LAUSR

BACKGROUND A country's health expenditure significantly improves its population health status. This study aims to examine the determinants of health expenditure in dictatorships. METHODS We designed a mixed methods research approach. First, we used panel data from 1995 to 2014 covering 99 countries (n = 1488). Fixed effects regression models were fitted to determine how different types of authoritarianism relate to health expenditure. Second, we chose Ivory Coast to apply the synthetic control methods for a case study. We constructed a synthetic Ivory Coast, combining other dominant party regimes to resemble the values of health expenditure predictors for Ivory Coast prior to a regime change from a dominant party system to personalist dictatorships in 2000. RESULTS We found that dominant party autocracies, compared with non-dominant party regimes, increased health expenditure (% of GDP) (1.36 percentage point increase, CI = 0.59-2.12). The marginal effect, however, decreased when an autocrat in this type of regime held elections (0.86 percentage point decrease, CI = 0.20-1.52). Furthermore, we found the difference in health expenditure between the actual Ivory Coast and its synthetic version starts to grow following the regime change in 2000 (in 2000, actual: 6.00%, synthetic: 6.04%; in 2001, actual: 4.85%, synthetic: 5.99%), suggesting a pronounced negative effect of the government transition on Ivory Coast health expenditure. CONCLUSION The findings suggest that different forms of dictatorship are associated with varying levels of health expenditure. Where dictatorships rely on popular support, as is the case with dominant party dictatorships, health expenditure is generally greater.