Hostname: page-component-78c5997874-g7gxr Total loading time: 0 Render date: 2024-11-07T23:24:05.516Z Has data issue: false hasContentIssue false
  • English
  • Français

Domestic political regime changes and Third World voting realignments in the United Nations, 1946–84

Published online by Cambridge University Press:  22 May 2009

Joe D. Hagan
Joe D. Hagan
Affiliation:
Associate Professor of Political Science and Director of the International Studies Program at the University of Wyoming, Laramie.
Get access

Abstract

This article presents a cross-national analysis of the relationship between domestic political regime changes and voting realignments of Third World nations in the United Nations (UN). It seeks to move beyond existing research that has assumed that foreign policy is rooted in political and economic structures and changes only when a political revolution occurs. It argues that a wider variety of regime changes—ranging from those involving mainstream political parties to milder ones such as factional shifts in single-party regimes—can also provoke major realignments. Using a new data set on Third World regimes, the article examines the impact of regime changes for eighty-seven nations on their UN voting patterns during the period from 1946 to 1984. Although the findings indicate that revolutions are most likely to provoke major voting realignments, they also show that the more frequent, nonrevolutionary types of regime change are associated with many voting realignments. A major implication of these findings is that foreign policy changes reflect a complex set of domestic regime factors, including leadership belief systems and internal political constraints, as well as aspects of political structure.

Type
Articles
Information
International Organization , Volume 43 , Issue 3 , Summer 1989 , pp. 505 - 541
Copyright
Copyright © The IO Foundation 1989

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. See, for example, Wittkopf, Eugene, “Foreign Aid and United Nations Votes: A Comparative Study,” American Political Science Review 67 (09 1973), pp. 868–88CrossRefGoogle Scholar; Richardson, Neil, “Political Compliance and U.S. Trade Dominance,” American Political Science Review 70 (12 1976), pp. 1098–1109CrossRefGoogle Scholar; Richardson, Neil, Foreign Policy and Economic Dependence (Austin: University of Texas Press, 1978)Google Scholar; and Richardson, Neil and Kegley, Charles, “Trade Dependence and Foreign Policy Compliance,” International Studies Quarterly 24 (06 1980), pp. 191–222CrossRefGoogle Scholar.

2. See Midlarsky, Manus, “The Revolutionary Transformation of Foreign Policy: Agrarianism and Its International Impact,” in Kegley, Charles and McGowan, Patrick, eds., The Political Economy of Foreign Policy Behavior (Beverly Hills, Calif.: Sage, 1981), pp. 39–62Google Scholar; and Moon, Bruce, “Consensus or Compliance? Foreign Policy Change and External Dependence,” International Organization 39 (Spring 1985), pp. 297–328CrossRefGoogle Scholar. For a cross-national assessment of dependence and regime change effects which looks at shifting foreign trade patterns, see Andriole, Stephen and Hopple, Gerald, “The Process, Outcomes, and Impact of Regime Change in the Third World, 1959–81,” International Interactions 12 (04 1986), pp. 363–92CrossRefGoogle Scholar.

3. The importance of domestic political regimes as well as other factors is suggested by several chapters in Holsti, K. J., ed., Why Nations Realign: Foreign Policy Restructuring in the Postwar World (Boston: Allen & Unwin, 1982)Google Scholar; and in Boyd, G. and Hopple, G. W., eds., Political Change and Foreign Policy (New York: St. Martin's Press, 1987)Google Scholar. Weinstein's, FranklinIndonesian Foreign Policy and the Dilemma of Dependence (Ithaca, N.Y.: Cornell University Press, 1976)Google Scholar remains the most detailed analytic case study of perceptual and political sources of foreign policy changes across different regimes. For two recent theoretical works on American foreign policy that point (at least implicitly) to the importance of regimes, see Krasner, Stephen, Defending the National Interest (Princeton, N.J.: Princeton University Press, 1978)Google Scholar; and Gaddis, John Lewis, Strategies of Containment (New York: Oxford University Press, 1982)Google Scholar. For a cross-national assessment of regime effects on foreign policy behavior, see Hagan, Joe, “Regimes, Political Oppositions, and the Comparative Analysis of Foreign Policy,” in Hermann, Charles, Kegley, Charles, and Rosenau, James, eds., New Directions in the Study of Foreign Policy (Boston: Allen & Unwin, 1987), pp. 339–65Google Scholar.

4. A brief review of Moon's findings supports this argument. He finds that regime changes statistically accounted for changes in UN voting behavior for seventeen of the forty-three nations in his sample. However, for only nine nations could realignments be traced to revolutions: Bolivia, Burma, Chile, Cuba, Egypt, Guatemala, Indonesia, Iraq, and Peru. Other cases of realignment had to have resulted from less dramatic types of regime change. Most were cases involving the exchange of power between mainstream political actors operating within the established political arena: either shifts between military and civilian leaders or competition between different political parties (as in Colombia, Haiti, Panama, the Philippines, and Venezuela). For other countries, voting shifts had to have evolved from still milder types of regime change. For example, Lebanon, Morocco, and Saudi Arabia each had only shifts in the coalition of leadership factions or parties within the same basic ruling group continuously in power. Finally, several countries with revolutions did not have realignments in UN voting patterns, including Libya from King Idris to Qaddafi and Yemen with the 1962 overthrow of the monarchy. See Hagan, Joe, “What Types of Regime Changes Lead to Foreign Policy Realignments? A Further Analysis of Moon's Findings, 1946–1974,” paper presented at the annual meeting of the International Studies Association, Washington, D.C., 04 1987Google Scholar.

5. Only the very small Third World states are excluded from the sample. Full documentation on each regime change can be found in Hagan, Joe, Handbook of Political Regimes and Oppositions in the Postwar Era—Version II, University of Wyoming, monograph, 1988Google Scholar.

6. This definition of regime is based on that found in Salmore, Barbara and Salmore, Stephen, “Political Regimes and Foreign Policy,” in East, Maurice, Salmore, Stephen, and Hermann, Charles, eds., Why Nations Act: Theoretical Perspectives for Comparative Foreign Policy Studies (Beverly Hills, Calif.: Sage, 1978), pp. 103–22Google Scholar. Although at variance with more structurally oriented notions of regimes, this definition is consistent with the term as it is popularly used, and it offers flexibility in the conceptualizations of regime properties, particularly nonstructural ones such as leadership beliefs and political oppositions. It is also compatible with Moon's operationalization of regime change, despite his theoretical emphasis on political and economic structures.

7. On the impact of leadership belief systems and the dynamics of perceptions, see Jervis, Robert, Perception and Misperception in International Politics (Princeton, N.J.: Princeton University Press, 1976)Google Scholar; and Holsti, Oli R., “The Belief System and National Images: A Case Study,” Journal of Conflict Resolution 6 (09 1962), pp. 244–52CrossRefGoogle Scholar. For a parallel argument on the relative importance of the interests of state versus societal actors, see Stephen Krasner's “statist” interpretation of American foreign policy in his Defending the National Interest. The juxtaposition of structural and perceptual/political explanations of foreign policy might be considered to be a central area of theoretical overlap between comparative foreign policy (CFP) and international political economy (IPE) perspectives, two fields that have thus far shown relatively little concern for each other's recent theoretical development. On the latter point, see Rosenau, J. N., “CFP and IPE: The Anomaly of Mutual Boredom,” International Interactions 14 (01 1988), pp. 17–26CrossRefGoogle Scholar.

8. Weinstein, , Indonesian Foreign Policy and the Dilemma of Dependence, p. 28Google Scholar.

9. For discussions of the perceptual and political roots of policy changes in these countries, see Weinstein, Indonesian Foreign Policy and the Dilemma of Dependence; Robinson, Thomas W., “Restructuring Chinese Foreign Policy, 1959–76: Three Episodes,” in Holsti, , Why Nations Realign, pp. 134–72Google Scholar; Terrill, Ross, The Future of China After Mao (New York: Delta, 1978)Google Scholar; Baker, Raymond William, Egypt's Uncertain Revolution Under Nasser and Sadat (Cambridge, Mass.: Harvard University Press, 1978)CrossRefGoogle Scholar; and Taylor, Alan R., The Arab Balance of Power (Syracuse, N.Y.: Syracuse University Press, 1982)Google Scholar.

10. On the politics of policymaking in Third World states and its implication for foreign policy, see Good, Robert, “State-Building as a Determinant of Foreign Policy in the New States,” in Martin, Lawrence, ed., Neutralism and Non-Alignment (New York: Praeger, 1962), pp. 3–12Google Scholar; Heeger, Gerald, The Politics of Underdevelopment (New York: St. Martin's Press, 1974)Google Scholar; Wriggins, W. Howard, The Ruler's Imperative: Strategies for Political Survival in Asia and Africa (New York: Columbia University Press, 1969)Google Scholar; and Weinstein, Indonesia and the Dilemma of Dependence, chap. 8. There remains in this literature the issue of whether oppositions constrain leaders from initiating changes in policy or whether they actually push leaders to take highly assertive and aggressive actions. An analysis of this issue is beyond the scope of my article; here, I simply assert that internal political dynamics underlie regime change effects on voting alignments either by diminishing or by amplifying leaders' predispositions to act. For a discussion of this theoretical issue and some empirical evidence, see Hagan, Joe, Political Opposition and Foreign Policy in Comparative Perspective (Boulder, Colo.: Lynne Reinner, 1989)Google Scholar.

11. Because this article focuses on UN voting data, the original sample size of ninety nations and over six hundred regimes has been considerably reduced. In most cases, the deleted regimes were of very short duration and were not in power during the major part of any session of the UN General Assembly. (This problem was compounded for regimes that were in power in 1964 but not during the General Assembly's one-vote 19th session.) Two countries in the regime data, North Korea and South Korea, were deleted because they have never been UN members. Surinam was deleted for technical reasons. Several other Third World nations were not included for all years, because they did not participate in all sessions of the General Assembly; these included the People's Republic of China (admitted in 1971), Taiwan (evicted in 1971), South Africa (barred from the General Assembly in 1974), and Kampuchea (whose UN seat as of 1979 was controlled by insurgents and not the government in power).

12. For complete information on coding rules, see Hagan, Handbook of Political Regimes and Oppositions.

13. Unfortunately, Europa does not publish a Latin American volume, so for this region several readers were used extensively for historical background. The most useful of these were Wiarda, Howard and Kline, Harvey, eds., Latin American Politics and Development, 2d ed. (Boulder, Colo.: Westview, 1985)Google Scholar; Anderson, Thomas, ed., Politics in Central America (New York: Praeger, 1982)Google Scholar; Kantor, Harry, ed., Patterns of Politics and Political Systems in Latin America (Chicago: Rand McNally, 1969)Google Scholar; and Burnett, Ben and Johnson, Kenneth, eds., Political Forces in Latin America (Belmont, Calif.: Wadsworth, 1969)Google Scholar.

14. Two other nations did not have any leadership changes during the time they were in the UN and therefore had no regime effects in the analysis reported here. These countries were Vietnam, which did not have a regime change since its 1977 admission to the UN, and Taiwan, over which Chiang Kai-shek ruled until several years after that country's eviction from the UN.

15. Voting data were taken from the Inter-university Consortium for Political and Social Research (ICPSR) annual data collections of General Assembly plenary sessions for 1946–84, the entire period for which data are now available. Votes from special sessions and in General Assembly committee settings have not been included in my analysis. As is customary, only votes in the General Assembly plenary sessions on so-called “controversial resolutions” have been included here; this excluded the many resolutions on which over 90 percent of the yes and no votes were cast on the winning side.

16. Only votes for which the nation was present were counted in computing its annual voting scores; therefore, the number of votes for each nation in each session varies.

17. This procedure, as recommended by Moon in “Consensus and Compliance,” helps control for annual variations in the voting due to the shifting agenda of issues before the General Assembly and for varying responses to that agenda across the four basic Third World regions. Although it enhances data controls, the procedure presents problems in making comparisons across regions and years; therefore, the reader should recognize that the voting scores across regions can be compared only roughly. An examination of raw rates of opposition votes shows important temporal and regional variations. There has been a gradual increase in anti-U.S. positions among Third World countries, from roughly 50–60 percent opposition to U.S. positions in the early 1950s to 70–80 percent opposition in the early 1980s. Regional variations have occurred bcause of changes in the General Assembly agenda. Although these have occurred across all four regions, Latin American votes have historically deviated the most from Third World norms. Only by mid-1970 did the Latin American states on average approach the level of opposition characteristic of other Third World nations.

18. These labels as used here are not intended to identify directly the orientation of the regime itself. Again, I emphasize that they merely refer to the level of agreement with the United States shown by a given regime as compared with that shown by other Third World nations in the region. The measure was constructed by dividing the regimes within each region into five groups and classifying the 20 percent with the highest voting scores as strongly pro-U.S., the next 20 percent as moderately pro-U.S., the middle 20 percent as centrist, the next 20 percent as moderately anti-U.S., and the final 20 percent (the nations with strongly negative voting scores) as strongly anti-U.S. in orientation. As noted later in the text, although most pro-U.S. regimes appear to side with the United States quite often by Third World standards, they in fact by the 1980s voted against the U.S. positions over 60 percent of the time. They had a strongly pro-U.S. orientation merely in the sense that the other 80 percent of the Third World membership voted against the United States at higher rates.

19. Explaining the classification of regimes is complicated because voting data are normalized by region and by year, and presenting a full discussion of this data and a list of regime voting scores and orientations is precluded by space limitations. The examples that follow in this section provide some basic information about the meaning of voting orientations. In addition, Tables 2 and 3 show changes in voting orientation that resulted from statistically significant regime changes. Readers wishing further information may contact the author or consult appendices in Hagan, Joe, “Domestic Political Regimes Changes and Third World Voting Realignments in the United Nations, 1946–1984,” paper presented at the annual meeting of the Northeast Political Science Association, Providence, R.I., 11 1988Google Scholar.

20. To cite the extreme cases, for example, in the 1980s the most pro-Soviet regimes (such as Afghanistan, Cuba, and Vietnam) voted against the United States at rates of almost 100 percent, while such “outcasts” in the UN as Israel and South Africa had for some time voted with the United States at rates of about 80 percent. Differences between most other regimes were not as extreme but still represent meaningful differences, even in the 1980s when the Third World agenda was strong. Compare, for example, the voting percentages of Syria's strongly anti-U.S. regime under Assad with those of Saudi Arabia's moderately pro-U.S. regime under Khalid and Fahd: the Syrian rates were 7 percent agreement with the United States, 82 percent opposition, and 11 percent abstention, while the Saudi rates were 26 percent agreement, 61 percent opposition, and 13 percent abstention.

21. Take, for example, the rates of voting by two strongly pro-U.S. regimes from the Philippines and by two strongly anti-U.S. regimes from Syria. In the Philippine's Magsaysay regime of the mid-1950s, rates were 57 percent agreement with the United States, 17 percent opposition, and 26 percent abstention. The Marcos regime remained in the strongly pro-U.S. category despite the fact that it had rates of only 38 percent agreement, 42 percent opposition, and 21 percent abstention. At the other extreme, Syria had strongly anti-U.S. regimes in the early 1950s under Shishakli and since 1969 under Assad, but the raw voting rates of these two regimes differed considerably owing to the changing character of UN politics. The Shishakli regime, one of the most anti-U.S. regimes in the 1950s, opposed the United States at a rate of 49 percent and had 13 percent agreement and 38 percent abstention. The Assad regime, in contrast, opposed the United States at a much higher rate (82 percent), reflecting a diminished level of agreement (7 percent) and far fewer abstentions (11 percent).

22. Two other cases—Israel and South Africa—are puzzling in that they had large standard deviations but certainly have not experienced recent foreign policy realignments and are widely assumed to have had a consistently pro-U.S. orientation in the UN. Examination of their voting trends suggests, however, a more complicated picture. While much of the Israeli and South African voting variation appears to reflect their responses to growing isolation in the UN, both countries had a regime change in the postwar years (1950s) that did affect voting patterns. These were changes in which more moderate political elements were replaced by relatively hard-line leaders. In Israel, this was Sharett's replacement by the more hawkish Ben-Gurion; in South Africa, it was Strijdom's replacement by the extremist Verwoerd.

23. These results are somewhat stronger than those reported by Moon, who found that seventeen of his forty-three countries (39 percent) had statistically meaningful regime effects. His Cuban illustration is not particularly exceptional in my analysis. The eta squared value for Cuba (.6932) was exceeded by that of fifteen other nations (Benin, Bolivia, Chile, Ghana, Guatemala, Indonesia, Iran, Israel, Kampuchea, Laos, Madagascar, Peru, South Africa, Syria, and Turkey). In each, over 70 percent of voting variance was explained by regime differences.

24. Across all eighty-seven nations in Table 1, there was a moderate correlation between the strength of regime effects and the amount of variance in their UN voting: the Pearson correlation between the eta squared values and the standard deviations listed in Table 1 was +.315 and was significant at the p =.001 level, indicating that regime effects are present mainly for nations that have been most prone to shift their voting alignments.

25. Nevertheless, I make no claim that regime changes exclusively accounted for voting behavior in the UN. While this general test points to the overall importance of regime effects, it is clear from even a preliminary examination of annual voting changes that there were numerous cases in which a nation's voting pattern shifted in a way unrelated to regime effects, although, interestingly, the shifts were often for only a single year.

26. In three cases—one each for Costa Rica, Haiti, and Israel—a sizable voting shift (a shift of one or more standard deviations) was spread across several regimes and involved more than three years. In these rare cases, caution suggested that the long-term shifts be deleted from further analysis because it could be argued that they occurred for reasons not entirely related to regime changes. In subsequent stages of my analysis, Haiti was excluded from the tables because it had no direct, sizable, short-term voting shifts between adjacent regimes, but Costa Rica and Israel were included because they did. The eta squared value reported for Israel in Table 1 is probably somewhat inflated because of the rare circumstance noted above.

27. The picture is further complicated by the fact that in three cases the revolutionary effects were not immediately apparent. Only with qualification can the 1978 Afghanistan, 1952 Egyptian, and 1975 Laotian revolutions be said to have had significant effects, since their realignments occurred only after factional infighting was resolved, with a later regime change marking the rise to power of a single faction (Karmal's Parcham faction) or a predominant leader (Nasser after evicting Naguib) or marking the eviction of temporary, nonrevolutionary allies (Kaysone's dominance after the Pathet Lao coalition took complete control of the revolutionary regime).

28. This does not mean that the Deng and Mubarak regimes were as pro-U.S. as their predecessors in terms of the raw percentage rates of agreement, opposition, and abstention; these regimes had higher rates of opposition to the United States than did the Chiang Kai-shek and Farouk regimes. However, in terms of the standards in the UN during the 1980s, the Deng and Mubarak regimes were at least as pro-U.S. as their prerevolutionary predecessors.

29. The volatility of the political systems of these countries should not be taken to suggest that they have been politically unstable. Some of them (for example, Costa Rica, Israel, and, until recent years, Sri Lanka) have had well-institutionalized democratic political systems. Their systems have been volatile only in the sense that their party systems have been intensely competitive, with the result that polarized parties on the left and the right have been able to achieve power at times.

30. Added to these unsurprising cases are countries with no regime changes prior to 1985 or during the country's years in the UN: the Ivory Coast, Malawi, Mozambique, Singapore, Taiwan, Tanzania, Tunisia, Vietnam, and Zambia.

31. These combine with the five revolutions that had no statistically significant effects in Afghanistan, Indonesia, Ghana, Mali, and Syria (see Table 2).

32. When the same analysis of variance was done using the voting scores for each nation without regrouping them by voting orientation, the results were similar. There were statistically significant impacts on the size of voting realignment across the five types of regime change and in roughly the same ordering presented for the data regrouped by voting orientation. The eta squared statistic was also about the same.

33. Moon, “Consensus or Compliance”; and Midlarsky, , “The Revolutionary Transformation of Foreign Policy,” p. 56Google Scholar.