Few puzzles in international relations research have received as much attention as the relationship between political regimes and interstate conflict. Here we examine a particular aspect of this puzzle: the systemic relationships between democracy, democratization, and interstate conflict. We test hypotheses aimed at exploring three dimensions of this general inquiry. First, that democracy, democratization, and war are endogenous phenomena. Second, that the relationships between these three processes vary across time with respect to their strength and direction. Lastly, that these relationships also vary spatially, paying particular attention to geographic regions. Using time-series techniques with data for the period 1816-1992, we employ Granger causality and vector auto-regression (VAR) techniques to test our hypotheses. Briefly, we find evidence confirming the endogeneity of these processes, though this endogeneity is not as extensive as we anticipated and is conditional on the spatial and temporal sample selected. Perhaps more importantly, we find evidence suggesting that the statistical strength and sign of the relationships of interest exhibit considerable spatial and temporal heterogeneity. We conclude that the regional level, rather than the global level, holds more promise for understanding the interrelationships between systemic democracy, democratization, and interstate conflict.
Crescenzi, Mark J. & Andrew J. Enterline (1999) Ripples from the Waves? A Systematic, Time-series Analysis of Democracy, Democratization, and Interstate War, Journal of Peace Research 36 (1): 75–94.