Targeting the Key Player: An Incentive-Based Approach

We consider a network game with local complementarities. A policymaker, aiming at minimizing or maximizing aggregate effort, contracts with a single agent on the network to trade effort change against transfer. The policymaker has to find the best agent and the optimal contract to offer. Our study shows that for all utilities with linear best-responses, it only takes two statistics about the position of each agent on the network to identify the key player: the Bonacich centrality and a weighted measure of the number of closed walks originating from the agent. We also characterize key players under linear quadratic utilities for various contractual arrangements.