When general equilibrium theorists attempted to extend the resource allocation mechanisms to nonconvex environments they realized that new issues of commu- nication and incentives arose:
In a broader perspective, these findings suggest the possibility of a more systematic study of resource allocation mechanisms. In such a study, unlike in the more traditional approach, the mechanism becomes the unknown of the problem rather than a datum. . . . The members of such a domain (of mechanisms) can then be appraised in terms of their various “performance characteristics” and, in particular, of their (static and dynamic) optimality properties, their informational efficiency, and the compatibility of their postulated behavior with self-interest (or other motivational variables).
—Hurwicz (1960, p. 28)
Hurwicz (1960) dedicated his paper to Jacob Marschak. Marschak was the only major economist aware of incentive problems in the fifties, problems that he chose not to study:
This raises the problem of incentives. Organization rules can be devised in such a way that, if every member pursues his own goal, the goal of the organization is served. This is exemplified in practice by bonuses to executives and the promises of loot to besieging soldiers; and in theory, by the (idealized) model of the laisser-faire economy. And there exist, of course, also negative incentives (punishments). I shall have to leave the problem of incentives aside.
—Marschak (1955, p. 128)
Marschak was familiar with the literature of statisticians who became aware of incentive problems quite early. The problem of moral hazard arose in sampling theory for quality control. Whittle (1954) and Hill (1960) understood that the distributions of quality were endogenous and dependent on the care taken in the production process. They studied how to take into account this noncontrollable effort level in their analysis of quality from a sample. Adverse selection appeared when forecasting probabilities of some events. Good (1952), McCarthy (1956), and later Savage (1971) looked for payment formulas leading forecasters to announce their true estimated probabilities and discovered the incentive constraints for the revelation of information.
Economists working with Hurwicz developed a general framework, the mech- anism design approach, which treated the competitive markets as just one partic- ular institution in a much more general family of mechanisms run by benevolent planners. During the sixties the emphasis of the research was on the commu- nication costs required by nonconventional environments, until Groves (1973), influenced by Schultze (1969),30 called for considering incentives in public policy and constructed incentive-compatible mechanisms in a team problem.
The next major step was the understanding of the Revelation Principle,31 which shows that, with adverse selection and moral hazard, any mechanism for organizing society is equivalent to an incentive-compatible mechanism by which all informed agents reveal their private information to a planner who recommends actions. The Revelation Principle provided the appropriate framework for the nor- mative analysis of economies with asymmetric information and contracts that can be written on all observable variables. It delivered a neat methodology to study incentive theory that we will use in most of this book.
Source: Laffont Jean-Jacques, Martimort David (2002), The Theory of Incentives: The Principal-Agent Model, Princeton University Press.