Optimal Delay in Committees
Hao Li*, Ettore Damiano, Wing Suen
Date: 2012-05-04 3:15 pm – 3:45 pm
Last modified: 2012-04-20
Abstract
We consider a symmetric two-member, two-alternative, three-state committee problem of information aggregation. A tight bound on maximum delay that can be imposed limits what can be implemented with a single round of delay. In a dynamic delay mechanism, each round is a direct revelation mechanism with delay uniformly bounded from above. We characterize optimal sequences of delays for any given uniform upper bound on each delay. Any optimal sequence is finite, inducing in equilibrium both a ``deadline play,'' with no attempt made to reach the efficient decision just before the deadline, and ``stop-and-start,'' with maximum attempt alternating with no attempt before the deadline play kicks in. The latter is achieved by binding and slackening the bound on delay in turn.