author
qiang fu department of strategy and policy, national university of singapore, 15 kent ridge drive, singapore, 119245. email: bizfq@nus.edu.sg
zenan wu school of economics, peking university, beijing, china, 100871. email: zenan@pku.edu.cn
youji lyu school of economics, peking university, beijing, china, 100871. email: lyjecon@pku.edu.cn
yuanjie zhang school of economics, peking university, beijing, china, 100871. email: zyj economic@pku.edu.cn
abstract
this paper studies a multi-player tullock contest in which contestants exhibit reference-dependent loss aversion `a la k˝oszegi and rabin (2006, 2007). contestants may differ in their prize valuations. we verify the existence and uniqueness of purestrategy choice-acclimating personal nash equilibrium (cpne) under moderate loss aversion and fully characterize the equilibrium. the equilibrium in our setting sharply contrasts that in the usual two-player symmetric case. loss aversion can lead contestants’ individual efforts to change nonmonotonically, while the total effort of the contest must strictly decrease. further, it always leads to a more elitist distributional outcome, in the sense that a smaller set of contestants remain active in the competition and stronger contestants’ equilibrium winning probabilities increase. our results are robust under the alternative equilibrium concept of preferred personal nash equilibrium (ppne).
keywords:
loss aversion; contest; reference-dependent preference; choice-acclimating personal nash equilibrium (cpne); preferred personal nash equilibrium (ppne).
jel: c72, d72, d81, j31.
download：
expectation-based loss aversion in contests.pdf