Does local sporting success affect sport participation? An examination of Danish professional soccer's effect on club membership

ABSTRACT

Research question: The body of literature examining potential trickle-down effects from elite sport success to mass sport participation has grown considerably over the years. The evidence is mixed but generally suggests that when a nation's athletes win medals at international tournaments – for example the Olympic Games or the FIFA World Cup – it cannot be expected to inspire amateurs to take up sport themselves. So far, existing research has primarily focused on international elite success. Effects from local elite clubs playing in national leagues have only been examined once.

Research methods: Since recent research indicates that national and cultural contexts can influence whether a potential trickle-down effect materialises, this paper aims to test these factors in another setting. By deploying panel data regression models on data from Danish top-tier professional soccer, we test whether local club performance affects local membership levels in Denmark.

Results and Findings: We do not find direct positive effects from sporting success. However, we do identify negative effects on membership figures due to club relegation indicating that sporting failure can affect mass participation.

Implications: A key recommendation drawn from this study is that politicians, sport managers and civil servants should be careful using the double pyramid/‘virtuous circle of sport’ metaphor as the basis of their decision-making because it is an imprecise theoretical interpretation of reality. Instead, stakeholders should focus effort and resources on other relevant factors that potentially can lead to higher – or even lower – levels of participation in order to stimulate such factors (or prevent those that affects negatively). Future studies should examine these implications further.

KEYWORDS:

• Trickle-down effects
• club level sporting success
• mass participation
• professional soccer
• Denmark

Acknowledgements

The authors would like to thank the anonymous reviewers and the associate editor for their constructive comments on earlier drafts of this paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 According to the study, the average German triathlete spends more than nine hours a week on triathlon participation.

2 In the analysis we have led the dependent variable with one and two years resulting in the loss of one/two years of observations – the average number of observations per municipality in the final models is therefore 7.5/6.3.

3 We anticipate that there is some kind of correlation between membership levels and broader participation patterns.

4 In some years in the period covered, the Danish first tier had 14 teams; inmost years it had 12.

5 In Copenhagen Municipality, in some years of the period covered, there were two to three teams in the first and second tiers. In these cases, point values have been aggregated into a single data point for Copenhagen. Nine data points – representing four municipalities in total – in the Point variable therefore have values above 26 (28).

6 We did not include a measure for education because we anticipated problems related to multicollinearity due to the correlation between education and income.

7 This law is called ‘The Act on Non-formal Education and Democratic Voluntary Activity’ (Danish: Folkeoplysningsloven). https://ec.europa.eu/epale/en/resource-centre/content/act-non-formal-education-and-democratic-voluntary-activity

8 It should also be noted that the ‘Leisure law’ does not determine the proportion of a budget municipalities should allocate for sport; this varies across the country. Each municipality can determine the level of expenditure allocated to sport.

9 This being said, the initial AR-B estimate with the first lag of Membership rate showed signs of second order serial correlation. To avoid this problem, also a second lag of the dependent variable was included in the model in addition to the variables described above.

10 An effect is found only at the 10 percent level due to participation in the UEFA Cup in FE Model 2.

11 Although they are related, they are not so closely correlated that multicollinearity was found to be an issue in the models.

Additional information

Funding

The research presented here has not received any funding.