I have a new paper co-authored with Todd Jewell from the University of North Texas and Rob Simmons from the University of Lancaster about the impact of hooliganism on the performance of English clubs (hooliganism_JSE_jewell_simmons_szymanski), which got a nice write up from Sean Ingle in the Guardian today.
First the disclaimer- the paper has been submitted to a peer reviewed journal but we have not received any reviews yet and we have to wait and see if it will be accepted for publication. Some academics think you should not disseminate research to the wider public until it has been through the peer review process, but obviously we don’t agree with that. My own view about any research in social science is that you always have to make your own mind up about it, no matter how many times it has been reviewed.
In the paper we do something that doesn’t appear to have been done before- to look at the impact of “hooliganism”- which we measure by the number of arrests per thousand fans at league matches over a season. A higher figure, we reason, means a higher level of disturbances, which will be highly correlated with hooliganism- though we spend some time in the paper discussing precisely what this word actually means.
We mainly look at two aspects of club performance- league position, and revenues. We argue that hooliganism might affect league position via intimidation- of opposing fans, match officials and players. There is now a great deal of evidence that match officials are influenced by the crowd (next summer Ignacio Palacios-Huerta, one of the leading researchers on this issue, has a great book coming out which devotes quite a lot of space to the evidence on this), so it seems reasonable to think that violence at football grounds can have this effect.
On the other hand, consistent hooliganism is likely to have an adverse effect o club finances- directly through the cost of policing which is partly borne by the club, but also because demand is likely to suffer: more fans would like to avoid violence than would like to participate in it.
Our data consists of annual arrests figures for the 92 English league clubs between 1984 and 1994, and between 1999 and 2009 (the data for 1995-1998 seems to have been lost – no one involved in collecting the data seems to know what happened to it). The earlier period covers the high water mark of domestic hooliganism, while during the latter period hooliganism as we define it has fallen dramatically.
Our results show that in the earlier period hooliganism did have the predicted effects, but in the more recent periods these effects seem to have disappeared. We think this makes sense. From the perspective of crowd control, the clubs and police have intervened more systematically to snuff out incidents as they start, and the advent of all seater stadiums and new investments, including CCTV, have made the job easier. On the demand side, rising ticket prices have priced out a lot of young men who were predominantly (but not exclusively) associated with the unrest. An aging fan profile, and the desire the bring young children to games, has changed the nature of fandom to some extent. As the phenomenon has has diminished, so has its potential to impact club performance.
One interesting aspect of this research is that we seem to be the first to link hooliganism to club performance, even though to us it seems a fairly obvious point. There is a huge sociological literature on hooliganism, but we were almost completely unable to find anyone who entertained this possibility. Since the effect seems to be there in the data, at least in the heyday of English hooliganism, we hope our paper might lead to some re-evaluation of the phenomenon.