Does democracy cause growth? A meta-analysis perspective

The relationship between democracy and economic growth has long been investigated both in the political science and in the economic literature with inconclusive outcomes. By adopting a multi-level meta-analysis framework, we tried to shed lights on this conundrum. Our hierarchical sample includes 103 studies containing 942 point-estimates. Our random effects model suggests that the sign of this relationship, albeit positive, is statistically weak. We then address the high between-studies heterogeneity by adopting meta-regression analysis models. Results are striking: the effect sizes’ variance is largely driven by spatial and temporal differences in the samples, indicating that the democracy and growth nexus is not homogeneous across world regions and time periods. Conversely, the large number of control variables included in the papers, do not impact the reported results. At the same time, models estimated by means of the within estimator have a significant, albeit negative, impact on economic growth. This seems to suggest that scholars have not yet found the appropriate control variables - or their suitable proxies - to explain such widely debated relationship.