Beyond Winning

by Robert H Mnookin

  Different Forecasts: Parties may have different beliefs about what the future will hold. In the entertainment industry, for example, performers, agents, and concert halls often have different predictions about the likelihood of various attendance levels. Performers are often convinced of their ability to draw huge crowds, while concert halls may be much less sanguine. By trading on these different forecasts—perhaps through contingent fee arrangements—the parties can resolve these differences to mutual advantage. A singer who expects to draw a standing-room-only crowd might agree to a guaranteed fee based on 80 percent attendance, plus a percentage of any profits earned from higher attendance. Such arrangements allow the parties to place bets on their different beliefs about the future.

  Different Risk Preferences: Even if the parties have identical forecasts about a particular event, they might not be equally risk-tolerant with regard to that event. My life insurance company and I might have similar expectations about what the odds are that someone my age will die within the next year. But we will probably have very different risk preferences regarding that possibility. I will be risk-averse, knowing that my family will face financial hardship if I die. Therefore, I might pay the insurance company to absorb that risk. The insurance company, by pooling my risk with the risk of others, can offer me insurance based on costs averaged over the entire pool. In effect, I have shifted the risk of my early demise to the more efficient risk carrier—the insurance company. Negotiators often create value in this way. A car buyer might purchase an extended warranty, or a start-up company might sell shares to a wealthy investor in exchange for needed capital. In each case, by allocating risk to the more risk-tolerant party for an acceptable price, the parties create a more beneficial agreement.

  Different Time Preferences: Negotiators often value issues of timing differently—when an event will occur or a payment will be made. For example, a law school graduate and his wife fell in love with a condominium in Washington, D.C. Because he was going to be clerking for a federal judge for two years, his salary during that time was not sufficient to cover the mortgage payments. After the clerkship, however, he knew that he would be joining a large D.C. law firm, at more than twice his clerkship salary. He could then easily afford the house. The solution lay in structuring a mortgage schedule so that there were small payments for the first two years—less than even the interest costs—and larger payments thereafter. Although he had to pay a premium for agreeing to this tiered payment schedule, in the meantime he was able to “afford” his dream home.

  Similarly, Jim and Sara might have different preferences about when Jim moves into the apartment. Although a standard lease would begin on the first of the month, Jim may need to move in earlier. If it is worth more to Jim to move in early than it costs Sara to move out early, they may agree to accommodate Jim’s schedule in exchange for compensation to Sara.

  These five types of differences—in resources, relative valuations, forecasts, risk preferences, and time preferences—are all potential sources of value creation. They all support the same basic principle: trades can create value.

  NONCOMPETITIVE SIMILARITIES

  In some instances, parties have similar interests that truly do not compete, in that one person’s gain does not mean the other’s loss. For example, negotiators often have a shared interest in a productive, cordial working relationship. To the extent that they can improve their relationship, both gain. Likewise, parents generally share an interest in the well-being of their children. If a child flourishes, both parents derive satisfaction. Thus, even for divorcing parents, arrangements that benefit a child create joint gains for both adults.

  Jim and Sara may share several interests that do not compete. For example, they may both hope that Jim gets along with the downstairs neighbors. Sara may value them as friends and neighboring property owners; Jim may simply believe that getting along well with them will make his year in Sara’s apartment more enjoyable. If Jim and Sara identify this shared interest, they might arrange for Sara to introduce Jim to the neighbors before she moves.

  ECONOMIES OF SCALE AND SCOPE

  Economies of scale—in either production or consumption—can also create value. For example, two firms that each have a small plant may be able to reduce the unit cost of production by having a joint venture that builds one large production facility. Or a group of friends who share the same commute can organize a car pool to save money on gas and tolls. Families are perhaps the most natural beneficiaries of economies of scale; they share food, shelter, a car, and a television set, which lowers the cost per member of such basic living expenses. Jim and Sara have also identified a potential economy of scale: sharing Jim’s storage unit, which will reduce storage costs for each of them by exploiting Jim’s excess capacity. Creating or preserving such scale economies is a rich source of value creation.

  Economies of scope can also create value. These arise when more than one good or service can be produced using the same basic resources, thus reducing the cost of each. A restaurant supplier who is selling and delivering fresh vegetables may be able to offer fresh fruits at very little additional cost. A law firm that’s handling a client’s corporate work may be able to more effectively offer legal advice concerning employment law because the firm may already know a great deal about the client’s business and its practices.

  THE PROBLEM: DISTRIBUTIVE ISSUES AND STRATEGIC OPPORTUNISM

  Why don’t negotiators just share all their information, search for value-creating trades, and both walk away happy? The answer is that as negotiators share information in order to attempt to create value, they increase the risk of being exploited. A negotiator who freely discloses information about her interests and preferences may not be met with equal candor from the other side. Herein lies the core of our first tension: without sharing information it is difficult to create value, but when disclosure is one-sided, the disclosing party risks being taken advantage of.

  Two classic stories from the negotiation literature capture this dilemma. The first story concerns two siblings who had what they perceived as a purely distributive dispute over how to divide an orange.3 Each claimed the right to the entire orange, and after much haggling they decided to compromise and cut the orange in two. Each went her separate way with half an orange. One ate the fruit of her half and threw the peel in the trash. The other went home to the kitchen, peeled her half of the orange, used the peel to flavor a cake, and tossed the juicy pulp in the garbage. The point of this story is that when negotiators focus myopically on distributive issues and don’t share any information, they may squander a lot of value.

  The second story involves Nancy and Bob.4 Nancy has ten oranges and no apples. Bob has ten apples and no oranges. Apples and oranges are otherwise unavailable to either. Bob loves oranges and doesn’t much like apples. Nancy likes them both equally well. Bob suggests to Nancy that they both might gain from trading. Before the bargaining begins, neither knows the preferences of the other. If Bob discloses to Nancy that he loves oranges and hates apples, Nancy might exploit him. She might say that she has the same preferences as Bob, which would be a lie. Or she might simply propose that Bob give her nine of his apples in exchange for one of her oranges. In either case, she knows that Bob would probably prefer having just one orange to ten apples. This story illustrates that the disclosure of preferences—particularly if unreciprocated—can invite exploitation with respect to the distributive aspects of bargaining.

  Lurking distributive issues may inhibit the disclosure needed to find value-creating trades. For example, Sara might initially have been reluctant to volunteer that she was going to have to spend $1,200 to store her furniture for the year she was going to be in Paris, because Jim might exploit her need by pretending that he didn’t really want her furniture around but would tolerate it if she insisted on leaving it behind. Conversely, when Jim disclosed that he needed her furniture, Sara might have tried to extract a more substantial premium for a partially furnished apartment. Jim might rent
the apartment unfurnished because they never discover the option that could make them both better off. More fundamentally, as we will see, concern about distributive issues may lead to no deal whatsoever.

  Distributing Value

  For many, distributing value—as opposed to creating it—is the essence of negotiating. Consider the negotiation between Sara and Jim. Rent is a key term in their agreement. Every extra dollar of rent represents a dollar more for Sara and a dollar less for Jim. If the monthly rent were the only term under discussion, their negotiation would be almost purely distributive. But because they are willing to explore a deal involving other elements as well, their negotiation has value-creating potential. Sara is willing to lend Jim some furniture, for a price. Jim might be willing to share his storage space, if he gets some credit for it. Of course, no matter how much value is created, at some point they will still have to divide the larger pie and price the deal by setting the rent.

  To explore the distributive aspects of bargaining, consider a more straightforward negotiation where the key element is simply the price of a single item. Imagine that Sara says: “By the way, you don’t need a car, do you? I’m selling my 1992 Honda Accord.” As it happens, Jim recently changed jobs and does need a car. Their negotiation turns from the apartment to the Honda. What will this negotiation be about?

  Begin by considering Sara’s situation. She received the car as a graduation gift from her parents. The eight-year-old car now has 58,000 miles on it. Sara has taken the car to three used car dealers to see what she can get. The local Honda dealer offered her the best price: $6,900. But Sara is starting to get nervous. She is leaving for France in six days. One way or another, she has to do something with the car before she leaves. She knows that the Honda dealer would sell the car for about $9,800, and she has advertised the car in the local newspaper for $9,495. She tells Jim that this is her asking price.

  Box 1

  Jim needs a car to get to work. He once owned a Honda Accord, so he likes them and is confident of their reliability. He takes Sara’s car for a test drive and does a little research. Based on the age and condition of Sara’s car, he estimates that a dealer would charge about $10,000 for it. He has already visited several dealers and has found only two other used Hondas for sale: a 1994 with lower mileage than Sara’s, for which the dealer’s firm price is $11,500, and a 1990 with much higher mileage, which Jim could buy for $6,500. Faced with these alternatives, Jim would much prefer to buy Sara’s car than the 1990, even if it costs him more.

  Should we expect Sara and Jim to make a deal? To explore this question and unpack the distributive issues involved, let’s consider the alternatives available to each party. Alternatives are those things that Sara or Jim might do if they don’t reach agreement. Sara has a number of alternatives: she can sell the car to a dealer; wait and see if another buyer comes along; lend the car to a friend; donate it to a charity; or take it with her to France. She can do all of these things without Jim’s agreement. Jim, too, has alternatives: he can buy either of the used cars at the dealership, or he can investigate the ads in the local paper.

  Our colleagues Roger Fisher, Bill Ury, and Bruce Patton have coined a phrase to denote a negotiator’s best course of action away from the table: the Best Alternative to a Negotiated Agreement, or BATNA.5 Which alternative would Sara choose if she makes no deal with Jim? Sara decides that her best alternative to a negotiated agreement with Jim is to continue trying to sell the car to another private party for a few more days, and, failing that, to sell it to the dealer for $6,900.

  Knowing her BATNA is not enough, however. Sara needs to translate it into a reservation value, which is the minimum amount she would accept from Jim rather than pursue her BATNA. Suppose Sara is mildly optimistic that in the next six days she will find another buyer who would pay more than the $6,900 offered by the dealer. In this case, she might set a reservation value of $7,000. This is the lowest price she would accept from Jim rather than take another course of action. (Sara’s reservation value could also be lower than the cash value of her BATNA. If she doesn’t want to go to the trouble of seeking out other buyers or taking the car to the dealer, she might decide that her reservation value with Jim is $6,800.)

  What is Jim’s best alternative if he doesn’t buy Sara’s car? He will buy the 1994 Honda for around $11,500. Does this mean that he is willing to pay $11,500 for Sara’s 1992 Honda? No; it is an older model with more mileage. To determine Jim’s reservation value, we need to know the highest amount he would pay Sara and still prefer buying Sara’s car to pursuing his BATNA. Suppose this amount is $9,000. If Jim can get Sara’s car for $9,000 or less, he’d rather buy it than the 1994 model. Otherwise, he’d prefer the newer car.

  Given these assumptions, Sara and Jim could make a deal somewhere between $7,000 and $9,000, and both parties would be better off than with no deal at all. This is the Zone of Possible Agreement, or ZOPA (Figure 2), and in this simple transaction we might expect the parties to settle somewhere in this range.

  At stake in this negotiation is a surplus of $2,000, which must somehow be divided. If Jim pays $8,900, Sara captures most of the surplus. If Jim pays $7,100, Jim gets most of the surplus. If they decided to split the difference between what a dealer would pay Sarah ($6,900) and what Jim would have to pay a dealer ($9,800), the price would be $8,350. Or, if they truthfully disclosed their reservation values and split that difference, the price would be $8,000.

  Figure 2

  Because of the distributive issue, however, Jim and Sara might not reach a deal at all. Neither of them knows that a ZOPA even exists. Sara’s asking price of $9,495 is higher than Jim’s reservation value of $9,000. Although it would be efficient for Sara and Jim to reach an agreement at any price between $7,000 and $9,000, they might fail to do so. Two factors help explain this conundrum: information asymmetries and strategic behavior.

  Information Asymmetries

  In most negotiations, each party has at least some material information that the other party doesn’t have. Such information asymmetries exist here. Sara knows nothing about Jim’s job or the fact that he is under time pressure to buy a car. Nor does she suspect that Jim has a fondness for Hondas. Nor does she know that Jim, having now set his heart on a Honda, has little choice but to spend $11,500 for a newer model than hers.

  Jim does not know that Sara must sell the car in the next six days. Nor does he know that if necessary Sara is prepared to sell the car to the dealer for $6,900.

  The condition or quality of the goods to be traded raises another potential information asymmetry. A seller typically knows far more about the quality of what is being sold than the buyer. This is true whether it is a car or a corporation being sold. Sara is in a better position than Jim to know the condition of her car. Some defects may be obvious, such as a dented fender, but other latent problems are not readily apparent. A mechanic may have told Sara to expect the transmission to need replacement within the next few months, for example. Jim knows that sellers often exaggerate the quality of what is being sold and fail to disclose latent defects. Even if Sara states truthfully that to the best of her knowledge the car is in great shape (and refuses to drop her price), Jim might be quite suspicious if he can’t verify her claim. A skeptical buyer has little way of knowing whether a stranger is an honest seller. Ironically, the more successful a buyer is at negotiating a bargain price, the more suspicious he should be that he’s being sold a lemon.

  Strategic Opportunism

  The desire for distributive gain may not simply inhibit value creation—it can also lead to other sorts of negotiation failures. Parties may not reach an agreement at all even though both might benefit. And even if agreement is reached, they may unnecessarily waste a lot of time and resources playing hardball. The strategic problem is that neither negotiator knows how far it might be possible to push the other side.

  Negotiators rarely honestly reveal their reservation value, and are often reluctant to talk about their BATNAs. Thus,
Jim is unlikely to know that Sara will accept anything above $7,000, and Sara is unlikely to learn that Jim will pay up to $9,000. Moreover, it may be difficult for either of them to obtain and confirm such information independently. If Jim were well prepared, he might have consulted the “Blue Book,” which lists approximate retail and wholesale prices for used cars, but this would have given him only an estimate of what a dealer would pay for Sara’s car. He still wouldn’t know how long Sara was prepared to search for a buyer who would pay substantially more.

  Consider the strategic difficulties that Sara and Jim face. With respect to the distributive dimension of bargaining, each negotiator is trying to assess two things. First, what’s the best agreement that I can reasonably hope to get? Second, can we make a deal here at all? (In other words, is the other side willing to accept an agreement that is at least minimally acceptable to me?) If Jim only cared about finding the answer to the second question, he might simply offer Sara $9,000—saying that this is the most he’s willing to pay. Sara, for her part, can’t be sure that Jim might not be willing to pay more. Should she hold firm at her initial asking price of $9,495—which is still less than what a dealer would charge Jim? Sara might not believe him and might counter for $9,200 or more. In all events, by making this his initial offer Jim has given up any opportunity to explore whether he might make a deal that’s better than minimally acceptable. On the other hand, if Jim pushes too hard for distributive gain by firmly making a lowball initial offer, Sara may conclude that it’s not worth her time to negotiate further. The parties may never make a deal, even if there is a zone of possible agreement.