The removal of economic and social constraints on voting rights and inclusion of females in the voting franchise represent key events in the development of democracy in Western countries. Yet, an important question is to what extent these changes in the composition of the electorate really changed actual public policies. Theoretically, the effect on spending on public services are ambiguous. Previous studies from the US consistently find that female suffrage increased education and health spending, while evidence from European countries are more mixed. This paper studies the effect of introduction of universal female suffrage in local elections in 1910 on public school expenditure and school resource use in local governments in Norway. The 1910 reform removed previous limitations on female voting rights in terms of property ownership and taxable income requirements.

We exploit the 1910 national reform combined with the substantial pre-reform differences in the share of females in the electorate across local governments as a source of exogenous variation in a difference-in-differences approach to identify causal effects. This empirical strategy circumvents the endogeneity problem facing earlier studies using temporal and cross-sectional variation in voting rules across countries or states. We find that introduction of universal female suffrage did not increase local government spending on education or the teacher-student ratio. This absence of systematic positive effects of the franchise extension are robust to different sample definitions and model specifications despite a positive and statistically significant effect on female turnout as measured by the share of total votes cast by females.