Psychologists have long known that people form impressions of others purely based on their appearance. Although it is well known that appearance-based impressions affect labor market and election outcomes, little is known about the role appearance plays in financial transactions. In a P2P platform, can photographs of borrowers affect the loan performance?
Duarte et al. (2012) employed data from Prosper.com and found that borrowers who appear more trustworthy have higher probabilities of having their loans funded. Moreover, borrowers who appear more trustworthy indeed have better credit scores and default less often. Overall, the findings suggest that impressions of trustworthiness matter in financial transactions as they predict investor, as well as borrower, behaviour.
To analyse the pictures that borrowers post on Prosper, the authors employed Amazon’s Mechanical Turk (MTurk) service. MTurk is an online service through which individuals, referred to as workers, perform highly standardised tasks, often related to text or image processing, for a preset, per-task compensation rate.
The results show that a borrower’s trustworthiness are positively and usually significantly related to the funding probability. Furthermore, the economic size of the effect appears to be significant. For those loan requests that are fully funded, borrowers who are perceived as trustworthy pay significantly lower interest rates than do less trustworthy borrowers. Furthermore, borrowers who appear more trustworthy have better credit grades than do less trustworthy borrowers, and the trustworthiness proxies are negatively related to default.
Full paper see Here