Hostname: page-component-586b7cd67f-t8hqh Total loading time: 0 Render date: 2024-11-25T21:34:28.142Z Has data issue: false hasContentIssue false
  • English
  • Français

On the operations of the International Coffee Agreement

Published online by Cambridge University Press:  22 May 2009

Robert H. Bates  and
Da-Hsiang Donald Lien
Get access

Extract

Next to oil, coffee is the most valuable commodity traded in world markets; it is produced in the developing world and consumed within the developed nations. Moreover, the International Coffee Agreement is the major successful international commodity agreement. In this research note we analyze the allocation of quotas to export coffee in 1982 under the terms of the agreement. The way in which entitlements to export are allocated offers insights into significant features of the international political economy.

Type
Research Notes
Information
International Organization , Volume 39 , Issue 3 , Summer 1985 , pp. 553 - 559
Copyright
Copyright © The IO Foundation 1985

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1. The most notable recent application in strategic studies is de Mesquita, Bruce Bueno, The War Trap (New Haven: Yale University Press, 1981)Google Scholar. For extensions to areas of agreement and cooperation, see Axelrod, Robert, The Evolution of Cooperation (New York: Basic, 1984)Google Scholar. Coming closest to the approach explored here is Keohane, Robert, After Hegemony (Princeton: Princeton University Press, 1984)Google Scholar.

2. For background, see Fisher, Bart S., The International Coffee Agreement (New York: Praeger, 1972)Google Scholar; Geer, Thomas, An Oligopoly: The World Coffee Economy and Stabilization Schemes (New York: Dunellen, 1971)Google Scholar; Krasner, Stephen D., “Manipulating International Commodity Markets,” Public Policy 21, 4 (1973), pp. 493–513Google Scholar; Rowe, J. W. F., The World's Coffee with Special Reference to Control Schemes (Stanford: Food Research Institute, Stanford University, 1943)Google Scholar; and Bates, Robert H. and Lien, Da-Hsiang Donald, “Coffee: An Essay on the Origins and Operations of Political Intervention in International Markets,” unpublishedGoogle Scholar.

3. For a variety of practical reasons we have had to assume that there is a one-year lag in revising the votes. Although this assumption may not be realistic, because of the stability on market shares in the short run, these regression results should still apply even if we assume not lag in the allocation of votes.

4. A useful introduction to the Shapley value, its problems, and other solution concepts is contained in Riker, William H. and Ordeshook, Peter C., An Introduction to Positive Political Theory (Englewood Cliffs, N.J.: Prentice-Hall, 1973)Google Scholar. For the approximation to the Shapley value employed here, see Owen, Guillermo, Game Theory (New York: Academic, 1982)Google Scholar. For our problem, the Shapley value for the i-th player is denned as:

where the summation is taken over all winning coalitions (i.e., a collection of members for which the summation of votes exceeds two-thirds of the total votes) T such that T-{i} is not winning. The number of elements in T is “t” and N is the number of players. A major problem with the Shapley value is that in calculating it all coalitions are treated as equally likely, which under many circumstances does not hold.

5. One reader queried whether alternative specifications of the model should not be reported. We resist that invitation because our purpose here is to model the institution. The rules of the institution translate quite literally into the relationships specified in the model in the text. Reporting alternative specifications would, in other words, amount to modeling some other institution.

6. We were asked to report the values of R2 for these equations and have done so. It should be noted, however, that the values of ‘S’ are estimated within the system of simultaneous equations. They are therefore stochastic. The advantage of this procedure is that we thereby obtain estimates of the coefficients that are unbiased. But as a consequence it is difficult to argue that the right-hand side variables in these equations generate the element of “explained variance” employed in calculating the R2 statistic; they import elements of randomness as well. We therefore stress that in this case R2 does not communicate the same meaning as it does in single equation models.

7. We can also estimate the system by GLS, which yields the following results:

Using two-stage least-square estimation, the first equation is

The two methods yield almost the same estimation and maintain the same qualitative results. In general, when error terms across equations are correlated, GLS will be biased. But for small sample cases, there are some situations in which GLS will dominate two-stage least squares. See Mariano, Robert S., “Analytical Small-Sample Distribution Theory in Econometrics: The Simultaneous-Equations Case,” International Economic Review 23 (10 1982), pp. 503–33CrossRefGoogle Scholar. We therefore provide both estimates for comparison.

8. For a review of much of this literature, see Riker and Ordeshook, Introduction to Positive Political Theory.

9. An exception would be the analysis of the United Nations. See, for example, ibid., pp. 170ff.

10. Bueno de Mesquita, The War Trap.

11. Schelling, Thomas, The Strategy of Conflict (New York: Oxford University Press, 1963)Google Scholar.

12. See, for example, Keohane, After Hegemony.

13. See, for example, Bates and Lien, “Coffee.”