Government Vastly Undercounts Defaults

Photo illustration by Ron Coddington

July 11, 2010

The share of borrowers who default on their student loans is bigger than the federal government's short-term data suggest, with thousands more facing damaged credit histories and millions more tax dollars being lost in the long run.

According to unpublished data obtained by The Chronicle, one in every five government loans that entered repayment in 1995 has gone into default. The default rate is higher for loans made to students from two-year colleges, and higher still, reaching 40 percent, for those who attended for-profit institutions.

The numbers represent thousands of students like Lourdes Samedy, of Boston, who ended up defaulting on about $7,000 in student loans after completing a nine-month-long medical-assistant program at Corinthian Colleges Inc. Everest College, and now cannot get a job.

They also show that the government's official "cohort-default rate," which measures the percentage of borrowers who default in the first two years of repayment and is used to penalize colleges with high rates, downplays the long-term cost of defaults, capturing only a sliver of the loans that eventually lapse.

While the data obtained by The Chronicle are not directly comparable to the two-year rate, which reports defaults by borrowers rather than loans, they reveal that default rates continue to climb years after borrowers have left college, particularly among students who attended two-year and for-profit colleges.

For loans made to community-college students, the 15-year default rate is 31 percent. David S. Baime, senior vice president for government relations at the American Association of Community Colleges, called that number "shockingly high."

"It's really just a tragedy given the consequences of student loan default," he said.

Borrowers who default on their student loans face significant personal and financial burdens. They become ineligible for additional federal aid and may have their wages and tax refunds seized by the government. Their negative credit records make it harder for them to obtain car loans, mortgages, and credit cards, and even apartments or jobs. When they can get loans, they pay higher interest rates.

But it's the high rates of default at for-profit institutions that are likely to get the most attention from members of Congress, who have recently raised concerns about the cost and quality of for-profit higher education. Fifteen years into repayment, two out of every five loans made to students who attended two-year for-profit colleges are in default.

Last month the Senate education committee held what is expected to be the first in a series of hearings on the growth and risks of for-profit higher education. The hearing, which was stacked with critics of for-profit colleges, came a week after the education committee of the House of Representatives held a hearing focused on accreditors' oversight of online learning.

"We have a responsibility to ensure that taxpayer dollars are being spent wisely, and that for-profit colleges are serving students, not just shareholders," said Sen. Tom Harkin, chairman of the Senate committee, in his opening remarks. Shown the 15-year default data for for-profits after the hearing, he appeared taken aback. "Whoa," he said.

While for-profits educate less than 10 percent of students, those colleges' students received close to a quarter of Pell Grant and federal-student-loan dollars in 2008, according to the College Board. And they accounted for 44 percent of defaults among borrowers who entered repayment in 2007, according to the Institute for College Access and Success, a nonprofit organization that advocates making higher education more affordable. When the government can't collect on those loans, taxpayers pick up the tab.

For-profit colleges have long blamed the sector's higher-than-average default rates on the sociodemographics of their students. According to the Career College Association, 43 percent of students attending for-profits are members of minority groups, and almost half are the first in their families to attend college. More than three-quarters are employed.

Presented with the long-term data, Harris N. Miller, the association's president, stuck to that explanation. "Four-year public and nonprofit colleges and universities have more affluent populations," he said. "Four- and two-year career colleges have less affluent populations."

Mr. Miller added that the data may be skewed in favor of traditional colleges by the inclusion of PLUS loans, which have a higher repayment rate and most often go to parents of students attending traditional colleges. The two-year cohort-default rate that the Education Department releases annually does not include PLUS loans.

He argued that for-profits "more than hold their own" against community colleges in the long-term data, given that community-college students borrow less, and that many community colleges have opted out of the federal loan program to avoid the government penalties that a high default rate would bring. In theory, smaller debts should be easier to repay than larger ones.

But Gail O. Mellow, president of LaGuardia Community College, of the City University of New York, said comparing community colleges with their for-profit counterparts ignores the fact that far fewer community-college students borrow. So while the percentages of defaults may be similar, the number of defaulted loans of former community-college students is smaller.

Only 10 percent of community-college students took out federal Stafford loans—the most common type of federal education loan—in the 2007-8 academic year, and most borrowed less than $10,000, according to the College Board. At for-profit colleges, 88 percent of students took out Stafford loans, and nearly 20 percent of associate-degree recipients graduated with more than $30,000 in debt. Those borrowing rates reflect the higher cost of attending a for-profit college. In the 2009-10 academic year, the average for-profit institution charged $14,174 in tuition and fees, according to the College Board, and the average community college only $2,544.

"I think you're looking at very different numbers," Ms. Mellow said. "They're numbers that for-profit colleges do not want us to look at."

In fact, Ms. Samedy, the former Everest College student, now says she's "going straight to a community college," chiefly to avoid accumulating even higher levels of debt. She plans to attend Bunker Hill Community College for a degree in nursing, to try to improve her employability.

The defaults picture isn't likely to improve anytime soon. After falling steadily through the 1990s and reaching a low of 4.3 percent in 2003, the national two-year cohort-default rate began climbing in 2004. The rate for the 2008 cohort of borrowers was provisionally put at 7.2 percent, the highest since 1997.

At the end of the 2008 fiscal year, $39.1-billion worth of loans were in default, according to the Education Department. By the end of the 2009 fiscal year, that total had swelled to $50.8-billion, an increase of nearly 30 percent. As of the end of April, the government had recovered only $6.2-billion of that money.

Disproportionate Defaults

By any measure, for-profit colleges account for a disproportionate share of student-loan defaults. Two years into repayment, 11.9 percent of borrowers who attended for-profit colleges have defaulted on their federal loans, compared with 6.2 percent of those who attended public colleges and 4.1 percent who attended private colleges, according to provisional data that the Education Department released in February. Three years out, for-profit colleges' default rate has nearly doubled, to 21.2 percent of borrowers, and the gap between the sectors has widened.

That trend continues as the loans mature, according to the Education Department data provided to The Chronicle. For-profits accounted for 16 percent of all the loans (other than consolidated loans) issued from 1995 to 2007, but 34 percent of the defaults. Thirty percent of loans made to students attending four-year for-profit colleges have defaulted within 15 years of entering repayment, more than twice the default rates at public and private nonprofit four-year colleges, which are 15.1 percent and 13.6 percent, respectively.

The differences are smaller at the two-year level, where 40 percent of loans made to students attending for-profits have gone into default within 15 years of entering repayment, compared with 31.3 percent of loans made to community-college students and 29.3 percent of those made to students who attended two-year nonprofit private college.

One reason for that narrowing may be demographics. Compared with four-year public and private nonprofit colleges, four-year for-profits tend to enroll more students who are older, lower-income, from minority groups, and of the first generation of their families to attend college, characteristics that correlate with higher dropout rates. For-profit and nonprofit two-year colleges tend to serve similar high-risk populations.

Recent studies lend some support to that interpretation but also conclude that demographics do not fully explain the differences in default rates between for-profit and nonprofit colleges. One analysis, by Education Sector, a nonprofit public-policy group, found that the biggest factor in explaining differences in those rates among two-year colleges is the percentage of students receiving Pell Grants.

But the group also said that only 15 percent of the variation in two-year colleges' default rates could be explained by measurable demographic differences—evidence, the analysts wrote, "of the influence institutions have over whether students stay in good standing on their loans."

A Chronicle analysis of three-year default rates published by the Education Department illustrates how the differences among colleges shrink, but do not disappear, when for-profit colleges are compared with nonprofit institutions serving similar kinds of students.

Still, for-profits say it's unfair for the government to penalize them for serving a riskier population. Under the Higher Education Act, colleges with consistently high default rates can lose their eligibility to award federal financial aid, a penalty that is a death sentence for most colleges. At some publicly traded higher-education companies, the proportion of revenue coming from federal aid approaches 90 percent, the maximum allowed under law.

To account for differences in student populations, Mark Kantrowitz, who runs Finaid, a Web site that provides student financial-aid information, suggests that the Education Department publish separate cohort-default rates for Pell Grant recipients and students who do not receive Pell Grants, and use the non-Pell number only to determine which colleges should be cut off from federal aid.

He also recommends that the department develop an independent measure of institutional quality that could yield an "objective ranking of for-profit colleges" and serve as a guide for prospective students. One possibility, he said, is a standardized test of achievement for graduates from each degree program, administered by one of the major test developers.

"If such tests evaluated mastery of material from a real-world perspective, it would provide value not just for public policy but also for employers," Mr. Kantrowitz said.

Mr. Baime, of the community-college association, agreed that default rates are a "crude proxy" for quality, particularly because relatively few students at community colleges borrow to attend. "The link between default rates and institutional quality is a questionable one," he said.

Asked who was to blame for rising default rates, if not colleges, Mr. Baime hesitated.

"I don't know how to answer that question," he said. "But whoever is ultimately responsible, it seems like it's a really severe problem."

Toward a 'Truer' Default Rate

The new, long-term data provided to The Chronicle don't capture all defaults, since many borrowers take longer than 15 years to repay their loans. But the information provides a more complete picture of student-loan defaults than do any publicly available data.

Under federal law, the Education Department is required to publish only a two-year cohort-default rate, which is the percentage of borrowers who default in the first two years of repayment. However, because it takes so long for borrowers to default, the statistic really reflects only those who do so in the second year of repayment.

The short-term rate also undercounts defaults because it puts borrowers who have received deferments and forbearances on their loans into the repayment category, even though they may have never made a payment on their debt and are likelier than other borrowers to default later on.

The proportion of borrowers in such plans grew from 10 percent in fiscal 1996 to 22 percent in fiscal 2007, as lenders and colleges have pushed students into deferment or forbearance to keep default rates down, so as to avoid federal penalties. Institutions with default rates greater than 25 percent for three consecutive years, or 40 percent for a single year, lose their eligibility to award federal student aid.

Acknowledging those statistical flaws, Congress passed legislation in 2008 that extended the time frame over which cohort-default rates are calculated by a year and raised the three-year penalty threshold to 30 percent. While that change, which will take effect in 2011, will capture more defaults, it still won't prevent colleges and lenders from pushing defaults beyond the window being measured, since borrowers can receive deferments for up to three years and forbearances for up to five.

James Kvaal, U.S. deputy under secretary of education, said the Education Department had known that cohort default rates were lower than lifetime rates. "Student-loan-default rates are a serious concern," he said. "They're something we're watching closely."

He cited efforts by the Obama administration to make student loans more affordable, such as by improving the benefits of income-based repayment options for borrowers and increasing funds for federal student aid. The Education Department will continue to look for more ways to make sure students can afford the loans they take out, he said.

Over time, though, the Education Department has resisted calls to publish a lifetime default rate, saying it would be too administratively complicated, as well as unfair, to hold colleges responsible when their students default years after they drop out or graduate.

When the department's own inspector general suggested that it publish a lifetime rate in 2003, officials initially agreed to do so in terms of dollars in default (rather than per borrowers) but then brokered a compromise. In addition to the two-year cohort-default rate it already provided, the department would publish an annual projection of default rates for the next 20 years, by dollar amount, and an account of actual defaults, by number of loans, for each of the past five cohorts.

The result is a hodgepodge of data that confuses as much as it illuminates, says Alan Collinge, founder of StudentLoanJustice.Org, an advocacy group for borrowers.

"They are putting on an elaborate dog-and-pony show designed to baffle, not elucidate," said Mr. Collinge, who had defaulted on his own student loans.

Asked why the department would want to mask the true default-rate picture, he said, "The Department of Education makes a lot of money off of defaulted loans."

At first glance, the federal budget suggests that's the case. According to estimates by the White House Office of Management and Budget, the government is expected to collect roughly $111 on every $100 of defaulted direct loans and $122 on every $100 of defaulted guaranteed loans in 2011.

But those numbers do not take into account collection costs or inflation. In 2009 the federal government paid $258.3-million to collection agencies, according to President Obama's budget plan.

Including those and other costs in the budget estimates could reduce the government's recovery rate to as little as half of the dollars lost through default, according to a 2007 study by Damien Moore, an analyst at the Congressional Budget Office, and Deborah Lucas, a professor of finance at Northwestern University.

When borrowers default on their student loans, taxpayers are on the hook for 97 to 100 percent of the losses, depending on whether the default is on a guaranteed or direct loan. For example, if a student with $30,000 in debt were to default without making a single payment, taxpayers would be responsible for $29,100 to all of it. In practice, though, many borrowers pay down a portion of their loan before defaulting, and the government collects on some of the remainder.

Some of the recent rise in defaults may be a result of the economic downturn, which has left a growing number of students unable to find jobs or earning too little money to pay back their loans. The unemployment rate among 16- to 24-year-olds was 18.1 percent in May, almost twice as high as the national rate of 9.7 percent, according to the Bureau of Labor Statistics. That month there were 1.2 million unemployed Americans between the ages of 18 and 24 who had completed some college or a college degree.

Some of the increase in defaults may also be due to the expansion of the for-profit higher-education sector, in which enrollments more than doubled between 1998 and 2008, according to a report by Senator Harkin's office.

Critics of for-profit colleges say a growing share of those new students are unqualified to do college-level work and are likely to drive defaults even higher. As evidence, the critics point to the Education Department's data, which show that nearly a quarter of the loans awarded to students at two-year for-profit colleges were already in default only three years into repayment.

"While the data from 15 years ago is very interesting, we are learning how much the industry has changed over that time period, including the major institutions, the composition of students, and the type of degrees being offered," said a Senate Democratic aide. "It just underscores how little we understand the real risk."

Jeffrey Brainard contributed to this article.