Incentive Design with Spillovers

Economists Krishna Dasaratha, Benjamin Golub, and Anant Shah have published a significant paper titled 'Incentive Design with Spillovers' (arXiv:2411.08026v2), introducing a formal mathematical model for optimizing team compensation. The research addresses a classic principal-agent problem: how a manager (the principal) should design payment contracts to motivate a team when individual efforts produce stochastic outcomes with 'spillover' effects on colleagues. The authors develop a multi-agent generalization of contract theory by applying methods from network games, moving beyond simple individual performance metrics to account for complex team dynamics.

The core finding is a precise formula for optimal incentive allocation. The model demonstrates that pay should be structured to equalize, across all team members, a specific product of three factors: (i) the agent's individual productivity, (ii) their organizational centrality within the collaboration network, and (iii) their personal responsiveness to monetary incentives. This means the 'star performer' isn't necessarily the highest paid; an employee who is highly central to team workflows and highly motivated by money might warrant greater incentive investment.

The paper specializes this general framework to answer practical management questions. It explores whether compensation should reward raw individual ability or 'collaborativeness' (a function of network position), and how the strength of complementarities between team members' work shapes overall pay dispersion. This provides a data-driven, theoretical foundation for moving from intuition-based to algorithmically-informed compensation design, particularly in knowledge-work and R&D teams where output is collaborative and hard to measure individually.