مقاله ای راجع به انسان اقتصادی و اخلاقی بودن با رویکرد نظریه بازی های تطوری/تکاملی که با همکاری ویبول - که در زمینه نظریه بازی های تطوری/تکاملی آثار فراوانی دارد - در اینجا معرفی شده است. برای بارگیری متن کامل مقاله بر عنوان آن کلیک کنید. بخش انتهایی مقاله در ادامه آورده شده است.
Alger, I., & Weibull, J. W. (2013). Homo moralis—preference evolution under incomplete information and assortative matching. Econometrica, 81(6), 2269-2302.
بخش انتهایی (نتیجه گیری) مقاله
We have here tried to contribute to the understanding of ultimate causes for human motivation by proposing a theoretical model of the evolution of preferences when these preferences remain private information and when the matching process is random but may involve correlation between types in the matches. Our approach delivers new testable predictions. Although we permit all continuous preferences over strategy pairs, we find that a particular one-dimensional parametric family, the preferences of homo moralis, stands out in the analysis. A homo moralis acts as if he or she had a sense of morality: she maximizes a weighted sum of her own payoff, given her expectation of the other’s action, and of the payoff she would obtain if the other party were also to take the same action.We show that a certain member of this family, homo hamiltonensis, is particularly viable from an evolutionary perspective. The weight that homo hamiltonensis attaches to the moral goal is the index of assortativity in the matching process. Our theory further predicts that if one and the same individual is engaged in multiple pairwise interactions of the sort analyzed here, perhaps with a different index of assortativity associated with each interaction (say, one interaction taking place within the extended family and another one in a large anonymous market), then this individual will exhibit different degrees of morality in these interactions, adapted to the various indices of assortativity.
While the general predictive power of preferences à la homo moralis remains to be carefully examined, the behavior of homo moralis seems to be compatible with some experimental evidence (for preliminary calculations in this direction, see Alger and Weibull (2012b)). What is more, the goal function of homo moralis appears to be consistent with the justifications many subjects offer for their behavior in the lab, namely, saying that they wanted to “do the right thing” (see, e.g., Dawes and Thaler (1988), Charness and Dufwenberg (2006)). While we leave analyses of the policy implications of such moral preferences for future research, we note that, in our model, the evolutionarily stable degree of morality is independent of the payoffs in the interaction at hand. Hence, the degree of morality cannot be “crowded out” in any direct sense by economic incentives. For instance, if one were to pay people for “doing the right thing” or charge a fee for “doing the wrong thing,” this would change the payoffs and thus also the behavior of homo moralis, in an easily predictable way, but evolutionary forces would not change her degree of morality (as long as the matching process remains the same).
While the self-interested homo oeconomicus does well in nonstrategic interactions and in situations where there is no assortativity in the matching process, natural selection wipes out homo oeconomicus in large classes of other situations. Arguably, assortativity is common in human interactions. Allowing for arbitrary degrees of assortativity in the matching processes, our analysis suggests that pure selfishness, as a foundation for human motivation, should perhaps be replaced by a blend of selfishness and morality. Such a change would affect many predictions in economics.
This is but a first exploration, calling for extensions and applications in many directions and areas, such as multi-player interactions, asymmetric and repeated interactions, signals and cues about others’ types, partner choice, public-goods provision, environmental policy, institution building, voting, and political economy.