Democracy and Economic Growth: A Causal Analysis

Since Seymour Martin Lipset's seminal work on the socioeconomic prerequisites of democracy, a substantial number of social scientists has theorized and empirically tested theories about the relationship between democracy and economic development.1 Democratization's third wave has renewed interest in this relationship.2 This interest is evident in the increasing number of studies that explore the relationship of democracy to economic development.3 Nevertheless, a clear consensus has yet to emerge on the causal relationship between democracy and economic performance. Some scholars argue that economic growth drives social mobilization, which in turn leads to political mobilization and eventually regime change to democracy.4 Logically, new ideas obtained as a result of economic growth and/or industrialization generate new social classes and strata which drive a social transformation. Other scholars, in contrast, suggest that, as a nation democratizes, the economy develops faster.5 Individuals want to be confident that property they accumulate will be respected, and only democratic societies provide such confidence. Still other scholars find no systematic relationship between economic development and democracy, for two reasons. First, economic development does not necessarily lead to higher levels of democracy because democratization is affected by multiple factors. Second, by the same token economic development is affected by many factors; a democratic government by itself can have only limited impact on economic development.6 Thus, previous studies have not provided a dominant theoretical framework to explain the democracy-development relationship. In particular, the question of causation remains to be answered. The issue of causal direction between democracy and economic development is particularly important in developing theory since clarification of the theoretical relationship between the two variables eases model specification and empirical testing of the theory. To answer this question we employ a Granger causal approach. This approach offers several benefits over the methods used in previous studies. First, many previous studies were based on cross-sectional, multinational evidence. According to Jung and Marshall, however, a cross-sectional approach assumes high levels of similarity among different countries, which may not be true.7 Moreover, a longitudinal design is necessary to see the dynamic processes of economic development and democratization. Second, most previous studies implicitly assumed that either economic development