This chapter evaluates a more general attack on optimal contract and mechanism design stressing cases of adverse selection, which makes use of the revelation principle. One should be clear about the uses to which the revelation principle is put. It can be thought of as a statement about how actually to implement contracts. But it may be better to use it with greater circumspection as a tool of analysis for finding the limits of what outcomes can be implemented, without reference to how best to implement a particular outcome. In some contexts of direct revelation, there will be situations ex post where the party in the role of the government knows that it can obtain further gains from trade from one or more of the parties who participated. Meanwhile, in many applications of the revelation principle, the party in the role of mechanism designer must be able to commit credibly to no subsequent (re)negotiation once it learns the types of the parties with which it is dealing.
Computational criticisms of the revelation principle