Measuring National Power

Abstract

Power's central role in international relations theory is unsurpassed, yet considerable debate persists over the quality of its most commonly used indicator, the Correlates of War project's Composite Indicator of National Capabilities (CINC). At issue is whether CINC's main feature, its ability to measure a nation's power relative to other nations' power levels, inadvertently creates errors when membership in the comparison group fluctuates. Using mathematical proofs and an empirical investigation of the major power system, we show that Organski and Kugler (1980) and Gleditsch and Ward (1999) are correct: changes in the comparison group do create errors in CINC. In particular, CINC inadvertently mismeasures dyadic power distributions. Using power transition theory as a context within which to evaluate CINC, we find that it creates artificial power transitions, masks actual transitions, changes the timing of transitions, alters the magnitude by which one state overtakes another, and produces specious relationships between transitions and conflict. We also offer a viable alternative measure, called the Geometric Indicator of National Capabilities (GINC), and demonstrate how its use of the geometric mean retains CINC's notion of systemically-based relative power and immunizes it from the problems afflicting CINC. GINC is strongly recommended for dyadic analyses, especially when membership in the comparison group fluctuates frequently.