Association for Public Policy Analysis & Management

Tertiary education is highly supported by government expenditures (Winter-Ebmer & Wirz, 2002). This is justified by the fact that a better educated population, as a high quality human resource, plays an important role in facilitating economic growth and increasing competition capability in the global market (Kaiser, Vossensteyn and Koelman, 2001). Furthermore, a more educated population also contributes to civic participation, a stable democracy and a richer cultural life (UNESCO, 1998). Finally, public funding for education helps in maintaining social equality and reducing social exclusiveness by providing opportunities for students from poorer backgrounds to access a better education (Winter-Ebmer & Wirz, 2002).

However, most current literature focusing on the impact of government expenditure on tertiary education are reports rather than empirical studies. And the few empirical papers that explore the relationship between government expenditures and tertiary education are conducted only in European countries and the United States. Moreover, most of these studies only utilized a linear model, which may not be the optimal model to capture the relationship between government expenditures and the enrollment rate of tertiary education. Finally, government expenditure is likely to have a delayed impact on the enrollment rate, the mechanism of this delay was not found in former studies.

In response to these literature gaps, this paper examines the impact of government expenditure on the enrollment rate of tertiary education in context of more than 200 countries. More specifically, this study first utilizes a nonlinear model to precisely capture the relationship between government expenditure and the enrollment rate of tertiary education. Secondly, we will explore the delay mechanism by examining how an increase in government expenditure affects the enrollment rate of tertiary education in the next few years. Lastly, this paper compares the impact of government expenditure on the enrollment rate of tertiary education between genders, developed and developing countries, and across different regions all over the world.

Raw data used in this study is extracted from the World Bank dataset, which includes data from 268 countries and regions from 1970 to 2014. The World bank dataset is a free and open database of global development data provided by the World Bank (World bank, 2017).

The quantitative models which will be used in this paper include correlation, panel regression and polynomial regression. The dependent variable of the regression models in this study is the enrollment rate of tertiary education. The independent variable is government expenditure. The regression models in this study also include control variables, which are namely GDP growth, GDP per capita, fertility rate and unemployment rate.