary Student Aid Study and the Internal Revenue Service's 2002
Statistics of Income. Estimates from both data sets are subject to
sampling errors and the estimates we report are surrounded by a 95 percent
confidence interval. The following tables provide the lower and upper
bounds of the 95 percent confidence interval for all estimate figures in
the tables in this report. For figures drawn from these data, we provide
both point estimates and confidence intervals.

 Table 7: Description of Federal Student Aid Programs Authorized under Title IV
                          of the Higher Education Act

               Number of         Total award       Average award Median       
               recipients                                        income       
                Lower Upper                          Lower Upper  Lower Upper 
Type of      bound bound      Lower bound Upper   bound bound  bound bound 
assistance                                bound               
Dependent                                                     
students                                                      

Pell Grant   2,026,011 2,115,312  5,201,091,600 5,452,845,564  2,543 2,573 24,165 24,999 
Supplemental   530,408   577,316    466,079,305  522,325,472    857   892  22,022 23,484 
Educational                                                                       
Opportunity                                                                       
Grant                                                                             
Federal      1,023,755 1,089,687  1,927,247,135 2,090,819,033  1,856 1,901 45,000 48,231 
Work- Study                                                                       
Federal        472,640   517,207    907,800,538 1,004,290,295  1,887 1,932 37,623 40,814 
Perkins Loan                                                                      
Subsidized   2,505,118 2,604,668  7,962,531,788 8,329,729,995  3,155 3,188 43,834 45,446 
FFEL or                                                                           
Direct                                                                            
Stafford                                                                          
Loan                                                                              
Unsubsidized 1,578,160 1,664,757  5,173,481,648 5,505,576,910  3,244 3,293 74,263 77,439 
FFEL or                                                                           
Direct                                                                            
Stafford                                                                          
Loan                                                                              
FFEL or        609,125   659,071  5,458,550,634 5,979,275,038  8,787 9,019 69,547 73,439 
Direct PLUS                                                                       
Loan                                                                              
Independent                                                                       
students                                                                          
Pell Grant   2,967,340 3,087,638  7,212,123,299 7,540,282,035  2,409 2,436 12,614 13,262 
Supplemental   684,528   745,839    368,492,546  415,343,758    526   548  10,425 11,626 
Educational                                                                       
Opportunity                                                                       
Grant                                                                             
Federal        676,216   766,317    933,916,755 1,084,530,206  2,192 2,303 9,808  11,525 
Work- Study                                                                       
Federal        522,918   595,499    839,749,704  970,851,318   2,648 2,752 9,181  11,628 
Perkins Loan                                                                      
Subsidized   3,658,692 3,869,237 15,604,880,694 17,068,144,196 4,244 4,340 18,754 20,148 
FFEL or                                                                           
Direct                                                                            
Stafford                                                                          
Loan                                                                              
Unsubsidized 3,154,948 3,359,231 17,728,962,613 19,212,909,259 5,531 5,671 21,190 23,095 
FFEL or                                                                           
Direct                                                                            
Stafford                                                                          
Loan                                                                              
FFEL or              0         0              0              0     0     0      0      0 
Direct PLUS                                                                       
Loan                                                                              

Source: GAO analysis of 2003-2004 National Postsecondary Student Aid Study
data.

                        Appendix II Confidence Intervals

            Table8: Selected Postsecondary Education Tax Preferences

             Number of returns       Total benefits        Average     Median income
                                                           benefit     
               Lower     Upper                             Lower Upper Lower   Upper 
Type of        bound     bound Lower bound   Upper bound   bound bound bound   bound 
assistance                                                                    
Hope       3,115,595 3,414,023 3,064,601,005 3,399,426,275   965 1,016 37,506 41,004 
Credit                                                                        
Lifetime                                                                             
Learning   3,307,354 3,612,179 1,560,825,683 1,740,857,453   462   493 38,060 41,001
Credit                                                                        
Student                                                                              
Loan       6,432,399 6,849,170   848,115,632   937,085,664   129   140 42,378 44,657
Interest                                                                      
Deduction                                                              
Tuition                                                                              
Deduction  3,295,741 3,599,012 1,226,452,349 1,370,953,823   364   391 51,808 56,842

          Source: GAO analysis of Statistics of Income data for 2002.

Table 9: Tax Filers Claiming an Education Tax Credit or Tuition Deduction

                                 1998      1999      2000      2001      2002 
Hope Credit,      Lower  4,482,106 6,233,732 6,606,583 6,997,019 9,319,692 
Lifetime Learning bound                                          
Credit, and       Upper                                                    
Tuition Deduction bound  4,827,719 6,639,576 7,024,049 7,428,088 9,809,833

          Source: GAO analysis of Statistics of Income data for 2002.

Table 10: Percentage of Aid Recipients and Dollars of Aid by Income
Category for Dependent Students Served by Selected Title IV Programs

              Dependent          $0- $20,001- $40,001- $60,001- $80,001-     More 
                                                                             than 
Program      students         20,000  40,000   60,000   80,000  100,000  $100,000 
Pell Grant   Recipients Lower  36.66    45.41    13.17     1.41        0        0 
                        bound                                            
                        Upper  38.89    47.72    14.76     2.02        0        0 
                        bound                                            
             Dollars    Lower  46.29    42.41     7.38     0.65        0        0 
                        bound                                            
                        Upper  48.82    44.89      8.5     1.04        0        0 
                        bound                                            
Stafford     Recipients Lower  15.41    26.79    22.45     16.1     8.38     6.23 
                        bound                                            
Subsidized              Upper  16.94    28.73     24.3    17.72     9.61     7.33 
Loan                    bound                                            
             Dollars    Lower  15.32    27.14    22.83    15.68     7.92     5.87 
                        bound                                            
                        Upper  17.07    29.35    24.94    17.51      9.3     7.08 
                        bound                                            
Stafford     Recipients Lower   6.51    12.83    13.15    17.69    16.68       27 
                        bound                                            
Unsubsidized            Upper   7.88    14.76    15.21    19.94    18.84     29.5 
Loan                    bound                                            
             Dollars    Lower   6.22    11.05    11.31    16.69    17.55     30.3 
                        bound                                            
                        Upper   7.75    12.99    13.41     19.2    20.15    33.37 
                        bound                                            

Source: GAO analysis of 2003-2004 National Postsecondary Student Aid Study
data.

                        Appendix II Confidence Intervals

Table 11: Percentage of Aid Recipients and Dollars of Aid by Income
Category for Independent Students Served by Selected Title IV Programs

                                   $20,001- $40,001- $60,001- $80,001-     More 
                                                                           than 
Program                        $0-   40,000  60,000   80,000   100,000 $100,000 
                              20,000                                   
Pell Grant   Recipients Lower  66.28  26.59     4.59        0        0        0 
                        bound                                          
                        Upper  68.35  28.57     5.62        0        0        0 
                        bound                                          
             Dollars    Lower  71.68  23.62     2.32        0        0        0 
                        bound                                          
                        Upper  73.77  25.65     2.96        0        0        0 
                        bound                                          
Stafford     Recipients Lower  49.67  27.54    10.78     4.04      1.3     0.86 
                        bound                                          
Subsidized              Upper  52.62  30.38    13.48     5.36     1.98     2.38 
Loan                    bound                                          
             Dollars    Lower  49.93  25.26    10.05     3.87      1.2     0.46 
                        bound                                          
                        Upper  54.61  29.79    14.73      5.4     2.05     2.65 
                        bound                                          
Stafford     Recipients Lower  44.65  26.59    12.09     5.48     2.31     2.26 
                        bound                                          
Unsubsidized            Upper  47.82  29.75    15.18     6.87     3.18     4.08 
Loan                    bound                                          
             Dollars    Lower  44.28  22.51    11.96     6.22     2.86     3.42 
                        bound                                          
                        Upper  48.37     26    14.78     8.49     4.12     6.99 
                        bound                                          

Source: GAO analysis of 2003-2004 National Postsecondary Student Aid Study data

Table 12: Percentage of Tax Filers Claiming Hope and Lifetime Learning
Credits and Tuition Deduction and Tax Preference Dollars by Income
Category, Tax Year 2002

                                   $20,001- $40,001- $60,001- $80,001-     More 
                                                                           than 
 Type of                 $0-20,000  40,000    60,000   80,000 100,000  $100,000 
 Aid                                                                   
 Hope      Tax     Lower     16.63    30.43    18.06     16.2     9.45      0.1 
 Credit    filers  bound                                               
                   Upper     20.29    34.75    21.67     19.7    12.27     0.62 
                   bound                                               
           Dollars Lower      9.14    30.19    20.37    21.75     8.66        0 
                   bound                                               
                   Upper     11.82    35.06    24.77    26.52     11.7     0.03 
                   bound                                               
 Lifetime  Tax     Lower     14.59    32.04    23.06    14.74     6.73     0.14 
           filers  bound                                               
 Learning          Upper     17.89    36.25    26.85    18.03     9.14     0.67 
 Credit            bound                                               
           Dollars Lower     10.73    31.78    22.56    18.58     5.98        0 
                   bound                                               
                   Upper     13.97    37.02    27.24    23.32     8.79     0.02 
                   bound                                               
 Tuition   Tax     Lower     18.38    16.56    14.81    11.03    14.08    15.23 
           filers  bound                                               
 Deduction         Upper     21.91    20.04    18.14    14.01    17.33    18.49 
                   bound                                               
           Dollars Lower      9.41     8.83    11.55    11.53    20.79    26.16 
                   bound                                               
                   Upper     11.93    11.63     15.1    15.52    25.91    31.65 
                   bound                                               
                             Source: GAO analysis of 
                       Statistics of Income data for 
                                               2002. 

                        Appendix II Confidence Intervals

Table 13: Percentage of Form 1098-Ts with Postsecondary Expense
Information in 2002: Point Estimates

Number of returns                                       Percent of returns 
1098Ts with expense information               1,795,180                 13 
1098Ts without expense information           12,356,444                 87 

          Source: GAO analysis of Statistics of Income data for 2002.

 Table 14: Percentage of Form 1098-Ts with Postsecondary Expense Information in
                           2002: Confidence Intervals

Number of returns: Lower bound  Number of returns: Upper bound Percent of
                                   returns: Lower bound Percent of returns:
                                   Upper bound          
1098Ts with                                                                
expense            1,687,744.88     1,902,614.62       11.97          13.4
information                                                     
1098Ts without                                                             
expense           12,087,410.46    12,625,476.86        86.6         88.03
information                                                     

          Source: GAO analysis of Statistics of Income data for 2002.

Table 15: Percentage of Taxpayers Apparently Eligible to Claim an
Education Tax Credit or Tuition Deduction in 2002: Point Estimates

                                        Number of returns  Percent of returns 
Total                                        1,795,180                 100 
Potentially eligible                         1,386,659                  77 
All Other                                      408,521                  23 

          Source: GAO analysis of Statistics of Income data for 2002.

Table 16: Percentage of Taxpayers Apparently Eligible to Claim an
Education Tax Credit or Tuition Deduction in 2002: Confidence Intervals

Number of Returns: Lower bound  Number of         Percent of   Percent of  
                                   returns: Upper    returns:     returns:    
                                   bound             Lower bound  Upper bound 
Total              1,795,176.75      1,795,179.75          100         100 
Potentially        1,290,394.34      1,482,923.26        74.83       79.66 
eligible                                                       
All other            360,292.26        456,749.64        20.34       25.17 
                  Source: GAO analysis of Statistics 
                            of Income data for 2002. 

                        Appendix II Confidence Intervals

Table 17: Percentage of Apparently Eligible Taxpayers to Claim an
Education Tax Credit or Tuition Deduction That Failed to Do So in 2002:
Point Estimates

                      Number of returns Percent of returns

                           Failed to claim 373,595 27

          Source: GAO analysis of Statistics of Income data for 2002.

Table 18: Percentage of Apparently Eligible Taxpayers to Claim an
Education Tax Credit or Tuition Deduction That Failed to Do So in 2002:
Confidence Intervals

         Number of returns:     Number of       Percent of         Percent of 
                                returns:         returns:            returns: 
                Lower bound       Upper bound     Lower bound     Upper bound 
Failed to                                                                  
claim         323,504.26        423,686.08           23.85           30.04

          Source: GAO analysis of Statistics of Income data for 2002.

Table 19: Amounts by Which Apparently Eligible Taxpayers Failed to Reduce
Their Tax Liability: Point Estimates

                                                    Inaction led to increased
                                                                tax liability
Median                                                               52.45 
Mean                                                                168.66 
10th percentile                                                       4.34 
25th percentile                                                      10.94 
75th percentile                                                      207.2 
90th percentile                                                     532.96 
Maximum value                                                        1,116 

          Source: GAO analysis of Statistics of Income data for 2002.

                        Appendix II Confidence Intervals

Table 20: Amounts by Which Apparently Eligible Taxpayers Failed to Reduce
Their Tax Liability: Confidence Intervals

                                                    Inaction led to increased
                                                                tax liability
Median: Lower bound                                                  34.69 
Median: Upper bound                                                  73.57 
Mean: Lower bound                                                   136.57 
Mean: Upper bound                                                   200.76 
10th percentile: Lower bound                                          3.01 
10th percentile: Upper bound                                          6.57 
25th percentile: Lower bound                                          8.66 
25th percentile: Upper bound                                         16.72 
75th percentile: Lower bound                                        137.73 
75th percentile: Upper bound                                        312.14 
90th percentile: Lower bound                                        429.22 
90th percentile: Upper bound                                        729.58 

          Source: GAO analysis of Statistics of Income data for 2002.

Table 21: Percentage of Apparently Eligible Taxpayers That Claimed the
Tuition Deduction but Would Have Been Better off Claiming the Lifetime
Learning Credit in 2002: Point Estimates

                                       Number of returns   Percent of returns 
Would have been better off                     50,908                   21 
claiming Lifetime Learning                            
Credit                                                

          Source: GAO analysis of Statistics of Income data for 2002.

Table 22: Percentage of Apparently Eligible Taxpayers That Claimed the
Tuition Deduction but Would Have Been Better off Claiming the Lifetime
Learning Credit in 2002: Confidence Intervals

Number of Returns: Number of returns: Percent of returns: Percent of
returns: Lower bound Upper bound Lower bound Upper bound

Would have been better off claiming Lifetime 34,819.89 70,274.77 14.53
29.33 Learning Credit

Source: GAO analysis of Statistics of Income data for 2002.

Appendix II Confidence Intervals

Table 23: Amounts by Which Apparently Eligible Taxpayers Could Have
Reduced Their Tax Liability in 2002: Point Estimates

                                                     Lifetime Learning Credit
                                                    produced larger reduction
Median                                                               50.67 
Mean                                                                 83.22 
10th percentile                                                       7.35 
25th percentile                                                      26.23 
75th percentile                                                      119.6 
90th percentile                                                     157.91 
Maximum value                                                          556 

          Source: GAO analysis of Statistics of Income data for 2002.

Table 24: Amounts by Which Apparently Eligible Taxpayers Could Have
Reduced Their Tax Liability in 2002: Confidence Intervals

Lifetime Learning Credit produced larger reduction

                           Median: Lower bound 32.89

                           Median: Upper bound 84.27

                            Mean: Lower bound 49.76

                            Mean: Upper bound 116.68

                         10th percentile: Lower bound .

10th percentile: Upper bound

25th percentile: Lower bound

25th percentile: Upper bound

75th percentile: Lower bound

                      75th percentile: Upper bound 148.53

                      90th percentile: Lower bound 106.35

                         90th percentile: Upper bound .

          Source: GAO analysis of Statistics of Income data for 2002.

                        Appendix II Confidence Intervals

Table 25: Percentage of Apparently Eligible Taxpayers That Claimed the
Lifetime Learning Credit but Would Have Been Better off Claiming the
Tuition Deduction in 2002: Point Estimates

                                       Number of returns   Percent of returns 
Would have been better off                     22,469                    8 
claiming the Tuition                                  
Deduction                                             

          Source: GAO analysis of Statistics of Income data for 2002.

Table 26: Percentage of Apparently Eligible Taxpayers That Claimed the
Lifetime Learning Credit but Would Have Been Better off Claiming the
Tuition Deduction in 2002: Confidence Intervals

                    Number of Returns: Number of    Percent of     Percent of 
                                       returns:     returns:         returns: 
                           Lower bound  Upper bound  Lower bound  Upper bound 
Would have been better    12,228.08     37,165.3         4.48        13.61 
off claiming the                                              
Tuition Deduction                                             

          Source: GAO analysis of Statistics of Income data for 2002.

Table 27: Amounts by Which Apparently Eligible Taxpayers Could Have
Reduced Their Tax Liability in 2002: Point Estimates

                                            Tuition Deduction produced larger
                                                                    reduction
Median                                                              108.05 
Mean                                                                137.68 
10th percentile                                                       17.3 
25th percentile                                                      36.42 
75th percentile                                                     191.55 
90th percentile                                                     237.42 
Maximum value                                                          456 

          Source: GAO analysis of Statistics of Income data for 2002.

Appendix II Confidence Intervals

Table 28: Amounts by Which Apparently Eligible Taxpayers Could Have
Reduced Their Tax Liability in 2002: Confidence Intervals

Deduction produced larger reduction

                           Median: Lower bound 37.39

                           Median: Upper bound 190.77

                            Mean: Lower bound 77.08

                            Mean: Upper bound 198.28

10th percentile: Lower bound

10th percentile: Upper bound

25th percentile: Lower bound

                      25th percentile: Upper bound 108.84

75th percentile: Lower bound

                      75th percentile: Upper bound 244.85

                      90th percentile: Lower bound 154.73

                      90th percentile: Upper bound 350.13

          Source: GAO analysis of Statistics of Income data for 2002.

Table 29: Percentage of Apparently Eligible Taxpayers That Claimed a Hope
Credit but Would Have Been Better off Claiming a Lifetime Learning Credit
                            in 2002: Point Estimates

                                       Number of returns   Percent of returns 
Total                                         271,494                  100 
Would have been better off                          0                    0 
claiming Lifetime Learning                            
Credit                                                
All other                                     271,494                  100 

          Source: GAO analysis of Statistics of Income data for 2002.

                        Appendix II Confidence Intervals

Table 30: Percentage of Apparently Eligible Taxpayers That Claimed a Hope
Credit but Would Have Been Better off Claiming a Lifetime Learning Credit
in 2002: Confidence Intervals

                    Number of Returns: Number of    Percent of     Percent of 
                                       returns:     returns:         returns: 
                           Lower bound  Upper bound  Lower bound  Upper bound 
Total                    271,491.04   271,494.04          100          100 
Would have been better            0            0            0            0 
off claiming                                                  
Lifetime Learning Credit                                      
All other                271,491.04   271,494.04          100          100 

          Source: GAO analysis of Statistics of Income data for 2002.

Table 31: Percentage of Suboptimal Choices Made by Paid Tax Preparers:
Point Estimates

                                           Taxpayers making suboptimal choice
                                                                    Number of
                                                          returns     Percent 
Total                                                  446,972         100 
No preparer                                            219,139       49.03 
Paid preparer                                          223,011       49.89 
IRS prepared/reviewed                                        0           0 
VITA/self help/outreach/elderly assistance               4,822        1.08 

          Source: GAO analysis of Statistics of Income data for 2002.

     Table 32: Percentage of Suboptimal Choices Made by Paid Tax Preparers:
                              Confidence Intervals

                                           Taxpayers Making Suboptimal choice
                            Number of returns: Number of    Percent: Percent: 
                                               returns:      Lower      Lower 
                                   Lower bound  Upper bound    bound    bound 
Total                               392,039      501,905    99.72      100 
No preparer                         179,777      258,500    42.87    55.19 
Paid preparer                       184,952      261,070    43.74    56.05 
IRS prepared/reviewed                     0            0        0     0.28 
VITA/self help/outreach/elderly       1,131        9,328     0.26     2.91 
assistance                                                        

          Source: GAO analysis of Statistics of Income data for 2002.

Appendix III

Postsecondary-Education-Related Tax Preferences

We analyzed the following postsecondary-education-related tax preferences
in detail in this review.

Lifetime Learning Credit: Income-based tax credit claimed by tax filer on
behalf of students enrolled in one or more postsecondary education
courses.

Hope Credit: Income-based tax credit claimed by tax filer on behalf of
students enrolled at least half-time in an eligible program of study and
who are in their first 2 years of postsecondary education.

Student Loan Interest Deduction: Income-based tax deduction claimed by tax
filer on behalf of students who took out qualified student loans while
enrolled at least half time.

Tuition and Fees Deduction: Income-based tax deduction claimed by tax
filer on behalf of students who are enrolled in one or more postsecondary
education course and have either a high school diploma or a General
Educational Development (GED) credential.

Section 529 Qualified Tuition Programs-College Savings Programs and
Prepaid Tuition Programs: Non-income-based programs that provide favorable
tax treatment to investments and distributions used to pay the expenses of
future or current postsecondary students.

Coverdell Education Savings Accounts: Income-based savings program
providing favorable tax treatment to investments and distributions used to
pay the expenses of future or current elementary, secondary, or
postsecondary students.

The following postsecondary-education-related tax preferences were not
included in this review.

Scholarships, Fellowships, Grants, and Tuition Reductions Income
Exclusion: Scholarships and fellowships paid directly to degree-candidate
students or to their educational institutions for tuition and fees are not
taxed as income. However, scholarships and fellowships covering room and
board or transportation or paid in return for services, such as teaching,
are taxable. Also, tuition reductions, for example discounts given to
employees of an educational institution or their children, are not counted
as income for tax purposes.

Appendix III Postsecondary-Education-Related Tax Preferences

Employer-Provided Education Benefits Exclusion: Financial assistance
provided by employers to employees up to $5,250 in 2004 to pay for
employee educational expenses is not counted as income for tax purposes.
Only funds used to pay for tuition, fees, books, equipment, and similar
expenses qualify. Funds from an employer and used to pay for meals,
lodging, or transportation count as income for tax purposes.1

Student Loan Forgiveness Exclusion: Student loan repayment assistance or
cancellation provided in exchange for working for a period of time in
certain professions for any of a broad class of employers is not treated
is taxable income.

Education Savings Bonds: Interest earned on U.S. savings bonds is not
taxed if the bond holder is paying postsecondary education tuition and
fees or making contributions to a 529 qualified tuition program or a
Coverdell education savings account. The exclusion is available to tax
filers with modified adjusted gross incomes below $74,850 ($119,750 if
married filing jointly or qualified widow(er)).

Business Expense Deduction of Work-Related Education: Tax filers may
deduct the cost of work-related education if the education is required by
their employer or the law to maintain the tax filer's present salary,
status, or job and maintains or improves skills needed in the tax filer's
present work. Education to meet the minimum educational requirements of
the tax filer's present trade or business or education towards a new trade
or business does not qualify. The tax filer must itemize deductions on
form 1040 Schedule A, C, or F. The amount of the deduction is the total of
work-related education expenses plus other job and certain miscellaneous
expenses that is in excess of 2 percent of adjusted gross income.

Uniform Transfers to Minors: Money paid directly to an educational
institution for another person's tuition are not subject to gift taxes.2

1 Under the Working Condition Fringe Benefit Exclusion, employer-provided
educational assistance that exceeds $5,250 may still not be counted as
income, provided it is used to pay for any educational expenses that are
required by the employer or the law to maintain the tax filer's present
salary, status, or job and maintain or improve skills needed in the tax
filer's present work.

2 This tax preference is not listed in the 2004 IRS Publication 970.

Page 52 GAO-05-684 Federal Student Aid Programs Appendix III
Postsecondary-Education-Related Tax Preferences

Early Withdrawals From Individual Retirement Accounts: The 10 percent
additional tax that applies to withdrawal of funds from an Individual
Retirement Account does not apply if the funds are used to pay for the
postsecondary education expenses of the account holder or his or her
dependent.

Parental Personal Exemption for Dependent Students: The tax code
definition of "dependent" for tax filing purposes involves 5 tests,
including whether the dependent is (1) a member of your household or
related to you, (2) a U.S. citizen or resident, (3) filing a joint tax
return, (4) earning less than $3,100, and (5) receiving more than half of
their support from the taxpayer claiming the dependent. In 2004, someone
over age 18, earning more than $3,100, and not living with the tax filer
throughout the year would likely not qualify as a dependent. However, the
tax code makes exceptions to these rules for students under age 24, thus
postsecondary education students are still dependents for tax purposes
while they are in school.3

3Ibid.

                                  Appendix IV

                   Comments from the Department of Education

Bibliography

Acosta, Rebecca J. "How Do College Respond to Changes in Federal Student
Aid?" Working Paper No. 808. Department of Economics, University of
California, Los Angeles, 2001.

Angrist, Joshua D. "The Effect of Veterans Benefits on Education and
Earnings." Industrial and Labor Relations Review, Vol. 46, No. 4 (1993):
637-52.

Avery, Christopher and Caroline M. Hoxby. "Do and Should Financial Aid
Packages Affect Students' College Choices?" Working Paper No. 9482.
Cambridge, Mass: National Bureau of Economic Research, 2003.

Bershadker, Andrew and Julie-Anne Cronin. "Winners (and Losers?) in the
Search for Higher Education Tax Subsidies." Unpublished draft, Office of
Tax Analysis, U.S. Department of the Treasury, 2004.

Bettinger, Eric. "How Financial Aid Affects Persistence." Working Paper
No. 10242. Cambridge, Mass: National Bureau of Economic Research, 2004.

Bound, John and Sarah Turner. "Going to War and Going to College: Did
World War II and the G.I. Bill Increase Educational Attainment for
Returning Veterans?" Journal of Labor Economics, Vol. 20, No. 4 (2002):
784-815.

Cameron, Stephen V. and James J. Heckman. "Can Tuition Policy Combat
Rising Wage Inequality?" in Marvin H. Kosters (ed.) Financing College
Tuition: Government Policies and Educational Priorities. Washington, D.C.:
The AEI Press, 1999.

Cornwell, Christopher, David B. Mustard, and Deepa J. Sridhar. "The
Enrollment Effects of Merit-Based Financial Aid: Evidence from Georgia's
Hope Scholarship." Unpublished manuscript, University of Georgia,
Department of Economics, 2004.

Cronin, Julie-Anne. "The Economic Effects and Beneficiaries of the
Administration's Proposed Higher Education Tax Subsidies." National Tax
Journal, Vol. 50, No. 3 (1997): 519-40.

Dynarski, Susan. "Hope for Whom? Financial Aid for the Middle Class and
Its Impact on College Attendance." National Tax Journal, Vol. 53, No. 3
(2000): 629-62.

Bibliography

-. "Loans, Liquidity, and Schooling Decisions." Cambridge, Mass.: Harvard
University, Kennedy School of Government and National Bureau of Economic
Research, 2002.

-. "Does Aid Matter? Measuring the Effect of Student Aid on College
Attendance and Completion." American Economic Review, Vol. 93, No. 1
(2003): 279-88.

-. "State Tuition Policy and Degree Completion." Unpublished manuscript,
Kennedy School of Government, Harvard University, 2004.

Hansen, W. Lee. "Impact of Student Financial Aid on Access" in Joseph
Froomkin (ed.) The Crisis in Higher Education. New York, N.Y.: Academy of
Political Science, 1983.

Kane, Thomas J. "Rising Public College Tuition and College Entry: How Well
Do Pubic Subsidies Promote Access to College?" Working Paper No. 5164.
Cambridge, Mass.: National Bureau of Economic Research, 1995.

-. The Price of Admission: Rethinking How Americans Pay for College.
Washington, D.C.: Brookings Institution Press, 1999.

-. "A Quasi-Experimental Estimate of the Impact of Financial Aid on
College Going." Working Paper No. 9703. Cambridge, Mass.: National Bureau
of Economic Research, 2002.

-. "Evaluating the Impact of the D.C. Tuition Assistance Grant Program."
Working Paper No. 10658. Cambridge, Mass.: National Bureau of Economic
Research, 2004.

Kitmitto, Sami. "The Effects of Pell Grants on Enrollment in Higher
Education." Unpublished manuscript, Department of Economics, University of
California-Davis, 2004.

Leslie, Larry and Paul Brinkman. The Economic Value of Higher Education.
Phoenix, Ariz.: American Council of Education and the Oryx Press, 1993.

Li, Judith Ann. "Estimating the Effect of Federal Financial Aid on Higher
Education: A Study of Pell Grants." Unpublished manuscript, Department of
Economics, Harvard University, 1993.

Bibliography

Linsenmeier, David M., Harvey S. Rosen, and Cecilia Elena Rouse.
"Financial Aid Packages and College Enrollment Decisions: An Econometric
Case Study." Working Paper No. 9228. Cambridge, Mass.: National Bureau of
Economic Research, 2002.

Long, Bridget Terry. "How do Financial Aid Policies Affect Colleges? The
Institutional Impact of the Georgia Hope Scholarship." Journal of Human
Resources, Vol. 39, No. 4 (2003a): 1045-66.

-. "The Impact of Federal Tax Credits for Higher Education Expenses."
Working Paper No. 9553. Cambridge, Mass.: National Bureau of Economic
Research, 2003b.

-. "How Does Availability of Loans Affect College Access?" Unpublished
presentation, Graduate School of Education, Harvard University, 2004.

McPherson, Michael S. and Morton Owen Shapiro. "Does Student Aid Affect
College Enrollment? New Evidence on a Persistent Controversy." American
Economic Review, Vol. 81, No. 1 (1991): 309-18.

Reyes , Suzanne. "Educational Opportunities and Outcomes: The Role of the
Guaranteed Student Loan." Unpublished manuscript, Department of Economics,
Harvard University, 1995.

Rouse, Cecilia Elena. "What to Do after High School: The Two-Year versus
Four-Year College Enrollment Decision" in Ronald G. Ehrenburg (ed.)

Choices and Consequences: Contemporary Policy Issues in Education.

Ithaca, N.Y.: ILR Press, 1994.

Seftor, Neil, and Sarah Turner. "Back to School: Federal Student Aid
Policy and Adult College Enrollment." Journal of Human Resources. Vol. 37,
No. 2 (2002): 336-52.

Van Der Klaauw, Wilbert. "Estimating the Effect of Financial Aid Offers on
College Enrollment: A Regression-Discontinuity Approach." International
Economic Review. Vol. 43, No. 4 (2002): 1249-1287.

  GAO's Mission

The Government Accountability Office, the audit, evaluation and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities and to help improve the performance and
accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and policies;
and provides analyses, recommendations, and other assistance to help
Congress make informed oversight, policy, and funding decisions. GAO's
commitment to good government is reflected in its core values of
accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no cost
is through GAO's Web site ( www.gao.gov ). Each weekday, GAO posts GAO
Reports and newly released reports, testimony, and correspondence on its
Web site. To

have GAO e-mail you a list of newly posted products every afternoon, go to
www.gao.gov and select "Subscribe to Updates."

                             Order by Mail or Phone

The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent of
Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more
copies mailed to a single address are discounted 25 percent. Orders should
be sent to:

U.S. Government Accountability Office 441 G Street NW, Room LM Washington,
D.C. 20548

To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061

Contact:

To Report Fraud, Web site: www.gao.gov/fraudnet/fraudnet.htm

  E-mail: [email protected]

Federal Programs Automated answering system: (800) 424-5454 or (202)
512-7470

Gloria Jarmon, Managing Director, [email protected] (202) 512-4400 U.S.
Government Accountability Office, 441 G Street NW, Room 7125 Relations
Washington, D.C. 20548

Paul Anderson, Managing Director, [email protected] (202) 512-4800

  Public Affairs

U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548
*** End of document. ***