Determinants of Social Expenditure Levels in the EU

The aim of the thesis is to identify the determiants of social expenditure levels in EU member states and estimate the sensitivities. The analysis is based on the fixed effects model with robust standard errors. Seven out of the 24 variables, that have been tested, are statistically significant. Social expenditure levels are positively dependent on population ageing, unemployment rate and economic level, and negatively on economic growth. The openness of economy and public debt ratio are also robust. The presence of a christian party in government is also statistically significant, however, the coefficient is not theoretically consistent. The sensitivity of social expenditure level to the share of population aged 65 years and over on the total population, is about 0,30. It means, that on average, the rise in the share of the old by 1 percentage point, is followed by the rise of social expenditure level by 0,3% of the GDP. In case of unemployment rate, economic growth and the openness of economy, the sensitivity estimates are 0,3, -0,1 and -0,03. The sensitivities to the other three robust variables cannot be quantified due to a strong dependence on the exact model specification or a bias, caused by extreme values. The expenditure on pensions are relatively more dependent on population ageing, whereas other social expenditure are more sensitive to current position in economic cycle. The determinants in the old and the new EU member states are similar, with only one exception, as the expenditure levels in the new member states are less dependent on demographic structure. The model explains international differences and the changes during the period between 1990 and 2010