I characterize the constrained efficient (or planner’s) allocation in a directed (competitive) search model with private information. There are sellers with private information on one side of the market and homogeneous buyers on the other side. They match bilaterally in different submarkets and trade. In each submarket, there are search frictions. In the market economy, homogeneous buyers enter different submarkets (i.e., post different contracts) and sellers with private information direct their search toward their preferred submarket. I define a planner whose objective is to maximize social welfare subject to the information and matching frictions of the environment. The planner can impose taxes and subsidies on agents that vary across submarkets while being subject to an overall budget-balance condition. I show that the planner generally achieves strictly higher welfare than the market economy. I also derive conditions under which the planner achieves the complete information allocation. I present examples in the context of financial and labor markets, explicitly solve for the efficient tax and transfer schemes and compare the planner’s allocation with the equilibrium allocation.
This a huge paper. Literally (64 pages), it indeed covers a lot of ground, but also in terms of its message. The abstract, while very precise, does not highlight the potential impact of this paper. We have learned from the Welfare Theorems that a planner can replicate a decentralized economy under perfect conditions. As socialist planned economies have demonstrated, these perfect conditions do not exist, one main reason being that the planner has less information than market participants. Here we have an economy where some market participants have more information than others and the planner has as much information as those that have the least of it. Yet, the planner is capable to improve on the decentralized outcome. It is not simple, as the planner needs to use subsidies and taxes, and to open and close markets. But it is possible.