Low level of GDP per capita is a robust and widely applied predictor of civil war. Yet, GDP is a crude macro-level indicator that masks considerable heterogeneity in economic structures, and it is less well able to explain variation in conflict risk among low-income countries. Here, we consider the merit of classic dual sector theory in improving common economic models of civil war. Two basic expectations are derived: the relative size of the traditional sector increases conflict risk via low opportunity cost and high share of immobile wealth, whereas high relative labor productivity (RLP) in the modern sector compared to the traditional sector facilitates labor mobility and wage growth, thus reducing the viability of rebellion. We evaluate these expectations via out-of-sample prediction analysis of civil conflict involvement, drawing on a unique 10-sector dataset of economic activity among 40 countries across the world since 1969. The analysis provides robust evidence that poor countries with a comparatively productive modern sector are less conflict prone than countries at similar income levels with lower RLP ratios. However, further probing into potential mechanisms producing this relationship does not provide decisive evidence in favor of any potential mechanism. We conclude that replacing GDP per capita with indicators of sector size and relative productivity improves the predictive performance of common civil war models, although more research is needed to assess the generalizability of these findings and to gain further insight into the underlying causal pathways linking relative labor productivity with reduced conflict risk.
Article is accessible for free through the journal's web page.