7/28/2019 Delinquency the Untold Story
1/44
BY AlisA F. CunninghAm
gregorY s. Kienzl, ph.d
March 2011
report BY
nstitute or Higher
ducation Polic
Accessa
ndSuccess
Accounta
bility
Diversity
Finance
GlobalIm
pact
Delinquency:The Untold Story ofStudent Loan Borrowing
7/28/2019 Delinquency the Untold Story
2/44
t i h eca pcy (ihep) is an independent, nonprot organization that is dedicated to access and success in
postsecondar education around the world. Established in 1993, the Washington, D.C.-based organization uses uniue research and innovative
programs to inorm ke decision makers who shape public polic and support economic and social development. IHEPs Web site, www.ihep.org
eatures an epansive collection o higher education inormation available ree o charge and provides access to some o the most respected
proessionals in the elds o public polic and research.
institute For higher eduCAtion poliCY
1320 19th Street, NW, Suite 400
Washington, DC 20036
202 861 8223 telephone
202 861 9307 FACsimile
www.ihep.orgWeB
7/28/2019 Delinquency the Untold Story
3/44
March 2011
A report prepAred BY
Institute or Higher Education Polic
Delinquency:The Untold Story ofStudent Loan BorrowingBY AlisA F. CunninghAm
gregorY s. Kienzl, ph.d
7/28/2019 Delinquency the Untold Story
4/44
Ackw
The authors would like to thank the Institute or Higher Education Polic (IHEP) sta and senior associates who contributed to this
report, including Michelle Asha Cooper, president; Brian A. Sponsler, research analst; Aleis J. Wesaw, research associate; Thomas
D. Parker, senior associate; Sand Baum, senior associate; Jill Jones, research intern, and Khadish Franklin, graduate ellow.
We appreciate the eedback and advice provided b all participants at the roundtable discussion held December 2010 in Washington
D.C. In addition, we beneted rom suggestions oered b a number o other reviewers, including Tia T. Gordon at TTG+PartnersLaura Perna at the Universit o Pennslvania, and Ken Redd at the National Association o College and Universit Business Ocers
Also, we thank the numerous individuals who agreed to be interviewed or this stud, and who provided helpul background
contet on the issues.
Finall, we are grateul or the nancial support rom the American Student Assistance, ECMC/CA, Great Lakes Higher Education
Guarant Corporation, Teas Guaranteed, and USA Funds, as well as the knowledge imparted b their sta.
The views epressed in this report are those o the authors and do not necessaril refect the views o IHEP, the reviewers, or the unders
DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING02
7/28/2019 Delinquency the Untold Story
5/44
ec say 04
ic 08
s la C 12
Caacc a ray Ba Bw 16
Wa d i A ma? sa F 26
o F dc 30
rc 32
A 36
Table of Contents
INSTITUTE FOR HIGHER EDUCATION POLICy 03
7/28/2019 Delinquency the Untold Story
6/44
Executive Summary
To better understand the impact o borrowing and student indebt-
edness, this report eamines the repament eperiences o
student loan borrowers using data provided b ve o the
largest student loan guarant agencies. It eamines more than
8.7 million borrowers with nearl 27.5 million loans who entered
repament between October 1, 2004 and September 30, 2009.
The primar ocus is on the nearl 1.8 million borrowers whoentered repament in 2005. This report is a snapshot o
borrower eperiences, but it can help inorm polic discussions
about student loan programs and the tools available to help
borrowers avoid delinuenc and deault.
Caac Bw Ba:ec rayBorrowers in the 2005 cohort aced a range o circumstances
and options as the started repaing their loans, and continued
to do so as the moved along the path o tring to meet their
repament obligations. The stud looks at whether these
borrowers became delinuent at some point during that period
or availed themselves o various options to postpone or dela
repament during their rst ve ears in repament.
t c a ay.About 37 perceno borrowers managed to make timel paments withou
postponing paments or becoming delinuent, representing
almost 667,000 borrowers in the 2005 cohort with nearl
$13.1 billion in loans. In other words, more than a third o the
borrowers in the 2005 repament cohort seem to be willing
and able to use the ederal student loan repament rame-
work in the intended wa.
Student nancial aidincluding grants and loansplas a ke role in supporting students access to
and success in college. yet, despite periodic increases in grant unding, students and their amilies
have increasingl relied on borrowing to cover more o the costs o higher education. As the number
o student borrowers has increased and their cumulative indebtedness has grown, so too has concern
about whether the resulting debt levels are manageable and about the long-term impact o student
loan debt on other lie choices and consumption patterns. Absent more complete data, policmakers
have oten ocused on deault rates, which are an incomplete measure o the range o eperiences o
contemporar students, including those who ma have diculties repaing their student loans. Deault
rates do not include the man borrowers who become delinuent on their ederal education loans, but
manage to avoid deault. These borrowers ace some o the same conseuences as borrowers who
deault, but until now, the size and signicance o this group has not been recognized or been part o
the polic discussion about deault prevention and nancial literac in general.
DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING04
7/28/2019 Delinquency the Untold Story
7/44
t aa a y ay a
. Other borrowers, about 23 percent, used the repa-
ment tools and options provided b the ederal government
to postpone their paments, thereb avoiding delinuenc.
Some o these borrowers11 percentused onl deerment,
mostl because the re-enrolled in college. But 12 percent o
borrowers used orbearance (oten in combination with deer-
ment) to postpone monthl paments. These borrowers were
aware o ederal repament options and used them or the
intended purpose.
t a cy w a. Although
repament options were available and could have been used
earlier, more than one ourth o the borrowers who entered
repament in 200526 percentbecame delinuent on their
loans at some point, but did not deault. Most o these
borrowers eventuall used deerment and/or orbearance as
tools to avoid deault (21 percent), while a smaller proportion
(5 percent) was able to resolve the delinuenc, presumabl
b making paments to get their account current.
t a. About 15 percent o borrowers not onl
became delinuent, but also had deaulted on their loan(s) at
some point during the rst ve ears o their repament term.
In total, 41 percent o the borrowers aced the negative conse-
uences o delinuenc or deault. It is important to recognize
that or ever borrower who deaults there are at least two
others who were also delinuent on their student loans, but
successull avoided deault. These data illustrate that man
more borrowers are having dicult repaing their loans in a
timel manner than is generall recognized when the ocus is
on deault rates alone. These patterns are both a cause or
concern and an opportunit or improvement.
Ky dc Bw Ba:W d Wa a W
Given the breadth o repament behaviors these borrowers
ehibited, it is important to better understand what tpes o
borrowers were or were not able to make paments on time.
Bw ba a w
bw aa.
Most o the borrowers who let postsecondar education withou
graduating had dicult in repaing their loans33 percent o
undergraduate borrowers who let without a credential became
delinuent without deaulting, and 26 percent deaulted.
Fort-eight percent o undergraduate borrowers who gradu
ated with a credential were repaing in a timel manner, but
21 percent became delinuent without deaulting and 16
percent deaulteda considerabl lower number than among
nongraduates, but still signicant.
Bw ay a a
y a a.
A third or less o borrowers at our-ear, public or private
nonprot institutions became delinuent or deaulted on thei
loans, while nearl hal or more (45 percent and 53 percent
respectivel) o their borrowers were making timel pamentson their loans.
In contrast, onl one-uarter to one-third o borrowers at or
prot and public two-ear institutions were making timel
paments on their loans, and more than hal o all borrowers
in these sectors were delinuent or had alread deaulted.
INSTITUTE FOR HIGHER EDUCATION POLICy 05
7/28/2019 Delinquency the Untold Story
8/44
Bw ay c a by
a aa.
Most borrowers who entered repament in 2005 last borrow-
ed ater onl a ew ears o enrollment37 percent ater just
one ear o college or less, and an additional 18 percent
ater two ears.
O those who last borrowed ater enrolling one ear or less,
two-thirds either became delinuent (30 percent) or deaulted
(34 percent), compared with 21 percent and 6 percent, respec-
tivel, o borrowers who last borrowed in their ourth ear.
Ca bw a acay ky bc -
, wc a ca c a acc
a a w cy a.
O borrowers who started repament in 2005, those who let
school without a credential, last borrowed ater attending onl
one ear o college or less, or attended a public two-ear or
or-prot institution were ar more likel than their counterparts
to become delinuent or deault during the rst ve ears o
the repament.
Man, i not most, borrowers who entered repament ater
leaving college without a credential became delinuent o
deaulted. For our-ear public and private nonprot institutions
the percentage o noncredentialed borrowers who were delin
uentbut did not deault (30 percent and 27 percent, respec
tivel)was twice that o those who deaulted (15 percent and
11 percent, respectivel). The opposite is true or two-ear or-
prot institutions, where hal o borrowers without a credentia
deaulted and 26 percent were delinuent without deault.
The rates o delinuenc and deault were generall much
lower or borrowers who had graduated than or those whohad not. However, even among borrowers who successull
completed their programs at two-ear or-prot institutions, 27
percent became delinuent without deault and 30 percent
had alread deaulted. Borrowers who graduated and las
attended our-ear public and private nonprot institutions had
much lower rates o delinuenc or deault. However, almost a
th o this group became delinuent at some point, although
5 percent or ewer deaulted.
DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING06
7/28/2019 Delinquency the Untold Story
9/44
o F dc
The goal o this stud has been to shine a light on the ull range o
borrower repament patterns, and particularl on students who
became delinuent on their student loans, but did not deault.
The number and percentage o borrowers in this stud who are
known to have had dicult in repaing their loans, particularl
those who became delinuent, but did not deault, is consider-
abl higher than the numbers usuall discussed in polic circles,
where the ocus is primaril on deault. The ull scope o the
problem is worrisome considering that more than two in ve
borrowers who entered repament in 2005 became delinuent
on one or more o their loans at some point during the repamentperiod covered b this stud. While nearl two-thirds o these
delinuent borrowers had not deaulted, this group is too large to
continue to ignore. This stud conrms that ar more students
than generall recognized enter repament and encounter a
range o nancial challenges with negative conseuences that
include delinuenc, damaged credit scores, and alternative
repament options that ma increase overall interest paments.
The initial ndings o this stud provide important rst steps to
understanding the broader scope o borrowers eperiences
with student loans. But there is much more to do, and lowering
rates o delinuenc and reducing deaults will reuire a
serious commitment rom man dierent stakeholders who
care about college access and success. From a public polic
perspective, student success should be viewed as not onl
access to college, but also persistence to a degree or certi-
cate, and the eective management o student loan debt. I, in
an era o limited resources, students must increasingl borrow
to help cover the cost o their education, then what additional
support do the need to help ensure that the have a successuleducational and repament eperience? Reraming the debate
about student loan debt to include the causes and conse-
uences o delinuenc could go a long wa toward improving
borrowers eperiences, enhancing the student loan program,
saving tapaers mone, and perhaps contributing more
broadl to higher education as a whole.
INSTITUTE FOR HIGHER EDUCATION POLICy 07
7/28/2019 Delinquency the Untold Story
10/44
Introduction
For man students, borrowing is essential to enroll in and
complete college. Man pa back their loans without incident or
interruption, but or some, loans can become unmanageable
and the all behind on paments or stop paing altogether. I
borrowers become delinuent (i.e., ail to make monthl
paments within 60 das o the due date), the delinuenc ma
be reported to credit bureaus and become part o their credit
record. Generall, i borrowers eceed 270 das o delinuenc
the will be considered in deault on their loans, with serious
conseuences to their nancial utures.1 In contrast, deermen
and orbearance provisions are designed to address repamen
diculties b allowing borrowers to temporaril suspend the
For decades, the benets o postsecondar education have been recognized as an increasingl
essential component o the nations economic and social well-being. Policmakers have called or majorincreases in educational attainment, both to compete with other nations and to reduce the participation
and graduation gaps between underserved students and their more afuent peers. However, man
challenges eist or meeting these goalsnot onl academic preparation, college awareness, and
institutional capacit, but also overcoming the nancial barriers created b rising college prices and
stagnating amil incomes, which have been eacerbated b the current economic downturn. In this
contet, student nancial aid, including grants and loans, plas a ke role in supporting students access
and success in college. yet despite periodic increases in grant unding, students and their amilies have
increasingl relied on borrowing to cover more o the costs o higher education. Federal student loans are
now the single largest source o nancial aid available to both undergraduates and graduate students.
1It is 360 das or Federal Famil Education Loans or Direct Loans held b the U.S. Departmen
o Education.
DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING08
7/28/2019 Delinquency the Untold Story
11/44
2See glossar or more inormation. Also, see http://ederalstudentaid.ed.gov/datacenter/cohort.html.
repament o their loans to avoid delinuenc. Participation in
these programs is oten an indication that current circumstances
make it dicult or impossible or borrowers to repa their debts.
As the number o student borrowers has increased and their
cumulative indebtedness has grown, so too has concern about
whether the resulting debt levels are manageable and what the
long-term impact o student loan debt will be on other lie
choices and consumption patterns. Without more complete
data, policmakers have oten ocused on cohort deault rates.
In scal ear (Fy) 2008, or eample, about 3.4 million ederal
student loan borrowers entered repament nationwide, and
almost 240,000 borrowers deaulted on their student loans b
the end o the net scal ear (U.S. Department o Education
2010a, 2010b). But cohort deault rate calculations are an incom-plete measure o the range o eperiences o contemporar
students, including those who ma be struggling to repa their
student loans. For eample, the understate deaults that occur
ears ater students leave college; to address this issue, the
Department o Education has recentl introduced measures that
are better able to capture the problems borrowers are having in
repament, including three-ear cohort deault rates.2
However, these measures still do not include the man
borrowers who become delinuent on their ederal education
loans, but manage to avoid deault. These borrowers ace
some o the same conseuences as borrowers who deault
including negative impacts on their credit records, but unti
now this group has not been part o the polic discussion
about deault prevention and nancial literac in general.
Eamining these issues raises a number o uestions:
How man borrowers become delinuent, but do not deaul
on their student loans?
Do borrowers use ederal repament options to postpone
paments and avoid delinuenc?
How man and what percentage o borrowers manage to
repa their student loans on schedule without having to post
pone or dela paments?
What are the characteristics o the borrowers in each o these
groups? How do the dier?
INSTITUTE FOR HIGHER EDUCATION POLICy 09
7/28/2019 Delinquency the Untold Story
12/44
To answer these and related uestions about the impact o the
reliance on borrowing and student indebtedness, this report
eamines the repament eperiences o student loan borrowers,
using data provided b ve o the largest student loan guarant
agencies on more than 8.7 million borrowers with nearl 27.5
million loans who entered repament between October 1, 2004
and September 30, 2009, with a ocus on the nearl 1.8 million
borrowers who entered repament in 2005. The data include
inormation about the specic loans taken out b each borrower;the tpes o loans, loan amounts, and specic repament
events; and select borrower characteristics such as age, gradu-
ation status, and last institution attended (see Box 1).
To complement the uantitative analsis, this report also discusses
the conseuences o delinuenc and deault on student loan
borrowers, including the eect on their credit scores and uture
abilit to borrow. In addition, it provides some contet on ederal
loans and repament options.
Although the stud is a snapshot o borrower eperiences, it can
inorm polic discussions about student loan programs and the
tools available to help borrowers avoid delinuenc and deault
Some borrowers nd it hard to make paments, but still manage
to do so in a timel manner; others, or one reason or another
do not. The act that some borrowers are able to avoid delin
uenc b using deerment, orbearance, or other repament
options indicates that the sstem is working or them. It seems
likel that more borrowers could be using those tools. There arerisks inherent in borrowing or college, especiall or disadvan-
taged students, but making sure borrowers have the inormation
on their repament options when the need it could help miti
gate those risks. The patterns o loan delinuenc, deerment
and orbearance revealed in this stud provide an importan
window into the challenges acing man borrowers.
10 DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING
7/28/2019 Delinquency the Untold Story
13/44
The data used in this analysis relied on borrower- and loan-level inormation or students who entered repayment between October 1
2004 and September 30, 2009. The data were provided by ve large student loan guaranty agencies: American Student Assistance, ECMC
(ormer CSAC/EdFund data only), Great Lakes Higher Education Guaranty Corporation, Texas Guaranteed, and USA Funds. Student loan
guaranty agencies are state agencies or nonprot organizations that insure student loans made through the Federal Family Education
Loan Program (FFELP) against deault. When a borrower deaults, the guaranty agency reimburses the lender or the balance remaining
on the loan and then may collect on the deaulted loans ater they have paid claims to the lender. Guarantors also play a role in providing
inormation to students and nancial aid oces on nancial literacy in general and on debt management and loan repayment options.
They also provide training and guidance to participating lenders and schools (U.S. Department o Education 2009a).
Box 1: daa sc
Together, the ve guarantors represented the majorit o ederal
Staord loan volume made through the FFEL program over the
ve-ear period (U.S. Department o Education 2009b). This
cohort was likel to be representative o the broader student
borrower population. However, in addition to the loan volume
held b other guarant agencies, the Direct Loan program was
responsible or a substantial amount o ederal student loan
volume overallalmost a uarter o new loan volume in Fy
2009 (U.S. Department o Education 2009a). These borrowers
were not included in the stud. The FFEL programs origination
o loans was discontinued in 2010; all new ederal loans are
now made through the Direct Loan program. Guarant agen-
cies will still manage eisting FFEL portolios until the under-
ling loans are paid in ull.
The guarantors provided borrower inormation that was not
individuall identiable; it included inormation on loan origina-
tion and repament dates; individual loan amounts; and loan
events, which occurred during the repament period, such as
deerment, orbearance, delinuenc, and deault. The data
include onl ederal loanssubsidized and unsubsidized Sta-
ord loansGraduate PLUS loans, and consolidation loans.
To answer the research uestions, multiple denitions were
devised based on the available data and the borrowers eperi-
ences in repament. For eample, one measure ocused on
whether a borrower had ever been delinuent, while another
eplored whether a delinuent borrower ever used options
such as deerment and orbearance. These events were aggre-
gated into a classication o borrowers according to their
repament status (see tABle 2). The eamination also includes
dierences based on a number o borrower characteristics
including loan tpe, institution tpe, age, and graduation status
To conduct the analsis, several assumptions were made. Fo
eample, some students ma have loans held with one o the
other guarantors or the Direct Loan program; unortunatel, this
cannot be uantied, although it is likel to be a airl smal
number, especiall or the 2005 repament cohort. This stud is
based on what is known about the loans in the portolios o the
ve guarantors. Thus, onl the last institution tpe is reported
regardless o where the borrower started or enrolled beore
initiating repament. Some outliers were ecluded, including
borrowers with 30 or more loans and those who ell outside the
probable college-age range. This eclusion aected a minis-
cule number o borrowers. Both undergraduate and graduate/
proessional borrowers are included in the data, and borrowers
could have varing durations o repament, depending on
when the rst entered repament. In large part, the repor
ocuses on borrowers who entered repament in 2005, as the
have the longest repament period.
In addition to the borrower data provided b the guarantors, anumber o phone interviews were conducted with guarant
agenc ombudsmen, eperts rom organizations that under
stand risk management and the calculation o FICO credit
scores, and eperts rom communit-based organizations tha
do credit counselingthese people provided background o
the analsis, helped put the uantitative analsis in contet
and made suggestions or uture research.
11INSTITUTE FOR HIGHER EDUCATION POLICy
7/28/2019 Delinquency the Untold Story
14/44
Student Loans
in Context
When borrowing, students have a number o choices, including
a range o ederal student loans as well as non-ederal alterna-
tives. The majorit o borrowers obtain loans through the ederal
Staord loan program; o these borrowers, 86 percent borrowed
through the subsidized loan program, 64 percent through the
unsubsidized program, and 50 percent rom both (National
Center or Education Statistics 2008). Some undergraduatestudents participate in the campus-based Perkins loan program;
others obtain private, state, or institutional loans; and some
parents borrow through the ederal Parent Loans or Undergrad-
uate Students (PLUS) program. In addition, graduate students
can borrow through the Grad PLUS program.
Each o these loan tpes has dierent terms and conditions, and
some are more avorable to the borrower than others. For
eample, with subsidized Staord loans, the ederal government
pas the interest or borrowers while the are in school and
during the si-month grace period ater the leave. For borrowers
with unsubsidized loans, interest starts to accrue when the loan
is disbursed, and the have the option o paing the interest odeerring it while the are enrolled. For both, repament o the
loan principal and interest begins si months ater leaving
school. The amounts students ma borrow in a ear and cumu-
lativel are limited b ederal statute.4 I students need more
unds, the can turn to private loans, which oten have less avor
able conditions (such as higher interest rates) and lack some o
The purpose o ederal student loans is to enable students to attend postsecondar education institu-
tions and move toward completing a degree or certicate. A substantial number o students depend on
loans to nance their postsecondar educationin 200708, 39 percent o all undergraduates borrowed
to help nance their education, up rom 34 percent in 200304 (U.S. Department o Education 2004,2008). Certain groups o students are more likel to borrow than others: those enrolled in higher priced
public, private nonprot, and or-prot institutions; those attending on a ull-time basis; and those with
greater nancial need (Cunningham and Santiago 2008).3
4Loan limits var depending on students dependenc status, class level, and some other actors. For
more details, see http://studentaid.ed.gov/PORTALSWebApp/students/english/studentloans.jsp.
3Other groups are less likel to borrow, but ma instead work ull time, drop their enrollment to
part time, or take other measures to nance their education.
12 DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING
7/28/2019 Delinquency the Untold Story
15/44
the repament protections o ederal student loans. However,
private loans are usuall onl available to students i the have a
avorable credit histor or a co-signer with good credit.
Ater borrowers leave school or are no longer enrolled at least
hal timewith or without completing a credentialStaord
loan borrowers have a si-month grace period beore entering
repament.5 Over the ears, the ederal government has created
a variet o options to encourage repament and make loan
paments more manageable (Department o Education, Student
Aid on the Web): 6
Cc ray pa. Various tpes o ederal student
loan repament plans are available to students, each with
dierent terms and structures. For eample, standard repa-
mentrepament o the loan over 10 earshas the highest
monthl pament, but the lowest amount o interest over the lie
o the loan. An etended repament plan lengthens the term othe loan to lower monthl paments; in the graduated pament
plan, paments begin at lower amounts and graduall increase
over time. Other options take the borrowers income into account
with pament levels tied to income and other actors. Income-
based or etended repament generall produces the lowes
monthl paments, but higher total interest over time.
la Ca. Federal loan consolidation makes repa
ment administration easier b combining several loans into a
single new loan.7 It usuall lowers monthl paments or the
borrower b providing access to the alternative pament plans
mentioned above that reduce monthl paments. These plans
increase the loan term, so the total interest paid over the term
is higher.8
d. Borrowers are able to deer (temporaril sus
pend) their loan paments i the meet certain criteria. These
include enrolling at least hal time in school or eperiencing
economic hardship or unemploment. Borrowers do not have
to make paments on the loan principal until the deermentends. The interest paments on subsidized Staord loans are
made b the ederal government during the deerment period
Unsubsidized Staord loan interest paments can be paid
monthl or deerred, but are tpicall added to the principa
balance at the end o the deerment period.5Repament can be dened as a period in which the loan is amortizing and the principal balance is
going down; thus, a deerred loan would not be considered in repament. However, repament can
also be thought o as borrowers entering repament ater the grace period. For this stud, we dier-
entiate between active repament and being in the repament term. Onl borrowers who are
making paments toward principal are in the active repament categor; those who are in deerment
or orbearance, or have become delinuent, are in the repament period without paing down the
principal on their loans.6See the glossar or more inormation.
7Consolidation was also used to lower the interest rate on one loan.8Subsidized Staord borrowers who consolidate lose the interest benet during a deerment.
13INSTITUTE FOR HIGHER EDUCATION POLICy
7/28/2019 Delinquency the Untold Story
16/44
Fbaac. At their discretion, lenders ma grant a orbear-
ance that temporaril suspends a borrowers paments.
Forbearances are tpicall granted in three- or si-month incre-
ments up to a limit o ve ears. A orbearance is generall a
more epensive option than deerment because interest
continues to accrue, even on subsidized loans. The borrower
does not make principal paments; he or she can make
interest-onl paments or have the interest capitalized and
added to the principal when the orbearance epires.
These repament options have dierent long-term and short-term
eects on students nancial situations. Students can use an
option to decrease their monthl paments or to postpone
pament or a certain period. Some borrowers reuire onl a short-
term solution, such as orbearance or deerment, while or others it
makes more nancial sense to restructure their loan entirel.
When borrowers are unable or unwilling to make their pament
on time each month, at some point the become delinuent on
their loans. When a borrower is 60 to 120 das delinuent, the
loan holder is reuired b ederal law to report the delinuenc
to a national credit bureau. Such a delinuenc can remain on a
borrowers credit report or up to seven ears ater it is reported,
making it dicult or the person to borrow in the uture (Amer-
ican Student Assistance 2010).
I delinuencies continue or nine months (270 das), the lende
will declare the borrower in deault and le a claim.9 Borrowers
deault or a number o reasons, rom unemploment to illness
to ailure to le deerment or orbearance reuests on time to
simpl reusing to meet their nancial obligations. The penalties
or deault are severe: Loan paments can be deducted rom a
borrowers wages, income ta reunds can be withheld, or the
account can be turned over or collection. In addition, deaul
has longer-term impacts on students credit ratings and eligi-
bilit or student aid.10(see Box 2.) Man studies have eamined
the actors that contribute to deault, but ew have eamined the
etent o delinuenc or the characteristics o delinuent borrowers
who do not deault.
9A claim would be led with the guarant agenc under FFELP. As o Jul 1, 2010, FFELP no longe
originates loans. Under the Direct Loan program, a claim is not led; rather, a demand letter is sent
directl to the borrower rom the Department o Education. For FFEL or Direct Loans held b the
Education Department, loans are considered in deault when the are 360 das past due.10For eample, the deault will show up on the credit histor or up to seven ears. In addition
deaulters can be sued or the entire amount o the loan; the are liable or an collection o
court costs; and, the might not be able to renew a proessional license.
14 DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING
7/28/2019 Delinquency the Untold Story
17/44
Signicant research has been conducted to examine the actors that predict or are associated with student loan deault (see Gross e
al. 2010 and McMillion 2004 or summaries o various research studies). Most o the literature is ocused on precollege, college eects
postcollege, and background characteristics o students who deault on education loans. Background characteristics include gender
race, ethnicity, amily income, age, and the students level o preparedness or college (Flint 1994; Herr and Burt 2005; Volkwein and
Szelest 1995). In addition, a plethora o studies have evaluated the between-college and within-college impact on a students likelihood
o deaultingincluding the number o semesters enrolled at an institution, college major, and employment statusor have ocused on
postenrollment variables such as income, personal and amily history, and nancial literacy (Steiner and Teszler 2003; Thein and Herr
2001; Volkwein and Szelest 1995; Woo 2002). Little research has been conducted on students who have diculty repaying their loans
and who become delinquent, but do not deault.
Box 2: Cc s la dc
The penalties or deault on ederal loans are serious and wide-
ranging. The include garnishment o a portion o the borrowers
wages or withholding o income ta reunds, Social Securit
benets, or other public benets. Unlike most other loans,
student loans are not dischargeable through bankruptc. The
borrower is not eligible or additional Title IV student aid until the
loan is repaid or ormal arrangements have been made to repa.
In addition, deault can have other long-term eects outside the
government penalties. These conseuences can be grouped
into a number o themes:
Cc. The ederal government and guarant agencies
that own deaulted loans ma turn them over to collection
agencies. Deaulters are liable or the original principal balance,
all accrued interest, court costs, and an collection ees, which
are all added to the outstanding balance.
iac Bw C sc. Delinuenc and
deault have a negative eect on credit ratings, such as the
FICO score. The eact impact on the credit score depends on
eactl what lenders report to the national credit bureaus,
which pull together inormation to create the credit prole that
goes to FICO. Delinuencies on student loans are onl one
part o the euation and ma be treated dierentl depending
on whether a borrower is chronicall late or generall has a
good pament histor. However, delinuencies o 90 das o
more are highl likel to have a negative impact.
iac F Bw. I a borrowers credit score is
negativel aected, it will have ramications on his or her abilit
to borrow in the uturemortgages, auto loans, and othe
consumer loansas well as on the terms o an uture borrowing
Generall, higher FICO scores signi less risk or lenders, which
usuall leads to more avorable terms or new loans, and vice
versa. I deault is refected in lower credit scores, a borrower
ma not be oered or be able to aord an new loans.
Borrowers who are delinuent on student loans, but do not
deault, ma not ace all these conseuences; or eample, theremain eligible or nancial aid and would likel not have their
income or benets withheld. However, depending on their etent
delinuencies aect borrowers credit scores and their abilit to
borrow in the uture.
Sources: FinAid.org, Oce o the FSA Ombudsman (www.ombuds
man.ed.gov), and background interviews.
15INSTITUTE FOR HIGHER EDUCATION POLICy
7/28/2019 Delinquency the Untold Story
18/44
Characteristics and
Repayment Behaviorof Borrowers
df Bw pa: dac,
e, a la Caacc
During the ve-ear period, about 8.7 million borrowers included
in the available data began repament; together, these borrowerstook out nearl 27.5 million loans totaling $148 billion. Borrowers
most oten received subsidized Staord loansabout 71
percent o borrowers obtained one or more o these loans.
About 60 percent o borrowers had received one or more unsub-
sidized Staord loans. In addition, 25 percent o borrowers had
consolidation loans.13 Man o the borrowers in the stud took
out more than one tpe o loan; in act, 53 percent obtained both
subsidized and unsubsidized Staord loans. Borrowers in the
stud took out almost our loans on average, with a median totaamount o over $9,500.14
Overall, borrowers in the dataset who had entered repament an
time during the ve-ear stud period had a range o demo-
graphic and enrollment characteristics (see tABle 1).
The borrower-level data available or this stud include inormation about borrowers who began
repament on their student loans between October 2004 and September 2009. The ocus is on
borrowers who began repaing their loans in 2005 and what happened to them in the rst ve earsater entering repament.11 To better understand these borrowers, selected characteristics were
eamined, including inormation about the loans the took out, the tpes o colleges the attended,
and other actors. This was ollowed b a detailed eamination o borrowers eperiences in repa-
ment so ar, including whether the ever used repament optionsspecicall deerment and
orbearanceor became delinuent during the ve-ear period.12
11 The 2005 cohort includes borrowers who entered repament throughout the ear.12 Unortunatel, the stud cannot take into account all tpes o repament plans that can be used
to avoid delinuenc, such as graduated repament or income-based repament.
13 The percentages do not sum to 100 because students oten took out more than one kind o loan14 I borrowers who consolidated are removed, the median total loan amount was approi
matel $7,040.
16 DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING
7/28/2019 Delinquency the Untold Story
19/44
A e i ray
Twent-nine percent o those entering into repament were
between the ages o 21 and 24 earsthe age at which tradi
tional students would tpicall have graduated or let college and
started to repa their loans. One uarter o borrowers began
repament when the were between 25 and 29 ears old; thusthe majorit o borrowers in the stud were less than 30 ears old
A substantial proportion o these borrowers refect non-traditiona
paths in college, with 27 percent between the ages o 30 and 44
and 11 percent 45 and older when the entered repament.
la i A B e ray15
About 45 percent o the borrowers had enrolled at public o
private, nonprot, our-ear institutions, 23 percent and 22
percent, respectivel.
Onl 12 percent o the borrowers in the stud had been enrolled
in public two-ear institutions beore starting repament. This isnot surprising, as communit college students are less likel to
borrow than students attending other tpes o schools, because
o lower tuition, part-time status, ailure to appl or nancial aid
or some other reason (Cunningham and Santiago 2008).
Twent-seven percent o borrowers in the stud had been
enrolled at a two- or our-ear or-prot institution. This ma
seem disproportionatel high given the relativel low percentage
o undergraduate students who attend or-prot institutions
overall, but a ver high proportion o students who enroll at or
prots borrowalmost 88 percent in 200708 (National Cente
or Education Statistics 2008).
h ga l16
At the time o their last loan, almost 68 percent o borrowers
reported that the were undergraduate students, with 31
percent obtaining their last loan ater onl one ear in college or
less, and an additional 14 percent ater two ears.
About 14 percent o these borrowers were graduate or proes-
sional students. The rest were either borrowers who had las
obtained a loan in the later undergraduate ears or those o
whom data was missing.
tABle 1
Age When entered repAYment 200409 2005
Under 21 ears old 8 7
2124 ears 29 28
2529 ears 25 26
3044 ears 27 27
45+ ears 11 12
institution tYpe lAst Attended
Public our-ear 23 25
Private nonprot our-ear 22 24
For-prot our-ear 16 11
Public two-ear 12 11
For-prot two-ear 11 9
Other 8 8
Missing 9 14
highest grAde level
First-ear undergraduate 31 24
Second-ear undergraduate 14 13
Third-ear undergraduate 7 7
Fourth-ear undergraduate 13 13
Fith-ear undergraduate 3 3
Graduate student 14 16
Missing 19 24
loAn tYpe (multiple loAns possiBle)
Consolidation 25 36
Subsidized Staord 71 67
Unsubsidized Staord 60 55
Grad PLUS 2 *
Number o borrowers 8,711,724 1,779,222
*GRAD PLUS WAS NOT AVAILABLE FOR 2005.
NOTE: INSTITUTION TyPE WAS BASED ON THE LAST INSTITUTION A BORROWER ATTENDED BEFORE ENTERING
REPAyMENT. OTHER INSTITUTIONS INCLUDED FOREIGN INSTITUTIONS, PRIVATE NONPROFIT TWO-yEAR AND
LESS-THAN-TWO yEAR, AND PUBLIC AND FOR-PROFIT LESS-THAN-TWO yEAR. GRADE LEVEL WAS BASED ON THE
yEAR CERTIFIED FOR FINANCIAL AID. MISSING PRIMARILy REFLECTS UNAVAILABLE DATA FOR MANy BORROWERS
WITH CONSOLIDATION LOANS.
SOURCE: GUARANTOR DATA FILES, AUTHORS CALCULATIONS
15 The dataset includes inormation or onl the last institution attended and not an other institu
tions that a borrower might have attended. This would mean, or eample, that borrowers who
started at communit colleges and transerred to our-ear institutions would be classied with
our-ear institutions, with a possible eect on classication o borrower behavior. Note tha
roughl 9 percent o the borrowers were missing data on the last institution attended. This
occurred in large part because o unavailable data or consolidation loans.16 Highest grade level is dened as the level certied b the institution or the purpose o the las
loan awarded to a borrower; it signies up to, but could be less than. In most cases, this variable
shows the highest level attained b borrowers beore leaving college. However, in a small numbe
o cases, a person borrows with another guarantor later or continues enrollment without borrowing
this would not be captured in the data. A substantial proportion o borrowers (19 percent, primaril
those with consolidation loans) did not have responses or this variable.
Select Characteristics of the Study Populationand the 2005 Cohort (%)
17INSTITUTE FOR HIGHER EDUCATION POLICy
7/28/2019 Delinquency the Untold Story
20/44
For the purpose o analsis, it is helpul to ocus on borrowers
who had the most ears o repament (or non-repament) histor.
Thus, the ollowing section ocuses on the almost 1.8 million
borrowers in the stud who started repament in 2005.17
Ca Bw Ba:
ec ray
A primar goal o this stud was to dene the number and propor-
tion o borrowers who have dierent repament eperiences, and
the data provided or this analsis allow a detailed eamination o
these eperiences. Borrowers in the 2005 cohort aced a range o
possibilities as the started repaing their loans and continued todo so as the moved along the path o tring to meet their obliga-
tions. This stud cannot capture all the nuances o borrower epe-
riences; rather, it is a snapshot o events that occurred during the
stud period and does not address the timing or recurrent
compleities o borrowers use o debt management options.
However, the available data can be used to classi borrowers into
a number o mutuall eclusive categories that give a broad sense
o their repament eperiences. The categories are based on the
events fagged in the data: Securing a deerment or orbearance
avoiding or becoming delinuent, and entering deault.
One wa o looking at these borrowers is to take into account
whether those in the stud period availed themselves o various
options to postpone or dela repament during their rst ve
ears in repament or became delinuent at some point during
that period.18
At one end o the spectrum are the active repaers, those who
managed to make timel paments without ever postponingpaments or becoming delinuent. About 37 percent o borrowers
were repaing their loans without taking an mitigating actions
representing almost 667,000 borrowers in 2005 with nearl $13.1
billion in loans. Whether the ound making timel paments eas
or dicult and whether the restructured their loans into other
repament plans to make the paments more manageable are
not captured in the available data.
Borrowers Who Entered Repayment in 2005 by Loan Repayment Status
tABle 2
% in speCiFiC BorroWer groups % in AggregAted BorroWer groups
Repament without event 37 37
Deerment onl (in-school enrollment)* 7
23Deerment onl (economic hardship)* 4
Forbearance onl 6
Forbearance and deerment 6
Delinuenc onl 5
26Delinuenc and deerment 5
Delinuenc and orbearance 8
Delinuenc with deerment/orbearance 8
Deault 15 15
* AN ESTIMATED TWO-THIRDS OF BORROWERS DEFERRED BECAUSE THEy WERE STILL IN SCHOOL; THE REMAINDER CITED ECONOMIC HARDSHIP. BASED ON ESTIMATES FROM THREE OF THE FIVE GUARANTORS.
NOTE: PERCENTAGES DO NOT SUM TO 100 DUE TO ROUNDING. INCLUDES BORROWERS WHO USED CONSOLIDATION LOANS. SEE APPENDIx TABLE 1 FOR NUMBERS OF BORROWERS IN EACH CATEGORy AND FOR THE DISTRIBUTION
OF BORROWERS WHO DID NOT HAVE CONSOLIDATION LOANS.
SOURCE: GUARANTOR DATA FILES, AUTHORS CALCULATIONS
17 Their characteristics are similar to those o the ull stud population (see tABle 1). 18 See Appendi Table 1 or the number o borrowers in each categor.
18 DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING
7/28/2019 Delinquency the Untold Story
21/44
The remaining 64 percentmore than 1.1 million borrowers with
over $25.3 billion in loanswere not activel repaing their loans
or at least a portion o the stud period and are likel to be a
source o concern to varing etents.
About 23 percent o the borrowers who started repament in
2005 were able to use orbearance and/or deerment to post-pone paments and avoid delinuenc during the stud
period. It is dicult to assess the eact circumstances that
prompted this group o borrowers to postpone repaing their
loans. For eample, 11 percent o borrowers used onl a
deerment. Although deerments can be used or economic
hardship, estimates rom three o the ve guarant agencies
that participated in the stud suggest that about two-thirds (7
percent) o deerments or these borrowers were or students
who had re-enrolled in college.19 Borrowers ma reuest and
be granted a orbearance or a multitude o reasons; these
borrowers might have needed some level o assistance to
postpone repaing their loans.
Twent-si percent o the borrowers in the 2005 cohort became
delinuent on their loans but did not deault. Most o these
borrowers (21 percent) used deerment or orbearance to avoid
deault, while a smaller proportion (5 percent) were able to
resolve their delinuenc, presumabl b making paments to
get their account current.20
At the other end, the snapshot shows that about 15 percent o
borrowers not onl became delinuent, but also had alread
deaulted on their loans at some point during the repament
term, despite the availabilit o mitigation tools.
The eperiences o this cohort o borrowers can be divided into
our broad categories: (1) Those who were activel repaing their
loans and making on-time paments; (2) those who used deer-
ment, orbearance, or both to postpone paments and did not
become delinuent; (3) those who became delinuent one o
more times during the ve-ear period, but did not deault; and (4)
those who deaulted.
The number and percentage o borrowers in this stud known to
have had dicult in repaing their loansparticularl those who
became delinuent, but did not deaultis considerabl highethan what is usuall discussed in polic circles, where the ocus is
primaril on deault alone. The ull scope o the problem is worri-
some: More than two in ve borrowers who entered repament in
2005 became delinuent on one or more o their loans (including
deault) at some point during the repament period covered b
the stud. Another set o borrowers (not as well dened) avoided
delinuenc onl b taking mitigating action to postpone pamen
on their loans or various reasons and periods o time.
Ky dc Bw Ba:
W d Wa a W
Given the breadth o repament behaviors these borrowers
ehibited, it is important to understand what tpes o borrowers
were able to make paments on time and what tpes were not
Man actors are known to be associated with deault behavior
including institutional characteristics, borrower background
and ailure to complete the program o stud (Dnarski 1994
Steiner and Teszler 2003; Volkwein and Cabrera 1998; Woo
2002).21 It is likel that at least some o those actors are associ
ated with other borrower behaviors, such as the use o mitiga
tion options and the likelihood o becoming delinuent. Some o
these actors were available in the stud data or eploration and
can provide contet or thinking about policies and practices
that could be used to increase the number o borrowers who
repa on a timel basis. This analsis o borrower characteristicsocuses on the 1.1 million borrowers entering repament in 2005
who did not have consolidation loans.22 The distribution and
numbers o all borrowers and those who did not consolidate
their loans is shown in Appendix tABle 1.
19 The percentage was not available or all guarantors, and the tpes o deerments could not be
separated out in a large proportion o the data. Thereore, subseuent analses do not break
out the tpes o deerments.20 Note that the options could have been used beore or ater the delinuenc.
21Also see McMillion (2004) and Gross and colleagues (2009) or summaries o relevant literature22 Borrowers with consolidation loans have been ecluded owing to the lack o ke data, such a
last institution attended and graduation. Thus, some o the numbers ma dier rom thos
presented earlier.
19INSTITUTE FOR HIGHER EDUCATION POLICy
7/28/2019 Delinquency the Untold Story
22/44
An initial eamination o the behavior o these borrowers reveals
that those who encountered problems in repament diered rom
their counterparts who did not. Some o these dierences are
outlined below. (see Appendix tABles For detAils.)
Bw ba a w
bw aa. The delinuenc and deault rates oborrowers diered between those who earned a credential and
those who dropped out o school.
Overall, 42 percent o undergraduate borrowers who started
repament in 2005 had graduated with a degree or credential
(see Appendix tABle 2).
Borrowers who let postsecondar education without gradu-
ating were more likel to eperience dicult in repaing their
loans59 percent o undergraduate borrowers who let withou
a credential became delinuent or deaulted (see tABle 3).
Fort-eight percent o those who graduated with a credentia
were repaing in a timel manner, while 38 percent became
delinuent or deaulteda considerabl lower number than
among non-graduates, but still signicant (see tABle 3).
Among graduate students, 68 percent o those who completed
their program were making timel repaments on their loanswithout using deerment or orbearance, and onl 12 percent
were delinuent or in deault (see tABle 3).
The proportion o graduate student borrowers who let withou
a credential and were delinuent or deaulted was 29 percen
(see tABle 3).
Percentage Distribution of 2005 Borrowers by Loan Status and Graduation Status
tABle 3
undergrAduAtes leFt Without degree/CredentiAl grAduAted All BorroWers
Timel repament 26 48 35
Deerment/orbearance without delinuenc 15 14 15
Delinuenc without deault 33 22 28
Deault 26 16 21
Total 100 100 100
grAduAte students
Timel repament 47 68 58
Deerment/orbearance without delinuenc 25 20 22
Delinuenc without deault 24 10 16
Deault 5 2 3
Total 100 100 100
NOTE: DOES NOT INCLUDE BORROWERS WITH CONSOLIDATION LOANS.
SOURCE: GUARANTOR DATA FILES, AUTHORS CALCULATIONS
20 DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING
7/28/2019 Delinquency the Untold Story
23/44
Bw ay ba a y a a.Previous studies have indi-
cated that the incidence o deault varies considerabl depending
on the tpe o institution a borrower attended (Gladieu and Perna
2005). This analsis provides support or that conclusion and
shows how it also applies or students who ma not have
deaulted, but had problems making timel paments.
A third or ewer o borrowers at our-ear public or private
nonprot institutions became delinuent or deaulted on their
loans, while close to hal (45 percent and 53 percent, respec-
tivel) were making timel paments, and an additional 20
percent had taken steps to secure a deerment or orbearance
to postpone paments without ever becoming delinuent (see
tABle 4).23 It is possible that these borrowers had better
economic prospects or had more knowledge o their options
than those who attended other kinds o institutions.
In contrast, onl one-uarter to one-third o borrowers at or-
prots and public two-ear institutions were making timel
paments on their loans, and more than hal o all borrowers in
these sectors were delinuent or had deaulted (see tABle 4).
These tpes o institutions are similar in that the tend to serve
disadvantaged and non-traditional populationsgroups tha
are oten dicult to reach with inormation about repamen
plans and nancial literac, in general. However, the are ver
dierent in the percentage o students who borrow. The
percentage o borrowers at two-ear public institutions is rela
tivel low, given airl low tuition and other epenses, and the
greater propensit o students to enroll part time while working
(Cunningham and Santiago 2008), whereas or-prot institutions
have a dierent business model that is reuentl refected in
higher costs o attendance and greater reliance on borrowing.
Bw ay c
a aa. The majorit o borrowerswho entered repament in 2005 last borrowed ater onl a ew
ears o enrollment37 percent ater one ear o college or less
and an additional 18 percent ater two ears. This could mean
that the borrowers let school at that grade level ater completing
a short-term program or dropped out beore inishing a longerprogram.24 O those who last borrowed ater enrolling or one ear o
less, two-thirds either became delinuent (30 percent) or deaulted
(34 percent), compared with 27 percent o borrowers who last
borrowed in their ourth ear (see Appendix tABle 4). Graduate
student borrowers were the least likel to have been delinuent (16
percent) or deaulted (3 percent) over this period.
Percentage Distribution of 2005 Borrowers Loan Status by Last Institution Attended
tABle 4
puBliC Four-YeAr privAte nonproFit
Four-YeAr
puBliC tWo-YeAr For-proFit
tWo-YeAr
For-proFit
Four-YeAr
Timel repament 45 53 24 32 35
Deerment/orbearance, but not delinuent 21 20 16 5 12
Delinuent, but not deaulted 24 20 36 27 29
Deault 10 8 24 36 24
Total 100 100 100 100 100
NOTE: DOES NOT INCLUDE BORROWERS WITH CONSOLIDATION LOANS. ALSO SEE APPENDIx TABLE 3.
SOURCE: GUARANTOR DATA FILES, AUTHORS CALCULATIONS
23 The data capture onl the last institution attended; borrowers could have been enrolled at a
dierent kind o institution beore the began repaing their loans. For eample, borrowers who
transerred rom a two-ear to a our-ear institution would be captured as our-ear students.
24 In a small number o cases, borrowers ma have borrowed with another guarantor or enrolled
without borrowing. This would not be captured in the data.
21INSTITUTE FOR HIGHER EDUCATION POLICy
7/28/2019 Delinquency the Untold Story
24/44
o bw a a c b
ay a.Si out o 10 o the oungest borrowers (thoseunder 21) either became delinuent or deaulted. This proportion
graduall decreases as borrowers get older, with about 33 percent
o borrowers who were 45 or older when entering repament
subseuentl becoming delinuent or deaulting (see Appendix
tABle 5).Research suggests that older, more eperienced borrowers
have more nancial literac skills, greater awareness o repament
options, and more marketable job skills and the resources to cover
short-term repament diculties (Gross et al. 2009).
s bw cc ay b aca w b-
ay a. It is possible that choices aboutthe number o loans or the amount borrowed are related to eperi-
ences during the repament period, whether positivel or nega-
tivel. In general, borrowers who started repament in 2005 and
deaulted had ewer loans and lower loan amounts than those who
did not deaultewer than three loans on average, compared with
slightl more than three or those currentl making timel paments
and more than our or those who used orbearance (see Appendix
tABles 6 And 7). Borrowers who deaulted had median total loanamounts o $6,600 compared with more than $8,000 or those
repaing without event and $9,000$11,000 or those using orbear-
ance. The pattern is similar to that ound in other studies (Steiner
and Teszler 2003; Woo 2002). Although it ma seem counterintui-
tive, the lower numbers are likel related to the number o ears
borrowers were enrolled beore entering repament, whether or
not the completed their educational program, and the tpe o insti-
tution the attended. The numbers also suggest that the degree o
debt burden is a relative, not an absolute, phenomenon.
Bw w a w a a
ay a, b a cOverall, borrowers who were delinuent, but did not deault and
those who deaulted, were similar in man respects. Both groups
were less likel to have last enrolled at a our-ear public or private
nonproit institution. However, borrowers who were delinuen
without deaulting were less likel to have last enrolled at a or-pro
institution than those who deaulted. Within that group, borrowers
who were delinuent and used both deerment and orbearance
were more likel than deaulters to have last attended a public two
ear institution and much less likel to have last attended a or-pro
institution. Indeed, the use o deerments or orbearance to avoid
delinuenc or deault was ver limited at or-prot institutions
Borrowers who were delinuent without using an repamen
options were almost as likel as deaulters to have last attended a
or-prot institution, but more likel to have graduated.
Another signicant dierence is in the long-term impact on
borrowers o becoming delinuent or deaulting. Delinuenc
(especiall multiple delinuencies) oten aects borrowers credi
score, limiting their abilit to borrow in the uture. But deaultcarries ar worse conseuences that can persist or decades
thus, it is important to ensure that borrowers who become delin
uent receive the inormation and counseling support the need
to resolve their delinuenc beore the deault.
22 DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING
7/28/2019 Delinquency the Untold Story
25/44
Ca bw a acay ky bc, wc a ca c a
acc a a w cy a.To decrease rates o delinuenc, programs need to target
borrowers who are most at risk. The data in this stud demonstrate
that among borrowers who started repament in 2005, those who
let school without a credential, last borrowed ater attending one
ear o college or less, or attended a public two-ear or or-prot
institution were ar more likel than their counterparts to become
delinuent or deault during the rst ve ears o the repament
period (see tABle 5). In most instances, the proportion o borrowerswho became delinuent without deaulting was signicantl higher
than the proportion who deaulted; eceptions were borrowers who
last attended or-prot two-ear institutions and those whose highest
grade level was one ear or less o college.
These three actorsinstitution tpe last attended, highest grade
level attained, and graduation statusare associated positivel or
negativel with delinuenc and deault. It is dicult to disentangle
them or uanti the impacts, because the are interrelated.
For eample, borrowers who had last attended public two-ear
or or-prot institutions were more likel to have taken out thei
last loan ater onl one ear or less o enrollment58 percent
or public two-ear institutions, or eample, and 74 percent o
or-prot two-ear institutions (see Appendix tABle 8). This makes
sense, as students at those institutions are oten enrolled in
shorter programs, such as certicates o a ear or less. The
proportion o borrowers entering repament ater enrolling o
one ear or less is much lower (20 percent or less) at public and
private nonprot our-ear institutions, which also would be
epected. Further, substantial proportions o borrowers at theseinstitutions last borrowed when the were graduate students
(20 percent and 37 percent, respectivel).
In addition, both institution tpe and highest grade level are
correlated with graduation. Collectivel, about 42 percent o
undergraduate borrowers who entered repament in 2005 had
graduated. For those who last borrowed ater onl one ear, the
gure drops to 36 percent. Borrowers who last attended our-
ear public or private nonprot institutions were more likel to
Percentage of Borrowers Who Started Repayment in 2005 Who Were Delinquent or Defaulted bySelected Enrollment Characteristics
tABle 5
lAst institution Attended
% oF BorroWers Who BeCAme
delinquent Without deFAult % oF BorroWers Who deFAulted
% oF BorroWers Who Were
delinquent or deFAulted
Public our-ear 24 10 34
Private nonprot our-ear 20 8 28
Public two-ear 36 24 60
For-prot two-ear 27 36 63
For-prot our-ear 29 24 53
highest grAde level
First-ear undergraduate 30 34 64
Second-ear undergraduate 33 18 51
Third-ear undergraduate 27 11 38
Fourth-ear undergraduate 21 6 27
Fith-ear undergraduate 22 6 28
Graduate student 16 3 19
grAduAtion stAtus
Graduated 22 16 38
Let without credential 33 26 59
NOTE: DOES NOT INCLUDE BORROWERS WITH CONSOLIDATION LOANS.
SOURCE: GUARANTOR DATA FILES, AUTHORS CALCULATIONS
23INSTITUTE FOR HIGHER EDUCATION POLICy
7/28/2019 Delinquency the Untold Story
26/44
graduate (42 percent and 49 percent, respectivel, compared
with 36 percent or our-ear or-prots and 23 percent or public
two-ear institutions). For-prot two-ear institutions had higher
completion rates (69 percent on average), perhaps owing to
short-term certicate programs (see Appendix tABle 9).
These gures refect substantial numbers o borrowers who
entered repament ater leaving college without a credential. As
might be epected (Steiner and Teszler 2003; Volkwein and
Cabrera 1998; Woo 2002), man, i not most, o these borrowersbecame delinuent or deaulted. This was true or all institutional
tpes, although the etent varied (see tABle 6). For eample, the
proportion o borrowers who let without a credential and became
delinuent or deaulted ranged rom 38 percent at private
nonprot our-ear institutions to 76 percent at or-prot two-ear
institutions. For our-ear public and private nonprot institutions,
the percentage o non-credentialed borrowers who were delin-
uent, but did not deault was twice that o those who deaulted.
The opposite is true or two-ear or-prot institutions, where hal
o borrowers without a credential deaulted (two out o ever three
who became delinuent). Other research (Volkwein and Szeles
1995; Woo 2002) suggests that the background characteristics o
borrowers who attend schools with higher deault ratessuch as
income, parent education, and nancial independencetend to
dier rom those o borrowers attending lower deault-rate schools
which ma account or some o the dierences. However, this
cannot be determined rom the stud data.
The rates o delinuenc and deault were generall much loweor borrowers who had graduated than or those who had not
suggesting that graduation ma be a crucial actor.25 Although this
analsis cannot show causalit, the ndings support previous
research that suggests that graduation (or, in this case, not gradu-
ating) is strongl correlated with loan deault (Gross et al. 2009).
25 The eception was two-ear or-prots, where rates o delinuenc were similar or borrowers
who graduated and those who did not.
Loan Status of 2005 Borrowers by Last Institution Attended and Graduation Status
tABle 6
oF those Who leFt
Without A Cred entiAl
% oF BorroWers Who BeCAme
delinquent Without deFAult
% oF BorroWers
Who de FAulted
% oF BorroWers Who Were
delinquent or deFAulted
Public our-ear 30 15 45
Private nonprot our-ear 27 11 38
For-prot our-ear 34 30 64
Public two-ear 39 27 66
For-prot two-ear 26 50 76
All 32 23 55
oF those Who grAduAted,
Public our-ear 15 4 19
Private nonprot our-ear 13 5 18
For-prot our-ear 21 14 35
Public two-ear 27 15 42
For-prot two-ear 27 30 57
All 20 13 33
NOTE: DOES NOT INCLUDE BORROWERS WITH CONSOLIDATION LOANS.
SOURCE: GUARANTOR DATA FILES, AUTHORS CALCULATIONS
24 DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING
7/28/2019 Delinquency the Untold Story
27/44
To acquire some insight to augment the quantitative analysis, a number o people who work directly with borrowers were interviewed to
get a sense o how delinquency and deault aect borrowers and how borrowers in trouble can get assistance in repaying their loans.
Several general themes emerged across the interviews; they are summarized below. Interviews included student loan ombudsmen
and people rom community-based organizations who work with thousands o borrowers in trouble, as well as experts in deault risk
management. Although these insights cannot be generalized to the total population o student loan borrowers, they help understand the
borrower experiences refected in the data and suggest uture avenues o research that can build on this study.
Box 3: pc F
In general, interviewees said that borrowers
were rarel amiliar with all the repament options available to
them beore the became delinuent or deaulted;
do not ull understand loan terms, interest accrual, and so
on, and reuired signicant assistance in selecting a loan
repament plan;
compounded their repament problems b ailing to ll out
paperwork in time to avoid deault; and
report that wage garnishment, Social Securit osets, and no
being able to use their earned income ta credit are conse-
uences o deault that concern them.
Man o the interviewees work with borrowers who are alread
in trouble; the help borrowers understand what is needed to
get back into good standing on their loans. But the all high
lighted the act that borrowers are oten not aware o options
that could have helped them avoid becoming delinuent in the
rst place.
The likelihood o delinuenc and deault or borrowers who grad-
uated also varied signicantl b tpe o institution, again with two-
ear or-prots having the highest rates. About 76 percent o all
borrowers who last attended or-prot two-ear institutions, but did
not graduate, became delinuent at some point, and over hal o
them deaulted on their student loans within ve ears o entering
repament. Even among borrowers who successull completed
their programs at these institutions, 57 percent became delinuentat some point, and 30 percent had alread deaulted. Considering
the eects o the recent global recession, the results are likel to be
even worse or students who tried to enter the job market rom
2008 onward. Borrowers who graduated and last attended our-
ear public or private nonprot institutions had much lower rates o
delinuenc and deault. However, even or these institutions
almost a th o graduating borrowers became delinuent at some
point, although onl 5 percent or ewer deaulted.26
In addition to the characteristics highlighted here, other actors
that could not be eamined with the stud data are likel impor-
tant and could shed more light on some o these repamen
patternsactors such as starting salaries, emploment statusand previous nancial knowledge. Man o these actors are
likel correlated with borrower behavior; deeper analsis migh
be able to identi the actors that most aect repament behavior
Understanding all the actors that aect borrower behavior is
essential to inorm polic discussions about improving college
completion, access, and aordabilit in an era o restricted
resources (see Box 3).26 These patterns among institution tpe, graduation status, and rates o delinuenc/deault holdor each age categor and highest grade level attained.
25INSTITUTE FOR HIGHER EDUCATION POLICy
7/28/2019 Delinquency the Untold Story
28/44
What Does It All Mean?
Summarizing the Findings
This stud enables us to better understand borrowers repa-
ment eperiences. O the nearl 1.8 million borrowers covered
b the stud who entered repament in 2005, we can see that
more than a third were repaing their student loans success-
ull without dela or delinuenc or the rst ve ears. The
remaining borrowers had either used one or more o the
options in the ederal loan program to postpone their paments
temporaril or had become delinuent at some point during
the stud period. This report provides a snapshot o the char-acteristics and loan behavior o borrowers who entered repa-
ment in 2005. To better understand the dierences among
these groups and the implications or uture polic, it is helpul
to group borrower behavior into the ollowing three themes:
1.Epected path through repament.
2. Appropriate and timel use o repament tools and options.
3. Magnitude and impact o delinuenc and deault.
ec pa t ray
Man borrowers were able to repa their loans on a timel
basis during the rst ve ears o entering repament
Borrowers who were making monthl paments on time during
the stud period account or 37 percent (almost 667,000) o
those who entered repament in 2005. Among the 1.1 million
who did not consolidate their loans, roughl hal last attended
our-ear public (45 percent) or private nonprot (53 percent)
institutions. The made up about 48 percent o undergraduateand 68 percent o graduate student borrowers who had
successull completed their degree programs. Some o these
borrowers ma have had dicult making timel paments
but used options not captured in the stud data, such as
income-based or graduated repament, to make thei
paments on time. Still, one in three borrowers in the 2005
repament cohort seemed willing and able to use the edera
student loan repament ramework in the intended wa.
The goal o this stud is to shine a light on the ull range o borrower repament patterns, particularl
those o students who became delinuent on their student loans, but did not deault. Historicall, public
debate about student loans and debt burdens has ocused primaril on deaulters and deault rates.
In realit, the majorit o student loan borrowers never deault. But while that is positive news, we shouldnot ignore the act that man students enter repament and encounter a range o nancial challenges.
26 DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING
7/28/2019 Delinquency the Untold Story
29/44
Aa a ty u ray o
Other borrowers used the repament options provided b the
ederal government to postpone their paments, thereb avoiding
delinuenc. Some o these borrowers (11 percent) used deer-
ment onl, usuall, although not alwas, because the re-enrolled
in college. Twelve percent used orbearance (oten in combina-
tion with deerment) to postpone monthl paments and avoid
delinuenc. Collectivel, this group totals more than 400,000 o
borrowers who entered repament in 2005.
Similarl to the group that took the epected path through
repament, these borrowers were more likel than all borrowers
to have last attended a our-ear institution, but there are some
dierences; or eample, borrowers who used both orbear-
ance and deerment tended to be ounger, more likel to
attend public two-ear institutions, and less likel to attend or-
prot institutions than those who used onl orbearance.
Although these borrowers needed a little etra assistance, it is
dicult to know the specic circumstances that prompted their
action because the repament options can be used or a variet
o reasons. Nevertheless, the were aware o the options and
used them or the intended purpose. However, these data raise
the uestion o wh the percentage o borrowers using these
options to avoid delinuenc was not higheran even larger
percentage o borrowers could have applied or a deerment or
orbearance but became delinuent beore doing so. Moreover
initial scans o the data reveal that some borrowers used multiple
orbearances when a deerment or some other, less costl loan
repament option might have been more advantageous.
ma a iac dcy
Despite the availabilit o repament options, over a uarter o
the borrowers who entered repament in 2005 (26 percent, or
approimatel 454,000 borrowers, representing $8.5 billion in
loans) were delinuent at some point, but had not deaulted
Much o the public debate over the past ew ears has ocused
on those who deault, without looking at the substantial popu
lation o borrowers who were delinuent, but did not deault. I
is important to recognize that or ever borrower who deaults
at least two others have been delinuent on their student loans
but successull avoided deault.
Most o the borrowers who became delinuent during the stud
period (21 percent out o 26 percent) used some combination o
deerment and orbearance to avoid the ar worse outcome o
deault. It is likel that these borrowers knew more about repa-
ment options or showed greater initiative in contacting lenders
guarant agencies, or consumer advocac organizations to
help them deal with their loan repament problems and avoid
deault. The remainder o the delinuent borrowers (onl 5
percent o all borrowers) managed to avoid deault without using
27INSTITUTE FOR HIGHER EDUCATION POLICy
7/28/2019 Delinquency the Untold Story
30/44
deerment or orbearance, presumabl b making paments
that brought their accounts current. Further research is needed
to better understand wh some borrowers were successul in
using available repament options to avoid delinuenc or
deault and others were not.
Then there are the deaulters. Fiteen percent o these borrowers
(258,000 borrowers, with $3.2 billion in loans) had alread
deaulted within ve ears o entering repament. Man o these
borrowers were either unaware o the range o repament
options available to them or ailed to act beore becoming delin-
uent and subseuentl deaulting. In total, 41 percent o the
borrowers (712,000 borrowers, $11.6 billion) aced the negative
conseuences o delinuenc or deault.
These data illustrate that man more borrowers are having di-
cult repaing their loans in a timel manner than is generall
recognized when the ocus is on deault rates alone. In this
stud, three-uarters o a million borrowers who entered repa-
ment in one ear alone had dicult. The stud period does not
capture what happened ater ve ears, so the actual rate odelinuenc and deault over time ma be understated. In addi-
tion, these gures do not include other FFELP or Direct Loan
borrowers who entered repament over this period, but were not
included in the stud data. These patterns are both cause or
concern and an opportunit or improvement.
This report has sought to highlight these issues, promote greate
understanding o the range o borrowers eperiences in repa-
ment, and thereb promote a vigorous polic discussion about
what steps or approaches might be appropriate to improve the
circumstances aced b all borrowers (For exAmple, see Box 4)
These challenges and opportunities have a heightened impor
tance as the impact o the economic recession continues to
aect the entire postsecondar education sstem.
28 DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING
7/28/2019 Delinquency the Untold Story
31/44
Research has shown that proactive debt management strategies can have a positive eect on borrower behavior. The strategies that
appear to be most eective include engaging borrowers early, beore payment problems occur; looking at a borrowers whole nancia
situation, including not only student loans, but other consumer debt; and delivering advice through a third party (American Studen
Assistance 2010; Steiner and Teszler 2003). Oten, not enough inormation fows to borrowers between the start o the six-month grace
period and their rst reported delinquency 60 to 90 days ater their rst payment due date. Yet borrowers ability to choose options to
deal with repayment problems depends on receiving good inormation early in the process.
Box 4: eca db maa
Student loan servicers, guarant agencies, nancial aid oces,
and other organizations are oten involved in counseling and
providing inormation to borrowers to help them avoid repament
problems. Counseling might involve helping students manage
budgets to prevent repament problems or helping borrowers get
their loans back in good standing ater delinuenc or deault.
Borrowers oten have trouble understanding their loan repament
optionsincluding loan consolidation, deerment, orbearance,
and so on. The availabilit o timel education debt management
services can be a tremendous help.
Historicall, loan servicers have generall oered borrower assis-
tance onl ater delinuent loans appear to be headed toward
deault, and the tpicall involve a series o attempts to contact the
delinuent borrower. The ederal loan program reuires that manda-
tor entrance and eit counseling inormation be provided to
students while the are enrolled, to ensure that the are generall
aware o their loan pament obligations and repament options
However, this inormation is not alwas well understood or timel
Other organizations are involved in providing inormation to
students, including communit-based organizations and guarant
agencies, which provide online counseling tools, publications, and
sometimes onsite assistance, but the nature and etent o this inor
mation and assistance var widel. It is not clear what eect the
recent dissolution o the FFEL program and the transition to Direc
Loans or all new student loans will have on the provision o these
services. Loan servicers might have an etra incentive to reach outo borrowers earl to help them avoid delinuenc and deault
Guarant agencies will continue to provide the services reuired
b the FFEL program, even as the consider how the will t into
the new structure. Colleges and universities also are evaluating
their options or assistance. Given current economic pressures and
the potential or higher deault rates, debt management and nan
cial literac tools are more important than ever (Lederman 2010).
29INSTITUTE FOR HIGHER EDUCATION POLICy
7/28/2019 Delinquency the Untold Story
32/44
Opportunities for
Further Discussion
These research ndings are onl a rst attempt to eamine the etent o repament problems beond
deaultmore work is needed to ull understand the etent and nature o borrowers eperiences.
However, this report has implications or a broader discussion o student loans and how the are
structured. For eample, on the polic level, the ocus needs to shit rom simpl lowering deault rates
to considering a wider spectrum o eperiences with student loan repament, including how to increase
the number o borrowers who successull repa their student loans and whether current programsadeuatel address borrowers needs. The stud raises the ollowing uestions:
With short-term solutions such as deerment and orbearance
available to help borrowers avoid delinuenc, wh do two
out o ve borrowers become delinuent during the rst ve
ears ater entering repament?
Wh is the proportion o borrowers who are activel repaing
their student loans without dela so low? How can we ensure that
this number is higher or uture cohorts o student loan borrowers?
The current ocus on deault rate measures does not ull
capture the etent o borrowers diculties in repaing student
loans. Is there a better wa to track students who are eperi-
encing diculties with loan repament, given a highl mobile
population and the challenges o reaching cell phone users?
Can we provide inormation about repament options in a
more targeted and timel wa?
The ollowing are important areas to eplore in the uture:
Bw caacc. The stud data had limited inor
mation on borrowers demographic and enrollment character-
istics. Other variables that might aect delinuenc or deaul
such as income or more detailed enrollment histor
should be included in uture analses. In addition, surves o
borrowers attitudes toward repament could provide critica
inormation to inorm policies and institutional practices.
Faca acy. Most people agree about the importanceo providing students with general nancial literac inorma-
tion, as well as specic inormation on loan repament options
However, man borrowers are unaware o their options o
choose not to use them. This situation suggests other topics
or discussion, including how and when inormation is deliv-
ered, what agencies should be providing it, and whether i
should be promoted broadl or targeted to borrowers who are
most likel to become delinuent.
30 DELINqUENCy: THE UNTOLD STORy OF STUDENT LOAN BORROWING
7/28/2019 Delinquency the Untold Story
33/44
ia . Some institutions are seeing positive
results with deault prevention eorts, including nancial
literac. For eample, IHEPs work with minorit-serving insti-
tutions has illustrated some promising eamples o institu-
tional practice (Institute or Higher Education Polic 2009 and
2010), and man guarant agencies have supported nan-
cial literac initiatives on college campuses. However, not
enough is known about such eorts in general or about the
interventions that work best.
pa a. The data or the stud included ederal student
loans, but private loans are also a part o the euation. Private
loans do not include man o the protections o ederal loans;
however, it is in the lenders interest to use debt management
practices to mitigate the risks o deault. Because these loans
were not included in the stud, it could be understating the
ull etent o delinuenc and deault.
o a ay . The data or this stud
covered onl deerment and orbearance; it was not able to
include participation in newer repament programs, such asincome-based repament, graduated repament, and public
service loan orgiveness (see glossArY). For students who are
currentl entering repament, restructuring their loans into an
alternative repament plan might be a better long-term solu-
tion than reling on multiple deerments and orbearances.
sc. This stud provided a snapshot o the tools
and events borrowers eperienced. An important ollow-up
would be to see the order o events (e.g., the patterns o
events occurring beore and ater delinuenc or deault), as
well as the options used over a longer period o time.
The initial ndings o this stud provide important rst steps to
understanding the broader scope o borrowers eperiences
with student loans. But there is much more to do, and lowering
rates o delinuenc and deault will reuire a serious commit
ment rom man stakeholders who care about college access