Effect of democratic reforms on child mortality: a synthetic control analysis

Background: The effects of political regimes on health are unclear because empirical evidence is neither strong nor robust. Traditional econometric tools do not allow the direction of causality to be established clearly. We used a new method to investigate whether political transition into democracy affected child mortality. Methods: We used a synthetic control method to assess the effects of democratisation on child mortality as a proxy of health in countries that underwent transition from autocracy to democracy that lasted for at least 10 years between 1960 and 2010. Democracy was indicated by a score greater than 0 in the Polity2 index. We constructed synthetic controls (counterfactuals) based on weighted averages for factors such as child mortality, economic development, openess to trade, conflict, rural population, and female education from a pool of countries that remained autocracies during the study period. Results: Of 60 countries that underwent democratic transition in the study period, 33 met our inclusion criteria. We were able to construct good counterfactuals for 24 of these. On average, democratisation reduced child mortality, and the effect increased over time. Significant reductions in child mortality were seen in nine (38%) countries, with the average reduction 10 years after democratisation being 13%. In the other 15 countries the effects were not significant. At the country level yhe effects were heterogeneous, but the differences did not correlate with geographic, economic, or political indicators. The effect of democratisation, however, was stronger in countries with above average child mortality before transition than in countries with below average child mortality. Interpretation: Our results are consistent with the interpretation that democratic reforms have the greatest effects when child mortality is a direct concern for a large part of the population. Future research could focus on identifying the precise mechanism through which the effects emerge. Funding: European Union 7th Framework Programme and KU Leuven Methusalem Fund