“Evaluation and Resource Allocation in Firms” by Man-Keung TANG
This paper develops a model to study the relationship between the headquarters and division managers in a setting of project evaluation with severe interest incongruence. Each division manager proposes a project to the headquarters, who makes the approval decisions based upon imperfect signals about project quality. I show that in response to her weak evaluation ability and ex post (efficient) bias, the headquarters may find it optimal to augment the division managers’ incentives by ex ante shifting resources from the strong to weak divisions, hence resulting in "socialism" in the internal capital market. In addition, in the presence of agency concerns, not only may the synergy possibilities across divisions be left unexploited, they may even lead the headquarters to choose to gather less information about each division, thereby endogenously lowering her evaluation ability. This latter result highlights a potential downside of centralization of control rights.