Randomization under ambiguity: efficiency and incentive compatibility (Q2201693)

The authors generalize the previous work made by \textit{L. De Castro} and \textit{N. C. Yannelis} [J. Econ. Theory 177, 678--707 (2018; Zbl 1417.91159)], where it was proved that when agents have the Wald's maxmin preferences, the conflict between efficiency and incentive compatibility no longer exists: all efficient allocations are also incentive compatible if and only if agents have the Wald's maxmin preferences. The generalization shown in the present manuscript is made by taking into account the use of randomization. In this direction, the authors analyze and answer the following questions: Is each efficient allocation of de Castro and Yannelis [loc. cit.] still Pareto optimal? Are all efficient allocations still incentive compatible under the Wald's maxmin preferences? Positive answers are provided and several applications are discussed.