Adverse Selection and Moral Hazard in Anonymous Markets

Informational asymmetries abound in anonymous markets, such as those opening in the internet almost every day. In particular, before trading takes place, the typical buyer does not know whether her anonymous counterpart, namely the seller she is confronted with, appropriately describes and prices the trading item; and whether he conscientiously conducts the transaction, so she receives the item in time and in good condition. Without remedies, these informational asymmetries invite adverse selection and moral hazard. Adverse selection may arise because conscientious sellers leave - or may not even enter - the market, as long as their behavioral trait, and their effort, are ex ante unobservable to the buyers, and thus the buyers' willingness to pay or even to trade is hampered. For complementary reasons, opportunistically exploitative and careless sellers tend to self-select into such a market, because they can cheat on buyers by incorrectly claiming to offer high quality products and good delivery service at high price. Moral hazard may arise because effort on both sides of the market is costly: sellers may package goods badly, or delay, or default on delivery. Likewise, buyers may delay, or default on payments. Whereas the consequences of adverse selection and moral hazard are well understood conceptually, we still lack empirical tests on the direction of the effects as predicted by theory, and evidence on the magnitude of these effects. Anonymous markets such as those organized in the internet provide a useful environment for collecting evidence and conducting tests. Faced with adverse selection and moral hazard in these markets, the market organizers designed remedies early on. In particular they constructed mechanisms under which buyers and sellers mutually evaluate their performance; and documented them, so that agents on both sides of the market could build reputation capital. In reaction to opportunistic reporting behavior on one, or both sides of the market, that was not anticipated by the mechanism designers, the reporting mechanisms were improved upon over time. We collected reporting data before, and after such changes to address a question we feel to be central in evaluating the performance of anonymous markets: What are the qualitative and quantitative effects of an unexpected improvement in the typical market participant's reporting possibilities about her counterpart's performance? We show that this leads to a significant increase in buyer satisfaction with the incumbent sellers. The effect is substantively and significantly stronger for sellers that had performed poorly before the change. In fact, rather than exiting from the market - which would have been the alternative reaction to the improvement in the reporting mechanism - , the sellers rated previously poorly improved on all dimensions reported on. Towards our preferred interpretation of these results, we developed a toy model from which we predict effects on seller adverse selection and moral hazard. Sellers differ by the dis-utility they suffer from engaging in effort towards satisfying a consumer at a given (competitive) price. Before the change reflected within this toy model, sellers are incentivized to announce low quality goods as of high quality, because that way, they can fetch a higher price; and to be careless in the delivery of the good – and still can expect to obtain a good buyer rating, because buyers fear retaliation by the seller to any negative rating. Removing that fear incentivizes the buyer to report truthfully rather than opportunistically, and in particular to report on negative experiences. Badly rated sellers react to this by either leaving the market, thus ameliorating adverse selection; or engage in more effort towards improving on buyer satisfaction, thus ameliorating moral hazard. We assess in detail the robustness of our empirical results against alternative interpretations and conclude that our interpretation fits best, that is: the reduction of the informational asymmetry generated by informational bias results in a reduction of seller moral hazard, rather than a reduction of adverse selection.

Klein, Tobias J., Christian Lambertz and Konrad Stahl (2013), Adverse Selection and Moral Hazard in Anonymous Markets, ZEW Discussion Paper No. 13-050, Mannheim.