Search

We study the role of reciprocity in markets where expert-sellers have more information about the severity of a problem faced by a consumer. We employ a standard experimental credence goods market to introduce the possibility for consumers to gift the expert-seller before the diagnostic, where the gift is either transferred unconditionally or conditionally on solving the problem. We find that both types of gifts increase the frequency of consumer-friendly actions relative to no gift, but only conditional gifts translate into efficiency gains when the consumer faces a high-severity problem. This suggests that partial alignment of incentives via conditional gifts may outweigh kindness motives when reciprocal actions are not directly observed. Using further treatments with surprise gift exchange, we show that withholding a gift that is expected by expert-sellers significantly reduces the likelihood of consumer-friendly behavior whereas sending a gift to expert-sellers who do not expect one has no effect. [ABSTRACT FROM AUTHOR]

One's willingness to accept an outcome or even to correct it depends on whether the underlying procedure is deemed legitimate. We examine a modified version of the dictator game, where dictatorship is assigned by a fair procedure that is linked to the participant actions but in effect is completely random, to illustrate that this belief is not independent of the outcome and is self-serving in its nature. We also discuss the perceptions of fairness and merit as potential drivers of the observed behavioral phenomenon. [ABSTRACT FROM AUTHOR]