Jessica Rinaldi / Reuters

For women and people of color, getting a loan from a bank can be an especially difficult proposition. That’s why many hope that the growth of alternative lenders will usher in a different, more equitable kind of credit system.

But a recent paper suggests that these new lenders may be in danger of repeating some of the same old mistakes.

The prevalence of discrimination in lending of all kinds is well documented. “Black testers, as opposed to white testers, are less likely to receive a quote for a loan, are often quoted higher interest rates, are given less coaching from loan officers, and loan officers are less likely to look for factors to justify accepting their applications,” writes Sarah Harkness, a sociology professor at the University of Iowa and author of the paper. She says that when assessing loan candidates lenders make judgments about a potential borrower’s reliability. Some of this judgment comes from hard data, such as income or credit score, but some of it can come from bias and stereotyping, Harkness theorizes. And that, in turn, leads to systemic discrimination that prevents millions of families from building wealth.

Harkness conducted a survey of users on a peer-to-peer lending site who all had the same credit ranking (a “B,” the second-highest level by site standards). She then assigned each applicant lender-facing avatars that differed by gender and race, and loan requests that totaled less than $10,000.

She finds that lenders consider black women and white men to be the most fundable. White women were less likely to receive funding than either black women or white men, and black men were the least likely of any group to obtain loan approval. While she stops short of assigning motives, Harkness does highlight research that may explain why potential borrowers are ranked in this order of preference. For example, recent studies have suggested that black women tend not to fit neatly into prototypical stereotypes for either gender or race. They are assigned traits such as assertiveness and self-confidence, which are often associated with men. These traits may also make lenders believe that a borrower is more likely to repay their loans. Black women are also often seen as more competent, self-reliant, and high-achieving than black men. This, may help explain why lenders in the study were more willing to give them access to capital despite their status as a double minority group.

But while this may seem like somewhat good news—at least for black women—it’s also true that white Americans generally have more access to capital in the first place. White families tend to have both more household and intergenerational wealth than their black peers. They also tend to have wealthier circles of friends. That means that even though white women might not be seen as a great bet in this particular study, they’re still more likely to be able to go to family or friends in order to round up funds for anything from a business venture to an emergency. That’s simply not the reality for black men, who arguably need even more help accessing credit.

It’s also true that this is a fairly small and narrow study that assesses only small-dollar loans doled out by peer-to-peer lenders—a relatively small share of the overall lending market. When it comes to the broader market, significant bias still exists that makes it harder for minorities, women, and minority women to get loans to start businesses, buy homes, or even build credit.

Harkness says that her findings provide support for the idea of removing identifying personal information from loan applications, which would decrease the ability for stereotyping, bias, and status assignment based on anything other than financial history. The good news here is that maybe, with some concerted effort, peer-to-peer networks may offer a more egalitarian option for lending than traditional financial institutions have in the past.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.