Who Graduates with Excessive Student Loan Debt?
www.studentaidpolicy.com/excessive‐debt/
MK Consulting, Inc. December 14, 2015
Mark Kantrowitz
President
Student Aid Policy Analysis Papers
‐ 1 ‐
Executive Summary
Student loan debt has been growing rapidly over the last decade. But, while milestones may be impressive, what matters more is the growth in excessive student loan debt at graduation.
A borrower has excessive student loan debt when the borrower graduates with more debt than he or she can afford to repay in a reasonable amount of time, such as within 10 years of graduation. This paper defines excessive debt as occurring when the
borrower’s debt‐service‐to‐income ratio – the percentage of monthly gross income devoted to repaying student loan debt – is 10% or more under a standard 10‐year repayment plan.
This paper provides a rational justification for using the 10%
debt‐service‐to‐income ratio as a cap on affordable student loan debt. It derives the threshold by assuming that part of the after‐tax increase in income for Bachelor’s degree recipients (as compared with high school graduates) is
available to repay student loan debt, assuming a 10‐year repayment term. The 10% threshold corresponds to using half of the additional net income to repay student loan debt. The paper also derives a 15% debt‐service‐to‐income threshold as a “stretch limit” by assuming that three‐quarters of the additional net income is available to repay student loan debt.
The 10% and 15% debt‐service‐to‐income thresholds are also consistent with the rule of thumb that total student loan debt at graduation should be less than the borrower’s expected annual starting salary.
This paper presents several other new results concerning students graduating with excessive student loan debt.
The percentage of Bachelor’s degree recipients graduating with excessive student loan debt has been growing for the last three decades. But, the percentage of Bachelor’s degree recipients graduating with student loan debt who graduate with excessive debt has remained at slightly more than a quarter of Bachelor’s degree recipients who graduate with student loan debt for the last two decades. This suggests that the growth in the percentage of Bachelor’s degree recipients with excessive debt is driven by the overall growth in the percentage of students who must borrow to pay for college.
Students who graduate with excessive student loan debt are more likely to delay major life‐cycle events, such as buying a home, getting married and having children, than students who
graduate with affordable debt.
Students who graduate with excessive debt are more likely to take a job outside their field, to work more than desired and to work more than one job. They are also significantly more likely to say that their education debt influenced their employment plans.
Students who graduate with excessive debt are just as likely to own a car as students who graduate with affordable debt, but are less likely to have a car payment of $350 or more. Thus, graduating with excessive student loan debt can be the difference between owning a new car and a used car.
Students who graduate with excessive debt are less likely to feel that their undergraduate education was worth the financial cost, as of one year after graduation.
A borrower has excessive
student loan debt when 10% or
more of a borrower’s gross
income must be devoted to
repaying the borrower’s
student loan debt, assuming a
10-year repayment term
Growt
Rapid gro and the p in 2010, a While the matters m (NPSAS), d graduate
But, the g their loan
$0
$10,000
$20,000
$30,000
$40,000
$50,000
$60,000
th in Stud
wth in the am ublic to the “ auto loans in 2 e growth in ou more. This cha demonstrates with student
growth in deb s. So, it is imp
45.5%
49.4%
53
1992‐93 1993‐94 1994‐95
0 0 0 0 0 0 0
Gro
dent Loa
mount of outs
“student loan 2011 and $1 t utstanding stu art, which is b
s the steady g loan debt an
bt at graduatio portant to as 3.7%
58.4%
59.3%
60.3%
61
1995‐96 1996‐97 1997‐98
wth in S for Ba
Average Stud
an Debt
standing stud problem.” O trillion in 201 udent loan de based on data growth in the nd the averag
on is not nece k: How many
% 1.2%
62.2%
62.7%
63.2
1998‐99 1999‐00 2000‐01 2001‐02
Student achelor's
dent Debt
‐ 2 ‐ dent loan deb
utstanding st 12, now secon ebt is notewo a from the Na e percentage
e debt at gra
essarily a pro y students are
2%
63.7%
64.2%
64.8%
65
200102 2002‐03 2003‐04 2004‐05
Loan De s Degree
% Gra
bt has drawn t tudent loan d nd only to hom orthy, the imp
ational Postse of Bachelor’s duation for th
blem, if most e graduating w
% 5.4%
65.9%
66.5%
67.1
2005‐06 2006‐07 2007‐08 2008‐09
ebt at Gr e Recipie
duating with
the attention ebt exceeded me mortgage pact on indivi econdary Stud s degree recip
hose who gra
t borrowers c with excessive 1%
67.7%
68.3%
69.0%
6
200809 2009‐10 2010‐11 2011‐12
raduatio ents
Student Loan
of news med d credit card e debt.
dual borrowe dent Aid Stud pients who aduate with d
can afford to r e debt?
% 9.6%
70.2%
70.9%
2012‐13 2013‐14 2014‐15
on
ns
dia debt
ers dy debt.
repay
0%
10%
20%
30%
40%
50%
60%
70%
80%
‐ 3 ‐
Defining Excessive Debt
Student loan debt is excessive when the borrower cannot afford to repay it in full within a reasonable amount of time, such as within 10 years or less after graduation.
The standard repayment term for federal student loans is 10 years. Borrowers can reduce their monthly loan payment by choosing a longer repayment term. For example, extended repayment and income‐
driven repayment plans can increase the repayment term to 20, 25 or even 30 years. But, increasing the repayment term also increases the total amount paid over the life of the loan.
Borrowers who increase the repayment term will still be repaying their own student loans when their children enroll in college. They will be less likely to have saved for their children’s college education and they will be less willing to borrow to help them pay for college. Thus, the burden of repaying excessive student loan debt will be severe enough to affect the next generation’s ability to pay for college.
Determining whether the student loan debt is affordable requires a comparison of debt with income, not debt by itself. Six‐figure student loan debt might be excessive for a borrower with just a Bachelor’s degree in a low‐paying liberal arts field, but not for a borrower who earns an advanced degree, such as an M.D. in a lucrative specialty like oncology, cardiology or orthopedics. To determine whether the student loan debt is affordable, the monthly loan payment must be compared with monthly income.
Many students pursue a college education in order to obtain a better‐paying job. Accordingly, it is reasonable to require that the incremental increase in net income after taxes from obtaining a college degree should be sufficient to repay the student loan debt. Moreover, one could argue that half of the increase in take‐home pay should be available for repaying student loans and half for other priorities, so that the college graduate derives an immediate financial benefit from earning a college degree.
Table 502.30 of the 2014 Digest of Education Statistics provides data on the median annual earnings of full‐time, year‐round workers age 25 to 34 by educational attainment for selected years from 1995 to 2013.1 This data is based on the U.S. Census Bureau’s Current Population Survey (CPS). In 2013, the median annual earnings for full‐time, year‐round workers with only a Bachelor’s degree in the decade from age 25 to age 34 was $48,530, compared with $30,000 for workers with just a high‐school diploma or GED. This yields an average $18,530 increase in annual gross income from obtaining a Bachelor’s degree. Assuming a 25% federal income tax rate, 7.65% in FICA taxes and up to 9.8% in state and local income taxes, this yields $10,655 after taxes, or 22.0% of gross income. Half of this figure is 11.0%, setting a reasonable limit on the percentage of income available to repay student loan debt.
Table P‐24 of historical income tables for the Current Population Survey (CPS)2 provides similar data concerning median earnings by educational attainment for full‐time, year‐round workers age 25 and older.3 However, unlike table 502.30, this data is not limited to the decade from age 25 to 34.
Nevertheless, this income data yields a similar result, capping the percentage of income reasonably available to repay student loan debt at 11.2% of gross earnings in 2013 and 2014 and around 10.2% in the early 1990s.
1 http://nces.ed.gov/programs/digest/d14/tables/dt14_502.30.asp
2 https://www.census.gov/hhes/www/income/data/historical/people/
3 https://www.census.gov/hhes/www/income/data/historical/people/2014/p24.xls
‐ 4 ‐
This chart shows limits based on percentages of historical income data derived from Table 502.30.
As this chart demonstrates, the debt‐service‐to‐income percentage has averaged about 10.0% for the last two decades. Using a similar approach to base the stretch limit on three‐quarters of the increase in net income after taxes would yield a percentage of gross income that averages around 15%.
Thus, student loan debt at graduation should be considered affordable if the monthly loan payments assuming a 10‐year repayment term are less than 10% of gross monthly income. The student loan payments will still be affordable, but more of a financial stretch for the borrower, if the payments are less than 15% of gross monthly income.
To set this in context, this table shows a histogram of debt‐service‐to‐income ratios for Bachelor’s degree recipients in 2007‐08 based on 2009 income, using the 2012 follow‐up to the 2007‐08 Baccalaureate & Beyond longitudinal study (BB12). The mode is at 7% and the median is at 8%.
9.4%
10.7%
9.8%
9.1%
10.2% 10.0%
9.6% 9.6% 9.6%
10.4% 11.0%
0.0%
2.0%
4.0%
6.0%
8.0%
10.0%
12.0%
1995 2000 2003 2005 2007 2008 2009 2010 2011 2012 2013
Percentage of Income Available to Repay Student Loans, 1995‐2013
0 2 4 6 8 10
1% 2% 3% 4% 5% 6% 7% 8% 9% 10% 11% 12% 13% 14% 15% 16% 17% 18% 19% 20% 21% 22% 23% 24% 25% 26% 27% 28% 29% 30% 31% 32% 33% 34% 35%
Distribution of Monthly Loan Payment as Percent of Monthly Income
‐ 5 ‐
This table shows the percentage of Bachelor’s degree recipients graduating in 2007‐08 with excessive student loan debt for various caps on the debt‐service‐to‐income ratio, based on BB12 data. Thus, the 10% threshold corresponds to about a quarter of Bachelor’s degree recipients graduating with excessive student loan debt.
Cap on
Debt‐Service‐to‐Income Ratio
Percent of
Bachelor’s Degree Recipients Graduating with Excessive Debt
5% 49.3%
8% 33.7%
10% 27.2%
15% 16.2%
20% 10.0%
25% 6.6%
Note that data for the first year after graduation tends to represent a ceiling on the amount of student loan debt and a floor on the amount of income, since student loan debt tends to decrease and income tends to increase over the borrower’s career. So, the debt‐to‐income and debt‐service‐to‐income ratios will tend to decrease over time.
This definition of excessive debt in terms of the debt‐service‐to‐income ratio represents a more direct measure of whether students are graduating with affordable student loan debt than the cohort default rate (CDR). The cohort default rate measures the percentage of students entering repayment during one federal fiscal year who default by the end of a subsequent federal fiscal year. Defaulting on a student loan is a potential consequence of excessive debt, but not the only possible consequence. Also, some borrowers default on their student loans for reasons other than affordability of the debt. So, the cohort default rate is, at best, an indirect measure of whether students are graduating with affordable student loan debt. The CDR does not measure other non‐paying statuses, such as deferments, forbearances and delinquencies. Moreover, borrowers who graduate with excessive debt might reduce the debt‐service‐
to‐income ratio by choosing alternate repayment plans that stretch out the repayment term instead of defaulting. Also, the cohort default rate is prone to manipulation by colleges that encourage borrowers to apply for deferments and forbearances, pushing the default outside the measurement window.
The derivation of a 10% to 15% cap on the debt‐service‐to‐income ratio is also more rational than the flawed reasoning used to justify an 8% cap on the debt‐service‐to‐income ratio in the 2014 gainful employment regulations.4 The regulations selectively cited papers that based an 8% cap on the
difference between mortgage underwriting standards for all debt and mortgage underwriting standards for mortgage debt. In addition to arbitrarily overlooking similar approaches that justified different caps on the debt‐service‐to‐income ratio, the regulations’ derivation of the student loan cap from mortgage underwriting standards assumes that mortgage lenders know more about affordable student loan debt than student loan experts. It also assumes that student loan debt is affordable only if the borrower is able to afford to buy a home instead of renting an apartment.5
4 Federal Register 79(211):64890 ‐65103, October 31, 2014. See specifically the discussion that begins on page 64917. https://www.gpo.gov/fdsys/pkg/FR‐2014‐10‐31/pdf/2014‐25594.pdf
5 The Administrative Procedure Act requires federal agencies to provide a reasoned basis for justifying regulations, but there is no requirement that the reasoned basis involve good reasoning.
Rule o
The rule o starting sa shown by consisten
Percen
Using the possible t This chart student lo 10% cap o term. Dat longitudin Survey of Bachelor’
0 2 4 6 8 10 12 14 16
of Thumb
of thumb that alary corresp y this table. T t with the dis
In
ntage of C
10% debt‐se o calculate th t shows how t oan debt has on the debt‐s ta for 1993‐94 nal studies. D
Recent Colle s degree reci
6.7
0.0%
2.0%
4.0%
6.0%
8.0%
0.0%
2.0%
4.0%
6.0%
1976
Pe
b: Total D
t total studen onds to a per This table assu
stinction betw
nterest Rate 0.0%
1.0%
2.0%
3.0%
3.4%
4.0%
4.29%
College G
ervice‐to‐inco he percentage
the percentag changed ove ervice to inco 4, 2000‐01 an ata for 1976‐
ge Graduates pients who a
7% 6.
6‐77 198
ercent o who Gr
Debt Les
nt loan debt a rcentage of gr umes a 10‐ye ween affordab
Percentage Gross Incom
10.
10.
11.
11.
11.
12.
12.
Graduate
me ratio as a e of college s ge of Bachelo r the last fou ome ratio, inc nd 2007‐08 ar
‐77, 1985‐86 a s (RCG). This c re graduating
5%
8
5‐86 198
f Bachel raduate
‐ 6 ‐
ss than A
t graduation ross monthly ar repayment ble and exces e of
me Interes .0%
.5%
.0%
.6%
.8%
.1%
.3%
es with E
threshold on tudents who or’s degree re r decades. Ex come the yea re based on th
and 1989‐90 chart shows a g with excessi
.3%
9
89‐90 19
lor's Deg with Exc
Annual In
should be les income with t term. Acco ssive student
t Rate Perc Gros 5.0%
6.0%
6.8%
7.0%
8.0%
8.5%
9.0%
Excessive
n affordable v graduate wit ecipients who xcessive stude
r after gradua he Baccalaure are based on an increasing ive student lo
9.8%
1
993‐94 20
gree Rec cessive D
ncome
ss than the ex in the range rdingly, the r loan debt.
centage of ss Income
12.7%
13.3%
13.8%
13.9%
14.6%
14.9%
15.2%
Debt
vs. excessive d th excessive s o graduated w ent loan debt
ation and a 1 eate and Bey n a similar ser trend in the oan debt.
11.9%
000‐01 2
cipients Debt
xpected annu of 10% to 15%
ule of thumb
debt, it becom student loan d with excessive
is based on t 0‐year repay ond (B&B) ies of studies percent of
14.4%
2007‐08
ual
%, as is
mes debt.
e the
ment s, the
This chart student lo recipients Bachelor’
loan debt
This demo excessive are gradu pay for co to familie Overall, a for 2007‐0 The avera Women a higher ave income th Caucasian American in income than for C parent ed
6 Catherine College Gra 0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
30.0%
35.0%
t uses the sam oan debt as a s. When those
s degree reci .
onstrates tha debt. It sugg ating with ex ollege, due to
s.
n average of 08 Bachelor’s age decreases are more likel
erage debt‐se han men afte n students are
students (14 e for minority Caucasian stu ducation debt
e Hill and Chris aduation, AAU 22.9%
%
%
%
%
%
%
%
%
1976‐7
Pe Stud
me data to ca percentage o e who gradua
pients who b
t a relatively ests that the xcessive debt a shift in the
12.6% of mo s degree recip s to 8.9% whe y to graduate ervice‐to‐inco r graduation, e more likely 4.6%) or Hispa
students wh dents. It coul t.
stianne Corbet UW, 2012. http
%
14.3
7 1985‐
ercent of dent Loan
lculate the pe of those grad ate with no de
orrowed for t
constant per recent growt is due to grow e burden of pa
nthly income pients who gr en students w e with excessi ome ratio (13
even when e to graduate w anic or Latino
o obtain colle d also be due
t, Graduating t ://www.aauw 3%
20
‐86 198
Bachelor ns who Gr
‐ 7 ‐ ercentage of
uating with s ebt are omitt their educatio
centage of st th in the perc wth in the pe aying for colle
e is used to re raduated with who graduate
ive debt than .2% to 11.6%
employed in t with excessiv o students (21
ege degrees, e to differenc
to a Pay Gap: T .org/research/
0.0%
2
89‐90 19
r's Degree raduate w
Bachelor’s de tudent loan d ted, slightly m on are gradua
tudents who b centage of Ba ercentage of s
ege from the
pay student l h student loan
d with no deb n men (29.1%
%), mostly due the same occ
e debt (30.0%
1.2%). This ma even though ces in the bala
The Earnings o /graduating‐to‐
28.9%
993‐94
e Recipie with Exce
egree recipien debt, not all B more than a q
ating with exc
borrow are gr chelor’s degr students who federal and s
loans the yea n debt, based bt are include to 24.3%). Th e to women e
upations.6
%) than Black ay be due to
the average ance between
of Women and
‐a‐pay‐gap/
27.5%
2000‐01
nts with essive Deb
nts with exce Bachelor’s de
uarter of cessive stude
raduating wit ree recipients o must borrow
state governm
r after gradu d on BB12 dat
ed.
hey also have earning lower
or African‐
a greater incr income is low n student and
Men One Year 27.2%
2007‐08
bt
ssive gree ent
th s who w to ments
ation ta.
e a
rease wer d
r after
‐ 8 ‐
Students majoring in theology (64.7%), law and legal studies (43.1%), communications (36.2%), agriculture (34.1%), education (33.9%), humanities (33.5%) and design (33.1%) are more likely to graduate with excessive debt, presumably due to the lower income in these fields of study. Students majoring in engineering (16.3%) and computer science (23.8%) are less likely to graduate with excessive debt, presumably due to the higher income in these fields of study.
Students who borrowed private student loans are more likely to graduate with excessive student loan debt (36.4% vs. 22.8%). This may be because it is more difficult for dependent students to borrow excessively with federal student loans alone, given the low annual and cumulative loan limits on federal student loans.
A college’s annual cost of attendance correlates with the percentage of students graduating with excessive student loan debt, as shown in this table based on BB12 data. College costs are a key driver of debt at graduation, so higher‐cost colleges are more likely to have more students graduating with excessive student loan debt.
2007‐08 Annual Cost of Attendance
Percent Graduating with Excessive Debt
Less than $10,000 22.2%
$10,000 to $19,999 24.7%
$20,000 to $29,999 29.5%
$30,000 to $39,999 34.6%
$40,000 or more 34.6%
The college affordability index7 correlates with excessive student loan debt. Students for whom the college affordability index is 75% or more are more likely to graduate with excessive debt than students for whom the college affordability index is less than 25% (32.7% vs. 24.8%).
Dependent students (as defined for federal student aid purposes) are more likely to graduate with excessive debt than independent students (30.2% vs. 23.2%). This is largely because students who are age 30 or older are much less likely to graduate with excessive student loan debt (16.4%), slightly more than half the rate for younger students.
College graduates who work for the military (22.3%) or government (22.9%) are less likely to have excessive student loan debt than graduates who work for a for‐profit company (27.6%), a non‐profit organization (31.2%) or who are self‐employed (36.6%). Students who work in science, technology, engineering and mathematics (STEM) occupations are much less likely to graduate with excessive debt (14.8%) than students who work in non‐STEM occupations (28.2%).
7 The college affordability index is the ratio of the net price to total family income. Colleges that are more affordable for a student have a lower college affordability index. The net price is the difference between the annual cost of attendance and gift aid (grants, scholarships and other money that does not need to be earned or repaid).
This chart loan debt
This chart have exce based on
Home‐sch high scho Students w 26.8%), pe
0 5 10 15 20 25 30 35 40 45
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
70.0%
t shows that h than low‐inc
t shows that B essive student
BB12 data.
hooled studen ol diploma (2 who receive erhaps becau
39
0.0%
5.0%
0.0%
5.0%
0.0%
5.0%
0.0%
5.0%
0.0%
5.0%
First (Less t
10.0%
Less than
$10,000
$1
$
high‐income B come Bachelo
Bachelor’s de t loan debt th
nts are more 27.1%) or GED
private schola use scholarsh
9.2%
Quintile than 20%)
Sec (2
P by An
20.6%
27.
0,000 to 19,999
$20,0
$29
P by Cumul
Bachelor’s de or’s degree re
egree recipien han Bachelor’
likely to grad D (25.9%).
arships are m ip recipients a
36.7%
cond Quintile 20% to 39%)
Percent wi nnualized S
.9%
35.9%
000 to ,999
$30,000
$39,99
Percent wi lative Loan
‐ 9 ‐ egree recipien ecipients, base
nts with highe
’s degree reci
duate with exc
more likely to are more like
27.9%
Third Quintil (40% to 59%
th Excessiv Salary Perc
% 38.5%
0 to 99
$40,000 to
$49,999
ith Excessi n Amount
nts are less lik ed on BB12 d
er student loa ipients with lo
cessive debt
graduate wit ely to enroll at
19.3%
le
%)
Fourth Qu (60% to 7
ve Debt centile, 20
36.8%
o $50,000 to
$59,999
$6
$
ve Debt through 20
kely to have e data.
an balances a ower student
(48.3%) than
h excessive d t higher‐cost
%
9
uintile 79%)
Fifth (80%
009
57.9%
54
60,000 to
$69,999
$70,0
$79
007‐08
excessive stud
re more likely t loan balance
students wit
debt (31.2% vs colleges.
9.7%
Quintile or more)
4.9%
61.1%
000 to 9,999
$80,000 more
dent
y to es,
h a
s.
%
0 or e
‐ 10 ‐
High school GPA, undergraduate GPA, SAT test scores, disability status, participation in study abroad, status as a transfer student, institutional selectivity, in‐state vs. out‐of‐state enrollment and Federal Pell Grant recipient status do not seem to have a significant impact on whether the student graduates with excessive debt or not.
Consequences of Graduating with Excessive Debt
Students who graduate with excessive student loan debt are more likely to have borrowed private student loans (42.8% vs. 27.3%), based on BB12 data. Dependent undergraduate students cannot graduate with excessive debt using only federal student loans.
Students who graduate with excessive student loan debt are more likely to feel that their undergraduate education was not worth the financial cost as of the year after graduation (35.8% vs. 22.2%) and more likely to say that the debt influenced their employment plans (64.3% vs. 41.6%), based on BB12 data.
Students who graduate with excessive debt are less likely to have a car payment of $350 or more as of four years after graduation (44.9% vs. 54.1%) and to be paying a mortgage (34.1% vs. 42.5%), based on BB12 data. But, borrowers appear to be equally likely to have a car payment. Rather, excessive debt seems to manifest itself in a lower car payment and a lower monthly rent or mortgage payment, so that the student loan payment is a greater percentage of overall household debt payments. Curiously, students who graduate from for‐profit colleges tend to have higher car payments and higher monthly rent or mortgage payments, despite their higher student loan debt.
These are some of the other reported consequences of graduating with excessive debt, based on BB12 data:
Delayed buying a home (49.8% vs. 38.1%)
Delayed getting married (27.1% vs. 20.9%)
Delayed having children (36.4% vs. 27.9%)
Took a job instead of enrolling in further postsecondary education (43.3% vs. 33.0%)
Took a job outside of field (50.8% vs. 36.4%)
Work more than desired (47.8% vs. 36.4%)
Worked more than one job (33.0% vs. 23.4%)
Potential Compliance with Gainful Employment
The average debt‐service‐to‐income ratio is 15.2% for private non‐profit colleges, 12.6% for private for‐
profit colleges and 11.0% for public colleges, based on BB12 data.
Using the 8% debt‐service‐to‐income ratio threshold used in the gainful employment regulations, 40.0%
of Bachelor’s degree recipients at private non‐profit colleges would have excessive debt, compared with 33.8% of Bachelor’s degree recipients at private for‐profit colleges and 30.0% of Bachelor’s degree recipients at public colleges, based on BB12 data. This suggests that private non‐profit colleges would have more problems complying with the gainful employment regulations than private for‐profit colleges, if the gainful employment regulations were applied to Bachelor’s degree recipients at private non‐profit colleges. Most Bachelor’s degree granting institutions would have problems complying with the gainful employment regulations even if the threshold for affordable debt were set at 15%.
Correl
Previous s likely to d graduatio financial s the previo A previou the debt‐s student lo percentag
8 Mark Kan www.finaid
9 Defermen Interest m
2
lation of
studies by oth default than b on presented stress – the fa ous studies.
s study by th service‐to‐inc oan default.8 ge of borrowe
ntrowitz, Relati d.org/educato nts and forbea ay continue to 1
0.0%
2.0%
4.0%
6.0%
8.0%
10.0%
12.0%
14.0%
16.0%
18.0%
20.0%
5%
Excessiv
her researche borrowers wit in this report ailure to keep
e author of th come ratio, su This chart sho ers in a defer
ionship of Defa rs/20100817a arances are tem o accrue during
12.6%
or more 8
Perc by Deb
ve Debt w
ers have repo th higher deb t is not consis p debt in sync
his paper dem uggesting tha ows that the ment or forbe
ault Rates to D ffordabilityme mporary suspe g a deferment o
14.2%
8% or more
cent in Def bt‐Service‐
‐ 11 ‐
with Debt
orted that bor t levels. The d tent with the c with income
monstrates th at graduating debt‐service‐
earance, an in
Debt and Incom asures.pdf nsions of the o or forbearance
15.0%
10% or more
ferment/Fo
‐to‐Income
t and Inc
rrowers with data concern ese previous s e – does not e
hat default rat with excessiv
‐to‐income ra ndicator of fi
me, August 17, obligation to m e.
15.7%
e 15% or m
orbearanc e Ratio, 20
come
lower debt le ing excessive studies. This s explain the re
tes increase w ve debt can b atio correlate nancial stress
2010.
make payments
%
17
more 20% o
e 009
evels are mor e debt at
suggests that esults reporte
with increase be a cause of
s with the s.9
s on a student .6%
r more
re
ed by
es in
loan.
This chart
This chart percent in
0 0 2 2 3 3
0 0 2 2 3 3 4 4 5
t shows that t
t shows that h n default in 20 2.2
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
3.0%
3.5%
5% or
1.1
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
4.5%
5.0%
Less
$10,
by
the debt‐serv
higher cumula 012, based on 2%
2.
more 8% or
by 200
1%
1.
than ,000
$10,0
$19
Cumulativ
vice‐to‐incom
ative undergr n BB12 data.
6% 2.
r more 10% o
Percent in 09 Debt‐Se
9% 2.
000 to 9,999
$20,0
$29
Percent in ve Undergr
‐ 12 ‐ e ratio correl
raduate debt .5%
2
or more 15% o
n Default in ervice‐to‐I
.0%
3
000 to 9,999
$30,
$39
n Default i raduate De
ates with def
through 200 .8%
3
or more 20% o
n 2012 ncome Ra
.6% 3
,000 to 9,999
$40,
$4
n 2012 ebt throug
fault rates.
7‐08 correlat
3.1% 3
or more 25%
tio
3.5%
4
,000 to 9,999
$50
$7
gh 2007‐08
tes with incre
3.2%
or more
4.7%
0,000 to 74,999
8
ased
This chart increasing
This chart increasing
0 10 20 30 40 50 60 70
0 5 10 15 20 25 30 35 40 45
t shows that t g debt, based
t shows that t g income, bas 11.4%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Less th
$10,00
0.0%
5.0%
0.0%
5.0%
0.0%
5.0%
0.0%
5.0%
0.0%
5.0%
Less t
the percentag on BB12 dat
the percentag sed on BB12 d
%
20.5%
han 00
$10,000 to
$19,999
P by Cumu
38.0%
than $25,000
Pe by
ge of Bachelo ta.
ge of Bachelo data.
26.2%
o $20,000 to
$29,999
$3
$
Percent wi lative Und
25.2%
$25,000 to $
ercent wit Annualiz
‐ 13 ‐ or’s degree re
or’s degree re 36.8% 35.
30,000 to
$39,999
$40,0
$49,
th Excessiv dergraduat
%
$49,999 $50,
th Excess zed Salary
cipients with
cipients with .6%
43.9%
000 to ,999
$50,000
$74,99
ve Debt te Debt, 20
10.4%
000 to $74,999
ive Debt y in 2009
excessive de
excessive de
%
62.5%
to 9
$75,000 to
$99,999
007‐08
1.3 9 $75,000 to
9
ebt increases
ebt decreases
66.0%
$100,000 or more
% o $99,999
with
s with
This chart service‐to
This chart higher de 0 5 10 15 20 25 30
0 5 10 15 20 25 30 35 40
t shows that B o‐income ratio
t shows that B bt‐service‐to 0.0%
5.0%
0.0%
5.0%
0.0%
5.0%
0.0%
Less t
7.6%
0.0%
5.0%
0.0%
5.0%
0.0%
5.0%
0.0%
5.0%
0.0%
Less th
$10,00
by
Bachelor’s de os, based on
Bachelor’s de
‐income ratio 23.5%
than $25,000
Average by
%
10.3%
han 00
$10,000 to
$19,999
Average y Cumulat
egree recipien BB12 data.
egree recipien os, based on B
9.8%
$25,000 to $
e Debt‐Se Annualiz
12.7%
o $20,000 to
$29,999
$3
$
e Debt‐Se tive Unde
‐ 14 ‐ nts with highe
nts with highe BB12 data.
$49,999 $50,
ervice‐to‐
zed Salary
14.3%
18.
30,000 to
$39,999
$40,0
$49,
ervice‐to‐
ergradua
er annualized
er cumulative 6.3%
000 to $74,999
‐Income R y in 2009
.2% 20.0%
000 to ,999
$50,000
$74,99
‐Income R te Debt,
salaries have
e undergradua 4.1
9 $75,000 to
Ratio 9
%
27.3%
to 9
$75,000 to
$99,999
Ratio 2007‐08
e lower debt‐
ate debt have
%
o $99,999
35.2%
$100,000 or more
‐
e
‐ 15 ‐
This chart shows that the distribution of debt is similar across income strata, so it is unlikely for borrowers with low debt to also have much lower income, thereby, yielding higher debt‐service‐to‐
income ratios. So, even if there is a minimum income threshold required for basic living expenses, it is unlikely to cause differences in ability to repay student loan debt for borrowers with low cumulative undergraduate debt.
This chart is similar, but shows the distribution of annualized salary in 2009 by cumulative
undergraduate debt. It shows that low‐income graduates aren’t any more or less likely to have lower cumulative undergraduate debt.
Perhaps, the results found in other studies depend more on college completion than the amount of student loan debt. After all, students who drop out of college tend to have less student loan debt than
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
30.0%
35.0%
Less than
$10,000
$10,000 to
$19,999
$20,000 to
$29,999
$30,000 to
$39,999
$40,000 to
$49,999
$50,000 to
$59,999 Cumulative Undergraduate Debt
Distribution of Cumulative Undergraduate Debt by Annualized Salary in 2009
Less than $25,000
$25,000 to $49,999
$50,000 to $74,999
$75,000 to $99,999
$100,000 or more
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
Less than
$25,000
$25,000 to
$49,999
$50,000 to
$74,999
$75,000 to
$99,999
$100,000 or more Annualized Salary in 2009
Distribution of Annualized Salary in 2009 by Cumulative Undergraduate Debt
Less than $10,000
$10,000 to $19,999
$20,000 to $29,999
$30,000 to $39,999
$40,000 to $49,999
$50,000 to $59,999
‐ 16 ‐
students who graduate because they are enrolled for fewer payment periods. Thus, the amount of debt may be a dependent variable and not an independent variable, influenced by college completion as a confounding factor.
Recommendations
It is increasingly important to understand the consequences of excessive student loan debt because of the increasing prevalence of student loan debt.
This paper discusses excessive student loan debt by Bachelor’s degree recipients only, due to the limitations of available data. It does not report on excessive debt by recipients of other degrees, such as Associate’s degrees, Certificates and more advanced degrees. It also does not report on excessive debt owed by students who drop out of college. It does not report on other reasons why students might fail to repay their student loans, such as dissatisfaction with the quality of their education.
Recommendation: The U.S. Department of Education should initiate a study of the consequences of excessive student loan debt. This study can be implemented by tracking outcomes for all degree levels and for non‐completers in a manner similar to that of the Baccalaureate & Beyond (B&B) longitudinal study, namely as a follow‐up to a subset of participants in the quadrennial National Postsecondary Student Aid Study (NPSAS). Congress should provide sufficient funding for this study to ensure that it occurs. Without funding, there is no guarantee that this study will be implemented, even if mandated by Congress.
Although the 2009 follow‐up to the 2003‐04 Beginning Postsecondary Students longitudinal study (BPS:04/09) may be used to calculate the percentage of college graduates with Associate’s degrees and Certificates who have excessive debt, the data is not comparable with the data used in this paper from the Baccalaureate & Beyond (B&B) study for Bachelor’s degree recipients. The B&B study provides income data one year after degree attainment. The BPS study provides income data income data in 2009 for cumulative degree attainment during the prior six years. Even if one restricts the degree attainment to students who received a degree in 2007‐08, the BPS study still involves an additional restriction to students who first enrolled in 2003‐04. The BPS study also does not address whether graduate and professional school students are graduating with excessive student loan debt. Thus, the data in this table is at best suggestive.
BPS:04/09
Cumulative Persistence and Degree Attainment as of 2008‐09
Percent with Excessive Debt
Percent of Borrowers with Excessive Debt
Attained Bachelor’s Degree 15.6% 24.4%
Attained Associate’s Degree 5.2% 8.9%
Attained Certificate 2.8% 4.7%
No Degree, Still Enrolled 2.4% 5.3%
No Degree, Left without Return 4.5% 9.5%
Colleges often refer to student loans as a form of financial aid, arguing that student loans make college more affordable. Student loans delay the repayment obligation, but do not reduce or eliminate it. But, despite widespread claims that student loans make college more affordable, few, if any, colleges track whether their students are graduating with affordable debt. Monitoring long‐term trends is necessary for anticipating and identifying problems when they occur. If too many students are graduating with
‐ 17 ‐
excessive debt, it can affect the college’s reputation. It can also have an impact on charitable contributions to the college. Increasing awareness is the first step in exercising restraint.
Recommendation: Each college and university should annually track the percentage of students graduating with excessive debt. The data should be disaggregated by degree level.
Recommendation: Alternately, the U.S. Department of Education could track the percentage of students graduating with excessive debt for each college and university. This will require tracking of private student loans in addition to federal student loans. It will also require an approach that protects the privacy of individual student data, especially data concerning income. Perhaps, it could be implemented in a manner similar to the one used for the gainful employment regulations.
Recommendation: Colleges and universities should include a discussion of excessive debt and the consequences of borrowing too much in their loan counseling programs. Students who are predicted to graduate with excessive debt based on borrowing patterns and academic major should be targeted for more aggressive loan counseling and financial literacy training.
Recommendation: Financial aid award letters should be standardized to clarify distinctions between loans and grants. Increasing awareness of debt is the first step toward exercising restraint. The financial aid award letters should include an “excessive debt alert” for students who are predicted to graduate with excessive debt based on past borrowing patterns.
