Abstract

We study competitive behavior in all-pay Tullock (1980) contests with identity-dependent externalities (IDEs) governed by a fixed network. First, we introduce a model of network contest games, in which the prize generates an externality---which may be positive or negative---that impacts each player directly connected by the network to the winner of the contest. We establish existence of Nash equilibria and provide sufficient conditions for uniqueness, building on recent theoretical advances for games played on networks. We then derive closed-form results, with an intuitive characterization, for regular networks and for a subclass of core-periphery structures. Second, using a controlled laboratory experiment, we provide robust empirical support for the comparative statics predictions of the model. Our experimental findings also suggest that observed patterns of mean over-investment relative to point predictions may be driven by both heterogeneous joy of winning and social efficiency concerns that emerge in the presence of IDEs. Altogether, our study provides a novel application for the theory of network games, and new insights regarding behavior in all-pay contests.

Full Text
Published version (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