Attendance at football games largely determined by economic trends

The uncertainty of outcome hypothesis

In previous research, a lot of attention has been given to the idea that the uncertainty of outcome hypothesis is a very important indicator of match attendance. However, notions of how competitive football has been over this period can only account for at most about 20% of the variation in attendances. Van Ours used unemployment rate data of a large dataset on attendances at football matches in England of over 130 years for his research.

The effect of unemployment on attendance rates

Attendances at football games in all divisions in England vary, but attendances at the top division vary most and are highest. High unemployment in the early 1930s coincides with a dip in attendances, and low unemployment in the post-war era coincides with a peak in attendances. Van Ours found that the responsiveness of attendances with respect to the unemployment rate in England was -0.31, meaning that a 1% rise in the unemployment rise is consistent with a 0.31% fall in attendances. When looking at lower division football, a rise in the unemployment rate decreases attendance rates even more.

Great value 

To control for pure coincidence in changes in attendance rates, van Ours uses time trends. He finds that around 80% of the variation in attendance can be explained by simple time trend variables and unemployment rates. The results of the research may be of great value when the football industry recovers from the pandemic. As the world returns to some kind of normality, how much hidden unemployment will turn into actual unemployment? This could have a huge impact on the income of supporters, and hence their ability to pay for attending soccer games.