'Redlining' or Reasonable Criterion?

Like most of the revelations and allegations that have poured out of Andrew M. Cuomo's office in the New York attorney general's investigation into the student loan industry, his latest is provocative and highly charged. Two weeks ago, in a powerfully worded letter to Sen. Christopher Dodd (D-Conn.) and Rep. George Miller (D-Calif.), Cuomo accused some providers of private student loans with engaging in the kind of racial discrimination that home mortgage lenders once practiced.

Some private lenders, he contended, have been using judgments based on the college a student attends -- in addition to typical criteria such as the credit ratings of borrowers and their parents -- as one factor in the evaluation that ultimately affects the interest rate on a loan. The practice, he wrote, is "analogous to 'redlining' in the home mortgage industry," a reference to charging higher rates based on a prospective homeowner's neighborhood, common a generation ago before federal laws changed to prohibit it.

Taking a specific college, or type of college, into account as a factor in determining a credit score could theoretically mean that loans to students at, say, Harvard could be seen by lenders as less risky and therefore more desirable than those made to students at community colleges, for-profit institutions and historically black colleges. John Dean, special counsel to the Consumer Bankers Association, said, for example, that "some underwriting criteria include future earnings prospects that may be reflected in the type of institution." That's where the allegations of discrimination have come in, because critics of the practice say there is a disparate impact that places disproportionately poor and minority students at a disadvantage when applying for loans.

Representatives of the private loan industry bristle at the contention that they are engaging in a present-day form of discrimination, but they maintain that the practice, for some lenders at least, can be useful as a way to make student loans available to borrowers who otherwise wouldn't have access to them at all.

Some on Capitol Hill, though, see things differently and are taking steps to ban extra factors such as a student's institution in calculating credit risk. Lenders counter that those institutions with the most financial need enroll many students who don't even have credit scores, and if they do, they tend not to be as good as those for students at more elite institutions. Taking an institution, or type of institution, into account is a way around the problem, allowing lenders to calculate rates -- even if they're higher -- for students who otherwise might pose too great a risk of defaulting on their loans.

But institutions with higher default rates and more credit risk tend to have more poor and minority students. According to the Project on Student Debt, which advocates capping interest rates and relieving student debt, a disproportionate percentage of African Americans leave four-year colleges in debt -- 84.7 percent, compared to 73.7 percent of Latinos, 64 percent of whites and 58.1 of Asians. They also borrow more, compared to other racial groups, with an average debt of $22,041. On average, underrepresented minorities also tend to have less income to repay their loans with.

By institution, the numbers are stark as well: About 78 percent of students at HBCUs took out federal loans in 2005, while the percentage drops to 44 percent for those at non-HBCU colleges and universities.

Cuomo illustrated his "redlining" allegations by describing practices he uncovered at an unnamed "large lender," one that separates schools into three different groups based on default rates. For example, colleges with default rates between 0 and 3 percent would get the best interest rates, from 8 to 9.25 percent. The next group, with default rates of 3 to 5 percent, would get interest rates from 9 to 12 percent, while the final group (default rates of 5 to 10 percent) received the highest interest rates (11 to 14 percent). A student's FICO score would determine whether the final rate was toward the low or high end of the range corresponding to his or her institution.

At Duke University, Cuomo continued, excellent credit will get a student an 8-percent rate, while the same credit rating will net an 11-percent rate at the for-profit University of Phoenix. And the worst interest rate students can receive at Duke -- 9.25 percent -- is still lower than the best rate at Phoenix.

A disproportionate 23 percent of the University of Phoenix's student population is African American, a fact that may have informed Cuomo's thinking in describing the lenders' practices using terminology with racial connotations.

With some of these facts surely in mind, Dodd has already proposed introducing legislation -- the Private Student Loan Transparency and Improvement Act -- that would, in the process of expanding federal oversight of the private loan industry, ban using some of the controversial criteria in determining the loans they offer to students. The bill would, in a draft of the language, "Prohibit lenders from using any data in their underwriting that may have disparate impact on the loan products, terms, or conditions available to student borrowers based on race, age, and other personal factors, or the institution they attend."

But that could change, according to a member of Dodd's staff who was familiar with the process of drafting the legislation, who suggested that a later version might shy away from an "out-and-out prohibition" of using extra factors. The focus would instead be on preventing any "disparate impact" on students, while acknowledging that lenders may have to resort to college-by-college distinctions to augment their credit analyses. The legislation is still being developed and could eventually be introduced as an amendment to the Higher Education Act or surface in a committee drafting session.

The staffer also said that there was "a lot of anecdotal evidence" that lenders are considering factors such as a student's institution. Lenders who use the college-by-college approach as part of their underwriting process have also acknowledged it publicly and privately in meetings with legislators -- and while they admit that the results could sometimes negatively affect students' interest rates, their overall argument centers on expanding students' access to credit, a point apparently not lost on the Dodd camp.

Lenders have explicitly denied that their loan calculations violate disparate-impact statutes. Dean said that his group was working with Dodd's office on the proposed bill to reinforce prohibitions on discrimination while at the same time establishing an allowance for potentially using other indicators beyond individual credit ratings, known as FICO scores, in the underwriting process.

"If Congress were to pass a statute that limited credit evaluation to FICO scores," Dean said, it could leave the lenders' hands tied "because most students don't have a credit record." Therefore, he said, "consideration of other factors that might help justify a lower rate is certainly justifiable.

"Conversely, a student who may not have all those characteristics may have to pay a little bit more."

Effects on the Institutions

Cuomo's allegations and Dodd's developing legislation still don't clarify the extent of the damage implied by allegations of disparate impact. And as the lawmaker continues to meet with interested parties that may be affected by any change in lending regulations, one constituency pointedly hasn't spoken out about discriminatory lending practices: the colleges themselves. Representatives of different kinds of institutions have varying assessments of the problem, and some have wondered whether the solution might have unintended effects.

Historically black institutions themselves have been fairly silent. While many students at the colleges depend on subsidized loans provided by the federal government, there is still some reliance on the private sector -- and with it, a realization that being too strict on lenders might ultimately reduce their access to loans.

And what if lenders couldn't factor in a student's institution, as proposed by Dodd? "It would still be possible" to offer loans to students at institutions such as HBCUs, suggested Dean, but "it would just simply be that they would have to take a look at the overall economics of their business," he said, adding that "certain students attending the HBC may not be able to get credit as a result of the fact that looking only at their FICO score, they would not be bankable. I would hope that no financial institution would walk away from HBCs."

There's also a distinction to be made, between students who are "credit ready" but don't have a credit history, and those who do have a credit history (which may be negative). "For people with bad credit, they may find themselves losing the opportunity to get a loan because Congress would have passed a statute" barring using some factors beyond FICO scores, Dean said. Students with negative histories enrolled at elite historically black institutions such as Howard University, Dean suggested -- which could count as a positive factor in credit underwriting -- might lose out in such a scenario.

But Cuomo's and Dodd's worry is that, conversely, students at historically black and other institutions seen as "high risk" would be punished for the college they attend, rather than judged on their own personal credit history -- what Cuomo, continuing the housing analogy in his letter, dubbed "a student's 'school neighborhood.' " A Dodd staffer suggested that if an upper-middle-class student attended a historically black university, he or she might not receive the best loan available to a comparable student enrolled at, say, Princeton. Hence Cuomo's allegations of disparate impact.

Officials of for-profit colleges have expressed support for continuing to use factors other than credit score -- but not if that means favoring elite schools based on a perception of quality or pedigree.

" 'Quality' shouldn't be based on whether an institution is a top-ranked institution," said Harris N. Miller, president of the Career College Association, which represents mostly for-profit colleges. "Quality should be based on very objective criteria" such as graduation rates and average salary, rather than evaluating schools based on their accreditation or membership in groups such as the Association of American Universities and the Career College Association, he suggested.

Miller supports "recourse loans," in which institutions share part of the risk with their students. But he stressed that he wanted to "make sure … Congress doesn't outlaw legitimate market responses that allow low-income students" to pay for college.

For community colleges, the effects of various lending practices -- or measures introduced to curb them -- will reach only as far as private loans are currently used. And for now, most students at two-year colleges who need to borrow take out federally subsidized loans, and at smaller amounts than students at other institutions.

"Given that community colleges enroll higher percentages of lower-income students who tend to have higher credit risks, lenders are going to be less interested in providing capital to them," said David Baime, the vice president for government relations at the American Association of Community Colleges. He suggested that community colleges are "toward the margins on this in terms of impact."

Cutting Subsidies

Tied into the issue of which factors are acceptable in calculating risk and interest rates is the looming threat to the student loan industry of massive cuts to their subsidies. Dean has suggested that slashing the federal government's funding could lead to a "flight to quality" -- in essence, a mass migration of financial institutions toward a tendency to lend only to students with lower risks of default. "What that means is that lenders may very well lose interest in making loans available to higher-risk borrowers, where it may no longer make economic sense," he said.

Dean even speculated that those cuts might lead some lenders to stop marketing their loans through schools and go directly to students with a risk level they are comfortable with. "Anybody who suggests that these budget cuts can be absorbed with no change … is crazy," he added. "People will try to get more loans from higher-quality schools [and] seek to avoid loans where they think the economics will be a net loss to the lender."

Robert Shireman, executive director of the Project on Student Debt, takes issue with that assessment. "The size of the reduction … is not the kind of reduction that's going to cause any kind of significant drop in availability of loans," he said. His evidence? As soon as the Senate announced its intention to slash subsidies, "the stock price of Sallie Mae went up, which suggests that the analysts who look at whether this is an enterprise that will be able to continue, their judgment is that [it will]."