Discussion Paper

No. 2016-36 | July 11, 2016
Broadcasting Revenues and Media Value in European Football
(Published in Special Issue Recent Developments in Applied Economics)


Sport talent and popularity are major assets on which some businesses develop their economic activities. Professional football is one of the markets that, being among the principal industries of entertainment, depend on the skills of players. This paper uses media value ratings to appraise the sport talent and potential economic contribution of players and teams in European football. The empirical analysis shows also that sport performance and attainments are keystone elements to procure visibility in the media and to attract potential revenues.

The goal of this paper is twofold. Firstly, it investigates the factors that enhance exposure and media value of professional football clubs. The media value status and popularity are mainly driven by past and current sport performance. Besides, our media value indexes for teams and leagues, which are derived from individual appraisals, inform about their potential capacity to generate income. Secondly, the paper uses media value appraisals to address if they explain some sources of teams’ revenues. The paper actually explores to what extent TV rights are in accordance to the share of interest that each club draws from the fans and the media. Among other results, we find a strong empirical relationship between media value scores and either total revenues or broadcasting revenues. The authors´ empirical analysis also indicates that the broadcasting revenues in some of the top European domestic leagues were in the past below the level that according to their media value status one might expect.

JEL Classification:

J24, J33, J71


  • Downloads: 1438


Cite As

Pedro Garcia-del-Barrio and Francesc Pujol (2016). Broadcasting Revenues and Media Value in European Football. Economics Discussion Papers, No 2016-36, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2016-36

Comments and Questions

Anonymous - Referee Report 1
July 11, 2016 - 13:23

This is an interesting piece of research which provides novel evidence on the (intangible) value of professional teams in European football leagues based on mass media exposure of teams and players, and the effect that such media value has on actual total and broadcasting TV revenues. The key element of ...[more]

... the study is the computation of an index of the media value of each team by examining a huge number of news and websites. It has also merit the compilation of information about team’s revenues for the major football leagues in Europe. Overall, and despite the shortcomings that I will mention next, in my opinion the manuscript deserves publication.

There are some issues that the authors may want to address in the final version of the paper:

• The structure of the submitted version is rather chaotic. The manuscript would benefit from a more logical (and standard structure). For instance, the review of the literature and the framework of the analysis should follow the introductory section. This may include some comments about previous contributions on the estimation of intangible assets in general and of talent in sports and football in particular. In turn, the detailed description of the methodology should be moved to a section before the discussion of results. On the other hand, the stylised facts about revenues in European football should precede the more detailed descriptive analysis in section 5. Finally, a concluding section must be added.

• Section 4, on the data source and characteristics should be improved. Considering that it is a major contribution in the paper, this section must pay much more emphasis on the media value index. Among other things, it must indicate the type of information that was considered to compute de index and a detailed description of the results. In addition, all the variables used in the analysis must be clearly defined, in the main text or in a table, and the most important ones must be described.

• Tables 2 to 7 of the submitted version must be formatted following a professional style.

• The manuscript would benefit from a (brief) discussion on how the broadcasting rights are negotiated in each country and how this affects differences in the distribution of total revenues across countries. This is likely to affect the impact of media value on revenues, although country dummies in the regression analysis absorb this effect.

• A crucial issue in the regression analysis is that of endogeneity of the media value index. I am aware that the solution to that problem is far from trivial and perhaps is beyond the scope of this paper. Still, in my opinion the manuscript must include a discussion about it and the consequence it might have on the estimate of the effect of media value on revenues. A simple robustness check could use the lag of media value (for instance 5 year ago). In any case, comments (direct or indirect) regarding causality (from media value to revenues) should be avoided unless strong arguments supporting the exogeneity of media value are provided.

• From Figure 4, I wonder whether the results may be driven by teams in the upper-right area (ManCity, Roma, …, Real Madrid, FC Barcelona). It would be interesting to check the robustness of results when these teams are excluded from the sample.

• The first sentence of section 4 says that the panel contains 1342 observations, for 122 clubs and 10 seasons. However, multiplying the last two figures results in a number of observations that is lower than 1342. The manuscript must be clearer about this.

• It would be interesting to distinguish between local and international sources when computing the media value index. Among other things it might be that, for instance, some clubs are mentioned in local newspapers regardless of their performance. If so, the effect of local and international sources on media value and, thus, on revenues may be different.

• How should we interpret the zero value for the minimum revenues in Tables 2 and 3? Is this correct?

• Specifications of the empirical models used to obtain the estimates in Tables 7 to 9 are rather adhoc. They do not result from clear arguments on the determinants of media value and revenues (total and broadcast) and, as a consequence, the interpretation of the estimated effects is far from straightforward. The manuscript would benefit from a discussion on the factors that affect the media value of European football clubs and their revenues. Besides, there are some decisions that need to be motivated further. For instance, the models include it is not discussed the reason for including an AR(1) structure in the error term (versus an alternative dynamic specification) and it is not mentioned if unobserved team effects are controlled for (either by fixed or random effects). In fact, the estimates should control for unobserved team heterogeneity to guarantee appropriate estimates of the impact of media value on club’s revenues.

• The interpretation of some results should be revised. For instance, the one regarding differences across countries in the impact of media value in the paragraph before Table 8. On the other hand, comments on the main results in the paper should be extended in a revised version. As an example there are just a couple of sentences discussing results in Table 9.

Anonymous - Report 2
July 29, 2016 - 06:58

Report on Broadcasting Revenues and Media Value in European Football

The paper estimate media values for individual football players in the most important European leagues for seasons 1999/2000 to 2008/2009. Using this information, the authors explore how media value is influenced by different features of the club and the ...[more]

... tournaments where the club compete. They also estimate the impact of media value on teams revenue.
The estimation of media value in football and its impact on club revenues is a very relevant topic of reseach. It is also remarkable that they use an innovative approach based on internet traffic.
My biggest concern with the empirical analysis is that there is a potential problem of reverse causality between clubs media value and revenue. This problem, together with the fact that clubs revenue and media value are non stationary, suggests the convenience of considering a panel vector error correction model. This would provide a richer environment to estimate the long-run relationship between the two variables of interest and to apprise whether club revenue reacts to previos changes in media value or it is the other way around.
Some other points are:
1) Sections 4&5, devoted to descriptive statistics, are far too long while there is only one page for the discussion of the empirical analysis in Section 7. I suggest to eliminate some descriptive tables and figures that do not add much value to the article such as Figure 2 and 3 and tables that contains aggregated mean values of series that are very heterogeneous both along time and geographical areas.
2) Table 3. Why the minimum value of revenue is zero. Do you mean it is a missing value?
3) Why are there only observations until 2008/2009?
4) Tables are not self contained and it is difficult in many cases to know what the variables mean.
5) The paper should clarify its contribution with respect to previous papers by the same authors in similar topics.
6) It is also strange that the authors do not take advantage from a very rich database that contains information at the player level. For example, it would be very interesting to estimate the impact of the value of each player to the total revenue of the club. However, I understand this experiment would be beyond the initial scope of the paper.
Overall, it is a promising paper but I suggest the authors to pay special attention to the potential problem of reverse causality.

Anonymous - Paper review
August 09, 2016 - 13:36

The paper is interesting and innovative in terms of new proxies created to evaluate and measure intangibles in sport teams. The paper would deserve a journal publication if the following points are taken into consideration:

1) the literature review should consider the prevision studies on intangibles and consider what ...[more]

... is the innovative contribution of the current research.

2) as the media attention is increasingly expanding with the presence of social media, the time span should be expanded to 2015 in terms of data collection.

3) the structure of the paper is not fluent and it is chaotic and confusing.

4) the authors should focus more on the pros and cons of their MERIT system and illustrate how their work can be improved and suggest further research.

5) causality and endogeneity need to be taken into consideration before submitting a final version to an academic journal.

In itself with all the potential limits, this paper has an innovative approach that should be taken into consideration by any journal editor.

Anonymous - Referee Report 2
August 22, 2016 - 09:48

see attached file

Anonymous - Invited Reader Comment
August 23, 2016 - 08:43

see attached file