Peace and Conflict Studies - Spring 2014

Peace and Conflict Studies Volume 21, Number 1 90 The dependent variable of this study is the occurrence of a militarized interstate dispute (MID) in a given dyad-year. Following Russett and Oneal (2001) and many others, onset and continuation of MIDs are treated the same. A MID is defined as an event where the government or citizens of at least one state threatened, displayed, or used forced against the government or citizens of at least one other state worldwide (Jones, Bremer, & Singer, 1996). The dependent variable equals 1 if in a given year a dyad involves a MID, 0 otherwise. I use Zeev Maoz’s (2005) dyadic MID data, which is a refined version of the Correlates of War (COW) data on MIDs (Ghosn, Palmer, & Bremer, 2004). Given the dichotomous nature of the dependent variable, I use logistic regression (logit) in my estimations. Explanatory Variables. Economic Development : Economic development is the major component of the “capitalist peace”. In measuring the economic development level of a country, I use its real gross domestic product per capita (GDPpc) measured in purchasing power parities (thousands) and constant (1996) dollars. Data availability becomes a serious problem in analyzing the economies of developing countries, though. In the most frequently used economic dataset Penn World Table (Version 6.1) (Heston, Summers, & Aten, 2002), economic data are unavailable sporadically for many countries and there is no GDP data at all for eleven countries until 1990s. Consequently, about 25% of observations in Penn World Table’s GDPpc dataset are missing. However, Gleditsch (2002) introduces some measures to reduce the number of missing observations in Penn World Table’s GDP as well as in International Monetary Fund’s trade data and manages to have a complete dataset of GDP and GDP per capita for years between 1945 and 2000. I follow Dixon’s (1993) “weak link” principle, which assumes that the likelihood of conflict is primarily a function of the degree of constraints experienced by the less constrained state in each dyad, and consider the level of development of the less-developed state for each dyad-year, a variable I call development low . To minimize the direction of causality problems, all data on economic development are lagged one year. Democracy : To determine national levels of democracy and autocracy, I use Polity IV (Marshall & Jaggers, 2004) data, which has become the standard for measuring institutional democracy, particularly in the study of international conflicts. The Polity IV dataset provides an 11-point scale (0-10) of autocracy and an 11-point scale (0-10) of democracy. To determine the “net” democracy score of a country, I subtract its autocracy score from its democracy score, which yields a range of -10 to 10. I add 11 points to each score and