How to Promote Trust: Theory and Evidence from China∗

Economists have long recognized the importance of trust, but how can trust be enhanced in a marketplace? We set up a rational expectation model that allows the marketplace to target buyer trusting (via a buyer protection program) and seller trustworthiness (via penalty on cheaters) separately. We show that these two policies have opposite effects on the equilibrium level of trust: buyer protection increases buyer willingness to trade and therefore invites strategic sellers to enter and cheat; in contrast, a penalty on cheating sellers raises the cost of cheating and leads to more honest behavior. Using transaction-level data from Eachnet.com (an eBay equivalent in China), which introduced different buyer protection policies and a warning system in its early time, we find evidence highly consistent with the theory. More generous buyer protection led to more seller cheating. In particular, sellers that entered the market after the introduction of buyer protection were more likely to cheat and became more trustworthy when the platform adopted the warning system. Reputable sellers were less likely to cheat, but this effect was weakened after the introduction of buyer protection and the warning system. These findings suggest that a trust-promoting policy aimed at buyer trusting may not be effective if it is not accompanied by additional incentives to improve trustworthiness. JEL: D8, L15, L81.