Designing Deliberation for Decentralized Decisions

I describe and analyze a model of strategic communication and deliberation in decentralized decision-making settings. I show that, in a cheap-talk environment, inclusion and exclusion of agents can affect the credibility of messaging between agents and, accordingly, the quality of policy decisions and overall social welfare. Somewhat surprisingly, the inclusion of agents can aid information aggregation and social welfare even when the added agents do not themselves communicate truthfully. Analogously, the results suggest an informational, social welfare–based rationale for excluding agents not only from observing policy-relevant deliberation but also from observing the product of the communication precisely because the excluded agents possess decision-making authority.