Contract theory I Patrick Bolton and Mathias Dewatripont. p. cm. Includes bibliographical references and index. ISBN 1. Contracts- Methodology. Contract Theory - Ebook download as PDF File .pdf), Text File .txt) or read book Patrick, Contract theory I Patrick Bolton and Mathias Dewatripont. p. cm. Preface xv. 1. Introduction. 1. Optimal Employment Contracts without Uncertainty, Hidden. Information, or Hidden Actions. 4. Optimal Contracts under.
|Language:||English, Spanish, Arabic|
|Genre:||Academic & Education|
|Distribution:||Free* [*Register to download]|
Bolton, P., and Dewatripont, M.: Contract Theory. Article in Journal of Economics 86(3) · February with Reads. *Bolton, P. and M. Dewatripont (), Contract Theory, chapter 1: Introduction. * ruthenpress.info Contract theory, by Bolton, P. and Dewatripont, M. MIT Press: Cambridge and London. , xvi+ pp., USD (cloth).
As we shall detail throughout this book. In practice. At least as a first approach. In reality it is likely that most contracts that we see partly reflect prior negotiations and each party's negotiating skills. One important dimension in reality is the extent to which employees are insured against economic downturns. Most existing unemployment insurance schemes are 4. Once this common description is determined. The reason is that entrepreneurs are natural "risk lovers" and are best able to absorb the risk that "risk-averse" employees do not want to take.
For example. They define a state of nature as any possible future event that might affect an individual's utility. To be concrete. The important conceptual leap is rather to suppose that rational individuals are able to form a complete description of all possible future events and.
A fundamental economic question concerning these insurance schemes is how much "business-cycle" and other "firm-specific" risk should be absorbed by employers and how much by employees.
Arrow and Debreu were the first to explore this extension. The state space is then simply the set of all possible future events. One theory.
Before doing so. The difficulty is not in defining all these notions. To be able to analyze this question of optimal risk allocation formally one must enrich the framework of section 1. More precisely.
Assume that there are only two possible future states of nature. Should employers take on all the risk. At one level this extension is extremely simple. V tH See Figure 1. It should be clear by now that the analysis of pure exchange under certainty can be transposed entirely to the case with uncertainty once one enlarges the commodity space to include contingent commodities.
Then the state of nature influences only the value of output each individual has as endowment. If the two individuals do not exchange any contingent commodities.
As before. Before the state of nature is realized each individual has preferences over consumption bundles tL' tH represented by the utility functions V tL' tH for the employer and V tL' tH for the employee. But they can also increase their ex ante utility by coinsuring against the economic risk. Note that in a "recession" aggregate outputis lower than in a boom It is this framework that is used in most contracting applications.
If we callpj E 0. In the setup considered by von Neumann and Morgenstern. The first complete framework of decision making under uncertainty Also let iij.
Although this basic assumption is rarely motivated. Let llL. In most contracting applications it is assumed that all parties share a common prior belief and that differences in posterior probability beliefs among the parties only reflect differences in information.
It is instructive to consider the optimal insurance conditions 1. This means that ex ante efficiency is achieved if and only if the contract is also ex post efficient. We shall see that when incentive considerations enter into the contracting problem there is usually a conflict between ex ante and ex post efficiency. Then the contract-maximizing joint surplus is fully characterized by the first-order conditions: VI llj.
For now. In general. Thus the simple Knightian idea that entrepreneurs perfectly insure employees is likely to hold only under special assumptions about risk preferences of entrepreneurs. This is exactly the solution that intuition would suggest. An individual's attitude toward risk is driven in part by initial wealth holdings. The rationality requirements imposed on the contracting parties and the enforcement abilities assumed of the courts should be kept in mind as caveats for the theory of contracting when faced with very complex actual contractual situations where the parties may have.
A risk-neutral individual has a constant marginal utility of money. Condition 1. In many contracting situations in practice it is possible that the contracting parties will be unable to agree on a complete description of the state space and that. To summarize. Another important simplifying assumption to bear in mind is that it is presumed that each party knows exactly the intentions of the other.
Thus it is generally accepted that individuals' absolute risk aversion tends to decrease with wealth. It should be clear from this brief overview of optimal contracting under uncertainty that the presumption of rational behavior and perfect enforceability of contracts is less plausible in environments with uncertainty than in situations without uncertainty.
What was less well understood. How far should employee insurance be scaled back to make way for adequate work incentives? How could adequate work incentives be structured while preserving job security as much as possible?
These remained open and hotly debated questions over the successive fiveyear plans. As a consequence. Another important reason employees are likely to get only limited insurance is that they need to have adequate incentives to work.
Much of Part I of this book will be devoted to a formal analysis of this question. If the output produced by employees tends to be higher when employees exert themselves more. These considerations are the subject of much of this book and are briefly reviewed in the next sections of this chapter. The first problem refers to a situation where the employee may have private information about her inability or unwillingness to take on.
One is the hidden-information problem and the other the hidden-action problem. Even in the egalitarian economic system of the former Soviet Union the provision of incentives was a generally recognized economic issue.
The reason is simply that the equilibrium price of such insurance would be too high if employers were also averse to risk. Two general types of incentive problems have been distinguished. This is a well-understood idea. Now the underlying contracting situation requires specification of the private information one of the parties might have and the beliefs of the other party concerning that information in addition to preferences. The second problem refers to situations where the employer cannot see what the employee does-whether she works or not.
These situations fall under the general descriptive heading of signaling problems. These chapters examine optimal bilateral contracts when one of the contracting parties has private information. The introduction of hidden information is a substantial break from the contracting problems we have already con. Chapter 3 considers the opposite situation where the informed party makes the contract offers. In these situations it is the employee's actions that are hidden from the employer.
Chapter 2 explores contracting situations where the party making thec. In the context of employment contracts the type of information that may often be private to the employee at the time of contracting is her basic skill. These situations are often referred to as screening problems. Problems of hidden information are often referred to as adverse selection. The revelation principle says that it is. Or at least they thought so. Prospective employers understand this and therefore are willing to pay educated workers more even if education per se does not add any value.
Another way for employers to improve their pool of applicants is to commit to pay greater than market-clearing wages.
The basic insight behind the revelation principle is that to determine optimal contracts under asymmetric information it suffices to consider only one contract for each type of information that the informed party might have. Such a policy naturally gives rise to equilibrium unemploYment.
Just how efficient can contracting under asymmetric information be? The answer to this question turns out to be surprisingly elegant and powerful. We review Spence's model and other contracting settings with signaling in Chapter 3.
It is generally referred to as the revelation principle and is one of the main notions in contract economics. With the introduction of asymmetric information in contracting problems economists have at last found plausible explanations for observed market inefficiencies that had long eluded them in simpler settings of contracting under complete information.
This tends to attract better applicants. More concretely. A first fundamental question to be tackled was. The basic idea behind his analysis is that more-able employees have a lower disutility of education and therefore are more willing to educate themselves than less-able employees.
Such a contract would maximize production efficiency. Suppose first that employee time and output enter additively in both utility functions: The variable 8 is thus the state of nature. As a way of illustrating a typical contracting problem with hidden information. When employee productivity is private. Assumptions similar to the nono. If that is the case. So the employer might as well pick that point directly.
Under that assumption. A slightly more realistic assumption serving the same purpose is that the employee's output may be random and that a "no-slavery" constraint prevents the employer from punishing the employee ex post for failing to reach a given output target.
It says that all the employer needs to determine is a menu of two "point contracts": If it were. The reason why the employer does not need to specify a full nonlinear contract t i is that each type of employee would pick only one point on the full schedule t l anyway.
Thus the optimal menu of employment contracts under hidden information can be represented as the solution to the optimal contracting problem under complete information: Although this class of contracts is much simpler than most real-world employment contracts. As this problem immediately reveals. The general economic principle that this chapter highlights is that hidden information results in a form of informational monopoly power and allocative inefficiencies similar to those produced by monopolies.
In the preceding example. It may therefore be attractive to lower skilled employment that is. This leaves informational rents for the unskilled employee relative to her outside opportunity.
Much of Chapter 2 will be devoted to the analysis of the structure of incentive constraints and the type of distortions that result from the presence of hidden information. One option is to have a contract with allocative efficiency. The trade-off between allocative efficiency and rent extraction will be detailed in the next chapter. It may not have had any official unemployment. Contracting problems with hidden actions involve a fundamental incentive problem that has long been referred to in the insurance industry as moral hazard: Chapters 7 and 13 will discuss at length the extent to which market outcomes under hidden information may be first.
One of the first and most striking empirical studies of. This behavioral response to better insurance arises in almost all insurance situations. In contrast to most hidden information problems. It is worth recalling here that the centrally planned economy of the Soviet Union was notorious for its overmanning problems. Chapter 2 also explores several settings where employers have private information about demand and the value of output.
In these problems the agent employee is not asked to choose from a menu of contracts. Our discussion has focused on a situation where the employee has an informational advantage over the employer.
When a person gets better protection against a bad outcome. As we highlight. Incentive problems like moral hazard are also prevalent in employment relations.
As they reluctantly found out. According to the theory. The most spectacular form of incentive pay seen nowadays is the compensation of CEOs in the United States. Actual CEO compensation packages have also shown some serious limitations. Moral hazard on the job was one of the first important new economic issues that Soviet planners had to contend with. How do insurers deal with moral hazard?
By charging proportionally more for greater coverage. As with insurance companies. For a time ideological fervor and emulation of model workers seemed to work. As is now widely understood. Employers typically respond to moral hazard on the job by rewarding. If they were to abolish unemployment and implement equal treatment of workers. Also the employer must now take into account the fact that will be chosen by the employee to maximize her own expected payoff under the output-contingent compensation scheme t 8j.
Since effort 1. For simplicity. Where PH['] respectively pd. Note that if output were to increase deterministically with effort then the unobservability of effort would not matter because the agent's hidden effort supply could be perfectly inferred from the observation of output. Determining the solution to the employer problem with both constraints is not a trivial matter in general. There is one situation. Chapter 4 provides an extensive discussion of the two main approaches toward characterizing the solution to this problem.
The application of this principle can give rise to quite complex compensation contracts ip. In that case it is efficient to have the employee take on all the output risk so as to maximize her incentives for effort provision. For now we shall simply point to the main underlying idea that an efficient trade-off between insurance and incentives involves rewarding the employee most for output outcomes that are most likely to arise when she puts in the required level of effort and punishing her the most for outcomes that are most likely to occur when she shirks.
They can also undertake high-expected-retum but high-risk investments. One reason why this simple theory may predict unrealistically complex incentive schemes is that in most situations with hidden actions the incentive problem may be multifaceted. We explore these ideas both in Chapter 6. These situations. Besides helping the readers to acquaint themselves with the core concepts of the theory.
All these problems raise new analytical issues of their own and produce interesting new insights. Part I of our book also discusses contracting situations with an intermediate form of asymmetric information.
This logic is so powerful that it is difficult to see why there should be mandatory disclosure laws. If an employee is not. We provide an extensive treatment of some of the most important contracting problems under multidimensional asymmetric information in the research literature.
Employers understand this fact and expect employees to be forthcoming. The basic logic. One of the main ideas emerging from the analysis of contracting problems with private but verifiable information is that incentives for voluntary disclosure can be very powerful. Chapter 5 discusses the main limits of the unraveling result and explains when mandatory disclosure laws might be warranted.
It is for this reason that employees have incentives to voluntarily disclose all but the worst piece of private verifiable information. Besides their obvious practical relevance. Because of this simplicity. The best way of dealing with this issue is in fact to design the contract in such a way that each player has a unique dominant strategy. Gibbard and.
Then the outcome of the game is easy to predict. Part II of this book is devoted to this question. It comprises two chapters. Rather than stick to predictable but inefficient contracts. While the general methodology and most of the core ideas discussed in Part I extend to the general case of multilateral asymmetric information.
Besides the obvious technical and methodological interest in analyzing these more general contractual settings. Chapter 7 deals with multilateral private information and Chapter 8 with multilateral hidden actions.
In the one-sided private information case the contract design problem reduces to a problem of controlling the informed party's response. One of the main new difficulties then is predicting how the game will be played.
Satterthwaite have shown that it is impossible in general to attain the full-information efficient outcome when there are more than two possible allocations to choose from and when the contracting parties' domain of preferences is unrestricted that is.
A first basic question of interest then is whether and how the theory of contracting with onesided private information extends to multilateral settings. In recent years there has been an explosion of research in auction theory partly because of its relevance to auction design in a number of important practical cases. Auction design with multiple informed bidders is by no means the only example of contracting with multilateral hidden information.
Chapter 7 devotes considerable space to a discussion of the main ideas and derivation of key results in auction theory. A major economic principle emerging from the analysis of contracting with one-sided hidden information is the trade-off between allocative efficiency and extraction of informational rents. The first result establishes that a number of standard auctions yield the same expected revenue to the seller when bidders are risk neutral and their valuations for the object are independently and identically distributed.
From a theorist's perspective this is a mixed blessing because the proposed efficient contracts or "mechanisms. Covering this research would require a separate book. Another leading example. The second idea refers to the inevitable disappointment of the winner in an auction where bidders value the object in a similar way but have different prior information about its worth: If the bargaining power lies Despite its relative fragility.
It must be stressed. Efficient trade can almost always be achieved if the parties' participation is obtained before they learn their information. Moral Hazard in Teams. Some prominent economic theorists of the firm like.
But note that if the informed party e. A fundamental insight highlighted in Chapter 7 is that the main constraint on efficient trade is not so much eliciting the parties' private information as ensuring their participation. It is for this reason that the leading application of contracting problems involving multisided moral hazard is often seen to be the internal organization of firms and other economic institutions. But labor market policies that try to intensify competitive bidding for jobs or for employees should lower inefficiencies caused by hidden information.
Applying this insight to our labor contracting. This possibility can be seen as one reason why firms like to provide incentives to their employees through promotion schemes. Chapter 8 covers multiagent moral hazard situations with a particular focus on firms and their internal organization.
A key distinction in contracting problems with multisided moral hazard concerns the measure of performance: Is each employee's performance measured separately.
In the former case. Many sports contests are of this form. A and B. They contend that the role of a firm's owner or manager is to mo: The reason why relative performance evaluation improves incentives is that when employees are exposed to the same exogenous shocks affecting their performance changes in demand for their output or quality of input supplies.
To illustrate some of the key insights and findings covered in this chapter. They argue that this is a major deviation from optimal incentive contracting and is evidence of a failure in corporate governance in most large U. PBj depends not only on individual effort 1-lA [resp. They also argue that the employer should be the residual claimant on the firm's revenues and that employees should be paid fixed wages to make sure that the employer has the right incentives to monitor.
Let us now tum to the second case. To see how this works.. Alchian and Demsetz proposed thatfree riding of employees can be prevented through monitoring by the employer. In this case. Then the employer faces a moral-hazard-in-teams problem. As is easy to understand. It is worth noting here that as compelling and plausible as the case for relative performance may be. Chapter 8 discusses extensively how to make the best use of relative performance measures in general problems with multisided moral hazard.
Multiple supervisors are then required. When the number of employees to be monitored is large. One of the reasons why there may be a loss of control as more supervisory tiers are added is that midlevel supervisors may attempt to collude with their employees against top management or the firm's owners. If the number of supervisors is itself large. Chapter 8 gives an extensive treatment of this theory of organizations.
The residual should then be sold to a third party. Chapter 8 provides an extensive discussion of some of the main models of optimal contracting with collusion. It emphasizes in particular the idea that beyond incentive and participation constraints.
Yet the theory we develop in the first two parts of the book deals only. Recent corporate scandals in the United States have painfully reminded investors of the risk of collusion between auditors and the agents they are meant to monitor.
An important insight of Holmstrom These examples vividly draw attention to the importance of considering the possibility of collusion in multiagent contracting situations.
A number of new fundamental economic issues arise w4en the parties are involved in a long-term contractual relation. In Chapter 9 we discuss dynamic adverse selection and in Chapter 10 dynamic moral hazard. How is private information revealed over time?
How is the constrained efficiency of contractual outcomes affected by repeated interactions? How does the possibility of renegotiation limit the efficiency of the overall long-term contract? To what extent can reputation serve as a more informal enforcement vehicle that is an alternative to courts?
We discuss these and other issues extensively in this third part of the book. The second class of problems is conceptually much closer to a static contracting problem.
In Part III we provide systematic coverage of long-term incentive contracting. What this conclusion implies in particular for contracting under hidden information is that the revelation principle still applies under full commitment.
In the first problem the main new conceptual issue to be addressed relates to learning and the gradual reduction of the informed party's informational advantage over time. The main novel economic question in this class of problems concerns the trade-off.
As we have already noted in the context of intertemporal coinsurance contracting problems. In the second period the employer would then know her outside opportunity. And suppose. In this situation the menu of contracts that we have described would no longer be feasible in the first period: But full commitment will not be feasible if the contracting parties are allowed to sign long-term contracts but cannot commit not to renegotiate them in the future if they identify Paretoimproving opportunities.
PH' As we have seen. The contract for the skilled employee would specify an inefficiently low level of employment. To see one important implication of learning of the informed party's type over time.
It was known for example to analysts of the Soviet system as the ratchet effect see Weitzman. Note that this commitment issue is a very general one that arises in many different contexts. Consequently more pooling of types is to be expected in early stages of the contracting relation.
Under full commitment to a comprehensive long-term contract the preceding problems disappear. The optimal long-term incentive-compatible contract would provide more within-period insurance against low-income shocks.
Once the high skill of the employee is revealed. The reason is that in a relation that is repeated twice or more. The starting point of this analysis is that there can be no gains from contracting at all in a one-shot contracting relation because the informed party will always claim to have had a low income realization in order to receive an insurance compensation. Chapter 9 provides an extensive discussion of the dynamics of contracting under adverse selection.
But even in a twice-repeated contracting relation there can be substantial gains from insurance contracting. As we highlight in Chapter 9. In very concrete terms an individual who gets a low-income shock in the first period can borrow against her future income to smooth consumption. While it is stretching our imagination to think that an employee's intrinsic productivity may change randomly over time.
Vice versa. One general lesson emerging from our analysis in this chapter is that there are no gains from enduring relationships when the type of the informed party is fixed.
The key insight of Townsend and the subsequent. But if this renegotiation is anticipated. It also illustrates the relevance of these ideas with several applications. If the employee were free to save any amount of her first-period income at competitive market rates. As with these contracting problems. A second potential positive effect comes from better information about the employee's choice of action obtained from repeated output observations. Chapter 10 begins with a thorough analysis of a general twice-repeated contracting problem with moral hazard and of the general result we just mentioned.
It then proceeds with a detailed discussion of the different effects in play in a repeated relation with moral hazard and identifies two important sources of gains and one important source of losses from an enduring relation. Offsetting these two positive effects.
It is this ability to modulate her effort supply in response to good or bad output changes that drives a striking insight due to Holmstrom and Milgrom concerning the shape of the optimal long-term incen-.
A first positive effect is that repetition of the relation makes the employee less averse to risk. In an enduring relation she can slack off following a good performance run or make up for poor performance in one period by working extra hard the next period. The broad intuition for this general result is that by inducing the employee to consume more in early periods the employer can keep her "hungry" in subsequent periods and thus does not need to raise her level of compensation to maintain the same incentives.
One might then fear that this extreme complexity could easily defeat the practical use of the theory. In short. One would think that when the relation between an employer and employee is enduring. A common practice is to let CEO. This has been viewed as an inefficient practice by some commentators e. In reality many long-. Instead they are sustained by implicit rules and incentives. Loosely speaking. Explicit long-term employment contracts may also take a simple form in practice because in an ongoing employment relation efficiency may be attained by providing a combination of explicit and implicit incentives.
This observation. The explicitly written part of the contract may then appear to be simple because it is supplemented by sophisticated implicit incentives.
This observation may be of particular relevance for evaluating long-term CEO compensation contracts. Reliance on such incomplete explicit contracts may often be a way of economizing on contract-drafting costs. Another important simplification that is available under fairly general conditions is that the incentive effects under an optimal long-term incentive contract may be replicable with a sequence of spot contracts. Chapter 10 provides an extensive treatment of so-called relational contracts.
Holmstrom and Milgrom. There are no contract-drafting costs. This issue is particularly relevant for CEO compensation where the "action" to be taken by the CEO may be the implementation of a new investment project or a new business plan. If this is true for dynamic decision problems. But if such renegotiation is anticipated. The chapter also deals with renegotiation. Our formulation of optimal contracting problems in the first three parts of the book abstracts from all these issues.
Once the project has been undertaken. Most people find it hard to think through even relatively simple dynamic decision problems and prefer to leave many decisions to be settled at a later stage when they become more pressing. This is clearly a drastic albeit convenient simplification. In the fourth and final part of the book we depart from this simple framework and explore the implications of contractual incompleteness.
Once a risk-averse employee has taken her action. This is indeed the prediction of optimal incentive contracting with renegotiation. As with dynamic adverse selection. It is easy to understand intuitively why this is the case. We provide an extensive discussion of this debate in Chapter The question then arises.
The first formal model of an incomplete contracting problem by Simon deals with a fundamental aspect of the employment relation we have not hitherto considered: In a nutshell.
We shall. The form of the incomplete long-term contract will be prespecified exogenously in at least some dimensions. Part IV is concerned with long-term contracts. It has been a matter of debate how much limitations on language are a constraint for drafting fully comprehensive contracts both in theory and in practice. Who makes these decisions? The principal focus of this part of the book will be to address this question. This part of the book also involves a fundamental substantive change: In the first three parts the focus was exclusively on monetary rewards for the provision of incentives.
When contracts are incomplete. But the focus is different. In Chapter 11 we specify exogenously which events the contract cannot be based on and focus on the implications of contractual incompleteness for institution design. The employment contract is preferred when the downloader is higWy uncertain at the time of contracting about which service he prefers and when the seller employee is close to indifferent between the different tasks the employer can choose from.
It is for this reason that Simon. So far we have described an employment contract like any other contract for the provision of an explicit service or "output. Grossman and Hart developed a simple theory and model of ownership rights based on the notion of residual rights of control.
They define a firm as a collection of assets owned by a common owner. In a pathbreaking article that builds on his insights. He can sell. Chapter 11 discusses the strengths and limitations of Simon's theory and provides an extensive treatment of a "modernized" version of his theory that allows for ex ante relation-specific investments and ex post renegotiation. Chapter 12 further builds on Simon's theory by explicitly modeling "orders" or "commands" given by the employer.
The notion that the presence of relation-specific investments creates the need for modes of exchange other than trade in spot markets has been articulated and emphasized forcibly in Williamson's writings A key new notion in their article is that ownership serves as a protection against future holdups by other trading partners and thus may give stronger incentives for ex ante relation-specific investments. Employment contracts thus' specify a different mode of transaction from the negotiation mode prevalent in spot markets.
Simon views the choice between the two modes of transaction as a comparison between two long-term contracts: Grossman and Hart are able to determine when it is optimal to integrate or not.
As Maskin and Tirole a hav. On the one hand. All they can do to improve their future negotiating position is to trade a very standardized contract: They are thus able to articulate for the first time a simple and rigorous theory of the boundaries of the firm.
Chapter 11 provides an extensive treatment of this theory. It exploits the idea that contracts. The owner can thus protect the returns from ex ante relation-specific investments. This agent may therefore invest less.
Grossman and Hart are able to provide a simple formal theory of the costs and benefits of integration and the boundaries of the firm. The advantage of integration is that the bargaining position of the owner of the newly integrated firm is strengthened.
A striking general result of implementation theory is that by designing suitable revelation games for the contracting parties it is often possible to achieve the same outcomes as with fully contingent contracts. This stronger position may induce him to invest more. This chapter begins by covering the theory of Nash implementation Maskin. This theory deals with issues of contract design in situations where an event is difficult to describe ex ante or identify by a third party but easily recognized by all contracting parties ex post.
And yet. Building on these notions. The drawback. Depending on the relative size of these costs and benefits of integration. We discuss the delicate theoretical issues relating to this basic logical tension in Chapter Thus a central focus of this chapter is the characterization of situations where bilateral contracting results in efficient outcomes. As is readily seen. If one assumes that one can contract on who controls these actions. This is not necessarily an abstruse theoretical issue.
Chapter 13 deals with another common form of contractual incompleteness: Chapter 12 provides an extensive discussion of these arguments. An important distinction that is drawn in the literature is whether the bilateral contract is exclusive or nonexclusive-that is.
When incomplete bilateral contracts are written in such multilateral contract settings. Employment contracts. It also explores another foundation for the theory of authority.
The equilibrium outcome of the contracting game may then be inefficient. One way the challenge has been taken up is to argue that Maskin schemes have limited power in improving efficiency over simple incomplete contracts in complex contracting environments where there may be many different states of nature or potentially many different services or goods to be traded see Segal.
In a landmark article Rothschild and Stiglitz have thus shown that if two insurers compete for exclusive bilateral contracts with one or several insurees who have private information about their likelihood of facing an adverse shock or accident. But this is not always the case for other contracts.
Bernheim and Whinston. Chapter 13 discusses this striking result and the vast literature it has spawned. This is a central theme in the common agency literature. Because of the presence of a potential externality. An early focus of the contracting literature has indeed been the potential nonexistence of equilibrium in such contracting games. But it is fair to say that Chapter 13 does not attempt to provide a systematic treatment of the existing literature on bilateral contracting.
Exactly why exclusivity is required for some types of contracts but not others involving externalities has not been fully explored.
Other important topics are touched on besides these two broad themes. Whether exclusivity is enforced or not. Chapter 13 provides an extensive treatment of this important contracting problem. While we have given careful consideration to what material to cover. As the reader may already have noted. In fact. As this introductory chapter makes clear. The book gives an overview of the main conceptual breakthroughs of contract theory.
The goal of this book is therefore to explain how existing contract theory allows one to incorporate these features. Nor do we cover more traditional general equilibrium analysis with moral hazard or adverse selection see Prescott and Townsend.
As contract theory has grown to become a large field. First and foremost. For this reason it may be worth leafing through this chapter even if the reader does not intend to try to solve any of the problems. The second purpose of this chapter is to cover some classic articles that we have not had the space to cover in the main chapters. Chapter 5. This is generally known as the moral hazard problem.
Chapter 3 turns the table and considers optimal contracts offered by the informed party. The first part of this book provides a systematic exposition of the general methods that have been developed to analyze this type of nested optimization problem. The fundamental conceptuai innovation in the contracting problems considered here relative to the classical decision problems in microeconomics textbooks is that one of the contracting parties.
Chapter 4 deals with the contracting problem involving hidden actions. Chapter 2 considers optimal contracts designed by the uninformed party. This involves a signaling problem and is generally referred to as the informed principal problem. The first two chapters are concerned with the general problem of contracting under hidden information. The next chapter Chapter 6 covers richer contracting problems that involve multidimensional private information or hidden actions.
This is generally referred to as the screening problem. Much of the material in these three chapters can also be found in advanced microeconomics textbooks. This chapter deals with a subclass of contracting problems under hidden information. Throughout this section we shall consider the following special and convenient functional form for the downloader's preferences: We do so by looking at the problem of nonlinear pricing by a monopolistic seller who faces a downloader with unknown valuation for his product.
Consider a transaction between a downloader and a seller. We then move on to other applications. T where q is the number of units downloadd. The methods we present will also be helpful in tackling multiage. This extension allows us to stress which results from the two-type case are general and which ones are not.
T is the total amount paid to the seller. This problem was first formally analyzed by Mirrlees In the last part of the chapter we extend the analysis to more than two types. This is only a partial list of economic issues where adverse selection matters. The downloader's preferences are represented by the utility function u q. We first explain how to solve such problems when the agent can be of only two types. For each of them we underline both the economic insights and the specificities from the point of view of contract theory.
We especially emphasize the continuum case. Screening In this chapter we focus on the basic static adverse selection problem. The probability 13 can also be interpreted as the proportion of consumers of type OL.
What is the best. The consumer is of type OL with probability 13 E [0. We treat in this section the case where there are only two types of downloaders: Perfect Price Discrimination To begin with.
The seller will try to maximize his profits subject to inducing the downloader to accept the proposed contract. The solution to this problem will be the contract qi' 7. The answer to this question will depend on the information the seller has on the downloader's preferences. The characteristics 0 are private information to the downloader. The seller can then treat each type of downloader separately and offer her a type-specific contract. The seller knows only the distribution of 0.
Assume the downloader obtains a payoff of u if she does not take the seller's offer. Cqi' The idea that. H From the first-order conditions we can derive the demand functions of each type: Note that it can be implemented by type-specific two-part tariffs. The assumed concavity of v. Given this contract the downloader chooses q to maximize eiv q -Pq. This ceases to be true in the presence of adverse selection.
Note that in a market context. Linear Pricing. We first look at two simple contracts of this kind. The contract set is potentially large. Without adverse selection. This idea also requires that surplus be freely transferable across individuals.
Note that. Note first that for any given price P. Can the seller do better by moving away from linear pricing? He will be able to do so only if downloaders cannot make arbitrage profits by trading in a secondary market: This is. Screening The downloader's net surplus can now be written as follows: Under this interpretation there are no arbitrage opportunities open to the downloader.
We can also observe the following: If Pm. The first-order positive effect dominates. But it has a first-order positive effect on consumer surplus. To see this point. If he decides to set an even higher fixed fee and to price the type-lh downloader out of the market. As the figure also shows Without loss of generality. Optimal Nonlinear Pricing In this subsection we show that the seller can generally do better by offering more general nonlinear prices than a single two-part tariff.
Since the seller does not observe the type of the downloader.. In the process we outline the basic methodology of solving for optimal contracts when the downloader's type is unknown. B H to the 8j["type downloader. These observations naturally raise the question of the form of optimal nonlinear pricing contract.. Step 1: Apply the revelation principle. This problem looks nontrivial to solve. H and The first two constraints are the incentive-compatibility IC constraints.
IRL 'f1? Such adverse selection problems can. Screening [ JR i means that the allocation that downloader of type 8i chooses gives her a nonnegative payoff]. Solve the relaxed problem without the incentive constraint that is satisfied at the first-best optimum.
Step 1 has thus already greatly simplified the problem. Step 2: Observe that the participation constraint of the "high" type will not bind at the optimum. We can now try to eliminate some of these constraints.
Remember that we now look at the problem. Note that the fact that only one incentive constraint will bind at the optimum is driven by the Spence-Mirrlees single-crossing condition. In order to choose which constraint to omit.
In step 3. Step 4: Observe that the two remaining constraints of the relaxed problem will bind at the optimum. It involves efficient consumption and zero rents for both types of downloaders. The strategy now is to relax the problem by deleting one incentive constraint. TH -e. This informational rent increases with QL' The following first-order conditions characterize the unique interior solution Q!. Step 5: Eliminate TL and TH from the maximand using the two binding constraints. And constraint IRL will also bind.
Substituting for the values of T L and THin the seller's objective function. The full text of this article hosted at iucr. Use the link below to share a full-text version of this article with your friends and colleagues. Learn more. Volume 28 , Issue 1. Please check your email for instructions on resetting your password.
If you do not receive an email within 10 minutes, your email address may not be registered, and you may need to create a new Wiley Online Library account. If the address matches an existing account you will receive an email with instructions to retrieve your username. Managerial and Decision Economics Volume 28, Issue 1.
First published: Tools Request permission Export citation Add to favorites Track citation. Share Give access Share full text access. Share full text access. Please review our Terms and Conditions of Use and check box below to share full-text version of article.