Roger Guesnerie

A complete solution to a class of principal-agent problems with an application to the control of a self-managed firm

This paper considers a class of principal-agent problems which have the following features. (1) There is adverse selection because the principal ignores the value of one parameter of the agent’s true characteristics. (2) Leaving aside the information parameter, the principal’s welfare as well as the agent’s welfare depend on two types of variables, observable to both of them. The first ones, possibly multidimensional, are called action variable(s), and the second one, which is one-dimensional, has in general the meaning of a money transfer. (3) The principal is a Stackelberg leader of the two-person game. He can commit himself to decision rules which are admissible on informational grounds. He optimizes within the adequate class, taking into account, besides the agent’s reaction, one constraint which has generally the meaning of an individual rationality constraint and sometimes of a feasibility constraint. The optimization is limited to the class of non-stochastic mechanisms. (4) The problem can also receive an alternative interpretation: the principal faces a continuum of agents of unknown characteristics (the distribution being known, however). Stylized principal agent problems of this type have often been considered in the economic literature. In particular, the reader will later be able to recognize that the standard income tax model a la Mirrlees (1971), the quality (or quantity) choice model of the monopolist a la Mussa and Rosen (1978) [or Maskin and Riley (1982)], and the model of government regulation of the private monopolist of Baron and Myerson (1982) all belong to

Planning under Incomplete Information and the Ratchet Effect

Central planning of production is usually performed under asymmetric information which leads to use of an incentive scheme. As the planner revises the scheme over time to take into account information provided by the firms performance, this induces firms to underproduce to avoid more demanding schedules in the future—the ratchet effect. This paper explores this phenomenon under the realistic assumption that the planner cannot commit himself to a revision procedure. We show that the ratchet effect exists, in the sense that the planner may choose a scheme which is suboptimal from a static viewpoint in order to induce revelation, with the marginal price of output exceeding its optimal static value.

Sunspots and Cycles

Because sunspot equilibria seem to be of central importance for an understanding of rational expectations, we seek here to characterize completely a limited class of sunspot equilibria (stationary ones with two possible natural events) in the simplest overlapping generations model of production. We present a sufficient condition for the existence of stationary sunspot equilibria, examine how these are related to strictly periodic equilibria of the same order, and investigate how deterministic stationary equilibria bifurcate to stationary sunspot equilibria. A concluding section examines how our results survive in more general settings.

Government Intervention in Production and Incentives Theory: A Review of Recent Contributions

This article reviews the recent literature on regulation under asymmetric information. We first develop the conceptual framework and offer a reminder of the techniques used in the field. Then we apply the framework and techniques to a variety of situations -- with or without commitment. We conclude with a discussion of desirable directions for research.

Second best taxation as a game

Second best theory usually lies in the realm of normative analysis. It focuses attention on the “best” solutions which still can be obtained in problems where constraints of various natures forbid the attainment of first best Pareto optimal situations. Its central concern is the characterization of such “second best” solutions, and a large part of the literature concentrates on the optimal design of taxes or of policies of public firms. Taking a second best problem, we try in this paper to shift the emphasis from “normative” to “positive” considerations. The solution we are interested in is not optimal with respect to some a priori given social welfare function embodying justice objectives; but it is supposed to reflect the power of the different agents in a negotiation process. Clearly, the conceptual tools required for such an analysis are to be found in game theory and it is actually a game theoretical approach that we follow here. The model we are looking at is presented in Section II. It formalizes one of the simplest second best problems that can be imagined. We have a twogood economy with one private good, and one public good. Agents have different wealth in private good and the public good is financed through a wealth tax. This latter fiscal system is clearly not flexible enough to adjust contributions with marginal willingnesses to pay, so that we actually face a second best situation. There is one decision variable in the society, the tax rate which determines the public good production, about which the agents have conflicting desires. To this problem, we associate a game which provides a precise framework for the analysis of the power of the agents. The “positive” outcomes of the negotiation process are supposed to belong to the core of this game. Sections III and IV concentrate on the analysis of the core. The results strongly contrast with those obtained in the classical studies of the core of 67 0022-053 l/81/040067-25

Sunspot Equilibria in Sequential Markets Models

02.00/O

Minimum wage legislation as a second best policy

Publisher Summary Equilibrium with rational expectations is a central construct of modern economic theory. The chapter studies surveys, which are primarily aimed at assessing the properties and relevance of this construct in a general equilibrium framework; in particular, they are related to the issue of multiplicity of rational expectation equilibria. The viewpoint discussed in the chapter is associated with the so-called concept of sunspot equilibrium. It focuses on aspects of general equilibrium with rational expectations that have been given a central role in recent literature. It presents the problem of sunspot equilibria in the broader perspective. It also discusses an example of sunspot equilibrium and a simple economic system—namely, a sequential economy with infinite horizon and time independent structure. Furthermore, this economy is one step forward looking and has no predetermined variable.

Noisy Observation in Adverse Selection Models

Mass unemployment is a striking phenomenon of the present economic situation of European market economies. Although the profession of economists has been criticized for not providing satisfactory cures, it has produced a number of different explanations, either complementary or contradictory for the occurrence of mass unemployment. In this paper, we examine the argument that unemployment is the consequence of minimum wage legislation which is socially desirable. The observation of current wage legislations suggests that the above argument is broadly (at least partly) accepted by policy-makers. Curiously enough, a defence of this argument meeting the rigorous requirements of modern theory seems to be missing. Symmetrically, attacks against minimum wage legislation often rely upon reasonings of dubious relevance, i.e., the obvious remark that unemployment is a waste with first best efficiency criteria. The present paper proposes a second best analysis where second best stems from informational asymmetries which prevent the implementation of lump-sum transfers needed for a first-best redistribution policy.’ The question under scrutiny can be reformulated in a more precise way as follows: Does a second best optimal redistribution policy lead to minimum wage legislation and unemployment? Our elements of answer are gathered in 3 sections. In section 2 the principles of the analysis of the desirability of quantity controls and existing results are recalled. In section 3 the analysis of section 2 is specified to a simple prototype model for the study of linear income taxation. In section 4

Sunspot Fluctuations around a Steady State: The Case of Multidimensional, One-Step Forward Looking Economic Models

We consider a principal-agent contracting problem under incomplete information where some of the agents actions are imperfectly observable. Contracts take the form of reward schedules based on the noisy observation of the agents action. We first review situations where the principal can reach the same utility as in the absence of noise. Then we focus on the use of linear reward schedules, which allow universal implementation, i.e. implementation of a given mechanism for any unbiased noise of observation, and on quadratic reward schedules, which only require the knowledge of the variance of the noise. We exhibit sufficient conditions under which linear reward schedules implement a given mechanism. Finally, we characterize necessary conditions for a mechanism to be implementable under noisy observation by a linear schedule, and by quadratic schedules. We give the geometric intuition behind all results.

Stationary sunspot equilibria in an N commodity world

The paper investigates the existence of stationary sunspot equilibria (SSE) in the vicinity of a steady state in a general, one-step forward looking economic model of dimension n. It is shown that, whenever the steady state is indeterminate, for the associated deterministic dynamics--i.e., there exists a continuum of perfect foresight paths converging to the steady state--then there exists a continuum of SSE of finite order in any neighborhood of the steady state. The proof relies upon bifurcation theory; it provides a characterization of the random processes for which SSE may appear and a description of the location of the support of the SSE close to the bifurcation. The results apply in particular to linear models. Copyright 1992 by The Econometric Society.