I am grateful to Martin McGuire, Hudson Meadwell, Glenn Palmer, and the reviewers and editor of International Organization for helpful comments on earlier drafts of this article. I also benefited greatly from discussions at NATO Headquarters, Brussels, and the International Institute for Strategic Studies, London, which were made possible by a Collaborative Research Grant from the Scientific Affairs Division of NATO. Of course, I alone am responsible for the analysis and conclusions drawn here.

1. See the more recent edition: Olson, Mancur, The Logic of Collective Action, rev. ed. (Cambridge, Mass.: Harvard University Press, 1971)Google Scholar.

2. Ibid., p. 3.

3. Hardin, Russell, Collective Action (Baltimore, Md.: Johns Hopkins University Press, 1982)Google Scholar.

4. Russett, Bruce M. and John D, Sullivan, “Collective Goods and International Organizations,” International Organization 25 (Autumn 1971), pp. 845–65CrossRefGoogle Scholar.

5. Thies, Wallace J., “Alliances and Collective Goods,” Journal of Conflict Resolution 31 (06 1987), pp. 298–332CrossRefGoogle Scholar.

6. Olson, Mancur, “Increasing the Incentives for International Cooperation,” International Organization 25 (Autumn 1971), pp. 866–74CrossRefGoogle Scholar.

7. Olson, Mancur and Zeckhauser, Richard, “An Economic Theory of Alliances,” Review of Economics and Statistics 48 (08 1966), pp. 266–79CrossRefGoogle Scholar.

8. There is an extensive literature on whether defense spending as a percentage of either GNP or GDP is a valid measure of the contribution the allies make to collective security. Nations vary in the efficiency with which they produce military capabilities; prices for military equipment are often not set by markets; and some contributions to collective security, such as the full value of conscripted personnel, are not reflected in defense budgets. In addition, not all of the military forces of several allies are deployed in or even dedicated to the defense of the North Atlantic region, and those that are may be used for purely national objectives. Regarding this last point, however, Allen and Diehl note that “NATO members have generally agreed to count all of a nation's military forces, regardless of mission or deployment, as a contribution to NATO. The rationale is that the strengthening of defense and deterrence anywhere in the world indirectly contributes to European security.” See Allen, Christopher S. and Diehl, Paul F., “The Defense Issue in West Germany,” Armed Forces and Society 15 (Fall 1988), p. 96CrossRefGoogle Scholar. For discussions of the defense burden as a measure of a nation's contribution to the alliance, see Lello, José, “Draft Interim Report on Burden-Sharing in the Alliance,” North Atlantic Assembly, Brussels, 11 1988Google Scholar; U.S. Department of Defense, “Report on Allies Assuming a Greater Share of the Common Defense Burden,” Report to the U.S. Congress by the Secretary of Defense, Washington, D.C., 02 1989Google Scholar; and Diehl, Paul F. and Goertz, Gary, “Measuring Military Allocations,” Journal of Conflict Resolution 30 (09 1986), pp. 553–81Google Scholar. Even Knorr, who provides one of the most complete critiques, concludes, however, that the defense burden is the “least unsatisfactory” measure available. See Knorr, Klaus, “Burden-Sharing in NATO,” Orbis 29 (Fall 1985), pp. 517–36Google Scholar.

9. See Olson and Zeckhauser, “An Economic Theory of Alliances”; Strihou, Jacques M. van Ypersele de, “Sharing the Defense Burden Among Western Allies,” Review of Economics and Statistics 49 (11 1967), pp. 527–36CrossRefGoogle Scholar; Pryor, Frederic L., Public Expenditures in Communist and Capitalist Nations (Homewood, III.: Irwin, 1969)Google Scholar; and Beer, Francis A., The Political Economy of Alliances (Beverly Hills, Calif.: Sage, 1972)Google Scholar.

10. Russett, Bruce M., What Price Vigilance? (New Haven, Conn.: Yale University Press, 1970)Google Scholar.

11. Sandier, Todd and Forbes, John F., “Burden Sharing, Strategy, and the Design of NATO,” Economic Inquiry 18 (07 1980), pp. 425–44CrossRefGoogle Scholar.

12. All tables and figures are based on data from the following sources: The defense expenditures of the NATO allies in current prices and national currencies are published annually in the NATO Review, usually the January issue. Germany's expenditures are counted as contributions to the alliance from 1950, though it did not become a member until 1954. This is NATO's convention. The alliance's data are preferable to national statistics because they conform to a common accounting standard. A serious problem that remains, however, involves conscripted personnel, whose value is greater than their pay because of the coercive character of the draft. To increase further the comparability of the data on defense spending, the expenditures of allies that did not rely on volunteers were increased by 10 percent to reflect the monetary value of conscription. See Oneal, John R. and Elrod, Mark A., “NATO Burden Sharing and the Forces of Change,” International Studies Quarterly 33 (12 1989), pp. 435–56CrossRefGoogle Scholar; and van Ypersele de Strihou, “Sharing the Defense Burden.” Soviet defense expenditures for the years 1963–84 are from the U.S. Arms Control and Disarmament Agency (ACDA), Yearbook of World Armament and Disarmament (Washington, D.C.: Government Printing Office, 1978 and 1987)Google Scholar. ACDA's estimates were extrapolated back to 1950 on the basis of data from the Stockholm International Peace Research Institute, Yearbook of World Armament and Disarmament (New York: Humanities Press, 1970)Google Scholar. The GDPs of all countries are from Summers, Robert and Heston, Alan, “A New Set of International Comparisons of Real Product and Price Level Estimates for 130 Countries, 1950–1985,” Review of Income and Wealth 34 (03 1988), pp. 1–26CrossRefGoogle Scholar. These estimates are based on purchasing power parities (PPPs), which are not only more stable than exchange rates in the years after the Bretton Woods system but are also theoretically preferable because they indicate the amount of a foreign currency required to buy the quantity of goods and services that can be purchased in the United States for one dollar. Specifically, the chain index of real domestic product (1980 base year) was used because the index number problem is least severe with its use. The military expenditures of the NATO countries were converted into U.S. dollars using the PPPs associated with this index. Ideally, PPPs specifically for military goods would have been used; but these are not available except for a few countries in a few years. See United Nations, Reduction of Military Budgets, UN General Assembly Report A/40/421, 40th Session, New York.

13. See Sandier, Todd and Cauley, Jon, “On the Economic Theory of Alliances,” Journal of Conflict Resolution 19 (06 1975), pp. 330–48CrossRefGoogle Scholar; Sandier and Forbes, “Burden Sharing, Strategy, and the Design of NATO”; Murdoch, James C. and Sandier, Todd, “A Theoretical and Empirical Analysis of NATO,” Journal of Conflict Resolution 26 (06 1982), pp. 237–63CrossRefGoogle Scholar; Murdoch, James C. and Sandier, Todd, “Complementarity, Free Riding, and the Military Expenditures of the NATO Allies,” Journal of Public Economics 25 (11 1984), pp. 83–101CrossRefGoogle Scholar; and Sandier, Todd and Murdoch, James C., “Defense Burdens and Prospects for the Northern European Allies,” in Denoon, David B. H., ed., Constraints on Strategy: The Economics of Western Security (Washington, D.C.: Pergamon-Brassey, 1986), pp. 59–113Google Scholar.

14. For a theoretical discussion of the difference between inclusive and exclusive public goods, see Olson, , The Logic of Collective Action, pp. 36–43Google Scholar.

15. See Oneal and Elrod, “NATO Burden Sharing and the Forces of Change”.

16. See NATO, Texts of Final Communiques, vol. 2, 1975–1980 (Brussels: NATO Information Service, n.d.)Google Scholar; U.S. Congressional Budget Office, Assessing the NATO/Warsaw Pact Military Balance (Washington, D.C.: Government Printing Office, 12 1977)Google Scholar; International Institute for Strategic Studies (IISS), The Military Balance (London: IISS, 1968, 1971, and 1977)Google Scholar; and Cusack, Thomas, “The Evolution of Power, Threat, and Security,” International Interactions, vol. 12, 1985, pp. 151–98CrossRefGoogle Scholar.

17. See Rogers, Bernard W., “The Atlantic Alliance,” Foreign Affairs 60 (Summer 1982), pp. 1145–56CrossRefGoogle Scholar; McNamara, Robert, “The Military Role of Nuclear Weapons,” Foreign Affairs 62 (Fall 1983), pp. 59–80CrossRefGoogle Scholar; Kaiser, Karl, “Nuclear Weapons and the Preservation of Peace,” Foreign Affairs 60 (Summer 1982), pp. 1157–70CrossRefGoogle Scholar; Mertes, Alois, “What Should Be Done to Change NATO,” Atlantic Community Quarterly 22 (Summer 1984), pp. 122–29Google Scholar; Hoffmann, Stanley, Janus and Minerva (Boulder, Colo.: Westview, 1987), pp. 251–57Google Scholar; and Sommer, Theo, “Through German Eyes,” The National Interest, no. 10, Spring 1988, pp. 3–12Google Scholar.

18. See Williams, Geoffrey, The Permanent Alliance (Leyden: A. W. Sijthoff, 1977), p. 75Google Scholar; Callaghan, Thomas A. Jr, “Can Europe Be Defended?” Atlantic Community Quarterly 21 (Summer 1983), pp. 110–17Google Scholar; and Canby, Steven and Dorfer, Ingemar, “More Troops, Fewer Missiles,” Foreign Policy, no. 53, Winter 1983, pp. 3–17CrossRefGoogle Scholar.

19. Deitchman, Seymour J., Military Power and the Advance of Technology (Boulder, Colo.: Westview, 1983), pp. 187–98Google Scholar.

20. See Brams, Steven J., Superpower Games (New Haven, Conn.: Yale University Press, 1985)CrossRefGoogle Scholar; Russett, Bruce M., “The Mysterious Case of Vanishing Hegemony; Or, Is Mark Twain Really Dead?” International Organization 35 (Spring 1985), pp. 845–65Google Scholar; and Oneal, John R., “Measuring the Material Base of the East-West Balance of Power,” International Interactions, vol. 15, 1989, pp. 177–96CrossRefGoogle Scholar.

21. Oneal and Elrod, “NATO Burden Sharing and the Forces of Change”.

22. Simon, Herbert A., “Human Nature in Politics,” American Political Science Review 79 (06 1985), pp. 293–304CrossRefGoogle Scholar.

23. Snidal, Duncan, “The Limits of Hegemonic Stability Theory,” International Organization 39 (Autumn 1985), pp. 579–614CrossRefGoogle Scholar.

24. See McGuire, Martin C. and Groth, Carl H. Jr, “A Method for Identifying the Public Good Allocation Process Within a Group,” Quarterly Journal of Economics 100 (Supplement 1985), pp. 915–34CrossRefGoogle Scholar; and Keohane, Robert O. and Nye, Joseph, Power and Interdependence (Boston: Little, Brown, 1977)Google Scholar.

25. Olson and Zeckhauser, “An Economic Theory of Alliances”.

26. Equal defense burdens are the most obvious and perhaps the most important sign of cooperation among allies; but cooperation can, of course, take other forms, notably in the standardization of equipment or specialization of contributions and roles according to comparative advantage. For discussions of the progress and continuing shortcomings of NATO in these areas, see U.S. Department of Defense, “Standardization of Equipment Within NATO,” Report to the U.S. Congress by the Secretary of Defense, Washington, D.C., 1988Google Scholar; and NATO, “Enhancing Alliance Collective Security,” Report by the Defence Planning Committee, Brussels, 1988Google Scholar.

27. See Borcherding, Thomas E., “Comment: The Demand for Military Expenditures,” Public Choice, vol. 37, 1981, pp. 33–39CrossRefGoogle Scholar; Comes, Richard and Sandier, Todd, “The Theory of Public Goods: Non-Nash Behavior,” Journal of Public Economics 23 (04 1984), pp. 367–79Google Scholar; and Keohane, Robert O., After Hegemony (Princeton, N.J.: Princeton University Press, 1984)Google Scholar.

28. Standardization was necessary because the sum of the unstandardized scores for any country and for all others as a group is not a constant; therefore, only comparisons within a single year would be strictly accurate. To permit comparisons through time, the burden of each ally relative to the NATO average each year was divided by the mean of the score for the United States and the score for all other countries combined.

29. Olson, , The Logic of Collective Action, p. 3Google Scholar.

30. See Forbes, Kevin, Korsun, George, and MaGuire, Martin, “Defense, Growth, and Allocation Behavior in the Alliance,” in Denoon, , Constraints on Strategy, pp. 114–51Google Scholar; and Calvocoressi, Peter, World Politics Since 1945, 4th ed. (London: Longman, 1982), pp. 181–95 and 413–14Google Scholar.

31. Oneal and Elrod, “NATO Burden Sharing and the Forces of Change”.

32. See Snidal, “The Limits of Hegemonic Staibility Theory”; and Olson, Mancur, “A Theory of the Incentives Facing Political Organizations,” International Political Science Review 7 (04 1986), pp. 165–89CrossRefGoogle Scholar.

33. Olson, , The Logic of Collective Action, pp. 28–30 and 49–50Google Scholar.

34. Russett, , What Price Vigilance? p. 110Google Scholar.

35. The three exceptional cases have been excluded in all years because they were unusually influenced by private interests during much of the postwar period. Although most significant after 1974, the rivalry between Greece and Turkey predated the invasion of Cyprus, of course; and Portugal's defense spending was skewed by its colonial obligations before 1964 and by its military governments after 1975. Excluding these countries in all years also ensures comparability in analyses across time.

36. Throughout this study, regression analyses have been corrected for an AR(1) autocorrelation process using the method recommended by Hibbs. See Hibbs, Douglas A. Jr, “Problems of Statistical Estimation and Causal Inference in Time-Series Regression Models,” in Costner, Herbert L., ed., Sociological Methodology, 1973–1974 (San Francisco: Jossey-Bass, 1974), pp. 252–308Google Scholar.

37. Hitiris, Theodore, “Convergence and the Relationship Between the Members of the European Community,” in Dosser, Douglas, Gowland, David, and Hartley, Keith, eds., The Collaboration of Nations (New York: St. Martin's Press, 1982), pp. 42–60Google Scholar.

38. McGuire and Groth, “A Method for Identifying the Public Good Allocation Process”.

39. Palmer, Glenn, “Guns, Vacations, and GNP: The Elasticity of Defense Spending,” paper presented at the annual meeting of the Southern Political Science Association, Atlanta, 1988Google Scholar.

40. Including Greece, Turkey, and Portugal would exaggerate the results of the tests for cooperation, since their increased expenditures on private goods in later years adds to their defense burdens and thereby raises the lower end of the regression line. It is particularly important to recognize this distorting influence because it is rivalry between Greece and Turkey that has caused their defense spending to be so high.

41. These results have been corroborated by pooled cross-sectional and time-series regression analyses. See Oneal, John R., “Testing the Theory of Collective Action: NATO Defense Burdens, 1950–1984,” Journal of Conflict Resolution 34 (09 1990)CrossRefGoogle Scholar.

42. Deutsch, See Karl W., The Analysis of International Relations, 3d ed. (Englewood Cliffs, N.J.: Prentice-Hall, 1988), p. 255Google Scholar; and Sloan, Stanley R., NATO's Future (Washington, D.C.: National Defense University Press, 1985), pp. 173–82Google Scholar.

43. Olson and Zeckhauser, “An Economic Theory of Alliances”.