The optimum size of public education spending: panel data evidence

The paper examines the presence of positive effect of public education spending in a panel of 50 developed, developing and transition economies (over the 1980-2012 period) on the level and growth of output, and, provided such effect holds, considers the optimal provision of public education spending. The econometric methodology relies on panel unit root and cross-sectional dependence tests, panel regression with fixed effects, and panel quantile model with fixed effects. It is demonstrated that public education spending is productive at the margin under alternative specifications, and has positive externalities on the private economy, while the factor productivity in the government sector is higher than in the private. For the panel as a whole, the public education tended to be under-provided (the optimal level of 5.05% of GDP compared to the actual average level of 4.14% of GDP); however, the over-provision is observed in the slow-growing economies in the lower quantiles.