Issue Brief

The Debt Explosion Among College Graduates

By Heather Boushey1

April 3, 2003


Executive Summary

This paper examines trends in indebtedness among recent college graduates over the last decade. College loan burdens increased among graduates all along the income distribution, at both public and private colleges and universities.

Student loan debt is 85 percent higher among recent college graduates who took on debt while attending public four-year colleges than among graduates from a decade ago. Recent graduates owed an average of $15,100 in 1999/2000, up from $8,200 in 1989/1990. Student loan debt increased by 55 percent among recent graduates of private four-year colleges with student loan debt. These graduates owed an average of $16,500 in 1999/2000, compared to $10,600 in 1989/1990 (all figures in 2002 dollars).

Lower income students tend to owe the most money, but the biggest increase in indebtedness over the decade has been among higher income students. In 1999/2000 recent graduates of public colleges from families in the poorest two quartiles owed an average of $13,300 and $13,400, respectively. This compares with an average debt burden for indebted students from the richest quartile of $12,000. However, this debt burden represented an increase of 85 percent for the students from the richest quartile, compared to increases of 67 and 62 percent for poorest and second poorest quartiles, respectively.

The percentage of recent graduates with student loan debt also rose substantially over this period. For example, the percentage of recent graduates from private colleges with student loans from families in the poorest quintiles rose from 63.7 percent in 1989/1990 to 75.4 percent in 1999/2000. Among private school graduates in the richest quartile the increase was from 26.8 percent to 50.1 percent.

The rise in indebtedness over this period was the result of federal government policies that favored loan aid rather than grants, and focused increasingly on students from relatively affluent families. Data from the late nineties indicates that the strong labor market at the time allowed most graduates to cope with higher debt burden. However, the recent weakness of the labor market could pose problems for heavily indebted students. In addition, sharp tuition increases of recent years (driven by state financial problems) are likely to lead to higher debt burdens for future graduates. Such burdens may seriously constrain students’ career choices, and could lead them to delay starting a family or buying a home. 


Today’s college graduates begin their careers burdened with a level of debt unknown to previous generations. Among graduating seniors in 1999/2000, two-out-of three (67.9 percent) borrowed money for their undergraduate degree (including family loans), up from less than half (46.3 percent) in 1992/93. On average, an indebted student who graduated in 2000 had loans totaling $17,785  (National Center for Education Statistics).2 Higher student loans are a function of both rapidly rising costs for higher education and policy changes during the early 1990s that made it easier for students to receive college loans as a part of their student aid package. These loans made attending college possible for millions of young Americans, however the burden they now face is substantial. Higher student debt burdens have implications for the career choices of graduates and their decisions to pursue further education, as well as their capacity to make other investments in their future, such as starting a family, or saving for a home.

Incurring student debt can be an excellent investment that will pay off in the form of higher lifetime earnings. The benefits to having a college degree have grown, even as college costs more than doubled. Over the past three decades, the gap in wages between workers with a high-school degree or less and those with a college degree has increased. Wages among recent college graduates have grown rapidly, especially over the late 1990s, while entry-level wages of high-school graduates are still below their 1973 peak  (Mishel, Bernstein, and Boushey 2003). Because they entered the workforce during a period of low unemployment and relatively strong wage growth, most graduates in the late 1990s found jobs that paid enough to allow them to make their monthly payments. These graduates were relatively well equipped to cope with their debt burdens.

            Students graduating today, however, may not be so lucky. Unemployment has spiked up over the past two years, and young workers have found that the job market is not nearly as rosy as it was just a few years ago. Wage growth has stalled, so even those who have found employment may find that their wages are not enough to cover their debt payments and their other expenses. One favorable condition is that interest rates are at historic lows and the rate on many student loans hovers just above 4 percent, lowering monthly payments. However, those attending college now face the prospect of an even larger debt burden. Over the past two years, tuition has increased rapidly and states have cut back considerably on their aid; some states even raised tuition mid-year in 2001-02. In addition, low current interest rates are not likely to persist into the future.

At no point in recent history have we required young people to shoulder so much of the burden of their post-secondary education through a lien on their future wages. At the same time, young people need a college degree more than ever to enable them to find a job at a decent wage. Whether or not that wage will cover their living expenses and their loan burdens is, however, another question. For many, rising loan burdens will mean abandoning their first career choice or graduate school in favor of a more financial stability while for others it will mean forgoing or postponing starting a family or investing in other major purchases.

Rapid Increases in College Costs Lead to Greater Student Debt

As college became critical to finding a good-paying job, college costs rose rapidly. Over the 1980s and 90s, college costs rose faster than median income and faster than financial aid. Between the 1980/81 and 2001/02 academic years, average tuition more than doubled at both public and private four-year institutions. Tuition and fees increased more during the 1980s than the 1990s, but even so, over the 1990s, tuition rose by 40 percent at public four-year institutions and 33 percent at private four-year institutions  (The College Board 2001). Over this same period, median family income grew relatively slowly, so that the share of income necessary to cover tuition grew substantially. For many families, affording college on their own became increasingly difficult and low-income families were increasingly unable to close the gap with higher-income families in college attendance rates (U.S. Department of Education 2001).

As college costs rose, student aid also increased, but not at pace with tuition levels. Whereas in the early 1970s, grants comprised the majority of student aid, loans now comprise the majority of student aid funds. While grant aid per full-time equivalent student increased by 62.1 percent between 1982-82 and 2000-01, loans increased by 153.8 percent. During the 2001-02 school year, the amount of grants per full-time equivalent student totaled $3,085, while loans totaled $4,200. Over this same time period, tuition at public four-year institutions increased by 86.5 percent and by 121.8 percent at four-year private institutions (The College Board 2002). Even with the increase in the amount of loan funds dispersed, more students receive a grant than a loan: among students in public, four-year institutions, for example, 39.6 percent took out a student loan in their own name while 46.3 percent received a grant (U.S. Department of Education 2002).

The increase in loans obscures the true decline in aid to students. Now, just over half of student loans are unsubsidized, a dramatic change from a decade ago when the majority of loans were subsidized by the federal government (Figure 1). Under the subsidized student loan program, the government pays the interest on the loans until the student graduates from college (after a grace period) and also during periods of unemployment. Interest rates are capped at favorable levels, below the market rate for personal loans. Unsubsidized loans have higher interest rate caps than subsidized student loans and interest begins accruing from the time of disbursement. Unsubsidized loans now make up 86.4 percent of student loans disbursed. Parent loans, which are called PLUS loans (Parent Loans for Undergraduate Students), which are also unsubsidized, made up a little over one-tenth of all loan funds disbursed in 2000/01, up from 7 percent in 1990/91.


 The increase in unsubsidized student loans over the 1990s was a direct result of higher education legislation in the early 1990s, the 1992 Reauthorization of the Higher Education Act. This legislation increased the amount of money students could borrow under the student loan program, changed the definition of need so that it was easier for dependent students to qualify, and made unsubsidized loans available to dependent students for the first time. The aim of these reforms was to increase the accessibility of higher education and close the gap in college attendance between students from rich and poor families  (Ellwood and Kane 2000) . Given budget constraints, the administration chose to increase loan availability because it was more cost-effective than increasing grant aid. Reforming the student loan program was only one aspect of the Clinton education reforms; other elements included small increases in the Pell grant program for low-income students, from $2,300 in 1993 to $3,300 in 2000 (nominal dollars), and new tax credits. Although the Pell grants increased $1,000 in nominal terms, in constant dollars, they only increased by $584. 

            One implication of the increased availability of student loans was that higher income families now had access to low-cost loans. Prior to the 1992 legislation, such families were unable to qualify for the subsidized loans and not offered unsubsidized loans. Now, they are able to borrow money through the unsubsidized loan program. This has led to a process of “catch-up” by higher income families. In 1999/2000, students from low-income families were still more likely to have taken out student loans, but those from higher-income families had closed that gap substantially over the 1990s (Figure 2).3 This is true among graduates from both public and private four-year institutions. Among graduating seniors from the lowest family income quartile, the proportion who took out loans went up slightly among those who attended private, four-year institutions and barely at all among those attending public four-year institutions. Low-income families did not increase their loans as much during the nineties because they were already borrowing at relatively high rates.



            For students across the income distribution and across both public and private schools, the amount of loans increased substantially between the early and late 1990s in response to higher tuition. Students from low-income families who attended public four year institutions saw their loans increase by 66.8 percent over the 1990s, while those who attended private four year institutions saw their loans increase by 55.3 percent. The amount of loans increased more so for students from higher-income families, where loans to students who attended public schools increased by 85.2 percent among students from families in the highest income quartile and 94.7 percent among students from families in the second highest income quartile  (U.S. Department of Education 2002). 

Even as students from high-income families began to accumulate more debt, the total value of loans remains higher among students from low-income families (Figure 3). In 1999/2000, graduating seniors from low-income families who attended public institutions had an average debt burden of $13,294, compared to $11,966 for students from high-income families (in 2002 dollars). The trend is similar for those who attended private schools: those from low-income families had an average debt of $15,926 while students from high-income families had $14,570.



             The Clinton-era educational reforms mostly affected middle and upper class students. As noted earlier, poorer students already had access to subsidized loans. Thus, the increased availability of student loans has not helped lower-income students to attend college more, but has increased the debt burden among students from wealthier families. The other elements of the reforms also benefited higher-income families: for example, the Hope Scholarships and Lifetime Learning tax credits, which took effect in 1998, are not refundable. Most low-income families do not owe federal income tax and therefore cannot benefit from a tax credit unless it is refundable. In addition, the amount of the tax credit is reduced by the amount of tax-free education assistance that the student receives (including Pell grants), further limiting the number of low-income families that can receive them.

The real issue with student loan debt burdens, however, is not just the amount, but whether students can pay them off in a reasonable time period once they begin working. Growing wages among recent college graduates made it possible for many graduates to earn sufficiently high incomes to pay off their loans. Table 1 shows the average debt burden in 1994 and 1997 (the latest year available) among those who graduated in 1992/93.4 The higher a graduate’s post-college income, the faster they paid off their loans and the lower their overall debt burden: among graduates who had taken out student loans and who were earning $50,000 or more in 1997, only 58.6 percent still had debt while 71.7 percent of those earning less than $20,000 still had debt. However, the average debt was only slightly smaller among the lower paid graduates: among those earning $50,000 or more, average debt was $8,829 in 1997, while among those earning less than $20,000 the average debt was $7,699 (in 2002 dollars). Upon graduation, students with high-paying jobs and those with low-paying jobs have similar debt levels. Those who earn less, however, take longer to pay these loans off.


Table 1: Debt burdens among bachelor's graduates from 1992-93 who took out student loans



Percent who still owed


Average debt burden


Average amount still owed













 $ 11,029 

 $  7,936 








By 1996 total income:







less than $20,000





 $ 10,407 

 $  7,699 






 $ 10,140 

 $  7,286 






 $ 11,226 

 $  7,734 






 $ 11,089 

 $  8,320 

$50,000 or more





 $ 11,592 

 $  8,829 








Source: Choy and Carroll (2000).

Note: Sample on includes those with no further enrollment bachelor's degree.

Dollar values are annual CPI-U 2002$.

Implications: Recent Graduates Start Off Life Burdened by Debt

As college costs skyrocketed, the federal government chose to provide relief to students and their families primarily through expanding the availability of unsubsidized student loans while grant aid has become a smaller share of the aid package. Taking on student debt can be a good investment for families -- as the returns to a college education have increased, this investment has paid off for millions of young Americans. However, the success of this policy hinges on a strong labor market for recent college graduates. The student loan program was expanded just before the long economic expansion of the 1990s. As the 1990s wore on, unemployment remained at historic lows and wages grew rapidly. Individuals who took on debt for college during that period were able to pay off their loans upon graduating because they had relatively good labor market opportunities.

Student loan debt has not yet posed significant problems. Even with historically high levels of debt, there was not a disproportional increase in bankruptcy among young college graduates. Those who had taken out loans in the early 1990s were just as likely as those who did not to make major purchases (such as a house or a car) and to marry and start families by 1997 (Choy and Carroll 2000). However, these is no data on graduates after the late 1990s and time will tell whether students who took out large loans in the mid- and late-1990s will fare as well as those who took out loans in the early 1990s. The sheer size of loans has grown so much that it is unlikely that this burden will not affect the career and lifestyle choices of today’s graduates.

            The problem of financing higher education is currently becoming even more difficult for families around the nation. Over the past two years, the states have begun to experience large deficits. In response, many have chosen to cut funding for higher education leading to big increases in tuition. Tuition is up by more than 10 percent for the 2002-03 academic year in 16 states (Idaho, Indiana, Iowa, Kansas, Kentucky, Massachusetts, Missouri, New York, North Carolina, Ohio, Pennsylvania, South Carolina, South Dakota, Texas, Virginia and Washington). Further, some states have had to resort to relatively dramatic steps as six states have implemented mid-year tuition increases, which is highly unusual and difficult for students to cope with (Johnson, Lav, and Ribeiro 2003). It is likely that student debt will increase substantially in response to tuition costs. Alternatively, many students will be forced to simply forgo or postpone getting a college education.

The implications of the policy of providing students mostly loans rather than grants, are only now being seen. The first classes of students with high, unsubsidized loan debt graduated in the mid-1990s during a very strong labor market. These graduates had starting wages that were large compared to earlier graduating classes and thus they were well equipped to pay off their loans in the short-term. Further, student debt increased during a period of low interest rates, which has reduced the value of loan payments and lightened the debt burden. Given the current recession and more limited opportunities for young graduates, the debt of more recent graduating classes may prove to be less manageable.


 Choy, Susan P., and C. Dennis Carroll. 2000. Debt Burden Four Years After College. Washington, DC: U.S. Department of Education, Office of Educational Research and Improvement, National Center for Education Statistics. 

Ellwood, David, and Thomas Kane. 2000. Who is Getting a College Education? Family Background and the Growing Gaps in Enrollment. In Securing the Future: Investing in Children from Birth to College, edited by S. Danziger and J. Waldfogel. New York: The Russell Sage Foundation.

Johnson, Nicholas, Iris J. Lav, and Rose Ribeiro. 2003. States are Making Deep Budget Cuts in Response to the Fiscal Crisis. Washington, DC: Center on Budget and Policy Priorities.

Mishel, Lawrence, Jared Bernstein, and Heather Boushey. 2003. The State of Working America 2002-3. Ithaca, NY: Cornell University Press. 

National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education. Table 129: Percent of graduating seniors who borrowed for undergraduate education. Vol. 2003. 

The College Board. 2001. Trends in Student Aid. Washington, DC. 

The College Board. 2002. Trends in Student Aid. Washington, DC. 

U.S. Department of Education, National Center for Education Statistics. 2001. The Condition of Education 2001. No. NCES 2001-072. <

U.S. Department of Education, National Center for Education Statistics. 2002. National Postsecondary Student Aid Study, 1989-90, 1992-93, 1995-96, 1999-2000. Washington, DC. <>

[1] Heather Boushey is an Economist at the Center for Economic and Policy Research.

[2] This figure includes debt from all loan sources, including family loans.

[3] The data for this analysis come from the National Postsecondary Student Aid Study, conducted by the National Center for Education Statistics, U.S. Department of Education, which is only conducted every 4 years and the next release will not be until 2004 or 2005. Figures only include subsidized and unsubsidized Stafford loans, which account for about 85 percent of all student loans. They do not include loans to parents under the PLUS program.

[4] The data for this analysis come from the Baccalaureate and Beyond Longitudinal Study that followed students who graduated in 1992/93 who are in the NPSAS survey. The NCES follows graduates for over a decade, beginning with an interview one year after graduation. The next scheduled Baccalaureate and Beyond cohort will be associated with the NPSAS 2000 and data from the first follow-up should be available this year.