Using a sample of 306 estimates drawn from 31 primary studies, this paper conducts an empirical synthesis of the link between economic growth and government expenditure on education or health using meta-analysis. We also explain the heterogeneity in empirical results. We find that the effect of government education expenditure on growth is positive, whereas the growth effect of government health expenditure is negative. Our meta-regression analysis suggests that factors such as econometric specifications, publication characteristics as well as data characteristics explain the heterogeneity in the literature. We also find no evidence of publication selectivity.
JEL code: O40, H50, C1Keywords: government education expenditure; government health expenditure; human capital; economic growth Acknowledgements: We would like to thank two anonymous reviewers of the Program Committee of MAER-Net Colloquium 2014 in Athens; and the participants in the conference for helpful comments.