Transparency, complementarity and holdout

This paper characterizes the conditions under which holdout (i.e. bargaining inefficiency) may, or may not be significant in a two-sided, one-buyer-many-seller model with complementarity. We address this problem in a very general setup with a bargaining protocol that is symmetric and allows for both publicly observable, as well as secret offers, and a technology that allows for variable degrees of complementarity. The central insight is that the transparency of the bargaining protocol, formalized by whether offers are publicly observable or secret, as well as the extent of complementarity, play a critical role in generating efficiency. Even with perfect complementarity, holdout seems to be largely resolved whenever the bargaining protocol is public (but not if it is secret). Further, irrespective of the bargaining protocol, holdout is resolved if the marginal contribution of the last seller is not too large.