Florian Hagn’s research while affiliated with Hamburg University and other places

This study analyses on the basis of a multivariate analysis ex post the effects on the jobs market of a soccer World Cup, in this case the 2006 World Cup held in Germany. In addition to three methods already used for other analyses in studies of sporting events, an extended 'Difference-in-Difference' estimate is used in order to compare the development of the numbers of unemployed in the 12 World Cup venues with the development of the numbers of unemployed in 63 other German cities. The results demonstrate that in none of the respective match venues did the effect of the sporting event on unemployment differ significantly from zero.

This study demonstrates that the Football World Cup 1974 in Germany was not able to generate any medium to long-term employment effects that were significantly different from zero. It is the first work to examine the employment effects of Football World Cup tournaments. It is also the first work to undertake a multivariate analysis of the employment effects of a major sporting event outside of the USA. In addition, this study does not arbitrarily determine the time period for the potential positive effects of a major sporting event but instead examines several alternative periods. Furthermore, the study tests for method sensitivity by analysing the data set in parallel with the approaches used in the studies of sporting events in the USA as well as in a fourth modifying estimation approach. In contrast to the conclusions reached in comparable studies, the results are not regarded as a clear refutation of the positive effects of major sporting events.

This study demonstrates that the Football World Cup 1974 in Germany was not able to generate any short to long-term employment effects that were significantly different from zero. It is the first work to examine long-term employment effects of Football World Cup tournaments. It is also one of the first work to undertake a multivariate analysis of the employment effects of a major sporting event outside of the USA. In addition, this study does not arbitrarily determine the time period for the potential positive effects of a major sporting event but instead examines several alternative periods. Furthermore, the study tests for method sensitivity by analysing the data set in parallel with the approaches used in the studies of sporting events in the USA as well as in a fourth modifying estimation approach. In contrast to the conclusions reached in comparable studies, the results are not regarded as a clear refutation of the positive effects of major sporting events.

Using a panel of international student test scores 1980 – 2000 (PISA and TIMSS), panel fixed effects estimates suggest that government spending decentralization is conducive to student performance. The effect does not appear to be mediated through levels of educational spending.