EVEN IF WE SAY IT AIN’T SO, AMERICA HAS A STUDENT DEBT PROBLEM
The main contention in this report is that student loans should not be the centerpiece of the U.S. financial aid system,
particularly for low-income students who turn to financial aid to bridge access to higher education. Instead, student
borrowing may work best as a strategy for upper-income families, who can use student loans strategically, to make it
possible to attend more expensive institutions than they could otherwise afford, reserve parental asset stores for
other purposes, and/or prevent students from having to work while in school. While a case could be made that the
student loan program should not be a part of the federal government’s investment in facilitating access to postsecondary education,
at a minimum we contend that the program should be gradually returned to the level seen in
the 1970’s, when loans constituted a much smaller portion of the financial aid packages for most students, as well a
smaller footprint within institutional budgets (Martin & Lehren, 2012). Relying on student loans as a primary
mechanism for financing post-secondary education is a recent development. Geiger and Heller (2011) found that
federal, state, and private grants were the largest form of financial aid until 1982, when loans began to outpace
grants. During most years, this has only escalated. For example, in 2000, student loans made up 38 percent of net
tuition, fees, room, and board;
by 2013 they made up 50 percent (Greenstone, Looney, Patashnik, & Yu, 2013). This
policy shift coincided with the solidifying expectation that students—not states or institutions—would bear most of
the burden of college costs (Hiltonsmith, 2014). This shift fell particularly hard on disadvantaged students, who
were least likely to have a cushion of parental support to sustain them (Elliott & Friedline, 2012) and the support of
well-positioned K-12 institutions that could help them navigate the non-loan aid (Dynarski, 2002; Marin, 2002).
While students in poverty continue to face the greatest obstacles to financing post-secondary education, today
student debt is a problem for the many, not the few. Fry (2012) found that 40 percent of all households headed by
individuals younger than 35 years of age have outstanding student debt. Further,
the proportion of undergraduate
students who took out federal loans increased from 23 percent in 2001‒02 to 35 percent in 2011‒12 (College Board,
2012). This borrowing adds up. Total borrowing for college hit $113.4 billion for the 2011‒12 school year, up 24
percent from five years earlier (College Board, 2012). As a result, households are faced with ever-growing debt, and
this heavy borrowing threatens the financial security of more than just young college students. According to the
Federal Reserve Bank of New York, about 2.2 million Americans 60 years of age or older were liable for repayment
of $43 billion in federal and private student loans in 2012,
up $15 billion from 2007 (Greene, 2012).
Some research correctly notes that about 20 percent of the growth in student debt since 1989 can be explained by
increases in the number of people attending post-secondary education (Akers & Chingos, 2014), a core objective of
the student loan program as it is framed in the policy debate. But while 20 percent is meaningful, that still leaves 80
percent left unexplained by growth in attendance, revealing it to be an inadequate explanation for the sharp upward
trajectory of student borrowing.
Some of these same researchers also correctly point out that growth in the number
of people continuing their educations into post-graduate study has played an important role in the rise in student
debt, by extending the educational path and, therefore, increasing the total cost of higher education (Akers &
Chingos, 2014; Delisle, 2014). As Akers & Chingos (2014) indicate, student debt levels among graduate degree
holders quadrupled ($10,000 to $40,000) between 1992 and 2010; however, during the same time period student
debt amounts among bachelor degree holders almost tripled ($6,000 to $16,000), again pointing to the inadequacy of
this supposed explanation for increasing debt. Certainly, looking at amount of borrowing as an indicator of the scope
of the problem, just because bachelor debt only tripled does not mean that it does not warrant serious concern.
As other forms of financial aid, particularly means-tested grant assistance, have eroded in value, while college costs
have risen, not only have students become more likely to borrow; they are borrowing more. The amount borrowed
by the average student increased only marginally between 2003-04 and 2007-08; however, debt loads increased by
$4,700 (19 percent) between 2007-08 and 2011-2012 for bachelor’s degree recipients and by $3,100 (23 percent) for
associate’s degree holders (Miller, 2014). So, judged by indicators of incidence and amount of debt, this perceived
‘crisis’ only appears to be growing. Critically, however, viewing America’s student debt problem using only an
accounting lens largely obscures the most significant issues. Particularly given the role higher education plays in
facilitating economic mobility, energizing the U.S. economy, and sustaining the American Dream, our real concern
8
should be about outcomes, issues of equity,
and the real risk that the U.S. might inadvertently imperil our most
treasured collective narrative in our almost unthinking adherence to a mantra that student debt is, at least basically, a
sound way to finance an advanced education.
A STANDARD OF “STUDENTS EVENTUALLY RECOVER” IS NOT GOOD ENOUGH
When the college debt conversation takes into consideration outcomes and equity some media and researchers speak
as though there is no problem unless college debt rises to the level of creating a national financial crisis. This is too
narrow of a lens from which to view the student debt problem. A financial aid program that in 2011-2012 school
year cost Americans $70.8 billion (College Board, 2012),
from which the federal government earned $41.3 billion in
2013 (Jesse, 2013), and from which Sallie Mae (the nation’s largest private student loan lender) made $939 million
in net profit for 2012 (Hartman, 2013) should be held to a higher standard than ‘some students will eventually
recover’. It should actually have to be shown to be an equalizing force with regard to educational attainment and
financial wellbeing. It is not just our financial investment that makes this so important; we have also invested
considerable regulatory and legislative legitimacy in student borrowing, imbuing these loans with tremendous
significance in our policy structure—including preventing discharge in bankruptcy and deploying collection
agencies to recoup debts—in ways that ‘leak’ into other policy dimensions.
There is little evidence that we are holding student loans to this higher standard, however. Today, largely as a result
of media depiction, problematic debt has been roughly defined as $100,000 or more. This inaccurate definition
moves the goalposts, allowing some analysts, policymakers, and pundits to divert attention from the negative effects
associated with student debt by merely demonstrating that average student debt is far less (latest estimates $29,400
(Miller, 2014)) than $100,000 (also see, Sanchez, 2012; Edmiston, Brooks, & Shepelwich, 2012).
Significantly, this
accounting ignores compelling research that shows that amounts much smaller than $100,000 can create financial
hardship. For example, Akers (2014) finds that “high-debt borrowers face financial hardship at only slightly higher
rates than comparable households with less debt” (p. 4). What this suggests is that high debt does not necessarily
lead to hardship, because people with high debt often have higher earnings; equally important, it also means that low
debt does not necessarily mean absence of hardship. Additionally, this simplistic accounting glosses over the
significance of variables such as how people come to borrow and why, factors that may matter more than how much
they borrow, particularly when it comes to disadvantaged students, who borrow as their only option,
versus
wealthier students whose loans complement other financial aid and who likely have resources with which to
confront obligations post-graduation.
The fact that debt of any size may bring about financial hardship raises the question of whether making student loans
the centerpiece of the U.S. financial aid system is an inherently flawed idea. It also points to the potential futility of
policy reforms that merely seek to ‘tweak’ the student loan system in order to mitigate its worst effects. As we
understand student borrowing today, the factors that influence whether student debt will be harmful for a given
student include not only the amount of debt but other, more complex considerations, including the kind of postsecondary institution attended, the major pursued, the kind of job students secure after graduation, what the
economy will be like,
and the options they have before, during and after higher education. Beyond the fact that
answering these questions accurately is nearly impossible for a super computer, let alone a high school graduate,
even if answered correctly (i.e., college will pay off), the need for this calculus may itself weaken the ability for the
education path to act as an ‘equalizer’.
In fact, getting more information about how harmful student debt can be
could just further depress post-secondary education participation, particularly among low-income and minority
students who are more likely to be loan-averse, unless an alternative way to pay is provided.
THE STUDENT LOAN PROGRAM’S INFLUENCE ON THE DECISION TO BORROW
The economic model of human capital investment assumes that individuals decide to invest in postsecondary
education by making a subjective calculation that expected long-term benefits of attending would exceed expected
short-term costs (Becker, 1962).
Long-term expected benefits include both monetary and nonmonetary benefits
while short-term expected costs include the costs of attendance and foregone earnings. In addition to weighing costs
and benefits, economists recognize that this calculus also depends on the “demand for human capital and the supply
of resources for investing in human capital” (Perna, 2006, p. 107). In this section we will focus on how the student
9
loan program can complicate the subjective calculation of the return on receiving a post-secondary credential,
particularly in the case of lower-income children, and touch on the demand for human capital and resources for
paying for post-secondary education.
Student Loans Complicate the Calculation of the Return on Post-Secondary Education
Institutions influence human decision-making in nearly invisible ways, such that what is sometimes perceived as a
rational choice really is the product of institutional structures. In an analysis of institutions and rational choice,
North (2005) describes rational decision making this way, “much of what passes for rational choice is not so much
individual cogitation as the embeddedness of the thought process in the larger social and institutional context” (p.
24). The student loan program, as a type of institution,
provides high school graduates contemplating post-secondary
education with an embedded thought process that plays out in the subjective calculation these students make about
the return on college. While not often discussed, the role of student loans in this subjective calculation is likely as
much of a problem as the actual loan amount. It is important to emphasize that we recognize student loans are not
the only institution influencing student decision-making in this context; students may also be guided in their
decisions by parents, peers, and their secondary schools, among others. One of the reasons that student loans may
fail disadvantaged students, however, is that these students may not have the support of countervailing institutions to
mitigate loans’ negative effects.
Again, this subjective calculation is likely particularly salient for lower-income students.
Potential students from
upper-income families are much more likely to walk into a pattern of higher education, such that they may never
consciously and intentionally ‘decide’ to attend college. For instance, as children from upper-income families are
more likely to have parents who have attended a post-secondary institution (e.g., Terenzini, Cabrera, & Bernal,
2001), they have access to and support from institutions other than the student loan system as they face the prospect
of higher education. Significantly, educated parents are more likely to counsel their children on availability of
financial aid (Goldrick-Rab, 2006; Terenzini, et al., 2001) and provide more accurate information about the cost of
college (Horn, Chen, & Chapman, 2003), which mediates the role of student loans in these students’ accounting.
Moreover, for many upper-income students, alternative forms for paying for college exist (Sallie Mae, 2014), such
that student loans feature less prominently in their financial and mental preparation for higher education.
The Economic Context in which Decisions are Made
It is important to also point out that these calculations of subjective costs occur within an economic context that can
obscure the benefits of college. As the subjective calculation includes potential students’ image of the economy and
their role in it, today’s relatively adverse conditions, including for college graduates, may alter post-secondary
education decision-making in ways that have lasting repercussions. The Great Recession contributed to fewer jobs
for college graduates,
rising college costs as states reduced support to universities, stagnant incomes, and a shift
away from societal responsibility toward individual responsibility for financing post-secondary education
(Hiltonsmith, 2014). While college graduates inarguably continue to outperform their less-educated peers on many
economic and social measures, between 2000 and 2010 unemployment among college degree holders rose from 2.0
percent to 5.7 percent (Mishel, Bivens, Gould, & Shierholz, 2013). Between 2002 and 2012 wages were stagnant or
declining for the bottom 70 percent of U.S. families (Mishel & Shierholz, 2013).
Even more disturbing is the unequal distribution of risk exposure in this economy. Because different groups of
people face different economic conditions, the subjective calculations potential students must make are not the same.
For instance, the college-educated unemployment rate is higher among racial minorities.
Between 2000 and 2010 the
unemployment rate for White workers with a college degree increased from 1.8 percent to 4.9 percent, but for Black
college graduates it increased from 2.8 percent to 9.8 percent (Mishel, Bivens, Gould, & Shierholz, 2013). Recent
U.S. economic history complicates these decisions, too, particularly as prospective college students today cannot
rely on either earnings achieved by recent college graduates or their parents’ economic experiences as accurate
predictors of what they will be able to earn if they attain a college degree, or not. And even though economic times
change, perceptions of the current economic climate, good or bad, weigh into the decision.
10
The economic model’s focus on short-term costs—
particularly the time during higher education—has resulted in
neglect of the long-term costs of student debt in policy and research. Because the full effects of student loans,
including the constraints they place on asset accumulation and, potentially, earnings, post-college, do not show up as
short-term financial costs as currently defined (largely confined to the cost of attending and forgone earnings), this is
an unrealistically narrow conception of the ‘costs’ of college. This is important because it narrows the lens through
which many observers view the student loan problem. For example, Rothstein and Rouse (2011) state: “student debt
has only an income effect—proportional to the ratio of debt to the present discounted value of total lifetime
earnings—on career and other post-college decisions”
(p. 149). They go on to calculate that $10,000 in student debt
represents less than one percent of the present value of the average college graduate’s potential lifetime earnings.
The implication is that college pays off eventually, even if, in the short term, loans are expensive.
With this as the standard for success, with some exceptions, pundits and analysts have largely concluded that student
loans should not play a negative role in students’ decisions to attend college (i.e., it is rational for a potential student
who is disadvantaged to conclude that college is ‘worth it’). Our point is not to debate the calculation that adults
who have a college degree have higher lifetime earnings than adults without a college degree (e.g., Rose, 2014).
There is considerable evidence that college does pay off.
However, we aim to raise questions about the extent to
which anyone really can accurately predict costs and benefits; the extent to which the focus should be on costs
associated with the period of attendance and not outcomes before, during and after post-secondary attendance; and
the extent to which earnings alone are the right standard, rather than considerations of equity and economic mobility.
Does Anyone Really Think that Way? Information is Important, but Not Sufficient
It must be emphasized that the subjective calculation of the return on college is not a scientific analysis. The data
potential students collect to make this calculation are largely anecdotal, tested through life experience rather than a
statistical package. The decision to borrow is primarily based on prospective students’ ability to accurately predict
what their particular future earnings will be. Since little of the information needed for making the calculation is
known (e.g., whether he/she will get a job after graduation, whether that job will be in his/her field, if the field will
even exist 10 years from now, how much the borrower will earn over the course of his/her life, and so forth) at the
time the decision is to be made, it is a qualitatively different type of calculation than what the economist makes, for
example, when calculating lifetime earnings. Even in the best of economic times, there is no guarantee that past
returns will come to pass. This is further complicated by the fact that graduating with a four-year degree is almost a
50/50 proposition in America, so initiating degree pursuit does not at all guarantee that one will complete college.
For instance, Avery and Turner (2012) find that about 55 percent of dependent students who expect to complete a
bachelor’s degree and about 50 percent who expect to complete an associate’s degree do so within six years after
graduating high school. So, even looking at the very short term, prospective students cannot even be sure that they
will graduate.
In sharp contrast to this uncertain context facing borrowers, lenders have policy protections that allow debt
collectors to garnish wages, including huge penalties that balloon what is owed, which some worry may even
encourage practices that contribute to borrower default. They also benefit from polices that make discharging
student loans within bankruptcy nearly impossible and can obtain government assistance in collecting unpaid debts.
With these and many other protections, the risk to lenders is far less than for borrowers. Lenders also benefit from
having a large asset store not available to most borrowers, which further reduces their risk. And they benefit from
the changing landscape of financial aid, which essentially drives ‘captive’ customers to their doors, as non-loan aid
loses more of its capacity to meet rising college costs. Critically, this business model diverges dramatically from
how student loans are framed, as an almost altruistic intervention that puts college within reach for those who
otherwise would never be able to afford it.
Costs before Post-Secondary Attendance
The true costs of higher education include not just foregone earnings during post-secondary attendance and the
‘hard’ costs of tuition and fees, but also those prior to and well beyond college. For example, when two college-age
children, one who will have to rely on loans and one who will not, contemplate their opportunities, including what
college they will attend and what major they will pursue, they may view these decisions very differently. Research
11
on student debt aversion shines light on how student loans restrict opportunities for those who rely on them
(Campaigne & Hossler, 1998; Fenske, Porter, & DuBrock, 2000; Paulsen & St. John, 2002; Perna, 2000). For
example, Perna (2000) finds that student loans reduce the chance that Black students enroll in four-year colleges.
She attributes this, in part, to an aversion to borrowing. In essence, student loans may provide these students with an
embedded thought process that four-year college is out of their reach, not because they do not have the desire, effort,
or ability to succeed in more selective institutions but because they fear taking on debt in order to pursue this
education.
Beyond loan aversion, other embedded institutional thought processes may be at play, but these influences may also
be shaped by the reality of the debt-dependent financial aid system. For instance, field evidence suggests that
students facing the prospect of considerable debt may be steered toward two-year colleges as opposed to a four-year
college by high school personnel such as teachers and counselors, even when they are academically qualified to
attend more selective institutions (Elliott, 2013).
This field evidence is supported by survey data that indicates that
37 percent of school guidance counselors believe that low-income students should avoid student loans because of
risk of default (National Association for College Admission Counseling and the Project on Student Debt, 2007).
Similarly, McDonough and Calderone (2004) also find that counselors may steer low-income students toward
community colleges because they believe that they are more affordable. Critically, these same restrictions on the
attendance decision are not placed on students who do not require student loans, thus exacerbating the inequity of
institutional selection. And, again, since different types of institutions do not deliver the same educational outcomes,
these choices may have serious—even lifelong—effects. This is not how most people think that U.S. higher
education works for bright but poor students.
Costs during Post-Secondary Attendance
Student debt also appears to play a negative role in some children’s decisions to stay and complete college if they
go. For instance,
Baum and O’Malley (2003) find that 40 percent of students with student loan debt reported that
they did not return to school or transferred to a lower-cost school due to student loan debt. Significantly, loans’
depressing effects on educational attainment may be particularly salient for disadvantaged students. Kim (2007)
finds that higher student loan debt in the first year of college is associated with lower probabilities of graduating
from college among low-income and Black students but not high-income and White students. Kim (2007) suggests
that higher loan aversion by low-income and Black students might come from having less certain job prospects, less
familiarity with financial institutions, and a higher risk of not graduating from college than higher income and White
students. Additionally, there is evidence that students’ decisions do not unfold as economic models would suggest,
reinforcing the importance of making policy based on real conditions, not idealized abstractions. With respect to
student debt and college completion, Cofer and Somers (2000) state,
…rather than being incremental, the effect of debt is felt in lump sums. That is, a student
borrows in a lump sum…
at the beginning of the semester. When the next semester begins, the
student has to again make a decision to persist based, in part, on this new, higher level of debt.
Students view threshold levels as intimidating, especially when they move from one perceived
level to another. (p. 35)
Further, Millett (2003) finds that students with college debt are about 60 to 70 percent less likely to apply to
graduate school than students without college debt, potentially evidence of other ways in which student borrowing
distorts the educational paths of those who depend on loans.
Merely ‘counting’ total student borrowing cannot answer our most urgent questions about how to build a financial
aid system equipped to meet the challenges of the twenty-first century.
In the end, determining how many students
are borrowing exactly how much money is not the right evidence with which to measure what we most need to
know: whether the student loan program is delivering the best return on investment or whether other approaches
might be better. The question, then, is not whether or not student loans are creating the next financial crisis or even
whether students are crippled with unprecedented repayment burdens. Instead, the United States must determine
whether the student loan program is another source of inequality in the education system, compromising economic
mobility following college and, then, effectively closing off one of the only few remaining viable vehicles to upward
12
progress in the U.S. economy. Indeed, particularly in the context of limited government resources, a financial aid
program worth billions of federal dollars of investment should do more than only temporarily harm some students. It
should create greater economic equality, through the medium of the education system.
If there are good reasons to believe that student loans may not be fulfilling this function, it seems reasonable to ask
if there is a better way to provide access to college that more closely aligns with American values. However, the
current political environment of inaction with regard to funding education, the constant fear of deeper cuts, and an
emotional attachment to the student loan program as an access vehicle—including considerable defensiveness in
those who do now view their own use of student loans as having been detrimental—has created resistance to
acknowledging that they do not work for many students today. Instead of imagining alternatives more capable of
creating equity, the policy debate centers on the false counterfactual of whether students are better off taking on
loans to go to college than not attending college at all. To focus our collective attention where it belongs, we must
extend the time horizon on which we look for indicators of the success or failure of the student loan system. In the
next section we discuss evidence of the potential negative long-term costs of student loans.