Pareto gains of predonation in monopoly regulation

The revelation principle implies that given any admissible social welfare function, the outcome of Baron and Myerson's (BM) optimal direct‐revelation mechanism under incentive constraints cannot be dominated by any other mechanism in expected utilities. However, since the expected total surplus varies with a change in the social welfare function, Pareto improvements should be possible if the monopolist and consumers can agree, by means of side payments that reveal no additional information to the regulator, on the use of an alternative social welfare function which would generate a lower expected deadweight loss. We check the validity of this intuition by integrating the BM mechanism with an induced cooperative bargaining model where unilateral predonation by consumers or the monopolist is allowed. Under this new mechanism monopolist's predonation in the ex ante stage always leads to ex ante Pareto improvement while a certain amount of it eliminates the expected deadweight loss. Moreover, if optimally designed in the interim stage, the monopolist's predonation may also lead under some cost parameters to interim (and also ex post) Pareto improvement. Consumers, on the other hand, have no incentive to make a unilateral predonation, nor to reverse the optimal predonation of the monopolist.