Measuring state capacity: Theoretical and empirical implications for the study of civil conflict
This article identifies and addresses key conceptual and measurement issues raised by measures of state capacity in studies of civil conflict. First, it reviews competing definitions and operationalizations of state capacity, focusing specifically on those that emphasize (1) military capacity, (2) bureaucratic administrative capacity, and (3) the quality and coherence of political institutions. Second, it critically assesses these measures on the basis of construct validity, focusing attention on whether they accurately capture the theoretical concept of state capacity, and whether they allow the researcher to differentiate between competing causal mechanisms. Third, it employs principal factor analysis to identify the underlying dimensionality of 15 different operationalizations of state capacity. State capacity is characterized by low dimensionality, with three factors - or dimensions of state capacity -explaining over 90% of the variance in the 15 measures. While the first factor, rational legality, captures bureaucratic and administrative capacity, the second, rentier-autocraticness, and third, neopatrimoniality, capture aspects of state capacity that cut across theoretical categories. The article concludes by suggesting a multivariate approach to modeling state capacity, and that (1) survey measures of bureaucratic quality, and (2) tax capacity are the most theoretically and empirically justified.