The Debt Divide: The Racial and Class Bias Behind the "New Normal" of Student Borrowing

The Debt Divide: The Racial and Class Bias Behind the "New Normal" of Student Borrowing

May 19, 2015

Today, taking out loans is the primary way individuals pay for college—a major shift in how our nation provides access to higher education. While concerns about the growth in college costs and student debt are nearly universal, much of this concern focuses on how college debt is impacting the economic well-being of college graduates and our overall economy. What has been less understood, or examined, is how this shift to a debt-based system impacts our nation’s historical commitment to ensuring everyone—regardless of race or class—can afford to go to college. We need to understand whether or not the “new normal” of debt-financed college is having an impact on our ability to make good on that fundamental promise.

This report, The Debt Divide, provides a comprehensive look at how the “new normal” of debt-financed college impacts the whole pipeline of decision-making related to college. This includes, whether to attend college at all, what type college to attend and whether to complete a degree, all the way to a host of choices about what to do for a living, and whether to save for retirement or buy a home. In an America where Black and Latino households have just a fraction of the wealth of white households, where communities of color have for decades been shut out of traditional ladders of economic opportunity, a system based entirely on acquiring debt to get ahead may have very different impacts on some communities over others.

Our analysis, using data from three U.S. Department of Education surveys, the Federal Reserve’s 2013 Survey of Consumer Finances, and existing academic literature, reveals a system that is deeply biased along class and racial lines. Our debt-financed system not only results in higher loan balances for low-income, Black and Latino students, but also results in high numbers of low-income students and students of color dropping out without receiving a credential. In addition, our debt-based system may be fundamentally impacting the post-college lives of those who are forced to take on debt to attend and complete college. Our findings include:

  • Black and low-income students borrow more, and more often, to receive a bachelor’s degree, even at public institutions. A full 84 percent of graduates who received Pell Grants graduate with debt, compared to less than half (46%) of non-Pell recipients. While less than two-thirds (63%) of white graduates from public schools borrow, four-in-five (81%) of Black graduates do so. Latino graduates borrow at similar rates and slightly lower amounts than white students.
  • Associate’s degree borrowing has spiked particularly among Black students over the past decade. At public institutions, well over half (57%) of Black associate’s degree recipients borrow (compared to 43% of white students), and borrow nearly $2,000 more than white students.
    A decade ago, 38% of Black associate’s degree recipients borrowed (compared to 32% of white students). In other words, a six-point gap in borrowing between white and Black associate’s degree holders has turned into a 14-point gap.
  • Students at for-profit institutions face the highest debt burdens. Associate’s degree recipients at for-profit schools borrow almost the same amount (only $956 less) than bachelor’s degree recipients at public colleges.
  • Black and Latino students are dropping out with debt at higher rates than white students. At all schools, nearly 4-in-10 (39%) of Black borrowers drop out of college, compared to 29% of white borrowers. Around the same number (38%) of low-income borrowers1 drop out compared to less than a quarter of their higher-income peers. Nearly two-thirds of Black and Latino student borrowers at for-profit four-year schools drop out (65% and 67% respectively). Nearly half (47%) of Black student borrowers drop out with debt at for-profit 2, and less-than-2-,year institutions.
  • Graduates with student loan debt report lower levels of job satisfaction when initially entering the workforce. High debt borrows report levels of satisfaction around 11 percentage points lower than those who graduated from college debt-free.
  • Average debt levels are beyond borrowing thresholds that are deemed by research to be “positive.” Studies suggest that small amounts of debt—$10,000 or below—have a positive impact on college persistence and graduation, but amounts above that may have a negative impact. Unfortunately, average debt levels for both associate’s and bachelor’s recipients are now well beyond the “beneficial” threshold.
  • While those with a college degree are more likely to save or buy a home, student debt could be acting as a barrier. At every level of education, households without student debt are more likely to own homes, have slightly lower interest rates on mortgages, and have retirement and liquid assets that are considerably larger than those households with student debt.


In a gymnasium at Southwest Texas State Teachers College in 1965, President Lyndon Johnson remarked upon signing the Higher Education Act that “a high school senior anywhere in this great land of ours can apply to any college or any university in any of the 50 States and not be turned away because his family is poor.”2 The HEA, as it is known, created a system of grants for needy students, work opportunities for students, and interest-free loans as a backstop for students with unmet financial need. Rather than being seen as a partisan accomplishment of the Great Society, it was largely defended as a seminal piece of the American social contract. Rather than dismantling Johnson’s proud achievement, five years later, in 1970, Johnson’s successor Richard Nixon argued in a special address to Congress that “No qualified student who wants to go to college should be barred by lack of money. That has long been a great American goal; I propose that we achieve it now.”3

And so it went for a generation for aspiring college students, who could generally finance college from a combination of scholarships, part-time employment during the school year or summer, or family income. Student loans, while always nominally available, were reserved for middle-class families who used them as a cash-flow mechanism.

As more students entered college, however, our public officials began to renege on their promise to invest in the higher education system. States started cutting per-student funding at public institutions, and modest increases in grant aid were dwarfed by rising tuition. Meanwhile, working-class and middle-class incomes began to stagnate, leaving students with little recourse but to take on debt to reach their college dreams. With each successive reauthorization or rewrite of HEA, policymakers have done less to fulfill the public dreams of those who wrote it.

We have now entered a new phase where student borrowing is now the primary way young people pay for college. The heavy reliance on student loans has made the college-going process fundamentally different for some groups, notably Black and Latino students and students of modest means. And despite a growing body of research showing that need-based grant aid is the most effective mechanism to induce enrollment and completion, our public policy has led students to rely far more on loans—the effectiveness of which is mixed at best and actually harmful at worst.

This shift places an unequal burden on communities that have historically been denied an opportunity to gain and leverage wealth. While higher-income, predominately white, households can hope to minimize borrowing by using tax-advantaged savings and investment accounts, home equity, and other mechanisms, low-income households by and large cannot use these tools. For our entire history, public policies—from redlining, to inequitable state and local tax formulas that fund K-12 education, to the decline of defined-benefit pensions—have denied communities of color the same opportunities to build wealth and gain the same foothold in the middle class that whites have enjoyed. And despite the death of de jure Jim Crow-era segregation, gaps in wealth between white and Black, and white and Latino, households have actually increased. Two decades ago, white households had median net worth seven times higher than Black households, and six times higher than Latino households. In the aftermath of the recession, whites held 13 times more wealth than Black households and ten times more wealth than Latino households.4 These households are far less likely to have accumulated the wealth necessary to save for college and avoid borrowing to pay for rising costs of attendance.

The result is a burden of debt that is fundamentally unequal; low-income, Black and Latino students almost universally must borrow to attain a degree, while white, middle- and upper-class students are far less likely to need to borrow. This can distort choices about whether and where to go to school, and contributes to persistent gaps in attainment.

Reliance on loan debt also makes the consequences of dropping out of college far direr. A generation ago, the only consequence a college dropout faced was the loss of future earnings that could have come with the degree. Now, dropouts face loss of earnings as well as a debt burden that must be paid off in short order. The link between student loan defaults and dropping out is strong. In fact, a recent analysis by the New America Foundation shows that nearly two-thirds of those who default on student loans have no degree.5

Finally, student loan debt does not stop at the water’s edge—there is plenty of evidence that it can reduce lifetime wealth, affect important life decisions, and resonate long after a borrower is out of school. Analyses over the past few years from Demos6 and the New York Federal Reserve Bank7 have raised fresh concerns about the broad economic impacts of our debt-for-diploma system.

This report, The Debt Divide, outlines what we know about undergraduate student debt, using data from three U.S. Department of Education surveys as well as the Federal Reserve’s 2013 Survey of Consumer Finances, in addition to existing research on the topic. Where possible, we try to shine a light on students at public colleges and universities; after all, these institutions educate the vast majority of U.S. college students, and have a mission to remain affordable and maintain a student body that is representative of their state. What we find, unfortunately, is a system that not only overburdens low-income, Black and Latino students, but also may be fundamentally impacting the post-college lives of all students who are forced to take on debt to attend and complete college.

The Inequality of Student Debt, by Race and Class8

It is no secret that college costs have far outpaced inflation and growth in family income in recent decades, particularly (though not exclusively) at public institutions. Need-based grant aid, which is designed to defray costs for low-income students, has also dwindled as a percentage of college costs. It is disheartening but not surprising, then, that students who already have trouble financing school—namely, Black and Latino low-income students—have seen borrowing levels and amounts spike.

Indeed, low-income graduates (those who received a Pell Grant while in school) borrow at far higher rates—and in higher amounts—than their middle- and upper-income counterparts at both two- and four-year institutions, regardless of the type of institution attended, and despite receiving thousands of dollars in grant aid. Black students also borrow at much higher rates, and in higher amounts, to receive the same degrees as their white counterparts. Latino students borrow at higher percentages and in higher amounts than white students at private non-profit and for-profit institutions, but graduate with less debt on average than white and Black students at public institutions.

Borrowing for a Bachelor’s

Perhaps surprisingly, the gap in borrowing between Pell and non-Pell recipients, and white and Black students, is most pronounced at public institutions. A full 84 percent of graduates who received Pell Grants graduate with debt, compared to less than half (46%) of non-Pell recipients. Overall, borrowing rates are higher among bachelor’s recipients at private non-profit schools for every group, even though the gap may be smaller than one would think (see Figure 1).

In addition, Black bachelor’s degree recipients are more likely to borrow than white students at any type of institution (including for-profit schools, discussed below). While less than two-thirds (63%) of white graduates from public schools borrow, four-fifths (81%) of Black graduates borrow. While private non-profit schools command more frequent borrowing among Black students, the gap in the percentage of Black and white students who borrow is higher at public institutions.

Latino students, on the other hand, borrow at the exact same rate as white students (63%), and actually borrow an average of $2,400 less than whites to receive degrees from public colleges and universities (see Figure 2). This could be attributable to many factors, including whites attending slightly more expensive public institutions, or cultural attitudes towards debt and risk. However, borrowing rates are far higher for Latino students at private non-profit schools, where 87% borrow. Average debt at private non-profits is actually higher for Latino students than for Black and white students.

Debt Is Rising for Two-Year Degrees

Many students consider an associate’s degree as a low-cost, low-debt college option, either as a springboard for a bachelor’s
degree program or return to the workforce. Indeed, borrowing levels of all students at public 2-year schools are low (around 17%). But for those who are pursuing an associate’s degree, borrowing rates are far higher. In fact, 4-in-10 associate’s degree recipients at public institutions9 now must borrow in order to earn the credential (see Figure 3).10 Debt levels, while lower than those at four-year schools, average $13,970 at public institutions (see Figure 4).11

These numbers have jumped over the past decade. The mid-2000s saw substantial increases in the percentage of students who borrowed for associate’s degrees, which has held through today. In the midst of the recession, between 2008 and 2012, the percentage of borrowers increased slightly, but the average amount borrowed for an associate’s degree ballooned. Adjusted for inflation, today’s associate’s degree holders from public schools graduate with $3,000 more in debt than they did in 2004, and over $2,500 more than they did in 2008 (see Figure 5).

But, as with bachelor’s recipients, these figures mask substantial differences by race and income.

In fact, 57 percent of Black associate’s degree recipients borrow (compared to 43% of white students), and borrow nearly $2,000 more than white students. Black students also saw the largest spike in borrowing between the 2003-04 and the 2011-12 school years. A decade ago, 38 percent of Black associate’s degree recipients at public schools borrowed (compared to 32% of white students). In other words, a six-point gap in borrowing between white and Black associate’s degree holders has turned into a 14-point gap. On the other hand, only a third (35%) of Latino associate’s degree holders borrow to earn an associate’s, though that number is up from less than a quarter (23%) in 2003-04 (see Figure 6).

Additionally, despite the fact that the maximum Pell Grant often covers tuition and fees for associate’s degree programs at public schools, well over half (55%) of associate’s degree recipients who received Pell Grants graduated with debt. Pell recipients took on an average of over $14,500, nearly $2,000 more than those who never received the grant.

Perhaps more concerning, it seems that the fundamental transfer mission of community colleges is being undercut. A 2012 study from TG indicates that bachelor’s recipients who transferred from community colleges actually borrowed the same amount or more than students who started at public and private 4-year schools.12 In other words, contrary to intuition, transferring from a community college did not lower the cost of a degree.

Near-Universal Borrowing at For-Profit Schools

While three-in-four students attend public colleges and universities, for-profit institutions educate less than ten percent of all undergraduates.13 And yet, for-profit schools command media and policy attention precisely because of the outsized impact they have on overall student borrowing. For-profit institutions also enroll disproportionate numbers of Black and Latino students. In fact, Black and Latino students make up fewer than one-third (29%) of all college students, but nearly half (45%) of all private for-profit students.14

While for-profit schools graduate the lowest percentage of their students than any sector, those who do graduate almost certainly take on debt. Eighty-six percent of white students, 89% of Latino students, and 90% of Black students borrow to receive a bachelor’s degree at for-profit institutions, with debt averaging around $40,000 for each group. Ninety-six percent of Pell Grant recipients who graduate from for-profits incur debt (see Table 1).

Borrowing numbers are nearly identical at the associate’s degree level. As with bachelor’s degree programs, nearly all (94%) of associate’s degree holders at for-profit schools who received Pell Grants graduate with debt, averaging over $25,000. Nearly all students of color borrow as well, including 93% of Black students and 92% of Latino students (compared to 85% of white students). Although Black students at for-profit schools borrow around the same amount as white students, Latino degree holders actually borrow over $3,500 less than white students at for-profit schools.

To put for-profit borrowing in perspective—associate’s degree recipients at for-profit schools only borrow $956 less than bachelor’s degree recipients at public schools. The high debt that degree recipients must endure at these schools is one reason that for-profit institutions have come under extra scrutiny from both the federal government and state attorneys general. Another reason for scrutiny is the share of students at these schools that do not make it to the finish line, as mentioned below.

High Debt, No Diploma15

In some ways, the student borrowers described above may be in the best shape of all. After all, despite rising debt burdens, borrowers with degrees at least have a credential that remains valuable in the labor market. Unemployment rates remain lower and earnings remain higher for college graduates relative to their less-educated peers, even if the rise in overall debt threatens to consume more and more of their income and savings over time.

For dropouts, however, the story is different. In fact, dropping out of college is consistently the biggest predictor of whether or not someone will default on a student loan, and financial obligations (either the cost or the need to work to financially support oneself while in school) is the largest reason cited for dropping out.16, 17 And Black and Latino students are substantially more likely to cite financial reasons for dropping out. Around 7-in-10 Blacks dropouts cite student debt as a primary reason for not completing school, compared to fewer than half of white students.18 Essentially, as borrowing has increased in tandem with the importance of a degree, the consequences of dropping out have never been higher, and the burden of student debt may be making Black and Latino students less likely to complete their degree.

In a way, student debt would be a less worrisome issue if all students who entered college were essentially guaranteed to receive that credential, and that their degree always provided a labor market boost. Unfortunately, neither of those are the case. In fact, only 56 percent of degree-seeking students complete college within six years.19 Numbers are far worse for students who dip below full-time enrollment; less than half (43.2%) of students who enroll part-time at any point end up graduating within six years.20

In fact, evidence is mixed on whether student loans provide any positive impact on the ability to complete a degree. The research on the topic is complicated, since some consider student loans as financial aid while others do not. It’s also difficult to separate the reasons for a student dropping out. After all, while many students cite financial difficulties as a reason for leaving school, it’s unclear how much that interplays with academic preparation or other life obligations. Also, student loans could negatively impact graduation even when students do not rely on them. Among students with substantial unmet financial need, those that choose not to take out student loans are far more likely to simply enroll part-time.21 In other words, students are stuck with a Catch-22: take on loans, or engage in behavior—part-time enrollment or full-time work—that decreases the likelihood that they will complete a degree.

The picture is also complicated by the fact that extremely modest amounts of loans could be useful in helping students make ends meet. Two different studies suggest that small amounts of debt—$10,000 or below—have a positive impact on college persistence and graduation, but that amounts above that may actually have a negative impact on the ability to graduate.22, 23 This makes sense intuitively; loans may be useful to fill small gaps in need, but could become a burden when used as the primary financing tool. This is troubling, needless to say, when average debt levels for both associate’s and bachelor’s recipients are now well beyond the $10,000 threshold suggested by the research. Other studies also find that loans may have a negative impact only on students of color or students with few family resources to buffer against the risk of borrowing.24

It is telling, however, that the impact of grant aid on college persistence and completion is quite clear, while the impact of loans is far less so. Several studies suggest that grant aid positively impacts persistence25 and completion26 particularly for low-income students—the students who are forced to borrow far more today and graduate at much lower rates.

Indebted Dropouts Are More Likely to Be Low-Income, Black and Latino Students

The impact of student loan debt is more concerning when we examine the number of people who take on debt but do not graduate. Unfortunately, the ranks of indebted dropouts have grown in recent years. A recent Education Sector study indicates that nearly a third of borrowers are dropping out, up from about one-in-five in 2001. Student borrowers at for-profit 4-year schools are also far more likely to drop out than students at public and private non-profit 4-year schools.27

But understanding, and potentially remedying, this problem requires an understanding of exactly who is dropping out with debt. As with overall borrowing, nearly 4-in-10 (39%) of Black borrowers drop out, compared to 29% of white borrowers. A similar percentage (38%) of low-income borrowers28 drop out (see Figure 7). But these numbers are just the tip of the iceberg. In fact, nearly two-thirds of Black and Latino student borrowers at for-profit four-year schools drop out (65% and 67%, respectively) (see Figure 8). Over half of low-income borrowers drop out at these institutions as well. Nearly half (47%) of Black student borrowers drop out with debt at for-profit 2-, and less-than-2-, year institutions. Rates are worrisome at public institutions, if less so. Nearly a third of low-income student borrowers at public 4-year schools drop out, a rate 10% higher than student borrowers at those schools on the whole.

The link between dropping out and struggling to repay loans is strong, and helps explain why the average balance of a defaulted student loan is relatively low (around $15,00029). Students who borrow but drop out, by definition, do not have additional years to accumulate debt, but fall into trouble making monthly payments without the benefits of a degree. This explains how a law school student with six-figure debt can be in better financial shape than a dropout from an associate’s degree or certificate program, and speaks to the need for targeted policy solutions aimed at those most likely to struggle to repay.

Student Loans Cast a Post-College Shadow

Obviously, student loans stick with borrowers well beyond the time they leave school. In fact, one-third of all student debt is owed by borrowers over 40 years old. The average student loan balance for an indebted 60 year old is right around $20,000, likely due to accumulated interest (or borrowing for graduate school).30 The specter of debt, naturally, can last well into the age when workers could be saving for retirement or even a child’s education.

In 2013, Demos released At What Cost? How Student Debt Reduces Lifetime Wealth, which showed that relative to a college-educated household without debt, an indebted household stands to lose $208,000 over a lifetime, primarily from lost retirement savings.31 This figure stands to rise as debt levels, and thus the time it takes to offload student debt, extends into a borrower’s prime earning years. Even a 2014 Brookings Institution report that received wide attention for arguing that student debt is manageable for the average borrower noted that borrowers are now taking twice as long (13.4 years) to pay off their loans as they were nearly 20 years ago (7.5 years).32

Beyond potential lost savings, a recent poll from Gallup and the University of Purdue notes that indebted graduates—particularly those with high debt levels—report lower levels of financial worth as well as physical well-being.33

Student debt may also be impacting the decisions students make about future employment. Graduates with student loan debt also show less initial job satisfaction than those who did not borrow for undergraduate education (see Figure 9).

A 2008 study also found causal evidence—from a natural experiment at a highly-selective institution—that student debt causes graduates to choose highly-paid occupations and shy away from public-interest professions.34 And a recent study from researchers from the Federal Reserve Bank of Philadelphia and Penn State also recently noted that student debt has a significant negative impact on small business formation.35 Again, this makes sense; small businesses are more likely to be financed at least partially from personal debt.

A debate has also sprung up around the impact of student debt on this generation’s ability to purchase a home. According to the Federal Reserve, student borrowers continue to stay away from home purchases relative to their non-indebted peers. Whereas having student loan debt once made someone more likely to purchase a home, the opposite is now true: 27- to 30-year-olds with student debt have lower rates of homeownership.36 The same is broadly true of car ownership as well.

This may have something to do with the impact of student loans on credit scores. A 2014 Brookings paper notes that credit scores
for young households without student debt are higher than indebted households—a relatively new phenomenon over the past decade.37 And a 2012 study from Young Invincibles estimated that the typical single student borrower now has a debt-to-income ratio that would prohibit him or her from qualifying for a garden-variety home mortgage.38

Why Has This Happened? The Driving Factors Behind Rising Undergraduate Loan Debt

The overall dollar amount of student loans in the economy can also be attributed to increasing numbers of students attending college. This is most likely a positive phenomenon; enrollment in degree-granting institutions has grown from 25% of all 18- to 24-year-olds in 1979 to 41% today.39 Indeed, enrollment is up for all income groups—even half of all low-income high school graduates enroll in college the following fall, up from one-third in 1980. Despite a projected decline in the number of 18- to 24-year-olds, the U.S. Department of Education still projects college enrollment to grow by nearly 14% between now and 2022.40 Still, enrollment gaps persist, and the gap in college attendance between wealthy and low-income students has stayed basically the same over the past 30 years.41

But, as Demos has documented previously, in 2012’s The Great Cost Shift and 2014’s The Great Cost Shift Continues, a primary driver of student debt continues to be reduced state expenditures on higher education. In the past decade alone, state higher education funding per student dropped by 22%, and 2012 saw the lowest per-student expenditure on higher education in three decades.42 Even as the economy has rebounded from a bitter recession, state spending for higher education ticked upward by a negligible 1.4% and even then, 20 states still cut per-student funding.43 Gaps in funding have been made up primarily via tuition, shifting the cost away from the state and onto the student. Unsurprisingly, tuition makes up a far higher percentage of the cost of educating students. In 2000, tuition dollars covered 29%, with public support making up the rest. By 2013, tuition covered nearly half (see Figure 10).

As tuition has risen, grant aid has also failed to keep pace. The Pell Grant, the federal government’s cornerstone need-based aid program, covered over three-fourths of the total cost of attendance at public colleges and universities in the late 1970s and nearly 40% of the costs of attending a private non-profit. By 2014, it covered less than one-third, and less than 15% at private non-profit schools (see Figure 11). State grant aid programs have also failed to fill the gap while also moving toward rewarding a higher percentage of grants based on merit, rather than need. Meanwhile, many institutions of higher education are using grant aid on higher-income students, while low-income students face net prices that approach their entire family income.44

Meanwhile, family incomes for everyone but the wealthiest have remained relatively stagnant for the better part of three decades (see Figure 12).45

The crippling combination of stagnant incomes, state disinvestment, and insufficient and inefficient grant aid has led us to the point where student borrowing has become the norm even at public institutions, and the rise in average debt levels shows no signs of abating. Just two decades ago, fewer than half of bachelor’s recipients needed to borrow to finance a degree (see Figure 13).

Proponents of our current debt-based system often point out that borrowing provides students with funding for college when they are least likely to afford the cost of college, thereby providing access. And of course, very few borrowers could have paid the sticker price of college without loans.

But this presents a false choice; after all, loans are not an inevitable way to fund college. The alternative to loans could simply be increases in state appropriations that lower student costs, or increases in grant aid targeted at students who need it the most. Indeed, there is strong evidence that need-based grant aid contributes positively to college access,46, 47, 48 particularly for non-traditional students.49 On the other hand, evidence is mixed on whether or not student loans increase levels of college participation. To be sure, isolating the impact of student loans on the ability to attend college is difficult—it becomes quickly tangled in other questions, like family income, overall cost, the timing of when a student receives financial aid, not to mention academic or other non-financial factors. But while some find evidence that eligibility for loans drives up college attendance,50 others find that the prospect of borrowing51 or the prospect of excessive loan burdens can discourage college attendance.52 Cultural factors may come into play, as Latino students may be more averse to borrowing than other students.53 Rather than taking on loans, students may enroll in lower-cost institutions, which is only acceptable if those institutions have the resources to provide sufficient quality and support to help a student graduate.

The Lifelong Advantage of Attending College Debt-Free

As mentioned, Demos’ 2013 report At What Cost utilized the 2010 Survey of Consumer Finances (SCF) to determine the loss of lifetime wealth attributable to student loan debt. Others, including Richard Fry at the Pew Research Center, have also used the 2010 SCF to examine the economic well-being of households with and without student debt.54 Pew’s research found that college-educated households without student debt had a net worth seven times greater than those with student debt, and non-college educated households without debt had net worth nine times greater than those with student debt. In fact, net worth for non-college educated households without student debt was actually higher than college-educated households with student debt.

Thanks to new Federal Reserve data from the 2013 Survey of Consumer Finances, we can now take a post-recession snapshot of the debt and assets picture for households55 with and without student debt. Given the aforementioned impact of college completion on the ability to repay loans, we also compare those households with “some college” to those with college degrees (including dual-headed households). The full results56 are shown in Table 2 below.

We find, unsurprisingly, that at every level of education, non-indebted households are more likely to own homes, have slightly lower interest rates on mortgages, and have retirement and liquid assets that are considerably larger than those households weighed down by debt. The differences in retirement assets in particular are stark: Households with some college and no education debt have an average of over $10,000 more in retirement savings than indebted households; households with a college degree have over $20,000 more in retirement savings; and dual-headed households with college degrees have nearly $30,000 more in retirement savings.

Naturally, we also see the value of a college degree, as both homeownership rates and overall savings (both retirement and liquid) rise by education level, and spike in households in which both heads are college-educated. But it seems clear from the data that the burden of paying off student debt is taking away a sizeable portion of the ability to accumulate meaningful assets as workers enter their prime earning years. In other words, while a college degree provides many financial advantages, there is evidence that the debt needed to gain it is leaving some households behind.

Conclusion and Policy Recommendations

The debate around student debt often assumes that we have reached a “new normal” in requiring students to borrow substantial amounts of money for a degree. In fact, the broad assumption seems to be that student debt is a positive form of debt, one that allows students access to a system that will increase their earning power, thereby recouping the debt they initially face.

But these assumptions are difficult to reconcile with the impact that this system has wrought. Despite research strongly linking need-based grant aid to access, we have instead allowed a system to flourish in which need-based aid covers less and less of the cost of college. Despite ambiguity in whether or not loans provide more benefit than harm to college access and completion, we have forced more students to borrow. Despite the fact that we have not moved the needle on degree-completion rates in a generation, we have accepted a system in which a substantial portion of borrowers drop out. And despite bipartisan rhetoric around closing attainment gaps among students of color and low-income students, we have created a system in which more underrepresented students take on debt and drop out with debt, thereby saddling communities of color and those with modest means with substantial disadvantages as they enter the workforce.

In addition to the inequitable distribution of debt, we also see worrying signs around the impact of student debt on the ability to build wealth and assets, find a satisfying or civic-minded job, or start a business. It’s difficult to know how large the impact of this is on the broader economy, precisely because we have no historical comparison to this moment.

But that does not mean that this is irreversible. Demos has published several ideas on how to re-invigorate state investment in higher education, as well as how to simplify our system of federal financial aid that provides more benefits to students who need it.

In 2014’s The Affordable College Compact, we lay out a plan for a federal-state partnership that would allow the federal government to use its leverage to encourage states to increase state spending, and develop policies and plans to ensure the majority of poor-, working- and middle class-students can attend college without incurring debt or financial hardship. In our plan, states would be required to affirm that higher education is a public good—in other words, that tuition revenue does not exceed revenue from state appropriations (see Figure 14). This is historically consistent with public higher education in the U.S., and will prevent state institutions from excessively increasing tuition in tandem with federal help. States would also be eligible for two match tiers, depending on their level of commitment to providing debt-free college for low-income students in the state.

In 2012, Demos also developed the Contract for College, which would align federal student aid programs into one cohesive, guaranteed package for students (see Table 3). It would also simplify federal financial aid by providing low-income students with grants and work-study to cover the vast majority of college costs, and middle-income families with a guaranteed aid package of grants, work-study, and subsidized loans. Reforming financial aid could work in tandem with increased state investment—in fact, states that commit to debt-free college would have an easy guideline by which they could distribute their own support as well as federal subsidies.

These policies are developed on a principle of shared responsibility—by states, the federal government, and students—and are based in the historical promises by states and the federal government to provide an affordable, valuable degree to students regardless of race or class. As we have seen, from high borrowing to substantial numbers of indebted dropouts, we have yet to live up to that commitment.


  1. Defined as income at 200% of poverty or below.
  2. President Lyndon B. Johnson’s Remarks at Southwest Texas State College (now Texas State University-San Marcos) upon Signing the Higher Education Act of 1965. (1965, November 8).
  3. Richard Nixon: Special Message to the Congress on Higher Education. (1970, March 19).
  4. Rakesh Kochhar and Richard Fry, “Wealth Inequality Has Widened Along Racial, Ethnic Lines Since End of Great Recession, Pew Research Center, December 2014,
  5. Clare McCann and Jason Delisle, “ Student Loan Defaults Aren’t Who You Think They Are,” New America, October 2014,
  6. Robert Hiltonsmith, “At What Cost? How Student Debt Reduces Lifetime Wealth,” Demos, August 2013,
  7. Meta Brown, et al., “The Student Loan Landscape,” Federal Reserve Bank of New York, February 18th, 2014,
  8. The figures below were calculated from the U.S. Department of Education’s 2012 National Postsecondary Student Aid Survey (NPSAS), the last year for which data is available. Figures from 2003-04 were calculated from the 2004 NPSAS. Debt numbers are in 2012 dollars.
  9. While the majority of associate’s degrees at public schools are awarded by community colleges, some public 4-year schools also award associate’s degrees in addition to bachelor’s degrees. Strictly speaking we use the term “associate’s degree recipients at public institutions” since the data in this paper includes degree recipients from both public 2-year and public 4-year schools.
  10. Ben Miller, “The Student Debt Review: Analyzing the State of Undergraduate Student BorrowingNew America Foundation, February 2014,
  11. While some private non-profit schools do award associate’s degrees, there were not enough to meet reporting standards in the data sample.
  12. Jeff Webster, “A Brief Look at Transfer Students and Financial Aid,” TG, December 2012,
  13. National Center for Education Statistics (2014). The Condition of Education 2014. Total undergraduate fall enrollment in degree-granting postsecondary institutions, by attendance status, sex of student, and control and level of institution: Selected years, 1970 through 2012.
  14. National Center for Education Statistics (2014). The Condition of Education 2014. Table 306.50. Total fall enrollment in degree-granting postsecondary institutions, by control and level of institution, level of enrollment, and race/ethnicity of student: 2012.
  15. The figures below are primarily retrieved from the U.S. Department of Education’s 2009 Beginning Postsecondary Students survey (BPS 04:09), the last year for which data is available. 
  16. Jacob Gross, et al., “What Matters in Student Loan Default: A Review of the Research Literature,” Journal of Student Financial Aid, 39 no. 1 (2010): 19-29.
  17. Jean Johnson, et al., “With Their Whole Lives Ahead of Them: Myths and Realities About Why So Many Students Fail to Finish College,” Public Agenda, January 2009,
  18. Anne Johnson, Tobin Van Ostern, and Abraham White, “The Student Debt Crisis,” Center for American Progress, October 2012,
  19. Doug Shapiro, et al., “Completing College: A National View of Student Attainment Rates-Fall 2007 Cohort,” (Signature Report No. 6), National Student Clearinghouse Research Center, December 2013,
  20. Doug Shapiro, et al., “Completing College,” Four-Year Completions for First-Time Students Who Started at Two-Year Public Institutions,
  21. Alissa Cunningham and Debra Santiago, “Student Aversion to Borrowing,” Washington DC: Institute for Higher Education Policy & Excelencia in Education, December 2008,
  22. Rachel E. Dwyer, Laura McCloud, and Randy Hodson, “Debt and Graduation from American Universities.” Social Forces, 90 no.4 (2012): 1133–1155.
  23. Min Zhan, “The Impact of Youth Debt on College Graduation,” Journal of Sociology & Social Welfare, XLI no. 3 (2014).
  24. Dongbin Kim, “The Effects of Loans on Students’ Degree Attainment: Differences by Student and Institutional Characteristics, Harvard Educational Review, 77 no.1 (2007): 64-100.
  25. Eric Bettinger, “How Financial Aid Affects Persistence.” In College Choices: The Economics of Where to Go, When to Go, and How to Pay for It, ed. C. Hoxby, Chicago: University of Chicago Press (2004): 207–233.
  26. Benjamin L. Castleman, and Bridget T. Long. “Looking Beyond Enrollment: The Causal Effect of Need-Based Grants on College Access, Persistence, and Graduation.” Cambridge, MA: Harvard Graduate School of Education, 2012.
  27. Mary Nguyen, “Degreeless in Debt: What Happens to Borrowers Who Drop Out,” American Institutes for Research, February 23, 2012,
  28. Defined as income at 200% of poverty or below.
  29. U.S. Department of Education, Office of Federal Student Aid. Direct Loan Portfolio by Delinquency Status.
  30. Federal Reserve Bank of New York, “Student Loan Debt by Age Group,” March 29, 2013,
  31. Robert Hiltonsmith, “At What Cost? How Student Debt Reduces Lifetime Wealth,” Demos, August 2013,
  32. Most of the variables in the much-discussed Brookings study looked at households age 20-40 with education debt and at least $1,000 of student debt and repaying student loans. The Demos model in At What Cost used the same dataset from the 2010 Survey of Consumer Finances, creating a model for dual-educated households with average levels of student loan debt.
  33. Andrew Dugan and Stephanie Kafka, “Student Debt Linked to Worse Health and Less Wealth,” Gallup, August 7, 2014,
  34. Jesse Rothstein, “Constrained After College: Student Loans and Early Career Occupational Choices,” National Bureau of Economic Research, May 2007,
  35. Brent Ambrose, Larry Cordell, and Shuwei Ma, “The Impact of Student Loan Debt on Small Business Formation,” March 29, 2014,
  36. Meta Brown, Sydnee Caldwell, and Sarah Sutherland, “Just Released: Young Student Loan Borrowers Remained on the Sidelines of the Housing Market in 2013,” Federal Reserve Bank of New York, May 13, 2014,
  37. William Gale, et al., “Student Loans Rising: An Overview of Causes, Consequences, and Policy Options,” Tax Policy Center, May 2014,
  38. Jen Mishory and Rory O’Sullivan, “Denied? The Impact of Student Debt on the Ability to Buy a House,” Young Invicibles,
  39. National Center for Education Statistics, “Table 302.60: Percentage of 18- to 24-year-olds enrolled in degree-granting institutions, by level of institution and sex and race/ethnicity of student: 1967 through 2012,”
  40. William J. Hussar and Tabitha M. Baily, “Projections of Education Statistics to 2022,” 41st edition, National Center for Education Statistics, February 2014,
  41. Drew Desilver, “College enrollment among low-income students still trails richer groups,” Pew Research Center, January 15, 2014,
  42. State Higher Education Executive Officers (2014). State Higher Education Finance Report 2014.
  43. All state funding statistics, including educational appropriations, net tuition, and total revenue, accessed from State Higher Education Executive Officers [SHEEO] (2014). State Higher Education Finance 2013.
  44. Stephen Burd, “Undermining Pell: How Colleges’ Pursuit of Prestige and Revenue is Hurting Low-Income Students.” New America, 2014,
  45. College Board, Trends in College Pricing 2013. Figure 20A: Percentage Change in Inflation-Adjusted Mean Family Income by Quintile, 1982-1992, 1992-2002, and 2002-2012; U.S. Census Bureau, Current Population Survey, 2013 Annual Social and Economic Supplement.
  46. Susan Dynarski, “Does Aid Matter? Measuring the Effect of Student Aid on College Attendance and Completion.” The American Economic Review 93 no. 1 (2003): 279–288.
  47. Don Heller, “Making College a Priority: Promising Practices and Strategies.” Testimony before the U.S. Senate Committee on Health, Education, Labor and Pensions, 2012,
  48. Michael S. McPherson and Morton O. Schapiro, “Does Student Aid Affect College Enrollment? New Evidence on a Persistent Controversy,” The American Economic Review 81 no. 1 (1991): 309–318.
  49. Neil S. Seftor and Sarah E. Turner, “Back to School: Federal Student Aid Policy and Adult College Enrollment.” The Journal of Human Resources 37 no.2 (2002): 336–352.
  50. Susan Dynarski, “Loans, Liquidity, and Schooling Decisions,” Unpublished manuscript, Harvard University, August 2003.
  51. Boris Palameta and Jean-Pierre Voyer, “Willingness to Pay for Postsecondary Education Among Underrepresented Group,” Toronto: Higher Education Quality Council of Ontario, 2010.
  52. Sandy Baum, “The Financial Aid Partnership: Strengthening the Federal Government’s Leadership Role,” National Dialogue on Student Financial Aid Research Report 3 (2003).
  53. Alissa Cunningham and Debra Santiago, “Student Aversion to Borrowing.”
  54. Richard Fry, “Young Adults, Student Debt and Economic Well-Being,” Pew Research Center, May 14, 2014,
  55. We concentrated on households age 24-40 in order to filter out at least a sizeable number of borrowers who are not currently in school while also retaining a meaningful sample size.
  56. The Survey of Consumer Finances research comparing households was undertaken by Robert Hiltonsmith, Senior Policy Analyst at Demos and author of At What Cost?

​This publication was funded in part by the Kresge Foundation