Incentives for Information Sharing and Outcome Efforts in Networks

In social or organizational networks, it is often observed that different individuals put different levels of production effort depending on their position in the network. One possible reason is reward sharing, which incentivizes particular agents to spend effort in sharing information with others and increasing their productivity. We model the effort level in a network as a strategic decision made by an agent on how much effort to expend on the complementary tasks of information sharing and production. We conduct a game-theoretic analysis of incentive and information sharing in both hierarchical and general influencer-influencee networks. Our particular interest is in understanding how different reward structures in a network influence this decision. We establish the existence of a unique pure-strategy Nash equilibrium in regard to the choice made by each agent, and study the effect of the quality and cost of communication, and the reward sharing on the effort levels at this equilibrium. Our results show that a larger reward share from an influencee incentivizes the influencer to spend more effort, in equilibrium, on communication, capturing a free-riding behavior of well placed agents. We also address the reverse question of designing an optimal reward sharing scheme that achieves the effort profile which maximizes the system output. In this direction, for a number of stylized networks, we study the Price of Anarchy for this output, and the interplay between information and incentive sharing on mitigating the loss in output due to agent self-interest.