Two college students. Same credit profile. Same $10,000 loan. Same bank. The only difference is one attends a community college, while the other is enrolled at a four-year institution. But the community college student is charged more to borrow.
This scenario is at the heart of a report released Wednesday by the Student Borrower Protection Center examining how the use of education data in underwriting private student loans may exacerbate economic and racial inequality. The advocacy group, founded by former Consumer Financial Protection Bureau official Seth Frotman, suggests that Wells Fargo and Upstart could be engaging in educational redlining by raising the price of credit for historically marginalized groups.
Both financial-services firms dispute the findings of the report and question its methodology.
Researchers at the Student Borrower Protection Center say they stand by their findings, which are based on publicly available data, and encourage regulators and lawmakers to scrutinize the use of education data in consumer lending.
"Similar to banks' history of redlining in the housing sector, the use of education criteria in credit underwriting could result in borrowers of color receiving more expensive loans simply because of lenders' assumptions and prejudices regarding those who sit next to them in the classroom," said Katherine Welbeck, a civil rights counsel at the Student Borrower Protection Center. "What we found raises serious alarms and warrants immediate attention by lawmakers."
Researchers at the Student Borrower Protection Center submitted online inquiries for two private student loans offered by Wells Fargo: one for undergraduates attending four-year schools and the other for those pursuing two-year degrees or career training. Because the bank reports a range of interest rates for each of its student loans, researchers based their analysis on the average of the quoted rates to determine monthly and total payments over the life of the loans.
A community college borrower would pay $1,134.31 more on a $10,000 loan than a student with the same credit profile pursuing a bachelor's degree. Even though Wells Fargo offers shorter repayment terms for its community college loan, borrowers could still wind up paying more than their peers at four-year schools because of higher interest, according to the report.
Wells Fargo said where borrowers attend school has very little bearing on the terms of its loans.
"Wells Fargo has a long-standing commitment to providing access to financing for students attending community colleges," spokeswoman Vickee Adams said. "We follow responsible lending practices that take into account expected performance outcomes and are confident that our loan programs conform with fair lending expectations and principles."
Wednesday's report also takes a look at Upstart, an online lending platform that refinances student loans. Researchers searched for rates for applicants with identical credit profiles across a range of higher-education sectors, including Hispanic-serving institutions, known as HSIs, and historically black colleges and universities, known as HBCUs.
In one case, a hypothetical 24-year-old analyst seeking to refinance a $30,000 loan would pay nearly $3,500 more over five years if she graduated from Howard University compared with a similar student at New York University, according to the report. The Howard grad would also be hit with $729 in loan origination fees that the NYU alum would avoid.
Graduates of New Mexico State University, a Hispanic-serving school, in the same scenario were charged nearly $1,724 more than otherwise identical NYU graduates. That includes $631 more in origination fees.
Upstart chief executive and co-founder Dave Girouard called the findings "misguided." He said the analysis "flies in the face" of the company's own research that shows using a range of consumer information, including education, leads to higher approval rates and lower interest rates.
The company pointed to the results of its quarterly reports to the Consumer Financial Protection Bureau, which in 2017 raised questions about the fairness of Upstart's underwriting model. According to the bureau, the information it received from the company shows that Upstart approves 27 percent more applicants and yields 16 percent lower borrowing costs than traditional underwriting models.
"We've tested for bias over millions of applicants, not two or three, which can be very anecdotal," Girouard said, referring to the Student Borrower Protection Center report. "A system like ours uses 1,500 data points to assess creditworthiness. Education represents a few of them. [The report] was contrived and not statistically valid."
Girouard said that the use of FICO scores and income -- traditional variables in underwriting -- is "horribly biased" but that expanding the factors used to assess creditworthiness can mitigate the problem.
Frotman argues that the findings of his center's report tell a different story.
"Upstart's own data shows that consumers who attend HBCUs and HSIs can pay substantially more because of where they went to school," Frotman said. "It seems that Upstart's argument is with itself. I'm sure it is of little solace to those who pay thousands of dollars more that only a few of Upstart's inputs are discriminatory."