Released: November 23, 2010
The Rise of College Student Borrowing
III. Borrowing for Bachelor’s Degrees
The increase in borrowing is partly due to an increase in the share of students who decided to borrow. Bachelor’s graduates of private schools in 2008 were increasingly likely to have used loans to pay for their degrees. At for-profit schools, nearly every bachelor’s degree recipient in 2008 borrowed (97% in 2008), 12 percentage points higher than the share in 2004.9 Graduates at not-for-profit schools also increased their rate of borrowing, from 60% for the class of 1996 to 72% for the class of 2008. The rate at public schools slightly increased, from 59% to 62%.
Moreover, at all types of schools, bachelor’s degree borrowers in the class of 2008 had typically taken on more debt than those in the class of 1996. At public colleges and universities, 2008 graduates had borrowed $20,087, an increase of about $4,500 since 1996. At private not-for-profit schools, the average loan amount increased 41%, from $19,852 to $28,039. Borrowing at for-profit schools also increased between the class of 2004 ($30,106), the first year for which there are reliable data10, and the class of 2008 ($33,046).
For-profit schools have expanded very rapidly into the bachelor’s degree market in recent years. Although they account for a relatively small fraction of bachelor’s degrees (5% in 2009), for-profit schools are growing much more rapidly than the sector in general. From 2003 to 2009, the total number of bachelor’s degrees increased 19%; over the same time period, for-profit schools increased the number of bachelor’s degrees they awarded by 172%.
Borrowers at for-profit colleges and universities typically graduated with the largest loans ($33,046, on average), compared with $28,039 at private not-for-profit schools and $20,087 at public institutions.
Students at for-profit schools are also more likely than graduates of other types of schools to graduate with high levels of debt. One-in-four (24%) 2008 graduates of for-profit schools had borrowed more than $40,000. In comparison, 5% of graduates of public institutions and 14% of graduates of not-for-profit schools had that level of debt.
Given the generally lower cost of attendance at public colleges and universities, it is not surprising that graduates of these institutions borrow less. But there are other differences between private for-profit and private not-for-profit schools that may also contribute to the differences in student borrowing patterns.
Compared with graduates at public or private not-for-profit schools, bachelor’s degree recipients at for-profit schools are older, more likely to be from minority groups, more likely to be female, more likely to be independent of their parents, more likely to have their own dependents and more likely to have lower incomes. For example, in 2008, 53% of bachelor’s degree recipients at private for-profit schools were living independently with their own dependents, compared with 15% of bachelor’s degree recipients at public schools. Also, 42% of bachelor’s degree recipients at private for-profit schools were from lower-income households, compared with 34% at public schools.
Within each type of school, differences in amounts borrowed by bachelor’s degree recipients vary little by gender, race and ethnicity, or income. For example, men and women borrowed virtually identical amounts within each type of school in 2008. The most notable differences are that blacks at public schools borrowed considerably more than Hispanics ($23,155 versus $17,366) and lower-income graduates at public schools borrowed more than middle- and upper-income graduates at public schools. With respect to borrowing rates, there are some clear variations by demographic groups within each type of school. At public schools, for example, 72% of lower-income bachelor’s degree recipients in 2008 borrowed, compared with 49% of upper-income graduates.
Across types of schools, there are larger differences in both amounts borrowed and the rate of borrowing, even within individual demographic groups. For example, the shares of lower-income graduates who borrowed ranges from 72% in public schools to 99% in for-profit schools, and the amount borrowed by these graduates increases from $22,140 in public schools to $32,861 in for-profit schools. It is also worth noting that lower-income students in private not-for-profit schools were less likely to borrow than were either middle- or upper-income graduates in private for-profit schools.
For-profit schools also produce graduates with a different balance of fields of study than do private not-for-profit or public schools. For example, although 4% of bachelor’s degree graduates in the sample are from for-profit schools, they account for 13% of computer and information science graduates and only 1% of humanities and social science graduates. But for every field of study in which for-profit schools produce a significant number of graduates, borrowers with bachelor’s degrees in the same field of study borrow more at for-profit schools than at other schools.
Furthermore, although differences in borrowing patterns exist across fields of study, the differences are relatively small compared with the differences across the public, private not-for-profit, and private for-profit sectors. Consider graduates with bachelor’s degrees in business, who account for nearly one-quarter of graduates of private for-profit schools. Fully 95% of business graduates at for-profit schools choose to borrow, compared with 73% at private not-for-profit schools and 59% at public schools. Business graduates who borrowed at for-profit schools averaged $33,011 in loans, significantly more than business borrowers at either not-for-profit schools ($26,886) or public schools ($19,441).
- Sample size for private for-profit schools is too small for meaningful analysis before 2004. ↩
- The 2008 NPSAS sampled about 114,000 undergraduate students. This report focuses on the approximately 20% of those students who graduated in 2008. NPSAS does not report unweighted sample sizes for tabulations; all values in this report are based on estimated sample sizes larger than 100 individuals. ↩