Student Consolidation Loans: Further Analysis Could Lead to	 
Enhanced Default Assumptions for Budgetary Cost Estimates	 
(20-AUG-04, GAO-04-843).					 
                                                                 
The number of borrowers consolidating their federal student loans
has increased substantially in recent years, with the total	 
amount of loans being consolidated rising from $13 billion in	 
fiscal year 1999 to over $41 billion in fiscal year 2003. This	 
increase in consolidation loan volume and recent interest rate	 
trends have increased the overall estimated long-term cost to the
federal government of providing consolidation loans under the	 
Department of Education's (Education) two major student loan	 
programs--the Federal Family Education Loan Program (FFELP) and  
the William D. Ford Federal Direct Loan Program (FDLP). GAO is	 
providing information on (1) the differences that exist between  
FFELP and FDLP consolidation loans and borrowers, (2) the extent 
to which borrowers with student loans under one program obtain	 
consolidation loans under the other, and (3) how FFELP and FDLP  
borrower and loan characteristics and the movement of loans	 
between the two programs are incorporated into Education's	 
budgetary cost estimates for consolidation loans.		 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-04-843 					        
    ACCNO:   A11734						        
  TITLE:     Student Consolidation Loans: Further Analysis Could Lead 
to Enhanced Default Assumptions for Budgetary Cost Estimates	 
     DATE:   08/20/2004 
  SUBJECT:   Comparative analysis				 
	     Cost analysis					 
	     Direct loans					 
	     Future budget projections				 
	     Lending institutions				 
	     Loan defaults					 
	     Loan repayments					 
	     Student financial aid				 
	     Student loans					 
	     Consolidation					 
	     Dept. of Education Federal Family			 
	     Education Loan Program				 
                                                                 
	     Dept. of Education William D. Ford 		 
	     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-04-843

United States Government Accountability Office

 GAO	Report to the Chairman, Committee on the Budget, House of Representatives

August 2004

STUDENT CONSOLIDATION LOANS

 Further Analysis Could Lead to Enhanced Default Assumptions for Budgetary Cost
                                   Estimates

GAO-04-843

Highlights of GAO-04-843, a report to House Committee on the Budget.

The number of borrowers consolidating their federal student loans has
increased substantially in recent years, with the total amount of loans
being consolidated rising from $13 billion in fiscal year 1999 to over $41
billion in fiscal year 2003. This increase in consolidation loan volume
and recent interest rate trends have increased the overall estimated
long-term cost to the federal government of providing consolidation loans
under the Department of Education's (Education) two major student loan
programs-the Federal Family Education Loan Program (FFELP) and the William
D. Ford Federal Direct Loan Program (FDLP).

GAO is providing information on (1) the differences that exist between
FFELP and FDLP consolidation loans and borrowers, (2) the extent to which
borrowers with student loans under one program obtain consolidation loans
under the other, and (3) how FFELP and FDLP borrower and loan
characteristics and the movement of loans between the two programs are
incorporated into Education's budgetary cost estimates for consolidation
loans.

GAO recommends that Education consider the type of schools consolidation
borrowers attended in developing the risk categories for the department's
budgetary cost estimates. Education generally agreed with our
recommendation.

www.gao.gov/cgi-bin/getrpt?GAO-04-843.

To view the full product, including the scope and methodology, click on
the link above. For more information, contact Cornelia Ashby (202)
512-8403 [email protected].

August 2004

STUDENT CONSOLIDATION LOANS

Further Analysis Could Lead to Enhanced Default Assumptions for Budgetary Cost
Estimates

On average, during the 1995-to-2003 period, FFELP consolidation loan
borrowers had higher levels of consolidation loan debt than FDLP
consolidation loan borrowers. FFELP borrowers were also more likely than
FDLP consolidation borrowers to have attended a 4-year school versus a
2year or proprietary school. As a group, FFELP borrowers were less likely
to default on a student loan prior to consolidation than FDLP borrowers.
However, both FFELP and FDLP borrowers who had defaulted prior to
consolidation were more likely to default on their consolidation loan than
those who did not default prior to consolidation.

Over the 1998-to-2002 period, an increasing share of both FFELP and FDLP
underlying loan volume was consolidated into FFELP, while a decreasing
share of underlying loan volume was consolidated into FDLP. Defaulted
loans, however, whether from FFELP or FDLP, were much more likely to be
consolidated into FDLP.

In general, Education incorporates borrower and loan characteristics and
movement of loans between programs into its budgetary cost estimates by
(1) grouping loans with similar characteristics in risk categories, (2)
forecasting loan volume for each risk category, and (3) applying various
assumptions to each risk category based on historical and other economic
data. Education incorporates the default history of borrowers into its
cost estimates by grouping consolidation loans with underlying defaulted
loans in a risk category and applying higher default rate assumptions to
loans in this category. However, Education has not analyzed whether
borrowers with an underlying defaulted loan will default on their
consolidation loans at different rates based on the type of school
attended. Education does incorporate assumptions based on variations in
default rates by school type, but only for nonconsolidation loans. As
shown below, our analysis demonstrates that the extent to which borrowers
with an underlying defaulted loan default on their consolidation loan
varies according to the type of school they attended.

Default Rate of Consolidation Loan Borrowers Who Defaulted on a Loan
Underlying Their Consolidation Loan, by Program and School Type, Fiscal
Years 1995 to 2001

Percent

50.0 45.5 40.0

30.0

20.0

10.0

0 4-year 2-year Proprietary

2- and 4-year All other

FFELP FDLP

Source: GAO analysis of Education's National Student Loan Data System
data.

Contents 	

Letter                                                                   1 
                               Summary of Key Findings                      2 
           Further Analysis Could Lead to Enhanced Default Assumptions for 
                          Consolidation Loan Cost Estimates                 4 
                         Recommendation for Executive Action               10 
                                   Agency Comments                         11 
                                Scope and Methodology                      11 

Briefing Slides

Appendix I Comments from the Department of Education

  Figures

Figure 1: Default Rate of Consolidation Borrowers Who Defaulted

on a Loan Underlying Their Consolidation Loan, by

Program and School Type, Fiscal Years 1995 to 2001 6 Figure 2: Percentage
of FFELP Consolidation Loan Borrowers with an Underlying Default by School
Type 8 Figure 3: Percentage of FDLP Consolidation Loan Borrowers with an
Underlying Default, by School Type 9

Abbreviations

FDLP William D. Ford Direct Loan Program

FFELP Federal Family Education Loan Program NSLDS National Student Loan
Data System

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.

United States Government Accountability Office Washington, DC 20548

August 20, 2004

The Honorable Jim Nussle Chairman, Committee on the Budget House of
Representatives

Dear Mr. Chairman:

Consolidation loans-available under both of the Department of Education's
(Education) two major student loan programs, the Federal Family Education
Loan Program (FFELP) and the William D. Ford Direct Loan Program
(FDLP)1-allow borrowers who have multiple student loans, possibly from
different lenders and from different loan programs, to combine their loans
into a single new loan and extend their repayment period. Consolidation
loans can reduce borrowers' monthly repayments, which may lower default
risk and thereby reduce federal costs of loan defaults. Current provisions
of the program also allow borrowers to lock in a fixed interest rate on
their consolidation loans, unlike other FFELP and FDLP student loans,
which carry an interest rate that varies from year to year. As we reported
in October 2003 and in March 2004,2 the number of borrowers consolidating
their federal student loans has increased substantially, with the total
amount-or volume-of loans being consolidated rising from $13 billion in
fiscal year 1999 to over $41 billion in fiscal year 2003. This increase in
consolidation loan volume and the lower interest rates available to
borrowers in recent years have increased the

1Under FFELP, private lenders make consolidation loans to borrowers, with
Education guaranteeing lenders loan repayment and a minimum rate of
return. Under FDLP, Education uses federal funds to make direct student
loans.

2GAO, Student Loan Programs: As Federal Costs of Loan Consolidation Rise,
Other Options Should Be Examined, GAO-04-101 (Washington, D.C.: October
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.: March 29, 2004).

overall estimated long-term cost to the federal government of providing
consolidation loans.3

This report addresses your request that we expand upon the information
provided in our earlier reports on consolidation loans by determining (1)
what differences exist between FFELP and FDLP consolidation loans and
borrowers, (2) the extent to which borrowers with student loans under one
program obtain consolidation loans under the other, and (3) how FFELP and
FDLP borrower and loan characteristics and the movement of loans between
the two programs are incorporated into Education's budgetary cost
estimates. Our work is based on an analysis of a representative sample of
borrowers from Education's National Student Loan Data System (NSLDS), a
national database of student loan recipients. Our analysis primarily
focused on FFELP and FDLP borrowers in the sample who originated
consolidation loans from 1995, the first full year in which loans were
available under FDLP, through June 2003. On May 20, 2004, we briefed your
staff on the results of our work. This report transmits the slides we used
to brief your staff and conveys additional school type information
requested by your staff at the briefing. Our key findings provided at the
briefing are summarized below, followed by the additional information we
are reporting in response to your staff's request. At the end of this
letter, we provide additional details on the scope and methodology of our
work.

  Summary of Key Findings

In determining differences that exist between FFELP and FDLP consolidation
loans and borrowers, we found that on average, FFELP consolidation loan
borrowers, during the 1995-to-2003 time period, had higher levels of
consolidation loan debt than did FDLP consolidation loan borrowers. The
average consolidation loan amount among FFELP borrowers was about $26,400
versus about $20,000 for FDLP borrowers. FFELP consolidation borrowers
were less likely than FDLP consolidation borrowers to have attended a
proprietary (for profit) school prior to consolidation and were more
likely to have borrowed while attending

3Lower interest rates available to borrowers have increased the cost to
the federal government because FFELP consolidation loans carry a
government-guaranteed rate of return to lenders that is projected to be
higher than the fixed interest rate consolidation borrowers pay. Higher
loan volumes in the FFELP program also add to the estimated costs of
consolidation loans. FDLP consolidation loans are made by the government
and thus carry no interest rate guarantee to lenders, but changing
interest rates and loan volumes affect the costs in this program as well.

graduate school. Overall, FDLP borrowers had higher default rates in 4 out
of the 7 years between fiscal years 1995 through 2001. Additionally, for
borrowers who had defaulted prior to consolidation, borrowers from both
FFELP and FDLP were more likely to have defaulted on their consolidation
loan than those who did not default prior to consolidation.

From fiscal year 1998 to fiscal year 2002, the share of underlying FFELP
and FDLP loan volume consolidated into FFELP increased, while the share of
underlying loan volume consolidated into FDLP decreased.4 In fiscal year
2002 alone, 84 percent of FFELP loan volume that was consolidated was done
so under FFELP, while 16 percent was consolidated under FDLP. With regard
to FDLP loan volume consolidated, 58 percent was consolidated under FFELP,
while 42 percent was consolidated under FDLP. Defaulted loans, however,
whether from FFELP or FDLP, were much more likely to be consolidated into
FDLP. For example, in fiscal year 2002, 87 percent of defaulted underlying
FFELP loan volume and 92 percent of defaulted underlying FDLP loan volume
were consolidated under FDLP. According to Education officials, it is not
surprising that a larger share of defaulted underlying FFELP and FDLP loan
volume is consolidated into FDLP because requirements for consolidating
defaulted loans under this program are often less stringent than those
imposed by FFELP lenders. For example, FFELP lenders may chose not to
offer repayment plans based on income levels, while FDLP is required to
offer such a plan to eligible borrowers. In addition, an FFELP lender may
require that the borrower be employed, while FDLP does not have such a
requirement.

Education incorporates FFELP and FDLP borrower and loan characteristics,
and the movement of loans between the two programs, into its budgetary
cost estimates by (1) grouping loans that share similar characteristics in
risk categories, (2) forecasting loan volume for these categories, taking
into account the movement of loans between the two programs, and (3)
applying various assumptions to the categories, such as rates of interest,
estimates of loan prepayment, and rates of default. Among the risk
categories Education uses to estimate costs, for example, is one that
includes consolidation loans with underlying defaulted loans. Education
forecasts the expected loan volume for this risk category and

4The combination of declining interest rates and increased consolidation
loan marketing efforts by lenders has likely contributed to the increase
in the share of underlying loans consolidated into FFELP.

then applies various assumptions to derive an estimated budget cost for
the loans. For example, Education assumes that a certain proportion of
loans placed in this risk category will eventually go into default, thus
increasing the federal government's cost of the loans. Education sometimes
makes different assumptions for different groups. For example, Education
assumes that consolidation borrowers who have defaulted on an underlying
loan will default on their consolidation loans at a higher rate than will
consolidation loan borrowers who have not previously defaulted. In
estimating the costs of nonconsolidation loans-those that borrowers may
ultimately consolidate-Education groups loans based on the type of schools
borrowers attended (2-year, 4-year, proprietary, and so forth) because
experience has shown that these borrowers default on their loans at
different rates. However, Education does not group consolidation loan
borrowers in this way because consolidation loans could reflect multiple
underlying loans with different risk categories, and it believes that
other differences, such as default rates of underlying loans, are more
likely to significantly affect the estimated costs of consolidation loans.
For consolidation loans, however, Education has not analyzed whether
borrowers who consolidate a defaulted loan default again, or redefault, at
different rates based on the type of school they attended. Because of data
limitations, Education was, until recently, unable to link consolidation
loans to borrowers' underlying loans. Education can now do this, making it
possible to determine whether redefault rates vary by type of school. As a
result of the additional analysis we conducted after the briefing, we are
making a recommendation to the Secretary of Education that he direct
Education's Director, Budget Service, to consider the type of schools
consolidation borrowers attended in developing the risk categories for the
department's budgetary cost estimates.

  Further Analysis Could Lead to Enhanced Default Assumptions for Consolidation
  Loan Cost Estimates

Education's recently acquired ability to link consolidation loans to
borrowers' underlying loans allows it to conduct additional analyses,
which could be used to refine its budgetary cost estimates. In particular,
Education could expand its risk categories that currently segregate
consolidation loans by whether they have an underlying default to also
segregate by the type of schools borrowers attended. Our analysis
demonstrates that the extent to which borrowers redefault on consolidation
loans varies according to the type of school they attended. By analyzing
the extent to which borrowers will default on their consolidation loans
based on the type of school attended, Education could determine the
resulting impact on budgetary cost estimates. This could be important
since the proportion of consolidation loan borrowers by type of

schools attended has varied over time. As these proportions vary, the
total rate of default on consolidation loans will likely vary as well,
given that the type of school borrowers attended affects default rates. In
its report on internal control, Education's auditor recently recommended
that the department better monitor consolidation loan activity and conduct
studies of the assumptions used in estimating the budgetary costs of
consolidation loans.

    Extent to Which Consolidation Loan Borrowers Redefault Varies by Type of
    School Attended

According to our analysis, the extent to which borrowers consolidated
loans that included at least one loan on which they had defaulted, and
then subsequently defaulted on their consolidation loan, varies by the
type of school borrowers attended. For both FFELP and FDLP consolidation
loans originated from fiscal years 1995 to 2001, the overall default rate
for consolidation loan borrowers with an underlying default was higher for
borrowers who had attended a 2-year or proprietary school than for those
who had attended a 4-year school. As figure 1 shows, for FFELP, the rate
of default was 45.5 percent for borrowers who had attended proprietary
schools, compared with 29.6 percent for borrowers who had attended a 4year
school. We observed similar relative default rates with respect to FDLP
consolidation loans. The lower rate of redefault among consolidation loan
borrowers who attended a 4-year school is consistent with the lower risk
of nonconsolidation loan borrowers who attended a 4year school, compared
with borrowers who attended other types of schools.

Figure 1: Default Rate of Consolidation Borrowers Who Defaulted on a Loan
Underlying Their Consolidation Loan, by Program and School Type, Fiscal
Years 1995 to 2001

Percent

50.0

                                      45.5

45.0

40.0

35.0

30.0

25.0

20.0

15.0

10.0

5.0

0 4-year 2-year Proprietary 2- and 4-year All other

FFELP FDLP

Source: GAO analysis of NSLDS data.

Notes:

(1) Analysis based on borrowers in the NSLDS sample who originated a
consolidation loan from fiscal years 1995 through 2001, plus any
underlying loans these borrowers originated from January 1, 1980, through
September 2001.

(2) Four-year, 2-year, and proprietary represent borrowers whose
underlying loans were obtained exclusively while attending these types of
schools. Two-and 4-year category represents borrowers whose underlying
loans were obtained while attending both 2-and 4-year schools. All other
represents borrowers whose underlying loans were obtained while attending
some combination of 4year, 2-year, proprietary, and foreign schools other
than the categories listed above.

While figure 1 presents borrower rates of redefault, by school type, we
also analyzed borrowers' default rates by year of consolidation, by school
type. On this basis, we also observed differences by school type.
Moreover, we observed different borrower default rates by school type for
consolidation loans without an underlying loan default, and we observed
different dollar volume default rates by school type for both
consolidation loans with and without an underlying loan default. Because
Education assumes a similar rate of default among consolidation loan
borrowers without regard to the type of school borrowers attended,
Education's cost estimates may be excluding important risk factors
associated with specific

school types. However, the impact of this exclusion will not be known
until Education performs an analysis of the sensitivity of the cost
estimates to different school types.

    Share of Consolidation Loan Borrowers by Type of Schools Attended Varied
    over Time

Overall, the number of borrowers consolidating their student loans has
increased significantly in recent years, while the proportion of borrowers
by type of school attended has varied over time. As these proportions
vary, the overall rate of default on consolidation loans will likely vary
as well, given that the type of school borrowers attended affects default
rates. As shown in figure 2 for FFELP and figure 3 for FDLP, generally
there was an increasing share of consolidation loan borrowers with an
underlying defaulted loan who had attended a 4-year school. At the same
time there was a decreasing share of consolidation loan borrowers with an
underlying defaulted loan who had attended a proprietary school. For
example, for FFELP, the percentage of consolidation loan borrowers with an
underlying default who had attended a 4-year school increased from over 30
percent in fiscal year 1995 to almost 50 percent in fiscal year 2001. In
contrast, the percentage of consolidation loan borrowers with an
underlying default who had attended a proprietary school dropped from
almost 50 percent in fiscal year 1995 to about 26 percent in fiscal year
2001. Similar patterns are also observed with regard to FDLP consolidation
loans beginning in fiscal year 1996.

Figure 2: Percentage of FFELP Consolidation Loan Borrowers with an
Underlying Default by School Type

Percent 100

90

80

70

60

50

40

30

20

10

0 1995 1996 1997 1998 1999 2000 2001 Fiscal year

All other

2- and 4-year

Proprietary

2-year

4-year

Source: GAO analysis of NSLDS data.

Note: Analysis based on borrowers in the NSLDS sample who originated a
FFELP consolidation loan from fiscal years 1995 through 2001, plus any
underlying loans these borrowers originated from January 1, 1980, through
September 2001.

Figure 3: Percentage of FDLP Consolidation Loan Borrowers with an
Underlying Default, by School Type

Percent 100

90

80

70

60

50

40

30

20

10

0 1995 1996 1997 1998 1999 2000 2001 Fiscal year

All other

2- and 4-year

Proprietary

2-year

4-year

Source: GAO analysis of NSLDS data.

Note: Analysis based on borrowers in the NSLDS sample who originated a
FDLP consolidation loan from fiscal years 1995 through 2001, plus any
underlying loans these borrowers originated from January 1, 1980, through
September 2001.

While figures 2 and 3 present the variations in the percentage of
consolidation loan borrowers with an underlying default by school type, we
also analyzed volume changes by school type. On this basis, we also
observed similar variations in proportions by school type. Because default
rate assumptions are based in part on loans consolidated years ago that
had a different distribution of underlying loans by school type than
current consolidation loans, they may be less reliable than they could be.
This could result in less precise cost estimates for consolidation loans.

      Education's Fiscal Year 2003 Financial Statements Auditor Reported That
      Improvements Were Needed in Education's Cost Estimation Process