Information Design for Congested Social Services: Optimal Need-Based Persuasion

We study the effectiveness of information design in reducing congestion social services catering to users with varied levels need. In absence price discrimination and centralized admission, provider relies on sharing about wait times improve welfare. consider a stylized model heterogeneous users who differ their private outside options: low-need have an acceptable option the social service, whereas high-need no viable option. Upon arrival, user decides for the service by joining unobservable first-come-first-serve queue, or leave seek her To reduce outcomes, seeks persuade more to avail option, thus better serve users. characterize Pareto-optimal signaling mechanisms compare welfare outcomes against several benchmarks. show that if either type is the overwhelming majority population, does not provide improvement over sharing full information. On other hand, when population is sufficiently heterogeneous, information only Pareto dominates full-information no-information mechanisms, some regimes it also achieves same as “first-best”, i.e., admission policy knowledge users’ types.