Promoting Honesty in Negotiation: an Exercise in Practical Ethics*

In a competit ive and morally imperfect world, business people are often faced with serious ethical challenges. Harboring suspicions about the ethics of others, many feel justif ied in engaging in less-than-ideal conduct to protect their own interests. The most sophisticated moral arguments are unlikely to counteract this behavior. We believe that this morally defensive behavior is responsible, in large part, for much undesirable deception in negotiation. Drawing on recent work in the l iterature of negotiations, we present some practical guidance on how negotiators might build trust, establish common interests, and secure credibil i ty for their statements thereby promoting honesty. We also point out the types of social and institutional arrangements, many of which have become commonplace, that work to promote credibil i ty, trust, and honesty in business dealings. Our approach is offered not only as a specific response to the problem of deception in negotiation, but as one model of how research in business ethics might offer constructive advice to practit ioners. ...there is such a gap between how one lives and how one ought to l ive that anyone who abandons what is done for what ought to be done learns his ruin rather than his preservation... —Niccolo Machiavell i We must make the world honest before we can honestly say to our children that honesty is the best policy. —George Bernard Shaw F business ethics is to have a significant impact on business practice, many of us working in the field will need to take a more pragmatic approach to our craft (Dees and Cramton, 1991). Our work should help ethically sensitive business people establish stable institutional arrangements that promote and protect ethically desirable conduct, and it should help individuals to develop strategies for effective ethical behavior in a competitive and morally imperfect world. This paper is offered as one model of more practical business ethics To illustrate the model we have selected the topic of deception in negotiation. Negotiation is a pervasive feature of business life. Success in business typically requires successful negotiations. It is commonly believed that success in negotiation is enhanced by the skillful use of deceptive tactics, such as bluffing, exaggeration, posturing, stage-setting, and outright misrepresentation. As White (1980, p. 927) candidly states, “The critical difference between those who are successful negotiators and those who are not lies in this capacity both to mislead and not to be misled.” Some shrewd practitioners have advanced the art of deception beyond prudent concealment of preferences to more aggressive forms of strategic misrepresentation. Given the high value placed on honesty, the incentives for deception in negotiation create a serious moral tension for business people, as well as a public relations problem for business. The public relations problem is not new. In ancient Athens, Hermes, a trickster who stole his brother’s (Apollo’s) cattle on the day he was born and later lied about it, was the patron god of merchants. Anacharsis wrote in 600 B.C., “The market is the place set apart where men may deceive one another.” Not surprisingly, deception in negotiation is a widely discussed problem in business ethics. However, much of the attention has been devoted to the question of when (if ever) various deceptive tactics are ethically justified. Many writers have been highly critical of deception in business. Others, most notably Carr (1968) in his controversial piece on business bluffing, have argued that business has its own ethics, one that permits a wide range of deceptive practices that would not be acceptable outside of business. We see little benefit from joining this debate. It is not our intent in this paper, nor was it our intent in Dees and Cramton (1991). Boatright’s suggestion (1992) that our “mutual trust perspective” on practical ethics might best be understood as a two-tiered Hobbesian theory about the justification of deception misses our central point. We were not staking out a philosophical position, but rather attempting to articulate a deepseated view reflected in the attitudes and behaviors of many negotiators with whom we have had experience. The point was to provide a backdrop for developing practical recommendations that negotiators might actually take up. Debating the possible philosophical justifications for different types of deception is intellectually engaging, but unlikely to have a direct impact on practice. How many negotiators would change their behavior when it is pointed out that their views are essentially Hobbesian and that Kant, who is regarded as a superior moral philosopher, would find them objectionable? Even if this claim were accompanied by a philosophical refutation of Hobbes and an explanation of Kant’s categorical imperative, it seems doubtful that much negotiating behavior would change. I BUSINESS ETHICS QUARTERLY 2 We also avoid the standard debate because it focuses too much attention on justification, obscuring the fact that behavior may be justified, but nonetheless regrettable. A dramatic example is the killing of civilians during a war, even a just war. In some instances, it may be justified, but even when justified, it is regrettable. Rather than engaging in (perhaps unresolvable) philosophical debates about the exact conditions under which the killing of civilians is justified, it seems that one who is concerned about the loss of civilian life in war would be better served by putting resources and intellectual energy into developing political, economic, diplomatic, and military strategies that reduce the risk to innocent civilians. Our premise is that, outside of a few recreational contexts, deception is a regrettable feature of business negotiations, even when it is justified (or common-place). Machiavellian gap between what is done and what (ideally) ought to be done is real when it comes to deception in business negotiations. A purely moralistic (or philosophical) response is likely to be ineffective. A Machiavellian response is likely to make things worse. In the spirit of Shaw, we prefer to explore means of constructively narrowing the gap, thereby making the world more honest. Ethics, Opportunism, and Trust In an ideal world, people would do the right thing simply because it is right. In the world in which we live, morality is more complex. People often disagree about what is right. Even when a consensus on moral values is reached, many find that they do not consistently live up to moral standards. One reason for short is that most people place a high value on their own welfare They may have moral ideals and commitments, but concern about personal well-being is a powerful motivating factor. It is more powerful for some than it is for others, but few can claim to be indifferent to it. Any significant gap between the demands of ethics and the urging of self-interest, narrowly defined, creates incentive problems for individuals and for societies wishing to maintain high ethical standards. The problems arise on two levels. At the first level are the direct incentive problems of opportunism and desperation. Problems of opportunism arise when individuals willingly violate ethical norms in order to pursue opportunities for private gain. They yield to temptation. Problems of desperation arise when individuals violate ethical norms to avoid personal loss or hardship. Even if we grant that most people place some intrinsic value on doing the right thing, as they see it, sometimes the risk or the temptation is just too great. Philosophers refer to this problem as “weakness of the will.” Weakness of the will is not limited to moral deviants. Too often we are presented with evidence from our daily lives, from news stories, and from academic research, that well-educated, apparently normal individuals can be tempted or pressured into compromising ethical standards. The effects of opportunism and desperation are magnified by a second-level problem concerning trust and fair play. One of the reasons people are willing to behave ethically, even when their personal welfare is at risk, is that they expect others to behave likewise. It seems unfair for individuals with weaker ethical commitments to prosper materially, especially at the expense of individuals with stronger commitments. An atmosphere of mutual trust appeals to play an important role in grounding ethical behavior for many people. Suspicion that others are profiting from misconduct can destroy that atmosphere and spoil the sense of satisfaction that might be gained from principled behavior. A sense of fair play can motivate individuals with strong ethical commitments to engage in what they would otherwise consider unacceptable behavior. Individual integrity and social stability are difficult to maintain in a social setting in which there is serious conflict between ethics and personal welfare. Traditionally, moral philosophers have responded to this conflict in one of two ways. Some, particularly Kantians, acknowledge the gap between ethics and selfinterest, but assert on philosophical grounds the dominance of moral considerations over those of personal welfare. Others argue that the gap is only apparent. By refining the definition of self-interest, they attempt to reconcile ethics and self-interest (Kavka, 1984). As philosophically interesting as these views are, neither holds much promise of improving conduct. Practical, not conceptual, solutions are needed. Incentive for Deception in Negotiation To illustrate the practical approach to ethical incentive problems, we have chosen to concentrate on the phenomenon of deception in negotiation. Negotiation offers a familiar setting in which individuals often feel a tension between ethics and self-interest. In particular, individuals frequently face a temptation to deceive the other party, in hopes of bettering the outcome for themselves. We adopt the following definitions: A negotiation is any situation in which two or more parties are engaged in communications, the aim of which is agreement on terms affecting an exchange, or a distribution of benefits, burdens, roles, or responsibilities. PROMOTING HONESTY IN NEGOTIATION 3 Deception is any deliberate act or omission by one party taken with the intention of creating or adding support to a false belief in another party. 6 Honesty is the absence of deception. 7 Notice that lying is only one tactic that may be used to deceive a negotiation partner. Lying, strictly interpreted, requires making a false statement (or at least a statement believed to be false by its maker). The clever manipulation of verbal and non-verbal signals to create or support a false impression, without making a false statement, also counts as deception. Likewise, concealing information is a deceptive tactic if and only if the concealment is intended to create or support a false belief. In some cases, there is a fine line between allowing the other party to continue to hold a false belief and adding support to that belief. To understand the incentives for deception in negotiation, we use some basic game theory and a fictional world called Metopia. The simplified world of Metopia allows us to put aside temporarily some of the complexity of the real world, in order to analyze the incentives for deception. Metopia is a world much like ours, but populated exclusively with rational, self-centered individuals. We adopt the standard definition of self-interest from economics. An action is in a party’s self-interest if, given the party’s beliefs at the time of decision, the action yields greater expected utility for the party than any other available action. Metopians always act in their self-interest. Their interests are even more self-centered and material, focused on their own personal welfare. Metopians’ preferences are independent of the preferences of others. They have no specific preferences about the process of the negotiation. In particular, Metopians do not have an independent preference for honesty or dishonesty in the process of negotiation. They feel no guilt about deception, nor do they enjoy fooling others. Metopians have no interest in the opinions of others, except to the extent that such opinions are likely to inhibit or enhance their ability to satisfy their self-centered interests. Finally, Metopians do not have any religious belief system that provides moral rewards and punishments. These features characterize the narrow conception of self-interest in Metopia. Opportunities for Deception in Metopia The basis for deception in Metopia is the presence of (real or perceived) informational differences among the parties. Negotiators often have private information about the item under discussion, about their ability and willingness to take future actions, arid about their own settlement preferences. Private information, however, does not always present a profitable opportunity for deception. Often it is more prudent to be honest. In general, an opportunity for A to profitably deceive B arises only when B believes A has information that is of value to B in determining B’s negotiating position, B knows of no other cost-efficient way to get the information before making a commitment, and it is to A’s advantage if B acts on false beliefs about the matter. The opportunity will depend in part on the kind of information in question. Deception About the Matter Under Negotiation. It is common for at least one of the parties to have privileged access to information about the subject matter under negotiation. In an exchange, often the seller has the information advantage. The classic example is the seller of a used car who knows more about the history and mechanical condition of the car than the buyer. Sometimes, however, the buyer has an advantage over an unsophisticated or uninformed seller. An example of this would be an art dealer buying a dusty old painting at a garage sale, or a real estate developer buying parcels of property for an unannounced development project. In a barter situation, both parties may have informational advantages about their side of the exchange. Information asymmetry is not limited to product exchanges. It is also common when the negotiation is about a service to be delivered, a benefit (or burden) to be distributed, or a right (or responsibility) to be assigned. Even in Metopia, where negotiators recognize the other party’s incentives to deceive, the possibility of deception can cause serious problems. For example, suppose a seller has private information about the quality, broadly interpreted, of the good being sold. The higher the quality, the more the good is worth to the buyer, and consequently the more the buyer is willing to pay. In a one-shot situation the seller has an interest in overstating the product’s quality. If the seller is believed, she can get a higher profit by overstating, assuming it is more costly to produce high quality goods than low quality goods. A Metopian buyer, therefore, will not believe statements that the good is of high quality, even when it is. Thus, the incentive to deceive about quality leads to Akerlof’s (1970) “market for lemons.” The buyer, not believing statements of high quality, is only willing to pay a low quality price. The seller is only willing to produce low quality goods, even if both are better off with the exchange of high quality goods at a high quality price. BUSINESS ETHICS QUARTERLY 4 Deception About Future Actions. Statements about future actions play a key role in many business negotiations. Such statements fall into two categories: threats, and promises. Each of these presents an opportunity for deception about one’s true intentions. Threats are often used to place pressure on the other party to settle. They range from the simple threat of walking away from the negotiation to threats of causing harm to the other party. For instance, in a labor negotiation, management may threaten to close a plant if the union does not make concessions. Threats can be bluffed, but for a bluffed threat to work, it must be credible. Suppose A threatens B. The threat is credible if B thinks that it is in A’s interest to carry out the threat if B does not comply. The success of the threat, then, depends on how well B can assess A’s true incentives. In the above example, if labor does not have information about plant-by-plant profitability, management may get concessions by threatening to close the plant. B may not be able to tell a bluffed threat from a real one. Legitimate threats and warnings may be robbed of their informational content. Even if management offered to present labor with the financial information, labor would recognize management’s incentive to present deceptive information. Without a reliable source of information, B may have to discount all of A’s threats, legitimate or bluffed. Promises are an essential element in most negotiations. Often one patty performs, or makes an investment before the other. Even if the central exchange appears to be simultaneous, there may be understandings and expectations about future actions. Sellers may promise to protect buyers in certain contingencies, for instance, allowing the buyer to return clothing that does not fit. On the other hand, buyers may ask for credit and promise to pay sellers at a later date. In any case, a party making a commitment is likely to know much more about the probability of compliance than the party to whom the commitment is made. For instance, a computer manufacturer who is planning to go out of business, but wants to sell as much of her remaining stock as possible, might continue to offer a warranty that she knows she will not be in a position to honor. Again, lack of information about the true incentives, abilities, and resources of A, the party making the commitment, makes B vulnerable to deception. When A has an incentive to deceive, even A’s good faith promises are not believed. It is difficult for A to credibly reassure B of A’s willingness and ability to fulfill the promise. Deception About Settlement Preferences. The third area of opportunity for deception in Metopia involves the settlement preferences of the negotiating parties. Settlement preferences include the negotiator’s reservation price, time pressures, and the different values the negotiator places on specific terms of agreement. Sometimes it is useful to share these preferences, if the sharing does not make one vulnerable to exploitation by the other. Yet, often, negotiators find it in their interest to misrepresent their preferences. They may want to conceal a weak bargaining position, or to give the appearance of making a concession when they are getting what they want. It is possible for the other party to gather indirect evidence about these matters. For instance, if house buyer, B, discovers that house seller, A, has accepted a job in another part of the country and has made a deposit on a house there, this points to A’s eagerness to sell. If rug seller, A, sells many of the same rugs, rug buyer, B, may be able to get a history of the prices A has accepted in the past as an indicator of how low A is willing to go. Nonetheless, individual negotiators generally have privileged access to information about their own settlement preferences. To see why uncertainty about settlement preferences may lead to deception imagine the negotiation process between two traders, A and B, where each party is uncertain about the strength of the other’s bargaining position. Neither trader knows the other’s values or outside opportunities, or how these values and opportunities change as time passes. The negotiation consists of a sequence of offers and counteroffers until someone finally accepts the other’s offer. Agreement occurs at a point where the further benefits of bargaining (improved terms of trade) do not seem worth the costs to secure them (delayed agreement or a risk of disagreement). The parties bargain until the marginal cost of continuing exceeds the marginal benefit. B’s net gain of continuing is based on her expectation about A’s future concessions, which, in turn, is based on B’s belief about (he strength of A’s bargaining position. To do well, A would like to persuade B to accept A’s most recent offer by convincing B that only slight concessions will be made in the future. A natural justification for slight concessions in the future is a strong bargaining position—good outside opportunities arid a lack of eagerness to reach agreement. Hence, A benefits from convincing B that her bargaining position is stronger than it truly is. A has an incentive to deceive B about any facts, beliefs, or values that might expose the weakness in A’s position. For this reason, bargainers in Metopia are skeptical of information that the other party presents that bears on that party’s settlement preferences. This has its costs. To convince B of the strength of her position, A must take actions that a weaker A would not want to imitate, such as delaying agreement or risking disagreement, both of which involve bargaining inefficiencies. Indeed, so long as there is some uncertainty about whether gains from trade exist, some bargaining inefficiencies must occur regardless of the bargaining process adopted, if the parties act in their own self-interest and have only costly means of signaling strength. PROMOTING HONESTY IN NEGOTIATION 5 Factors Inhibiting Deception in Metopia One might think that Metopia would be riddled with attempted deception and bad faith, since Metopians exploit opportunities to misrepresent information in a negotiation. However, as Adam Smith observed, “The most notorious liar...tells the fair truth at least twenty times for once that he seriously and deliberately lies” (1759, p. 530). The same might be said of Metopians. Several mechanisms work to inhibit deception. Metopian’s would rationally invest in processes and mechanisms to protect themselves from deception and its untoward social consequences. Even potential deceivers recognize the effect that the possibility of deception has in undermining even their true statements. To create a climate of confidence that would facilitate negotiations, Metopians would investigate claims, construct contractual mechanisms to enforce honesty, and work to ease the availability of reputation information. Ex Ante Verification. One remedy for the risk of deception is to verify claims and assumptions before making commitments that depend on those claims or assumptions. Negotiators could gather independent evidence themselves, or they might hire others to do it. Gathering it themselves is, generally, the most reliable method. However, an individual negotiator does not always have the expertise to verify claims efficiently. Accordingly, just as the use of house inspectors has become commonplace in residential real estate transactions, Metopians would employ auditors, testers, inspectors, and private detectives to verify the truth of claims made or implied in negotiations. One problem is that verification of every significant claim made in a negotiation would be quite costly. Often, it would not be justified. Metopians would be creative in reducing the costs of verification. Two mechanisms for doing so are the use of economies of scale and random verification. It would be highly inefficient, for instance, for everyone buying a refrigerator to hire a private inspector to evaluate different models. However, because refrigerators are standard products, we would expect to see information gathered by a third party who would sell it to those who need it. When economies of scale are not possible, Metopians could adopt random verification procedures to lower verification costs. This would work so long as the parties caught in a deceptive act could be punished harshly. For example, if the maximum penalty that A could apply to B is ten times B’s individual net gain from deceiving A, then A’s verifying information one out of ten times is sufficient to prevent deception by B. Verification is economical in this case, so long as the cost of verification is no more than ten times the net efficiency gain from honest behavior. Contractual Mechanisms. Unfortunately, even with some creativity, ex ante verification is limited. Some claims are simply too difficult, or too costly to test ex ante with any confidence. Claims about the long-term reliability of a new product typically fall into this category, as do many claims about actions to be taken in the future and about settlement preferences. In such cases, Metopians would develop contractual mechanisms to add credibility to claims made or implied in a negotiation. Ideally, the mechanisms would be self-enforcing. However, such mechanisms are limited in their applicability, and may be more costly than alternatives, even though the alternatives require third-party enforcement. Consequently, we would expect to find in Metopia systems of third-party enforcement and a high level of explicit contracting. The third-parties could play a variety of roles from holding collateral to adjudicating disputes and exacting settlements. Metopia would benefit from an analogue to our court system, with coercive authority. Although complete contracts in many negotiations would be impossible to write, because of unforeseen contingencies, Metopians vulnerable to deception would press for clarity on important matters. Explicit claims are more easily verified to third-parties. Ambiguity and vagueness make enforcement problematic. 1. Warranties are one type of contractual device that would be used in Metopia to handle matters that ate not subject to ex ante verification. Warranties can be used to add credibility to claims about the item under negotiation that cannot be reliably or cost-effectively assessed until after the deal is struck. “Defective products will be replaced or repaired at the manufacturer’s expense.” They also can be used to make explicit and reinforce promises of future performance. “If we ever fail to plow the snow from your driveway before 7 am, your full year’s service contract is free.” They can even provide some assurance regarding settlement preferences. “If we sell this item to anyone at a lower price within six months, or if you can find another dealer who will beat our price on the item, we will pay you double the difference.” Metopians would use warranties to shift the risks inherent in a negotiation from the recipient of the warranty to the provider of the warranty. It is not easy to write an enforceable and cost-effective warranty in Metopia. To be enforceable, it needs to clearly specify the conditions under which it applies, the remedies available to the recipient, and the process for bringing the remedies about. Otherwise, it invites costly and hard-to-resolve disputes between selfinterested Metopians. Cost-effectiveness depends on the nature of the warranty’s specific provisions, the BUSINESS ETHICS QUARTERLY 6 ability and the incentives for the provider to make good, and tendencies for opportunistic behavior on the part of the recipient. For instance, a manufacturer’s warranty that allows for replacement or repair only when a manufacturing defect is detected may not be very effective in cases where it is difficult to determine whether a discovered defect is due to poor manufacture, or misuse of the product by the buyer. A warranty that requires that the product be returned in person to a remote regional service center for lengthy evaluation and repair could well make the transaction costs involved in seeking remedy under the warranty so high as to make its application a non-issue. Likewise wise, a product service contract offered by a company with limited financial resources and a significantly understaffed service department provides little value to the buyer. On the other hand, an unconditional warranty of consumer satisfaction that is easy to invoke and offers generous remedies (e.g., satisfaction guaranteed, or all your money back, no questions asked) is likely to create opportunistic behavior on the part of self-interested Metopian consumers. They will buy products, get some valuable use out of them, and then make warranty claims. Despite these difficulties, Metopians would work hard to construct enforceable and cost-effective warranties to secure their transactions, provide credibility to their claims, and reduce incentives for opportunism. Nonetheless, at its heart, a warranty is just another promise. Its credibility, in a Metopian world, rests on the availability of other enforcement mechanisms and on reputation effects (to be discussed shortly). 2. Bonds. Promises, including warranties, often need further reinforcement to be credible. Even if the promisor makes the promise in good faith, incentives can change before the promise is carried out. How can the promisee be reasonably confident of performance? Another way to provide credibility to claims about future action is to put in place mechanisms that will reinforce the promisor’s incentives for good faith ex ante and compliance ex post. This can be done by posting a bond. Posting a bond involves doing something at the time the deal is struck that allows the other party, or a third-party to either exact a penalty from a negligent promisor, or compensate the harmed promisee, if the promise is not kept. The most common form of bonding mechanism is collateral. The promisor gives the promisee, in effect, a contingent claim on some of the promisor’s assets. If the promise is not fulfilled, the assets revert to the promisee. In some cases, the promisee will hold the assets until the promise is fulfilled, but this need not be the case if a credible third-party can enforce the asset transfer when nonperformance is verified. What if the promisor does not have assets to compensate the promisee for nonperformance? One way of handling this would be through insurance schemes. An insurer would evaluate promisors and charge a premium based on the insurance company’s assessment of the probabilities of default. The insurer would pool premiums from a number of promisors to provide enough funds for paying expected claims. Such a scheme would have to be constructed carefully to avoid the problems of adverse selection and moral hazard. Another alternative for dealing with the problem of limited assets is the use of hostages. The agreement can be reinforced as long as the promisor has assets that she values as much as noncompliance with the promise. Hostages are items of value to the promisor that may not have value to the promisee. The incentive for compliance is the threat to destroy or transfer the hostages. Creating the right bonding agreement is also difficult in Metopia. A bond that requires too little from the promisor will result in nonperformance. A bond that provides too much value to the promisee may lead to a false claim of nonperformance, especially if the promisee is holding the collateral. Even with independent parties adjudicating disputes, it will be difficult to get the incentives just right for all three parties. Much rests on the reputation of the third party and on the clarity of conditions for nonperformance. Nonetheless, we would expect to find Metopians constructing a wide variety of bonding arrangements to off-set the problems raised by incentives to deceive. Contractual mechanisms, such as warranties and bonds, have limited application. In some cases, it will be impossible for Metopians to contract on relevant future actions. One reason it may not be possible is that one party’s action may be unobservable to the other. Hence, the party may make false claims about the unobservable action. For example, a consultant may be hired to complete a project that has uncertain time requirements, promising to work hard to complete the project in the shortest possible time. If the consultant’s effort is unobservable to the client, the consultant may have an incentive to renege on the promise of concerted effort. Likewise, a group sharing a common resource may agree to use the resource efficiently, but if the group cannot monitor individual use of the resource, individuals may overuse it. The difficulty of verifying compliance may make it impossible to write a contract that induces the efficient level of effort (Holmstrom, 1979). Even if it is possible to write a complete contract, it may prove too costly to enforce the contract, because the actions—while observable to the deceived party—may not be easily verified to the third-party attempting to enforce the contract. Furthermore, the effective application of contractual mechanisms often requires an independent third-party with the right incentives. Finally, these mechanisms can be costly. They can complicate contract writing, tie up assets that could be profitably used elsewhere, and PROMOTING HONESTY IN NEGOTIATION 7 divert value to third parties who must be paid. If not constructed with great care, these contractual mechanisms can create new incentives for deceptive behavior. Reputation Effects. To supplement ex ante verification and contractual mechanisms, Metopians would need to rely on reputations effects to induce honesty. A person’s reputation affects future opportunities by influencing the other’s belief about what the person will do in the future. A negotiator with a reputation for being deceitful is likely to be disadvantaged in future negotiations. She may have a hard time finding negotiating partners and when she finds them, they will be on their guard. This is not because Metopians are morally offended by deceit. It is simply that negotiations tend to go better, other things being equal, when one is dealing with a person who has a reputation for honesty to maintain. Thus, when deciding to deceive, a Metopian negotiator must consider not only the short-run consequence of the decision, but also how the decision affects her reputation and hence future negotiations. Adam Smith (1759, p. 350) was aware of the power of reputation, commenting, “The prudent man is always sincere, and feels horror at the very thought of exposing himself to the disgrace which attends upon the detection of falsehood.” Smith overstates the case. An intelligent Metopian will know that some attempts at deception pose greater risks to one’s reputation than others. Reputation works best when claims are explicit, ex post detection is likely, and information about deceptive practices can be credibly communicated to the culprit’s future negotiating partners. 1. Reputations in On-Going Negotiations Between Two Parties. The setting in which reputation is most likely to be effective is that of two parties, A and B, who negotiate repeatedly over time. Suppose both parties have an incentive to deceive if a particular negotiation is considered in isolation. The incentive problem in an isolated negotiation is captured by the following Prisoner’s Dilemma game: