Does appearance reflect creditworthiness on p2p lending sites?

Successful Prosper BorrowersA study conducted by Dr. Jefferson Duarte of Rice University sought to determine whether physiognomy-based prejudices about creditworthiness existed and, if they did, whether they were justified. To explore this issue, he used data provided by Prosper, a peer-to-peer lending hub, to determine whether analysis of a single photo provided by the borrower could help predict whether he received full funding of the loan and whether he would pay it back in time.

So do looks affect perceived creditworthiness? As the Economist profiles, the answer is yes. Even more interesting is that these perceptions correlated highly with the borrowers' actual credit ratings.

To explore the issue, the researchers paid assistants to analyze the pictures of thousands of aspiring borrowers on Prosper. The workers rated them, on a scale of one to five, on how trustworthy they seemed, and estimated the percentage probability that each individual would repay a $100 loan. They also made assessments of the borrowers' sex, race, age, attractiveness, family status, and obesity.

The researchers reached two main conclusions:

1. People flagged as untrustworthy by the assistants were less likely to be offered a loan. To have the same chance of getting one as those deemed most trustworthy they were required to pay an interest rate that was, on average, 1.82 percentage points higher, even when the effects of historical creditworthiness were statistically eliminated.
2. The assessments of trustworthiness, and of likelihood to repay a loan, made by the assistants correlated with potential borrowers’ credit ratings based on their credit history, even when the other variables, from beauty to race to obesity, were controlled for statistically.

So while Prosper is a relatively information-rich environment where potential lenders have access to complete financial profiles of borrowers including credit grades, income, employment, and assets, this study suggests that there is a substantial premium put on the subjective factor of perceived trustworthiness. For individuals who look untrustworthy, the "faceless bank" might be a better option after all.

An interesting aside offered by the researchers is that although attractiveness and trustworthiness are positively correlated, there is no evidence that attractiveness is related to the probability of a loan becoming fully funded. There are other physiology-based preferences at work.

And showing yet another emerging role for peer-to-peer platforms, the "assistants" used the researchers were in fact workers from the peer-to-peer market laborplace, Mechanical Turk (owned by Amazon). Peer-to-peer sites then provided both the source of data for the study as well as the independent evaluations and perceptions.