Abstract
Rabin's psychological game-theoretic model of ‘fairness’ has been the starting point for a literature about preferences for reciprocity. In this literature, reciprocity is modelled by defining an individual's ‘kindness’ or ‘unkindness’ in terms of the consequences of his actions for others, and assuming a motivation to reward (punish) other people's kindness (unkindness). Contrary to intuition, this form of reciprocity cannot explain mutually beneficial trust and trustworthiness in a simple Trust Game. We formalise and offer a diagnosis of this ‘Paradox of Trust’. We distinguish between two kinds of reciprocity. Rabin's concept of reciprocal kindness is a psychologically plausible motivation, and the paradox is an informative result about the implications of this motivation. However, trust is better understood in terms of reciprocal cooperation – the motivation to play one's part in mutually beneficial practices, conditional on others playing their parts. We show that a theory of reciprocal cooperation can avoid the paradox.
Highlights
We argue that Rabin’s conception of reciprocal kindness is a psychologically plausible motivation, and that the Paradox of Trust is an informative result about the collective consequences that follow when individuals are motivated in this way
The implication, we suggest, is that the reciprocity expressed in practices of mutually beneficial trust and trustworthiness must be something other than reciprocal kindness
We have argued that the Paradox of Trust is inherent to the psychology of reciprocal kindness
Summary
The Trust Game originates in the work of Berg et al (1995), but it has a much longer history as a theoretical model. Q1 = q2 = 1 in the Trust Game and q2 < 1 in P2’s Dictator Game; in both games, there is first- and second-order consistency of beliefs This scenario describes a world in which, if P1 and P2 play the Trust Game, P1 is fully trusting (he chooses send with probability 1) and P2 is fully trustworthy (she chooses return with probability 1). Contrary to this implication of his model, ‘it seems plausible that cooperation would take place’ His response to this variant of the Paradox of Trust is to say that his model is not intended to represent all psychological factors that can affect behaviour in games; theorists may need to consider modelling ‘additional emotions’. The implication, we suggest, is that the reciprocity expressed in practices of mutually beneficial trust and trustworthiness must be something other than reciprocal kindness
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
