A two-dimensional analysis of seventy years of United Nations voting

• Published in: Public choice Vol. 176; no. 1/2; pp. 33 - 55
• Main Authors: Bailey, Michael A., Voeten, Erik
• Format: Journal Article
• Language: English
• Published: New York Springer Science + Business Media 01-07-2018; Springer US; Springer Nature B.V
• Subjects: Arms control & disarmament; Dimensional analysis; Disputes; Economics; Economics and Finance; Foreign policy; Human rights; Intellectuals; International relations; Political Science; Public Finance; Voter behavior; Voting; International organizations; Ideal point estimation; United Nations; International relations
• DOI: 10.1007/s11127-018-0550-4
• EISSN: 1573-7101
• ISSN: 0048-5829

International relations scholars frequently use roll-call votes on resolutions in the United Nations General Assembly (UNGA) to measure similarity in the foreign policy ideologies of states. They then correlate those measures with consequential outcomes, such as development lending, trade, or military disputes. Dynamic ideal point models of UNGA voting thus far have been restricted to a single dimension. We examine the existence of a stable, important, and interpretable second dimension underlying contestation in the UN. From the mid-1960s to the mid-1980s, North–South conflict constitutes a stable second dimension, shaped heavily by the agenda-setting powers of the so-called Non-aligned Movement and the Group of 77. In the periods before and after, the second dimension neither is stable nor easily interpretable, though it is sometimes important. We suggest that in most applications, our original one-dimensional estimates have conceptual advantages with minimal losses in explanatory value. We illustrate that conclusion with an analysis that correlates ideal point changes with militarized interstate disputes. Yet, our findings also suggest that scholars interested in specific issues, such as the Middle East, human rights, or arms control, might benefit from more specifically tailored ideal point estimates.