I am grateful to Peter Haas, David Lax, John Odell, Howard Raiffa, Oran Young, and three referees for their insights and suggestions. I also acknowledge the support of the Charles Stewart Mott Foundation to the Harvard Negotiation Roundtable.

1. This approach is analyzed with particular clarity by Keohane, Robert O. in the following works: After Hegemony: Cooperation and Discord in the World Political Economy (Princeton, N.J.: Princeton University Press, 1984)Google Scholar; and “International Institutions: Two Approaches,” International Studies Quarterly 32 (12 1988), pp. 379–96.CrossRefGoogle Scholar

2. For a recent example of this argument, see Krasner, Stephen D., “Global Communications and National Power: Life on the Pareto Frontier,” World Politics 43 (04 1991), pp. 336–66.CrossRefGoogle Scholar

3. See the following works by Haas, Peter M.:“Do Regimes Matter? Epistemic Communities and Mediterranean Pollution Control,” International Organization 43 (Summer 1989), pp. 377–403CrossRefGoogle Scholar; and “Introduction: Epistemic Communities and International Policy Coordination,” in this issue of 10.

4. The negotiation-analytic approach has its origins in the work of Luce, Duncan, Raiffa, Howard, and Schelling, Thomas. In Games and Decisions (New York: Wiley, 1957)Google Scholar, Luce and Raiffa offered a brilliant synthesis and critique of game theory and began to raise questions about the limits of the game-theoretic approach in analyzing actual interactive situations. Perhaps the first work that could be said to be in the spirit of negotiation analysis was Schelling's The Strategy of Conflict (Cambridge, Mass.: Harvard University Press, 1960)Google ScholarPubMed, followed by his Arms and Influence (New Haven, Conn.: Yale University Press, 1966)Google ScholarPubMed. These works proceeded with less formal arguments and offered a broader direct scope of analysis than did conventional game theory. Nominally in the realm of collective bargaining, Walton, Richard and McKersie's, RobertA Behavioral Theory of Labor Negotiations (New York: McGraw-Hill, 1965)Google Scholar drew more broadly useful distinctions between distributive and integrative bargaining and highlighted the importance of intraorganizational bargaining. The first overall synthesis of the emerging field of negotiation analysis was Raiffa's, The Art and Science of Negotiation (Cambridge, Mass.: Harvard University Press, 1982)Google Scholar. An application and elaboration of some of these ideas appeared in my book Negotiating the Law of the Sea: Lessons in the Art and Science of Reaching Agreement (Cambridge, Mass.: Harvard University Press, 1984)Google Scholar, and Lax, David A. and I systematized the approach in the first part of The Manager as Negotiator (New York: Free Press, 1986)Google Scholar. Young's, H. Peyton edited volume entitled Negotiation Analysis (Ann Arbor: University of Michigan Press, 1991)Google Scholar furthers the evolving tradition. For examples of works in the general spirit of the negotiation-analytic approach, see Snyder, Glenn H. and Diesing, Paul, Conflict Among Nations: Bargaining, Decision Making, and System Structure in International Crises (Princeton, N.J.: Princeton University Press, 1977)Google Scholar; Mesquita, Bruce Bueno de, Rabushka, Alvin, and Newman, David, Forecasting Political Events: The Future of Hong Kong (New Haven, Conn.: Yale University Press, 1985)Google Scholar; and Putnam, Robert, “Diplomacy and Domestic Politics: The Logic of Two-Level Games,” International Organization 42 (Autumn 1988), pp. 427–60.CrossRefGoogle Scholar

5. See Haas, “Do Regimes Matter?”; and Haas, “Introduction.”

6. See, in particular, the following articles in this issue of IO: Peter M. Haas, “Banning Chlorofluorocarbons: Epistemic Community Efforts to Protect Stratospheric Ozone”; Emanuel Adler, “The Emergence of Cooperation: National Epistemic Communities and the International Evolution of the Idea of Nuclear Arms Control”; M. J. Peterson, “Whalers, Cetologists, Environmentalists, and the International Management of Whaling”; and G. John Ikenberry, “A World Economy Restored: Expert Consensus and the Anglo-American Postwar Settlement.”

7. See Oye, Kenneth A., “Explaining Cooperation Under Anarchy: Hypotheses and Strategies,” World Politics 38 (10 1985), pp. 1–24CrossRefGoogle Scholar; and Wagner, R. Harrison, “The Theory of Games and the Problem of International Cooperation,” American Political Science Review 77 (06 1983), pp. 330–46.CrossRefGoogle Scholar

8. Keohane, , After Hegemony, p. 51.Google Scholar

9. Keohane, . “International lnstitutions,” p. 381.Google Scholar

10. Jervis, Robert, “Realism, Game Theory, and Cooperation,” World Politics 40 (04 1988), pp. 329–32.CrossRefGoogle Scholar

11. See Krasner, , “Global Communications and National Power,” p. 342Google Scholar. For discussions of coordination games with distributional consequences, see Stein, Arthur A., “Coordination and Collaboration: Regimes in an Anarchic World,” International Organization 36 (Spring 1982), pp. 299–324CrossRefGoogle Scholar; Snidal, Duncan, “Coordination Versus Prisoners' Dilemma: Implications for International Cooperation and Regimes,” American Political Science Review 79 (12 1985), pp. 931–34;CrossRefGoogle ScholarJervis, , “Realism, Game Theory, and Cooperation,” p. 318;Google Scholar and Young, Oran R., “The Politics of International Regime Formation: Managing Natural Resources and the Environment,” International Organization 43 (Summer 1989), p. 359.CrossRefGoogle Scholar

12. This example is inspired by Fisher, Roger and Ury's, WilliamGetting to Yes (Boston: Houghton-Mifflin, 1981).Google Scholar

13. For a detailed analysis of an analogous case, see Sebenius, , Negotiating the Law of the Sea.Google Scholar

14. Keohane, . “International lnstitutions,” p. 380.Google Scholar

15. Young, , “The Politics of International Regime Formation,” p. 359.Google Scholar

16. See Walton, and McKersie, , A Behavioral Theory of Labor Negotiations.Google Scholar

17. Raiffa, , The Art and Science of Negotiation, p. 144.Google Scholar

18. Krasner, , “Global Communications and National Power,” p. 336.Google Scholar

19. Ibid., p. 342.

20. For a good list of references to “mixed-motive” bargaining, see Keohane, Robert O. and Nye, Joseph S., Power and Interdependence, 2d ed. (Glenview, lll.: Scott Foresman, 1989), p. 10;Google Scholar and Keohane, , After Hegemony, pp. 67–68.Google Scholar

21. This is not universally the case, however. For example, in After Hegemony, p. 53Google Scholar, Keohane states the following: “[Cooperation] is typically mixed with conflict and reflects partially successful efforts to overcome conflict, real or potential. Cooperation takes place only in situations in which the actors perceive that their policies are actually or potentially in conflict, not where there is harmony. Cooperation should not be viewed as the absence of conflict, but rather as a reaction to conflict or potential conflict. Without the spectre of conflict there is no need to cooperate.” Keohane's point—that agreements can be influenced by the prospect of a conflictual no-agreement alternative—is correct yet too limited. It still suggests a sharp divide between conflict and cooperation. Reality is more complex; for example, within the realm of cooperative agreements, there are generally conflicts of interest that can matter greatly to the outcome. Suppose agreements A and B are both Pareto-superior to agreement C (which is mutually preferable to discord) but that I prefer A and B and you prefer the reverse. I may make a binding conditional commitment to agreement C (still, “cooperation”) unless you agree with A.

22. Krasner, , “Global Communications and National Power,” pp. 361–62.Google Scholar

23. Lax, and Sebenius, , The Manager as Negotiator, p. 30.Google Scholar

24. For a discussion of the common inefficiencies in quasi-constant-sum bargains with complete information, see Schelling, , The Strategy of ConflictGoogle Scholar. Apart from the suboptimality resulting from prisoners' dilemma, the collective action problem, and similar structures, several authors have shown that the negotiators' rational self-interested behavior when bargaining with incomplete information can lead to ex post Pareto-inefficient equilibria. See, for example, Chatterjee, Kalyan. “Incentive Compatibility in Bargaining Under Uncertainty,” Quarterly Journal of Economics 97 (11 1982), pp. 717–26;CrossRefGoogle ScholarChatterjee, Kalyan, “Disagreement in Bargaining: Models with Incomplete Information,” in Roth, Alvin E., ed., Game-Theoretic Models of Bargaining (Cambridge: Cambridge University Press, 1985), pp. 9–26CrossRefGoogle Scholar; Myerson, Roger, “Incentive Compatibility and the Bargaining Problem,” Econometrica 47 (01 1979), pp. 61–74CrossRefGoogle Scholar; and Myerson, Roger and Satterthwaite, Mark, “Efficient Mechanisms for Bilateral Trading,” Journal of Economic Theory 29 (04 1983), pp. 265–81CrossRefGoogle Scholar. In “The Politics of International Regime Formation,” pp. 356–57, Young catalogs a wide range of causes of inefficiencies, including commitment tactics, internal bargaining, issue linkages, and compliance failures.

25. See Wagner, “The Theory of Games and the Problem of International Cooperation”; and Krasner, “Global Communications and National Power.”

26. As enjamin Franklin observed, “The worst outcome is when, by overreaching greed, no bargain is struck, and a trade that could have been advantageous to both parties does not come off at all.” Franklin is cited by Raiffa, in The Art and Science of Negotiation, p. 33.Google Scholar

27. Krasner, , “Global Communications and National Power,” p. 348.Google Scholar

28. See Ibid. Although Krasner characterizes the outcome as one of “where states would end up on the Pareto frontier, not how to get there,” there would seem to be the possibility of mutual gain from mutual forebearance.

29. See Raiffa, , The Art and Science of Negotiation, p. 144;Google Scholar and Lax, and Sebenius, , The Manager as Negotiator.Google Scholar

30. In “Global Communications and National Power,” pp. 353–60, Krasner persuasively documents the distributive conflict that has marked the emergence of present international telecommunications and communications satellite regimes. He correctly notes that the issue in each case was not market failure but conflicting preferences over outcomes. However, his assertion that the results in each case were somewhere “along the Pareto frontier” chosen as a function of relative bargaining power is extremely strong. Demonstrating the assertion would require showing that collectively preferable regimes do not exist in these vital areas, a controversial conclusion given the strife and inefficiencies apparent in the transitional international regimes.

31. Applbaum, Arthur, “Knowledge and Negotiation: Learning Under Conflict, Bargaining Under Uncertainty,” Ph.D. diss., Harvard University, Cambridge, Mass., 1987.Google Scholar

32. Krasner, , “Global Communications and National Power,” p. 364.Google Scholar

33. Considerably more detailed descriptions of the emerging realm of negotiation analysis, on which this and the following sections draw heavily, can be found in my summary essay entitled “The Negotiation Analytic Approach,” in Kremenyuk, Victor, ed., International Negotiation: Problems and New Approaches (San Francisco: Jossey–Bass, 1991), pp. 203–15Google Scholar, and a more technical assessment is offered in my “Negotiation Analysis: A Characterization and Review,” Management Science 38 (01 1992)Google Scholar. For elaborations of the main concepts, see Lax, and Sebenius, , The Manager as Negotiator;Google Scholar and Raiffa, , The Art and Science of Negotiation.Google Scholar

34. See Lax, and Sebenius, , The Manager as Negotiator.Google Scholar

35. In The Art and Science of Negotiation, Raiffa offers an extended discussion of assessing trade-offs in negotiation, building on the earlier work by Keeney, Ralph and Raiffa, Howard in Decisions with Multiple Objectives: Preferences and Value Tradeoffs (New York: Wiley, 1976)Google Scholar. See also Ralph Keeney and Howard Raiffa, “Assessing Tradeoffs: Structuring and Analyzing Values for Multiple-Issue Negotiations,” in Young, , Negotiation Analysis;Google ScholarBarclay, Scott and Peterson, Cameron, “Multiattribute Utility Models for Negotiations,” technical report no. 76–1, Decisions and Designs, Inc., McLean, Va., 1976Google Scholar; Lax, and Sebenius, , The Manager as Negotiator;Google Scholar and Wierzbicki, A. P., “Critical Essay on the Methodology of Multiobjective Analysis,” Regional Science and Urban Economics, vol. 13, 1983, pp. 5–29.CrossRefGoogle Scholar

36. Jervis and others have noted that game theory, for example, treats preferences as exogenously given and offers no account of their origin or formation. Negotiation-analytic accounts share this shortcoming, although not the presumption that interests are immutable. See Jervis, , “Realism, Game Theory, and Cooperation,” p. 336.Google Scholar

37. Krasner, , “Global Communications and National Power,” p. 91.Google Scholar

38. For an elaboration of this argument and a discussion of the various analytic techniques for assessing no-agreement alternatives, see Lax, David A. and Sebenius, James K., “The Power of Alternatives and the Limits to Negotiation,” Negotiation Journal 1 (01 1985), pp. 163–79CrossRefGoogle Scholar. See also Lax, David A., “Optimal Search in Negotiation Analysis,” Journal of Conflict Resolution 29 (09 1985), pp. 456–72.CrossRefGoogle Scholar

39. See Lax, and Sebenius, , The Manageras Negotiator, chap. 5.Google Scholar

40. Certain linkages and logrolling strategies fall within this category. For a discussion of these strategies, see Sebenius, James K., “Negotiation Arithmetic: Adding and Subtracting Issues and Parties,” International Organization 37 (Summer 1983), pp. 281–316CrossRefGoogle Scholar; and Tollison, Robert and Willett, Thomas, “An Economic Theory of Mutually Advantageous Issue Linkage in International Negotiations,” International Organization 33 (Autumn 1979), pp. 425–49CrossRefGoogle Scholar. If the various parties have different forecasts of the likelihood of future events, different attitudes toward risk, or different discount rates, for example, contingent agreements or agreements that vary over time may yield joint gains. For a more technical analysis of such joint gains, see Pratt, John and Zeckhauser, Richard, “The Impact of Risk-Sharing on Efficient Division,” Journal of Risk and Uncertainty 2 (09 1989), pp. 219–34CrossRefGoogle Scholar. For examples of studies that have taken various differences and analyzed their implied joint gains, see Sebenius, , Negotiating the Law of the Sea;Google Scholar Barclay and Peterson, “Multiattribute Utility Models for Negotiations”; Brown, Rex, Peterson, Cameron, and Ulvila, Jacob, “An Analysis of Alternative Mideastern Oil Agreements,” technical report no. DT/TR 75–6, Decisions and Designs, Inc., McLean, Va., 1975Google Scholar; Ulvila, Jacob and Snyder, Warren, “Negotiation of Tanker Standards: Application of Multiattribute Value Theory,” Operations Research 28 (01–02 1980), pp. 81–95CrossRefGoogle Scholar; and Raiffa, , The Art and Science of Negotiation.Google Scholar

41. See, for example, Walton, and McKersie, , A Behavioral Theory of Labor Negotiations;Google ScholarFisher, and Ury, , Getting to Yes;Google ScholarRaiffa, , The Art and Science of Negotiation;Google Scholar and Pruitt, Dean and Rubin, Jeffrey, Social Conflict (New York: Random House, 1986).Google Scholar

42. See, for example, Schelling, , The Strategy of Conflict;Google ScholarWalton, and McKersie, , A Behavioral Theory of Labor Negotiations;Google ScholarRaiffa, , The Art and Science of Negotiation;Google Scholar and Lax, and Sebenius, , The Manager as Negotiator.Google Scholar

43. Schelling, , The Strategy of Conflict, pp. vi–vii.Google Scholar

44. Considerable analytic effort has been directed toward the problem of managing the inherent tension between creating and claiming value. These can include individual styles, conditionally open “tit-for-tat” strategies in Axelrod's, RobertThe Evolution of Cooperation (New York: Basic Books, 1984)Google Scholar, “principled negotiation” from Fisher, and Ury's, Getting to YesGoogle Scholar, as well as various devices such as the single negotiating text and the “post-settlement settlement”; in general, see Raiffa's, The An and Science of Negotiation and chap. 7Google Scholar of Lax, and Sebenius's, The Manager as NegotiatorGoogle Scholar. For an extended example, see Antrim, Lance and Sebenius, James K., “Multilateral Conference Mediation: Tommy Koh and the Law of the Sea,” in Bercovitch, J. and Rubin, J. Z., eds., Mediation in International Relations: Multilateral Approaches to Conflict Management (London: Macmillan, 1991)Google Scholar. For valuable social-psychological insights into factors leading to stalemate conflict escalation and de-escalation, see Pruitt, and Rubin, , Social Conflict.Google Scholar

45. Conceptually, of course, one could argue that if the“supergame” of all possible interests, issues, and parties were specified at the outset, this phenomenon would not exist; this approach, however, would be to define a very real set of dynamics out of analytic bounds. As Oye noted, many linkages have been forged that were unanticipated by analysts or practitioners. One example is the Eisenhower link between the Suez affair and monetary policy, and another is the Third World link between development assistance and special drawing rights. See Oye, Kenneth A., “The Domain of Choice: International Constraints and Carter Administration Foreign Policy,” in Oye, Kenneth A. and Lieber, Robert J., eds., The Eagle Entangled: U.S. Foreign Policy in a Complex World (New York: Longman, 1979), p. 18.Google Scholar

46. See Sebenius, “Negotiation Arithmetic.”

47. See Zartman, I. William, “Negotiations: Theory and Reality,” Journal of Inteniational Affairs 29 (Spring 1975), pp. 69–77Google Scholar; and Zartman, I. William and Berman, Maureen, The Practical Negotiator (New Haven, Conn.: Yale University Press, 1982).Google Scholar

48. While the discussion thus far has been mainly in terms of bilateral bargains with extremely simple structures, these basic elements—parties and factions, interests, alternatives to agreement, creating and claiming value, and moves to change the game itself—are present in models of negotiations with markedly different structures as well as processes, communication possibilities, and decision rules. For example, “two-level games” have been studied in a number of settings. See Putnam, “Diplomacy and Domestic Politics”; Putnam, Robert and Bayne, Nicholas, Hanging Together (Cambridge, Mass.: Harvard University Press, 1987)Google Scholar; Lax, David A. and Meyer, Frederick W., “The Logic of Linked Bargains,” Harvard Business School Working Paper no. 89–012, Cambridge, Mass., 1988Google Scholar; Raiffa, , The Art and Science of Negotiation;Google Scholar and Lax, and Sebenius, , The Manager as NegotiatorGoogle Scholar. For analyses of more complex coalitional bargains, see Weeks, Thomas, “International Debt: From Debtors' Alliance to Global Bargain,” Ph.D. diss., Harvard University, Cambridge, Mass., 1988Google Scholar; Raiffa, , The Art and Science of Negotiation;Google ScholarLax, and Sebenius, , The Manager as Negotiator;Google Scholar and Lax, David A. and Sebenius, James K., “Thinking Coalitionally: Party Arithmetic, Process Opportunism, and Strategic Sequencing,” in Young, Negotiation Analysis, pp. 153–93Google Scholar. An important decision for the analyst seeking to explain a given case or cases of cooperation involves the choice of the most contextually appropriate model: monolithic parties in a bilateral or multilateral interaction, a principal-agent negotiation, internal negotiations, a two-level game. or a full coalitional negotiation with internal divisions and a dynamic evolution.

49. SeeYoung, , “The Politics of International Regime Formation,” pp. 357–58.Google Scholar

50. See Stein, , “Coordination and Collaboration”; Duncan Snidal, “The Game Theory of International Politics,” World Politics 38 (10 1985), pp. 25–57Google Scholar; and Krasner, , “Global Communications and National Power,” pp. 338–39 and 344.Google Scholar

51. See the next section for theoretical and empirical illustrations of these assertions.

52. Each side was given a confidential description of a realistic situation with a private multiattribute value function embedded in the instructions. The carefully structured exercise, set between a Middle Eastern country and a transnational corporation, was originally constructed by John Hammond and modified by Richard Zeckhauser to include an oil-for-aircraft swap and several additional technical, economic, and political issues. A similar plot for another such exercise is reproduced in Raiffa's, The Art and Science of Negotiation, p. 138.Google Scholar

53. When the results of exercises performed by a large number of experienced negotiators from the public and private sectors of many developed and developing countries were aggregated, they were found to spread out widely, rather than to cluster within a small region of the zone of possible agreement.

54. In The Strategy of Conflict, p. 22Google Scholar, Schelling offers the classic evaluation of the pitfalls and tautologies that plague many discussions of this concept:“‘Bargaining power,’ ‘bargaining strength,’ and ‘bargaining skill’ suggest that the advantage goes to the powerful, the strong, the skillful. It does, of course, if those qualities are defined to mean only that negotiations are won by those who win. But, if the terms imply that it is an advantage to be more skilled in debate, or to have more financial resources, more physical strength, more military potency, or more ability to withstand losses, then the term does a disservice. Those qualities are by no means universal advantages in bargaining situations; they often have a contrary value.”

55. Krasner, , “Global Communications and National Power,” p. 337.Google Scholar

56. Ibid., p. 342.

57. This is a formalized version of Schattschneider's well-known view that “the definition of the alternatives is the supreme instrument of power.” See Schattschneider, E. E., The Semisovereign People (Hinsdale, lll.: Dryden Press, 1975), p. 66.Google Scholar

58. For an extensive survey of definitions of the concept of bargaining power, see Bacharach, Samuel B. and Lawler, Edward J., Bargaining (San Francisco: Jossey-Bass, 1981), pp. 10–26Google Scholar. See also Dahl, Robert A., “The Concept of Power,” Behavioral Science 2 (07 1957), pp. 201–15;CrossRefGoogle ScholarWrong, Dennis, “Some Problems in Defining Social Power,” American Journal of Sociology 73 (05 1968), pp. 673–81;CrossRefGoogle Scholar and French, J. R. P. and Raven, B. H., “The Bases of Social Power,” in Cartwright, D., ed., Studies in Social Power (Ann Arbor: University of Michigan Press, 1959)Google Scholar. In The Manager as Negotiator, Lax, Google Scholar and I review and assess the unsatisfactory state of affairs in characterizing bargaining power and suggest an alternative based on perceived changes in perceptions of the zone of possible agreement.

59. See Lax, and Sebenius, , The Manager as Negotiator, pp. 249–58.Google Scholar

60. If a negotiator were considering the use of a tactic, his or her subjective distribution of beliefs about the negotiated outcome conditional on using the proposed tactic must be compared with the subjective distribution of beliefs about the outcome conditional on not using it. That the changed distribution of outcomes gives higher expected utility than the original still need not mean that the negotiator will get a better outcome than he or she would have gotten by eschewing the tactic. A random draw from the changed distribution will, with some probability, be worse than a random draw from the original one.

61. For experimental evidence, see, for example, Lax and Sebenius, “The Power of Alternatives and the Limits to Negotiation”; Lax, and Sebenius, , The Manager as Negotiator;Google Scholar and Neal, Margaret and Northcroft, Gregory, “Experience, Expertise, and Decision Bias in Negotiation,” in Sheppard, Blair, Bazerman, Max, and Lewicki, Roy, eds., Research on Negotiating in Organizations (Greenwich, Conn.: JAI Press, forthcoming)Google Scholar. In many settings, visibly improving one's no-agreement alternatives will improve one's outcomes; many case studies and laboratory experiments provide empirical support for this intuition. Yet there are instances in which this will not be the case. For example, in a marriage, finding a potential replacement mate during a dispute might seriously damage how one's spouse values the marriage. Or a disgruntled employee whose boss is known to value loyalty might be better off not seeking other offers before approaching the boss about improving the current job situation, since the breach of loyalty could reduce the boss's sense of obligation to keep the employee happy.

62. In “Global Communications and National Power,” Krasner offers three elaborations of the meaning of “power”—namely, the capacity to effect changes in the payoff matrix, the capacity to determine who plays, and the capacity to determine the rules. While his first meaning is directly compatible with my general characterization of“power” in terms of changes in the perceived zone of possible agreement, his other two meanings can also be understood in terms of a specific class of changes in the perceived zone of possible agreement. In particular, they can be subsumed under the category of improved alternatives to negotiated agreement. To see this with respect to determining the participants, imagine a tacit or explicit prior negotiation among potential parties to the later substantive negotiation of interest. The issue in the prior negotiation involves who can play in the later game. If one party's alternative to any negotiated agreement in this prior negotiation is simply to carry out the later negotiation absent another party, the party to be included later (whose prior no-agreement alternative was to exclude the other) can be said to be more “powerful” than if it had a less potent prior alternative to agreement. Similar reasoning holds with respect to rule changes that are imposed rather than agreed. Some payoff matrix changes also fall into this class of favorable changes to the perceived zone of possible agreement.

63. A formal measure of the conflict of interest, axiomatically derived by Robert Axelrod, equals the area above and to the right of the Pareto frontier inscribed on the unit square; a “distributive” game with purely conflicting interests has a conflict of interest measure equal to one-half, while a game with purely shared interests has a conflict of interest measure equal to zero. For a precise discussion of this concept, see Axelrod, Robert, Conflict of Interest (Chicago: Markham Publishing, 1970)Google Scholar. For an elaboration of the effects of changes in perceptions of joint gains, see Sebenius, , Negotiating the Law of the Sea, pp. 114–16Google Scholar. In a series of carefully controlled experiments, Malouf and Roth found that agreements took longer to achieve as the conflict of interest rose; see Malouf, Michael and Roth, Alvin, “Disagreement in Bargaining: An Experimental Approach,” Journal of Conflict Resolution 25 (06 1981), pp. 329–48CrossRefGoogle Scholar. In prisoners' dilemma games, Axelrod found a strong positive relationship between conflict of interest and the probability of defection versus mutually beneficial cooperation; see Axelrod, , Conflict of InterestGoogle Scholar. Reviews of the social-psychological literature lend support to the propositions that lower conflicts of interest (1) lead to speedier settlements; (2) have higher agreement probabilities, especially where aspiration levels are relatively high; (3) reduce the danger that one or more parties will repudiate the agreement; (4) tend to strengthen relationships among the parties, thereby facilitating later agreements; and (5) contribute to organizational effectiveness where actors (individuals, subunits, or departments) with distinct needs and values engage in intraorganizational bargaining. See Pruitt, Dean G., “Integrative Agreements: Nature and Antecedents,” paper presented at the Conference on Bargaining Inside Organizations, Boston University, 15 10 1982.Google Scholar

64. See Schelling, , The Strategy of Conflict;Google Scholar and Schelling, , Arms and Influence.Google Scholar

65. For analytic and experimental evidence concerning the conditions under which commitments can be effective, see Lax, and Sebenius, , The Manager as Negotiator;Google Scholar and Bacharach, and Lawler, , Bargaining.Google Scholar

66. See Ethan B. Kapstein, “Between Power and Purpose: Central Bankers and the Politics of Regulatory Convergence,” in this issue of IO; Kapstein, Ethan B., “Resolving the Regulator's Dilemma: International Coordination of Banking Regulations,” International Organization 43 (Spring 1989), pp. 323–47;CrossRefGoogle Scholar and Tobin, Glenn, “National Rules and Global Money,” Ph.D. diss., Harvard University, Cambridge, Mass., 1991.Google Scholar

67. The joint gains and reduction in perceived conflict of interest would be relative to earlier times, when some regulators favored, some opposed, and some were indifferent to common capital standards.

68. It should not be presumed, however, that once the distributive conflict was resolved, the outcome was necessarily Pareto-optimal from the point of view of the various regulators. Other mutually beneficial arrangements might have been discussed or imagined but proved unreachable.

69. See Lax and Sebenius, “Thinking Coalitionally.”

70. For classic discussions of game theory, see Neumann, John Von and Morgenstern, Oskar, Theory of Games and Economic Behavior (Princeton, N.J.: Princeton University Press, 1944)Google Scholar; and Luce, and Raiffa, , Games and DecisionsGoogle Scholar. For recent insightful assessments, with special regard to bargaining, see Roth, , Game-Theoretic Models of Bargaining;Google ScholarAumann, Robert, “Game Theory,” in Eatwell, J., Milgate, M., and Newman, P., eds., Game Theory (New York: Norton, 1989), pp. 1–53Google Scholar; Harsanyi, John, “Bargaining,” in Eatwell, Milgate, and Newman, Game Theory, pp. 54–67Google ScholarPubMed; and Rasmussen, Eric, Games and Information (Oxford: Basil Blackwell, 1989)Google Scholar. With nonspecialist audiences in mind, a number of analysts have recently described some of the most useful contributions of game theory for understanding negotiating behavior. See, for example, Brams, Steven, Game Theory and Politics (New York: Free Press, 1975)Google Scholar; Weber, Robert, “Negotiation and Arbitration: A Game-Theoretic Perspective,” in Lax, David A. et al. , eds., The Manager as Negotiator and Dispute Resolver (Washington, D.C.: National Institute for Dispute Resolution, 1985), pp. I–1 through I–31;Google Scholar Roger Myerson, “Analysis of Incentives in Dispute Resolution,” in Young, , Negotiation Analysis;Google Scholar and Siebe, W., “Game Theoretical Approach,” in Kremenyuk, International Negotiation, pp. 180–202.Google Scholar

71. In the realm of economics, see, for example, Tirole, Jean, The Theory of Industrial Organization (Cambridge, Mass.: MIT Press, 1988)Google Scholar; and Rasmussen, , Games and InformationGoogle Scholar. In the realm of political science, see, for example, Brams, , Game Theory and Politics;Google Scholar and Powell, Robert, Nuclear Deterrence Theory (Cambridge: Cambridge University Press, 1990).CrossRefGoogle Scholar

72. Rubinstein, Ariel, “Comments on the Interpretation of Game Theory,” Econometrica 59 (07 1991), p. 923.CrossRefGoogle Scholar

73. Snidal, , “The Game Theory of International Politics,” p. 44.Google Scholar

74. Critiques of game theory as a useful formalization of various realist and neorealist theories make up a veritable cottage industry. For insightful examples including numerous references, see Jervis, “Realism, Game Theory, and Cooperation”; and Snidal, “The Game Theory of International Politics.” See also Wagner, “The Theory of Games and the Problem of International Cooperation.”

75. For example, as central as the minimax criterion is to classic game theory in purely competitive situations, if the first player believes that the second player is not employing a minimax strategy, perhaps on the basis of experience or empirical studies, there are often strategies superior to minimax for the first player. See, for example, Kadane, J. B. and Larkey, Patrick D., “Subjective Probability and the Theory of Games,” Management Science 28 (02 1982), pp. 113–20CrossRefGoogle Scholar. More generally, Shubik, Mas-Colell, and others have described an impressive number of candidate solution concepts for cooperative games. See Shubik, Martin, Game Theory in the Social Sciences, vol. 1 (Cambridge, Mass.: MIT Press, 1982)Google Scholar; and Mas-Colell, A., “Cooperative Equilibria,” in Eatwell, Milgate, and Newman, Game Theory, pp. 95–102.Google Scholar

76. See Kreps, David, “Nash Equilibrium,” in Eatwell, Milgate, and Newman, Game Theory, pp. 167–77Google Scholar. Refinements include “sequential equilibrium,” as described by Kreps and Wilson; “perfection and subgame perfection,” as described by Selten; and “stability,” as described by Kohlberg and Mertens. See Kreps, David and Wilson, Robert, “Sequential Equilibrium,” Econometrica 50 (07 1982), pp. 863–94;CrossRefGoogle ScholarSelten, R., “Re-examination of the Perfectness Concept for Equilibrium Points in Extensive Games,” International Journal of Game Theory 4 (Winter 1975), pp. 25–55CrossRefGoogle Scholar; and Kohlberg, E. and Mertens, J. F., “On the Strategic Stability of Equilibria,” Econometrica 54 (09 1986), pp. 1003–1038CrossRefGoogle Scholar. Keohane and Snidal make limited versions of this point. See Keohane, , “International Institutions,” p. 388;Google Scholar and Snidal, , “The Game Theory of International Politics,” p. 54.Google Scholar

77. See Tirole, , The Theory of Industrial Organization, p. 446Google Scholar. In explaining why “we are now endowed with nearly a dozen refinements of perfect Bayesian equilibrium,” Tirole argues that the “leeway in specifying off-the-equilibrium-path beliefs usually creates some leeway in the choices of equilibrium actions; by ruling out some potential equilibrium actions, one transforms other actions into equilibrium actions. Hence it is not surprising that one often ends up with a continuum of perfect Bayesian equilibria.”

78. See Harsanyi, J. C. and Selten, R., A General Theory of Equilibrium Selection in Games (Cambridge, Mass.: MIT Press, 1988).Google Scholar

79. As Kadane and Larkey remark, “We do not understand the search for solution concepts that do not depend on the beliefs of each player about the others' likely actions, and yet are so compelling that they become the obvious standard of play for all those who encounter them.” This seems especially apt in light of their observation that “solution concepts are a basis for particular prior distributions” and their conclusion that “the difficulty in non-zero sum, N-person game theory of finding an adequate solution concept [is that] no single prior distribution is likely to be adequate to all players and all situations in such games.” See Kadane, and Larkey, , “Subjective Probability and the Theory of Games,” pp. 115–16.Google Scholar

80. Bell, David E., Raiffa, Howard, and Tversky, Amos, eds., Decision Making: Descriptive, Normative, and Prescriptive Interactions (Cambridge: Cambridge University Press, 1988), p. 9.CrossRefGoogle Scholar

81. Of course, this need not detract from the utility of a fully rational “baseline” analysis. After all, well-structured, repeated negotiations may penalize departures from rational behavior. Yet many negotiating situations are not well-structured, repeated, or embedded in a market context. And much behavior more than trivially departs from the canons of rationality. Recognizing this, researchers might hope that, after developing a better empirical understanding of negotiators' behavior, they can “redo” the game-theoretic program using a more behaviorally faithful theory, such as the prospect theory proposed by Kahneman and Tversky. Unfortunately, an even more fundamental problem with game analysis—namely, the problem of “common knowledge”—would carry over to and impede such a new quest. See Kahneman, Daniel and Tversky, Amos, “Prospect Theory: An Analysis of Decisions Under Risk,” Econometrica 47 (03 1979), pp. 263–91.CrossRefGoogle Scholar

82. Luce, and Raiffa, , Games and Decisions, p. 49.Google Scholar

83. See, for example, Harsanyi, John, “Games of Incomplete Information Played by Bayesian Players” (3 parts), Management Science, vol. 14, 1967–1968, pp. 159–82, 320–24, and 486–502.CrossRefGoogle Scholar

84. Aumann, Robert J., “Agreeing to Disagree,” Annals of Statistics 4 (11 1976), pp. 1236–39.CrossRefGoogle Scholar

85. On a personal note, in my own negotiation experience—which, apart from academics and consulting, includes time in the U.S. Commerce and State Departments, membership for three years on the U.S. delegation to the United Nations Law of the Sea Negotiations, and four years of full-time merger and acquisition work on Wall Street—relatively few negotiating situations have conformed even approximately to Harsanyi's exacting requirements of common knowledge, requirements that include common prior distributions of negotiator “types,” arrayed by reservation prices, utility functions, and so forth. See Harsanyi, “Games of Incomplete Information Played by Bayesian Players.” For a discussion of Harsanyi's doctrine of “common priors,” see Sebenius, J. K. and Geanakoplos, J. D., “Don't Bet On It: Contingent Agreements with Asymmetric Information,” Journal of the American Statistical Association 78 (06 1983), pp. 424–26.CrossRefGoogle Scholar

86. Aumann, , “Game Theory,” p. 31.Google Scholar

87. These factors do not invalidate the game-theoretic program, however. After all, the body of existing game theory is anything but static and will certainly make significant progress on these and other challenges. Nau, for example, has recently reported some interesting results on how the rules of the game become common knowledge. See Nau, R. F., “Joint Coherence in Games of Incomplete Information,” working paper no. 8916, Fuqua School of Business, Duke University, Durham, N.C., 1990.Google Scholar

88. See, for example, Max Bazerman, “Negotiator Cognition and Negotiator Rationality,” in Young, , Negotiation Analysis;Google ScholarKahneman, D., Slovic, P., and Tversky, A., eds., Judgement Under Uncertainty: Heuristics and Biases (Cambridge: Cambridge University Press, 1982)CrossRefGoogle Scholar; Schoemaker, Paul, “The Expected Utility Model: Its Variants, Purposes, Evidence, and Limitations,” Journal of Economic Literature 20 (06 1982), pp. 529–63;Google Scholar and Bell, , Raiffa, , and Tversky, , Decision Making.Google Scholar

89. Despite these contrasts, it would be an unfortunate mistake to characterize the negotiation-analytic approach as an “alternative” to game theory. Instead, negotiation analysis might be called either “nonequilibrium game theory with bounded rationality and without common knowledge” or “decision analysis with a strong interactive flavor.”

90. See Haas, “Introduction.”

91. See Emanuel Adler and Peter M. Haas, “Conclusion: Epistemic Communities, World Order, and the Creation of a Reflective Research Program,” in this issue of IO. In this sense, their work is in line with Oran Young's recent exploration of institutional bargaining in “The Politics of International Regime Formation,” which also expands the traditional reach of bargaining concepts.

92. For an excellent discussion of this point, see Lipson, Charles, “Why Are Some International Agreements Informal?” International Organization 45 (Autumn 1991), pp. 495–538.CrossRefGoogle Scholar

93. I use the terms “winning coalition” and “blocking coalition” in a looser sense than is common in parliamentary and other well-structured contexts. For traditional discussions of these concepts, see Luce, and Raiffa, , Games and Decisions;Google Scholar and Riker, William, The Theory of Political Coalitions (New Haven, Conn.: Yale University Press, 1962)Google Scholar. Here, winning coalitions are only defined with respect to a set of policy measures from the point of view of a particular actor or group of actors; such coalitions consist of sufficient numbers of adherents to render the policy effective, again from the point of view of the specific actor or actors. Blocking coalitions are those opposing interests that could prevent a winning coalition from coming into existence or being sustained. The term “actor” should be contextually obvious and can include states, domestic interests, and transnational groupings of either of these. See Lax, and Sebenius, , The Manager as Negotiator;Google Scholar and Sebenius, “The Negotiation Analytic Approach.”

94. In “Thinking Coalitionally,” Lax and I offer a more detailed analysis of the concept of coalescence without coalition and provide an extended discussion of the 1975 U.S.–Soviet grain deal and other examples in which the concept applies. In antitrust economics, the idea of tacit collusion is a useful analogy; see Loescher, Samuel, “Coalescence Without Coalitions,” in Groennings, S., Kelley, E. W., and Leiserson, Michael, eds., The Study of Coalition Behavior (New York: Holt, Rinehart & Winston, 1970), pp. 177–91.Google Scholar

95. This is distinct from but reinforced by the fact that the two bankers may be friendly competitors in a long-term game, while the buyer and seller are only one-time participants.

96. Haas, “Introduction.”

97. See, for example, Jervis, , “Realism, Game Theory, and Cooperation,” p. 228;Google Scholar and Snidal, , “The Game Theory of International Politics,” pp. 40–44.Google Scholar

98. Adler and Haas, “Conclusion.”

99. Multiple equilibria are mentioned by Adler and Haas, but almost as an afterthought. See ibid., p. 373.

100. Haas, “Introduction.”

101. Ibid.

102. Ibid.

103. See Haas, “Banning Chiorofluorocarbons.”

104. The following points are largely based on Benedick's, RichardOzone Diplomacy (Cambridge, Mass.: Harvard University Press, 1991)Google Scholar and on Haas's “Banning Chlorofluorocarbons.”

105. See Haas, “Banning Chlorofluorocarbons.”

106. See Peterson, “Whalers, Cetologists, Environmentalists, and the International Management of Whaling.”

107. See Adler and Haas, “Conclusion.”

108. See Peterson, “Whalers, Cetologists, Environmentalists, and the International Management of Whaling.”

109. See Ikenberry, “A World Economy Restored.”

110. See Adler, “The Emergence of Cooperation.”

111. Hirschman, Albert O., “How the Keynesian Revolution Was Exported from the United States, and Other Comments,” in Hall, Peter, ed., Political Power of Economic Ideas: Keynesianism Across Nations (Princeton, N.J.: Princeton University Press, 1989), p. 356.Google Scholar

112. See Lax, and Sebenius, , The Manager as Negotiator;Google Scholar and Schein, Edgar, Organizational Culture and Leadership (San Francisco: Jossey-Bass, 1985).Google Scholar

113. See Ingram, Helen, “Policy Implementation Through Bargaining: The Case of Federal Grants-in-Aid,” Public Policy 25 (Fall 1977), pp. 500–526.Google ScholarPubMed

114. See Haas, “Do Regimes Matter? Epistemic Communities and Mediterranean Pollution Control.”

115. See Sebenius, James K.., “Designing Negotiations Toward a New Regime: The Case of Global Warming,” International Security 15 (Spring 1991), pp. 101–45.CrossRefGoogle Scholar

116. Thacher, Peter, “Alternative Legal and Institutional Approaches to Global Change,” Colorado Journal of International Environmental Law and Policy 1 (Summer 1990), pp. 101–26.Google Scholar

117. Chayes, Abram, “Managing the Transition to a Global Warming Regime, or What to Do Until the Treaty Comes,” in Mathews, J., ed., Greenhouse Warming (Washington, D.C.: World Resources Institute, 1991), pp. 61–68.Google Scholar