Roger B. Myerson Prize Lecture

322

Townsend 1981, Holmström 1977, Myerson 1979, Rosenthal 1978) found in-

dependently, building on ideas of Gibbard (1973) and Aumann (1974). With 

the revelation principle, this feasible set essentially coincides with the set of 

incentive-compatible mechanisms, which satisfy certain incentive constraints. These 

incentive constraints express the basic fact that individuals will not share pri-

vate information or exert hidden efforts without appropriate incentives.

So mechanism theory expanded our general view of the economic prob-

lem to include incentive constraints as well resource constraints. Incentive 

constraints help us to explain many failures of allocative efficiency that we 

observe in the world. But in this new framework of economic analysis, we also 

have new concepts of incentive efficiency for evaluating the rules by which re-

sources are allocated (rather than specific resource allocations themselves), 

taking incentive constraints into account. These conceptual tools now allow 

us to analyze questions about efficient institutions that were beyond the ana-

lytical reach of economic theory in Hayek’s day.

2. ELEMENTS OF MECHANISM DESIGN THEORY

In society, people have information about their resources and desires, and 

people choose actions for producing, redistributing, and consuming re-

sources. In markets and other institutions of society, individuals’ actions may 

depend on others’ information as it has been communicated in the market 

or social institution. This is the perspective that Hayek recommended, that 

we should view social institutions as mechanisms for communicating people’s 

information and coordinating people’s actions. To decide whether we have 

a good social institution, we want to ask how it performs in this communica-

tion and coordination role. If we do not like the performance of our current 

institutions, then we may want to reform them, to get an institution that 

implements some desired social plan, where a social plan is a description of 

how everyone’s actions should depend on everyone’s information.

So the crucial question is, what kinds of social coordination plans are actu-

ally feasible? A feasible social coordination plan could be implemented by 

many different social institutions, but it is helpful to begin by considering a 

very centralized institution where every individual communicates separately 

and confidentially with a trustworthy central mediator. Suppose first that 

each individual confidentially reports all his or her private information to the 

mediator, and then, based on all these reports, the mediator recommends to 

each individual what actions he or she should take under the plan. But if we 

allow that individuals can be dishonest or disobedient to the mediator then, 

as Hurwicz (1972) observed, the social plan must give people incentives to 

share information and to act appropriately according to the social plan. 

First, to the extent that our social plan depends on individuals’ private 

information that is hard for others to observe, we need to give people an 

incentive to share their information honestly. This problem of getting people 

to share information honestly is called adverse selection. Second, to the extent 

that our social plan requires people to choose hidden actions and exert ef-

---

323

forts that are hard for others to monitor, we need to give people an incentive 

to act obediently according to the plan. This problem of getting people to act 

obediently to a social plan is called moral hazard. If it is a rational equilibrium 

for everyone to be honest and obedient to the central mediator who is imple-

menting our social coordination plan, then we say that the plan is incentive 

compatible.

There are two important things to say about such incentive-compatible 

coordination plans. First, they can be characterized mathematically by a set 

of inequalities called incentive constraints that are often straightforward to ana-

lyze in many interesting examples. Second, although we defined incentive 

compatibility by thinking about honesty and obedience in communication 

with a central mediator, in fact these incentive-compatible plans character-

ize everything that can be implemented by rational equilibrium behavior in 

any social institution or mechanism. This assertion of generality is called the 

revelation principle.

The revelation principle asserts that any rational equilibrium of individual 

behavior in any social institution must be equivalent to an incentive com-

patible coordination plan. Given any possible informational reports from 

the individuals, the equivalent incentive-compatible plan recommends the 

results of simulated lying and disobedience in the original institution or 

mechanism, as illustrated in Figure 1. Thus, without loss of generality, a trust-

worthy mediator can plan to make honesty and obedience the best policy for 

everyone. 

Adverse 

selection

reports

recommendations

Moral 

hazard

1’s type 

(private info)

1’s reporting 

strategy 

(lie?)

1’s reactive 

strategy 

(disobey?)

1’s action

...

...

...

...

n’s type 

(private info)

n’s reporting 

strategy 

(lie?)

n’s reactive 

strategy 

(disobey?)

n’s action

Incentive-compatible mechanism

General 

coordination 

mechanism

Figure 1. The revelation principle. 

(To prove the revelation principle, suppose that we are given a general co-

ordination mechanism and an equilibrium that describes rational individual 

strategies for reporting dishonestly and acting disobediently in this mecha-

nism. We need to describe how a mediator would implement the equivalent 

incentive-compatible mediation plan where honesty and obedience is an 

equilibrium. When the mediator has gotten a confidential report of every 

individual’s private information, the equivalent incentive-compatible me-

---