We consider a multi-sender cheap talk model, where the receiver faces uncertainty over whether senders have aligned or state-independent preferences. This uncertainty generates a trade-off between giving sufficient weight to the most informed aligned senders and minimizing the influence of the unaligned. We show that preference uncertainty diminishes the benefits from specialization, i.e., senders receiving signals with more dispersed accuracy. When preference uncertainty becomes large, it negates them entirely, causing qualified majority voting to become the optimal form of communication. Our results demonstrate how political polarization endangers the ability of society to reap the benefits of specialization in knowledge.

A principal designs a potentially random decision rule selecting between k outcomes. An agent has private information about the
state of the world, determining both von Neumann-Morgenstern utility functions. I place no further restrictions on preferences. The design problem reduces to selecting an optimal convex menu of lotteries from which the agent chooses his preferred one. I characterize the extreme points of the set of such menus as maximal and indecomposable subsets of the unit simplex. In particular, for three outcomes, there is always an optimal mechanism with at most a range of three, yet extreme points lie dense in the set of maximal menus for four or more outcomes. My results are related to previously observed phenomena in the multi-object monopolistic seller problem. My analysis rests on the literature on indecomposable convex bodies started by Gale [1954].
Applications include job design, the allocation of public housing, and hiring.

I study welfare maximizing and strategy-proof mechanisms in settings where agents have single-peaked preferences over a finite set of alternatives. My main finding is that restricting attention to unaninmous mechanisms can decrease welfare. However, the leftmost and rightmost alternatives are always in the image of a utilitarian mechanism. All other alternatives between these two are compromises. I call a compromise a bad compromise if the possibility to exclude it from the set of potential outcomes increases welfare. I find that for three alternatives up to 50% of utilitarian welfare can be lost if the compromise cannot be excluded. My analysis gives a new justification for constitutional provisions that limit the scope of negative majorities.