Star quality and competitive balance? Television audience demand for English Premier League football reconsidered

ABSTRACT

The aim of this article is to investigate the determinants of television audience for English Premier League (EPL) football matches during the 2013–2014 season (n = 154). The model is adapted from Buraimo and Simmons (2015) who concluded that star quality was an important determinant of television audiences for the EPL over the 2000–2008 period, whereas uncertainty of outcome and championship, Europe and relegation contention were not. Their measures for contention are replaced by indicators based on Scelles et al.’s (2013b, 2016) competitive intensity. Results show a significant positive impact of star quality but also championship and Champions League intensity and no significant impact of Europa League, potential Europa League and relegation intensity. Based on these results, the article suggests that the EPL should encourage both star quality and competitive balance so that all teams can be competitive in terms of the title or qualification for the UEFA Champions League. Given the differences in revenue generation between English teams, the best way to achieve both star quality and competitive balance would be through the introduction of a European Super League. However, this conclusion based on television audience contradicts that of Scelles et al. (2016) based on stadium attendance.

KEYWORDS:

• Star quality
• competitive balance
• competitive intensity
• television audience
• English Premier League football

JEL CLASSIFICATION:

• L83
• L82

Acknowledgements

I am grateful to Babatunde Buraimo, Stephen Morrow and an anonymous referee for their very helpful comments. Errors are my own.

Disclosure statement

No potential conflict of interest is reported by the author.

Notes

¹ Following Buraimo and Simmons (2015), the Heckman selection model was applied to ensure that OLS estimates are not biased by the broadcasters’ choice of games to televise. No selection bias was found (tests available upon request). Contrary to Buraimo and Simmons (2015), team and month effects were not controlled for.

² With a value equal to 3 instead of 4 for the next game, similar results were found.