Printer Friendly

TV Rights, Financial Inequality, and Competitive Balance in European Football: Evidence from the English Premier League and the Spanish LaLiga.


In the period between 2016-2017 and 2018-2019, the new TV deals signed by two of the top football championships in Europe come into force. These new TV contracts will bring about a significant increase in the revenues from broadcasting rights affecting the financial differences among the clubs. In the case of the English Premier League (EPL), this means an increase of 71% compared to seasons 2013-2016, and in the case of the Spanish LaLiga, it is expected to be even greater. The Ligue 1 (France) has also signed a new TV contract for the period of 2016-2020, and in the case of Bundesliga (Germany) and Series A (Italy), the new TV deals will apply later.

TV rights are one of the main sources of revenue for clubs. For example, these revenues accounted for 38.8% of total EPL club revenue in the season 2014-2015 (24.5% for the club with a lower percentage and 83.6% for the club with a higher percentage). In LaLiga's case, for the same season, the percentage was 35.4% (23.8% for the club with a lower percentage and 81% for the club with a higher percentage).

Following Sanderson (2002), we should expect the modification of club economic resources to affect both the results and the competitive balance (CB) in these championships (Michie & Oughton, 2004; Andreff & Bourg, 2006). The seminal work of Rottenberg (1956) and the later developments by El-Hoidiri & Quirk (1971) and Fort & Quirk (1995) point out that in a situation where incomes are asymmetric among clubs, the introduction of revenue sharing rules does not affect the distribution of player talent and, consequently, it does not impact the competitive balance. All clubs reduce their interest in investing in the talent of players by decreasing the income obtained by these assets when these incomes are shared, and competitive balance will be unaffected. This proposition is known as the invariance principle and it is based on some key assumptions. Several authors have discussed that when some of the assumptions behind this principle are not satisfied, then it does not hold (Kesenne, 2005). This is the case, for instance, when the goal for the clubs is to maximize the winning probability subject to some budget constraint, rather than maximizing profits (Fort & Quirk, 2004). Consequently, we face different views in the literature about how revenue sharing is affecting competitive balance: no effect, improvement (Atkinson et al., 1981; Kesenne, 2006) or worsening (Szymanski & Kesenne, 2004). (1)

There is also some empirical evidence of the negative relationship between the difference among clubs' revenues and CB. For instance, Pawlowski, Breuer, & Hovemann (2010) reported a significant decrease in CB for the five top European national competitions after the modifications of the Champion League payout system in the season of 1999-2000. This modification of the payout system increased the economic differences between domestic clubs involved in the Champions League and those that did not participate. Additionally, in the medium and long term, the effect of the new TV contracts on competitive balance may be even greater, if we consider the potential relationship between the sports performance of a club and its future incomes. Szymanski & Kuypers (1999) and Samagaio, Couto, & Caiado (2011) show this empirical relationship in the EPL. In the case of LaLiga, Barajas, Fernandez-Jardon, & Crolley (2005) also provide evidence supporting this relationship. In fact, some professional and academic experts have highlighted the benefits of a "virtuous circle" between sporting and economic results (Szymanski & Kuypers, 1999; Soriano, 2009; Sass, 2012), but this also depends on the ability to manage these relationships and exploit their full potential (Frick & Simmons, 2008; Allen & Chadwick, 2012; Bell, Brooks, & Markham, 2013; Dimitropolous & Limperopoulos, 2014).

As mentioned, these new TV contracts for EPL and LaLiga are going to affect the financial situation of clubs and, consequently, the financial inequality (FI) in the championship. This could have an effect on the capacity of capturing talent and, consequently, on the CB. According to the seminal contributions by Rottenberg (1956) and Neale (1964), an excessively unbalanced competition could have a negative effect on the football consumer demand, i.e., football demand is increasing in the uncertainty of outcome. But the empirical evidence about the significance and relevance of this hypothesis, based on stadium attendance and TV audience studies, is not conclusive and, if anything seems to indicate that fan interests are more related to aspects such as the talent of the players, the certainty that their team will win, the prestige of the rival, or the competitive intensity, than to the outcome uncertainty (Garcia & Rodriguez, 2002; Pawlowski & Anders, 2012; Scelles, Durand, Bonnal, Goyeau, & Andreff, 2016; Buraimo & Simmons, 2016; Scelles, 2017).

As pointed out by Fort & Maxcy (2003), the lines of research associated with the uncertainty of outcome hypothesis and the analysis of competitive balance are (weakly) complementary. The first one is related to how competitive balance is affecting fan welfare and the second one is concerned with how competitive balance changes over time and, in particular, in relation to changes in the business approaches. The latter is the niche where this paper should be placed.

We focus our attention on how financial inequality affects competitive balance. In that sense, when commenting on the new system of distribution of TV rights for LaLiga, Javier Tebas, its president, claimed that "the new system of distribution of TV rights will help to make the championship more balanced." In this paper, we bring evidence that contradicts Javier Tebas's prediction. The new system worsens the competitive balance despite, as a result of a more uniform distribution of TV revenues, the financial inequality among Spanish clubs has been reduced in relative terms.

We analyze two cases: LaLiga, where the new deal supposes an increase in the total TV revenue and the system of distribution of TV rights has moved from an individual to a collective bargaining system, in which there is a proportion of the total amount uniformly distributed, and the EPL, where there is just an increase in the amount, but the system is maintained similarly to the new one implemented in LaLiga. We show that in both cases the new scenario for the TV rights is associated with a decrease in FI among clubs in relative terms, but with an increase of this inequality in absolute terms. Consequently, since we provide evidence that CB is negatively related to FI in absolute terms and, as mentioned above, FI increases under the new TV deals, more competitive imbalance should be expected in both competitions.

The article is structured as follows. Section 2 describes the recent evolution of the FI and the CB in both championships. Section 3 discusses the expected effect of the new TV contracts on the FI and illustrates these effects using data from season 2014-2015. In section 4, we present the estimation of an econometric model which explains the result of a single match in terms of the difference of the financial power of both teams. Section 5 shows the results of a simulation exercise, based on the previous econometric model, analyzing the effect of the TV revenue under the new deals on the CB of the season 2014-2015, the last season with the old system. The paper ends with a summary of the main conclusions.

The EPL and LaLiga: Recent Evidence about Competitive Balance and Financial Inequality

In the case of European football, it is common to measure competitive balance (CB) by the degree of disparity among the total points obtained by each club participating in a domestic league at the end of the season. This is an alternative to the common measures of CB in American professional sports, usually based on the winning percentage of the teams (Kesenne, 2006; Fort & Quirk, 2011), since the draw does not play a relevant role there, as it has in European football. Following Pawlowski et al. (2010), in this paper we use three indicators to assess the competitive balance of the season for EPL and LaLiga. (2)

Herfindahl-Hirshman Index (HHI): defined as the summation of the square of the share of points won by each club in the league with respect to the total points obtained for all clubs. Let [p.sub.i]. represent the total points obtained by team i at the end of a season. The points share of team i ([s.sub.i]) in a tournament composed by n rival clubs is defined as

[mathematical expression not reproducible] (1)

The HHI index is defined as

[mathematical expression not reproducible] (2)

In a fully balanced competition with 20 teams, as the EPL and LaLiga, this index would take a value of 0.05 (1/n) and its value would be 0.0684 if the tournament was fully unbalanced. This value corresponds with a distribution of points in which the champion wins all its matches, the second team wins all its matches except those played against the champions, and so on.

Concentration Ratio for the Top Five Clubs ([CR.sub.5]): based on the percentage of the points won by the top five clubs in the final standings with respect to the total points obtained for all clubs. In this case, we consider the concentration ratio for the five most powerful financial clubs in the championship, not as the ratio corresponding to the top five teams in the ranking, given our interest in analyzing the relationship between financial inequality and competitive balance among clubs. Thus, if we rank clubs according to their financial power from the most powerful (1) to the weakest (20) and [p.sub.i]. is the total points obtained by team i at the end of a season, then [CR.sub.5] is defined as

[mathematical expression not reproducible] (3)

In a fully balanced competition with 20 teams, this index would take a value of 0.25 (5/n) and its value would be 0.447 if the tournament was fully unbalanced, such as the case described above.

Coefficient of Variation (CV): this is a standardized measure of dispersion, and is defined as the ratio between the standard deviation and the average points obtained by each club. Thus, if [bar.p] is the mean of the points obtained for clubs in the tournament, then the CV is defined as

[mathematical expression not reproducible] (4)

In a fully balanced competition with 20 teams, this index would take a value of 0 because all clubs would obtain the same points, and its value would be 0.601 if the tournament was fully unbalanced, such as in the case described above.

For all three indicators, we also calculate a relative measure of competitive imbalance (RCI), defined as

[mathematical expression not reproducible] (5)

where I is the value of the indicator, [I.sub.max] is the maximum value the indicator can take (fully imbalanced), and [I.sub.min] is the minimum value (fully balanced).

The same indicators, except RCI, are used in order to measure the evolution of the financial inequality among football clubs. The club's total current revenues are used as a club measure of its financial power (transfer player revenues are not considered in this study because they are not current revenues). This data has been obtained from publicly available sources, such as publications of annual accounts from clubs and the specialized press.

In Table 1 and Table 2, we report the evolution of the indicators on CB and FI for both the EPL and LaLiga, respectively. (3) In both cases, there exists some statistical evidence of a decline in the relative balance of the current financial power of the clubs, which is more marked in the case of LaLiga, competition for which the level of FI is clearly higher than in the EPL, according to the three indicators. For instance, for the last season with complete information (2015-2016) the [CR.sub.5] index is 16 percentage points higher in LaLiga, and the CV is more than twice that of the EPL, as it is the HHI index. In terms of CB and for the last seasons, it is clear that there has been a decline in CB in LaLiga, with the exception of the last season in Table 2 (2015-16). (4) This evolution is clearly showing a more balanced competition in the EPL's case in the last three seasons, where the indicators of the CB show a negative trend. This evidence about CB is also confirmed when we look at the RCI values of the different indexes.

Additionally, from the information in Table 1 and Table 2 we can obtain some preliminary evidence about the association between financial inequality and competitive balance by looking at the correlation coefficients between these two variables for the different indicators we have defined. It is important to point out that for LaLiga the correlation coefficients are positive and very high (0.78 for the [CR.sub.5], 0.80 for the CV and 0.79 for the HHI index), whereas there is no significant correlation between these variables for the EPL (-0.02, -0.01, and -0.01, respectively).

The three measures we have considered are basically taking into account relative differences between the analyzed magnitudes for different clubs. This means that the indicator does not change if there is a proportional increase in the corresponding variable for all the clubs. This is because the basic element of these measures is either a relative magnitude (a share in the HHI index), the definition is a proportion (the [CR.sub.5] ratio), or it is standardized by an element (the average) that is affected in the same way as the basic element of inequality (the standard deviation) in the case of the CV. But these measures do not capture that the absolute differences between the corresponding magnitudes for the clubs also affected, for instance, by a proportional change. In fact, they will change in the corresponding proportion.

This is particularly relevant for the case of the financial inequality where the amount (the total revenues or a component as the revenues from TV rights) to be distributed can change (usually increase) and is not limited, as in the case of the total points in a competition, when considering competitive balance.

This different pattern followed by absolute and relative differences in front of a change in the amount (total revenues) to be distributed makes relevant the consideration of two additional measures of financial inequality, which will be important in the empirical analysis to be discussed later. We will make use of the idea behind the concept of semivariance in the finance literature, in the sense of comparing how different one club's revenue is compared to those clubs that are poorer (smaller revenue). These measures are (5)

Absolute Positive Financial Difference (APFD(i,j)) between a richer club (i) respect to a poorer club (j): it is the difference between the current revenues of a richer club ([R.sub.i]) respect to those of a poorer club ([R.sub.j]): this is

APFD(i,j) = ([R.sub.i] - [R.sub.j]) with [R.sub.i]. [greater than or equal to] [R.sub.j] (6)

Relative Positive Financial Difference (RPFD(i,j)) between a richer club (i) respect to a poorer club (j): it is the ratio between the current revenues of a richer club ([R.sub.i]) with respect to a poorer club ([R.sub.j]): this is

RPFD(i,j) = [R.sub.i]/[R.sub.j] with [R.sub.i] [greater than or equal to] [R.sub.j] (7)

The indicators we will define are the means of the absolute and the relative differences between club revenues, and both are positive and different from zero and from one, respectively.

Table 3 shows the evolution of these indicators (mean), including the standard deviation and the maximum value, in the period 2002-2016 for both championships (the values are expressed in millions of pounds for the EPL and in millions of euros for LaLiga). The absolute financial differences among clubs have increased in a very significant way in both championships, but the size of the differences and the increase are more important in LaLiga than in the EPL, and, in particular, there is more variability in these differences, as the standard deviation and the maximum value show. However, in relative terms, the relative differences have increased less significantly, and this increase and the size of these relative differences have been greater in the case of LaLiga, for which the total revenues of the top team can be more than 25 times that of the smallest team in financial terms.

In the case of LaLiga, it is relevant for our analysis to point out that the increasing pattern of the relative indicator is broken in the season 2015-2016, when the new system of distribution of TV rights is implemented and, at the same time, the absolute difference indicator maintains the positive trend. This pattern of the relative indicator is corroborating the evidence in Table 2 for the three indicators associated with the revenue variable, given that, as mentioned above, these indicators are basically capturing relative differences.

Expected Effect of New TV Contracts on the Revenues of EPL and LaLiga

A relevant aspect for our analysis is to quantify how the increased revenues from the new TV contracts will affect the financial differences among the clubs in the EPL and LaLiga, in order to analyze later how competitive balance is affected by the new financial scenarios.

In the case of the EPL, revenues from overseas TV rights are distributed equally among all clubs, whereas those from domestic TV rights are distributed as follows: 50% is divided in an equal way between all clubs (lump sum), 25% is distributed among 20 clubs according to final standings in the championship (merit), and 25% is given out to clubs for live televised matches in the UK (a fixed amount for each match, with the guarantee of a minimum amount allocated for each club).

In the case of LaLiga, the Real Decreto-ley 05/2015 (6) on the distribution of TV rights was assumed to have an effect on the season 2016-2017, but this new system was already used in the season 2015-2016. Previously, TV rights were managed by each club individually. Under the new system, the clubs in the first division will receive 90% of the total revenues from TV rights, (7) distributed as follows: 50% will be divided in an equal way between all clubs (lump sum), 25% will be distributed among the clubs according to a weighted average of the standings in the last five seasons (merit), and 25% according to a weighting of the clubs' social implantation (revenues from season ticket holders and attendance) and the participation in the televised matches. There are also some legal limits on the maximum and minimum percentages that each club may receive to ensure both an equitable distribution and that no club loses out when compared to its previous situation, but imposing a temporal limit for this guarantee.

Notice that the new scenarios with respect to the TV rights imply different changes in both competitions. In the case of the EPL, the change is basically an increase in the total amount of TV revenues, whereas in LaLiga there is a change in the system from individual to collective negotiation, with the corresponding implications in terms of the distribution, as well as an increase in the total amount to be distributed.

As it is shown in the appendix, when there is a collective negotiation of TV rights with a distributional rule as in the EPL, increases in the total TV revenues are translated into greater absolute differences in revenues, if the merit indicator of the richest team is higher than that of the poorest team. Moreover, the relative differences decrease whenever the ratio of the revenues of the richest team, excluding TV rights, and the poorest team is greater than the ratio of the TV revenues component associated to merits.

In the case of LaLiga, where it is not only an increase in the TV total revenues but also a change in the system of negotiating and distributing TV rights, as it is shown in the appendix, the new system can imply both positive and negative absolute differences. With respect to the changes in relative differences, in general, we expect them to be negative; it does not matter whether the change in absolute differences is positive or negative, given that there is a proportion of the TV revenues equally distributed among clubs.

We illustrate the previous considerations for both the EPL and LaLiga as follows. We take the season 2014-2015 and we compare the actual situation in that season with a corresponding scenario where the TV revenues--and the distribution system in the case of LaLiga--correspond to the values for the season 2016-2017. Table 4 and Table 5 summarize these results for both championships, showing how the differences would change in both absolute (columns at the bottom of the main diagonal) and relative (rows at the top of the main diagonal) terms. Given that clubs are ranked from highest to lowest revenues, a positive value means that the difference between richer and poorer clubs would increase with the new TV income, and a negative value means that this difference would decrease. For instance, in Table 4 when we compare revenue differences between Manchester United and Everton, the first club is richer; the figures show in absolute terms that the total revenue difference between Manchester United and Everton is going to increase by 7.8 million pounds under the new scenario. However, in relative terms, this difference will decrease by 0.49 points.

In the case of the EPL (Table 4), the results show how, as expected, the absolute differences between richer and poorer clubs generally tend to increase, even though these differences in relative terms will be reduced the greater the difference is in revenues between the clubs. Both effects are more important in the case of LaLiga (Table 5) with a particular feature. According to the new distribution system, the new scenario implies a reduction in both the absolute and relative differences between the two richest clubs (Real Madrid and FC Barcelona) and the rest. In general, the reduction in absolute differences are more important with respect to the richer teams. On the other hand, the reduction in the relative differences follows the expected pattern of being more important the greater the difference is in total revenues. In particular, these figures for LaLiga are much more important than those for the EPL as a consequence of a change in the distribution system. However, for the rest of the clubs in LaLiga, there will be an increase in the absolute differences between a richer club compared to a poorer one, and they are more important the more different the ranking of the clubs in terms of total revenues is. The differences in relative terms will also be reduced.

As a summary of this simulation exercise, the mean of the APFD for EPL in season 2014-2015 would have been 5.7% higher if the new TV contracts had been implemented in this season. In the case of LaLiga, this increase has been estimated at 2.2%. However, the mean of the RPFD in the EPL season 2014-2015 would have been 14.3% lower if the new TV contracts had been implemented in that season. This effect would have been more intense for LaLiga, as a consequence of the new distribution system, where the decrease is estimated at 36.2%. These results are corroborated when we look at how the standard measures (HHI, [CR.sub.5], and CV) change, when considering the TV revenues associated with the new scenario. Since they capture changes in relative terms, all three measures decrease (less FI) compared to the real situation: between 13% and 29% in the case of LaLiga and between 8% and 16% for the EPL.

To summarize, the results of this exercise indicate that the increase in revenues from TV rights of the new contracts, in general, will increase absolute differences between the revenues of the clubs, but will also reduce the relative ones, as it happens for the EPL. In the case of LaLiga, given the change in the distribution system, the increase in TV revenues translates into reductions in the relative differences, but there is a mixed effect for the absolute differences, negative for the two richest clubs and positive for the rest.

Since it is commonplace to approach the relationship between competitive balance and financial inequality, expressing the latter in relative terms (Szymanski, 2001; Samagaio et al., 2009) (8), and the previous analysis has shown that financial differences can change in a different way or magnitude, depending on whether they are regarded in absolute or relative terms, it is relevant to analyze what type of difference is more significant in terms of explaining the probabilities of the different results in a particular match and, in explaining the expected competitive balance.

An Econometric Model Linking the Result of a Match and Financial Inequality

In the previous section we illustrated how increasing the total TV revenues translates, in general, into an increase in absolute differences between the total revenues of two clubs and a decrease in the relative differences. With a few exceptions, similar opposite findings for the two types of differences are estimated when there is also a change in the system of negotiating and distributing TV revenues. Both absolute and relative differences are two alternative ways of specifying a variable that captures how similar the quality of the two teams are, when modeling the probabilities of the different results for a match. In fact, this last concept (the result of a match) is the basic element behind any measure of competitive balance.

The approach we follow is to estimate an econometric model where the dependent variable is the final result of a particular match, i.e., a categorical variable with three possible outcomes: home team win, away team win, and draw. Since we are not interested in a model to predict the result of a match (Forrest, Goddard, & Simmons, 2005), but rather in estimating how financial differences between the two clubs participating in a match can affect the final result, we do not include explanatory variables associated with specific conditions previous to each match (recent performance of both clubs, standing, whether the teams have injured or suspended players, etc.).

To perform the econometric analysis, we have financial information on the revenues of each club at the end of the season, as well as information on the outcome of each match. These data are available for both championships (the EPL and LaLiga) and they correspond to five seasons (2011-2012 until 2015-2016). The database is made up of 1,900 observations (380 matches for each season: 20 teams, with each team playing twice--once at home and once away--against each of the other 19 teams).

Since the dependent variable is a qualitative variable with three categories, we estimated a multinomial logit model, previously used in this literature associated with the estimation of the probability of a particular result in a match (Carpita, Sandri, Simonetto, & Zuccolotto, 2015). (9) For explanatory variables, we use the comparison of revenues of both clubs in each match (either in absolute or relative differences) with a quadratic profile, also including two dummy variables to capture whether the home team or the away team has been promoted in the corresponding season, trying to proxy the inexperience of a team in the competition.

The variable capturing the financial imbalance between both contenders presents a potential endogeneity problem, because club revenues generated in a particular season depend on their sport performance during the same season. To address this problem, we will use an instrumental variable: the transfer market value of the clubs (10) at the start of the season. Following Scelles, Helleu, Durand, & Bonnal (2014), this variable is correlated with each club's operating revenues (the correlation coefficient is 0.919 for the EPL, and 0.984 for LaLiga); this variable is not affected by the match performance in the season. We follow a two-step approach. First, we run an OLS regression of the club revenues on their market value as regressor, obtaining the adjusted revenues for each club using the estimated parameters. (11) In the second step, we use these adjusted values in order to define the corresponding financial variable to be used instead of the actual revenues in the discrete choice model. As discussed above, we consider financial differences in both absolute and relative terms.

For both specifications (absolute and relative differences between the adjusted total revenues of the home and the away teams) we found an overall negative effect of the financial imbalance variable in the probability of a home win. But, since the two measures behave differently in terms of changes in the TV revenues or the distribution system, it is relevant to compare the performance of both specifications. Given that these two models are non-nested, we do this comparison by means of the Akaike Information Criterion, which, in this case, reduces to the comparison of the values of the log-likelihood functions. These values are reported in Table 6 and, for both the EPL and LaLiga, the model that measures financial imbalance by means of the absolute differences is performing better.

In Table 7, we report the results of the estimated models using the absolute adjusted revenues difference between the local and the away clubs for both the EPL and LaLiga. The multinomial logit model is nonlinear and is not used to assess how the different explanatory variables affect the probabilities of a home win, an away win, or a draw. But from the estimated coefficients we can conclude that the fact that the away team promoted the corresponding season is going to have a positive affect on the probability of a home win in the EPL, which will be equally compensated by a reduction in the probability of a draw or an away win. In the case of LaLiga, the home team having promoted in the corresponding season is going to have a significant negative effect in the probability of a home win.

As mentioned above, there is a negative significant relationship between the revenue differences of contenders and the probability of a home win. In order to illustrate the results, Figure 1 and Figure 2 show the estimated probabilities for the three outcomes, as a function of the absolute differences in total revenues between the home and the away club for both championships, respectively, in a situation where neither club has been promoted. The profile is very much the same for both winning probabilities, i.e., the probability of a home win increases with the absolute difference and the probability of an away win decreases and the probability of a draw decreases when the absolute differences (in absolute value) increase. (12)

We can also provide some evidence about the role of home advantage. First, we can observe that the same absolute differences in revenues in favor of the home team or the away team are associated with different probabilities of a win (higher in the case of a home win). For instance, in the case of LaLiga, an absolute difference of 300 million euros in favor of the home team has an adjusted probability of a home win of 0.81, but if this difference is in favor of the away team, the adjusted probability of an away win is 0.56. On the other hand, we can identify in both graphics a value for the absolute difference, which can be associated with a situation where the three probabilities are almost equal. This value is around -99 million euros for LaLiga and -45 million pounds for the EPL. In other words, the home advantage compensates for a difference in quality (talent) in favor of the visiting team, which can be associated with 99 million euros in the case of LaLiga and 45 million pounds in the case of EPL.

A Simulation Exercise: How the Competitive Balance in the 2014-2015 Season Changes when Considering the New TV Contracts of the 2016-2017 Season

The multinomial logit model estimated in the previous section is used to carry out a simulation exercise of how the change in TV revenues (amount or system) can modify the CB of both championships. For both the EPL and LaLiga, we will predict from the estimated model the outcome of each game under the new TV revenues the random errors generated by using the actual TV revenues, and the fact that the random component of the utility associated with each result of a game in a multinomial logit model is independently and identically distributed as an Extreme Value Type-I random variable. These random errors are generated in such a way that the model reproduces the observed results when using the actual revenues in season 2014-2015.

Once we have obtained these predictions, we can calculate the predicted number of points at the end of the season and, then, we can work out the different measures of competitive balance under the new situation and compare them with those corresponding to the initial scenario. In order to guarantee the robustness of this simulation exercise, we carried out a kind of Monte Carlo exercise by generating 1,000 replications of the random terms associated with the three alternatives of the dependent variable for all the observations.

For both competitions, the initial (or reference) scenario corresponds to the 2014-2015 season. The new scenario corresponds to a situation for which the total TV revenues are estimated for the 2016-2017 season. The associated distribution rule was applied to teams competing in the EPL and LaLiga in the season 2014-2015, keeping invariant revenues from other sources. This means that the estimated total TV revenues are increased by 66% for the EPL and by 79% for LaLiga.

In Table 8 we report the actual performance of the teams in terms of the number of points obtained in both the EPL and LaLiga at the end of the season 2014-2015 (Real) and the average of the expected number of points from 1,000 replications under the new scenario (TV revenues of season 2016-2017) (Simulation). It is immediately clear, by looking at the third and the sixth columns, where we report the changes in the points obtained (Difference), that the impact of the new situation is more significant for the EPL than for LaLiga. The total change in points in the EPL represents more than 10% of the total points obtained in the 2014-2015 season, whereas in the case of LaLiga, this percentage is just below 3%. This is because the change in the system of distribution in LaLiga to some extent compensates for part of the effect of the increase in the total amount of TV revenues.

When looking at the competitive balance measures for both competitions, we find an increase in the competitive imbalance, much more significant in the case of the EPL where the rates of change of the different measures are higher than in LaLiga. In particular, it is relevant to qualify the change in the CV, whose range of variation is between 0 (full balance) and 0.601 (maximum imbalance). This means that the change of this index for the EPL represents an increase of 6 percentage points in the ratio of relative competitive imbalance (RCI) defined in third section. In the case of LaLiga this change amounts to 2 percentage points. This different pattern in both competitions is explained by the fact than in the EPL the simulated change in TV rights is just an increase in the total amount but keeping the same distribution rule, as discussed before, and this kind of change translates into higher absolute differences. But in the case of LaLiga, the simulated change also implies a change in the distribution rule and the total effect in the absolute differences is not uniform, combining increases and reductions in the absolute differences.

Similar results are obtained if the simulation exercise is based on calculating the expected probabilities of each result from the previous model using both the total real revenues and the simulated ones, and calculating the expected points to be obtained in each game. The results go in the same direction if we carry out 1,000 replications based on random drawings for the final result based on the estimated probabilities for each result of a match.

Consequently, we can conclude that the change in the distribution of TV revenues proposed by LaLiga is not resulting in an increase of competitive balance as expected by its president. Just the opposite. Additionally, according to the simulations of the EPL, future increases in the TV revenues will translate into more competitive imbalance in both leagues. The reason behind this result is that the absolute, and not the relative, difference in quality of the teams, as approximated by the total revenues, is what seems to matter in explaining the result of a match and, consequently, the final result of the competition, which is the basic input to measure CB.


In season 2016-2017, both LaLiga and the EPL have signed new TV deals for the commercialization of TV rights of championships, starting in the 2016-2017 season. In the case of LaLiga, the new deal combines an increase in the total amount of TV revenue with the move from an individual to a collective bargaining of these rights, which implies that a proportion of the total revenue is shared uniformly among all clubs. In the case of the EPL, this kind of collective agreement was already in force and the new deal is basically associated with a substantial increase in the amount of TV rights.

Considering the distribution systems of LaLiga and the EPL in season 2014-2015, we have provided formal and empirical evidence to show that the new contracts coming into force during the season 2016-2017 are associated with an increase in the TV revenue of each club, but also to changes in the financial inequality among clubs. It is expected that, in general, the total revenue differences between richer and poorer clubs will increase in absolute terms and decrease in relative terms.

On the other hand, using data from both competitions for seasons 2010-2016, we have estimated an econometric model explaining how the result of a particular game (and, by aggregation, the final points and standings at the end of the championship) is affected by the difference in quality of both teams, approximated by their total revenue. We conclude that the performance of the estimated models is substantially better for both the EPL and LaLiga when we use absolute differences instead of relative financial differences as a measure of quality differences between the teams, and this relationship is negative, implying that the greater the financial differences, the more imbalanced the competition.

We use the estimated model to simulate how the new financial situation of the clubs with the new TV deals will affect competitive balance. Since the new deals increase financial inequality in absolute terms, the competitive balance will worsen in both competitions. In particular, the evidence from EPL is showing that future increases in TV revenue will further worsen the competitive balance. On the other hand, the change in bargaining system for TV rights in LaLiga is not producing the expected positive impact in CB, as the president of LaLiga announced, but rather a negative one, and will suffer in the future (the greater the TV revenue) the same negative effects in CB as the EPL has experienced.

Looking at the sources of revenue for the clubs, TV rights is one of them, but not the only one. Match day income and commercial revenues are also important, and each of them have different features in terms of how managers can proceed in order to try to increase them. In fact, TV revenue is less flexible in terms of the manager's incidence than the other two sources given the usual collective bargaining of these rights. In fact, when one looks at the distribution of total revenue among the top 20 football clubs in Europe in season 2015-2016 (Deloitte, 2017), we can see that commercial revenue accounts for 43% of the total revenue in front of the 39% associated with TV rights and 18% for match-day revenue. Consequently, an extension of this research is to look at how the evolution of this type of revenue for the different clubs will affect competitive balance, probably in a more intense way than TV rights. This type of analysis would be more relevant for LaLiga given that, for the same season 2015-2016, commercial revenue represented 40% of total revenue for the two top clubs (FC Barcelona and Real Madrid) and only 13% for the remaining clubs, and the gap, in absolute terms, is increasing (LaLiga, 2016).

On the other hand, we have focused on the differences in the financial power of clubs. Not only the size of these differences is important, however, the management of financial power is also a determining factor. Clubs with lower financial resources can perform more efficient management and limit the differences caused by the lower financial power. This is an issue to be addressed in future work.

Finally, this article has considered static (within season) measures of CB. It would be interesting to extend the analysis using dynamic (across seasons) CB measures in order to study long-term effects.


Allen, W. D., & Chadwick, C. (2012). Performance, expectations and managerial dismissal. Journal of Sports Economics, 13(4), 337-363.

Andreff, W., & Bourg, J. F. (2006). Broadcasting rights and competition in European football. In C. Jeanrenaud, & S. Kesenne (Eds.), The economics of sport and the media (pp. 37-71). Cheltenham, UK: Edward Elgar Publishing.

Atkinson, S.E., Stanley, L.R., & Tschirhart, J., (1988). Revenue sharing as an incentive in an agency problem: an example from the National Football League. RAND Journal of Economics, 19(1), 27-43.

Barajas, E., Fernandez-Jardon, C., & Crolley, L. (2005). Does sports performance influence revenues and economic results in Spanish football? (MPRA paper n[degrees] 3234). Retrieved from University of Munich, Munich Personal RePEC Archive https://mpra.ub.uni-muenchen. de/3234/

Bell, A., Brooks, C., & Markham, T. (2013). The performance of football club managers: Skill or luck? Economics & Finance Research, 1(1), 19-30.

Buraimo, B., & Simmons, R. (2016). Uncertainty of outcome or star quality? Television audience demand for English Premier League Football. International Journal of the Economics of Business, 22(3), 449-469.

Carpita, M., Sandri, M., Simonetto, A., & Zuccolotto, P. (2015). Discovering the drivers of football match outcomes with data mining. Quality Technology & Quantitative Management, 12(4), 561-577.

Deloitte (2017), Planet Football. Football Money League. Sports Business Group, Deloitte. Dimitropoulos, P. E., & Limperopoulos, V. (2014). Player contracts, athletic and financial performance of the Greek football clubs. Global Business and Economic Review, 16(2), 123-141.

El-Hodiri, M., & Quirk, J. (1971). An economic model of a professional sports league. Journal of Political Economy, 79(6), 1302-1319.

Forrest, D., Goddard, J., & Simmons, R. (2005) Odds-setters as forecasters: The case of English football. International Journal of Forecasting, 21(3), 551-564.

Fort, R., & Maxcy, J. (2003). Comment on "Competitive balance in sports league: An introduction." Journal of Sports Economics, 4(2), 154-160.

Fort, R., & Quirk, J. (1995), Cross-subsidization, incentives, and outcomes in professional team sports leagues. Journal of Economic Literature 33(3), 1,265-1, 299.

Fort, R., & Quirk, J. (2004), Owner objectives and competitive balance. Journal of Sports Economics, 5(1), 20-32.

Fort, R., & Quirk, J. (2011). Optimal competitive balance in a season ticket league. Economic Inquiry, 49(2), 464-473.

Frick, B., & Simmons, R. (2008). The impact of managerial quality on organizational performance: Evidence from German soccer. Managerial and Decision Economics, 29(7), 593-600.

Garcia, J., Perez, L., & Rodriguez, P. (2017). Forecasting football match results: Are the many smarter than the few? In P. Rodriguez, B. R. Humphreys and R. Simmons (eds.), The Economics of Sports Betting (pp. 71-91), Cheltenham, UK: Edward Elgar Publishing.

Garcia, J., & Rodriguez, P. (2002). The determinants of football match attendance revisited. Empirical evidence from the Spanish football league. Journal of Sports Economics, 3(1), 18-38.

Groot, L. F. M. (2008). Economics, uncertainty and European football: Trends in competitive balance. Cheltenham (UK) & Northampton MA (USA): Edward Elgar Publishing.

Humphreys, B. R. (2002). Alternative measures of competitive balance in sports leagues. Journal of Sports Economics, 3(2), 133-148

Kesenne, S. (2005), Revenue sharing and competitive balance--Does the invariance proposition hold? Journal of Sports Economics, 6(1), 98-106.

Kesenne, S. (2006). Competitive balance in team sports and the impact of revenue sharing. Journal of Sports Management, 20(1), 39-51.

LaLiga (2016), Informe economico-financiero del futbol profesional. LaLiga.

Michie, J., & Oughton, C. (2004). Competitive balance in football: Trends and effects. London: The Sports Nexus.

Montes, F., Sala-Garrido, A., & Usai, A. (2014). The lack of balance in the Spanish First Division football league. European Sport Management Quarterly, 14(3), 282-298.

Neale, W. C. (1964). The peculiar economics of professional sports: a contribution to the theory of the firm in sporting competition and in market competition. Quarterly Journal of Economics, 78(1), 1-14.

Pawlowsky, T., Breuer, C., & Hovemann, A. (2010). Top club's performance and the competitive situation in European domestic football competitions. Journal of Sports Economics, 11(2), 186-202.

Pawlowski, T., & Anders, C. (2012). Stadium attendance in German professional football: the (un)importance of uncertainty of outcome reconsidered. Applied Economics Letters, 19(16), 1553-1556.

Pawlowski, T., & Budzinski, O. (2013). The monetary value of competitive balance for sport consumers: A stated preference approach to European professional football. International Journal of Sport Finance, 8(2), 112-123.

Pedersen, C. S., & Satchell, S. E. (1998). An extended family of financial-risk measures. The Geneva Papers on Risk and Insurance Theory, 23, 89-117.

Rottenberg, S. (1956). The baseball player's labour market. Journal of Political Economy, 64(3), 242-258.

Samagaio, A., Couto, E., & Caiado, J. (2009). Sporting, financial and stock market performance in English football: An empirical analysis of structural relationships (CEMAPRE Working Paper n[degrees] 6). Retrieved from Technical University of Lisbon:

Sass, M. (2016). Glory-hunters, sugar daddies and long-term competitive balance. Journal of Sports Economics, 17(2), 148-158.

Sanderson, A. R. (2002). The many dimensions of competitive balance. Journal of Sports Economics, 3(2), 204-228.

Scelles, N. (2017). Star quality and competitive balance? Television audience demand for English Premier League football reconsidered. Applied Economic Letters, 24(19), 1,399-1,402.

Scelles, N., Durand, C., Bonnal, L., & Andreff, W. (2013). Competitive balance versus competitive intensity before a match: Is one of these concepts more relevant in explaining attendance? The case of the French Football Ligue 1 over the period 2008-2011. Applied Economics, 45(29), 4,184-4,192.

Sloane, P. (2006). Rottenberg and the economics of sports after 50 years: An evaluation. In P. Rodriguez, S. Kesenne, & J. Garcia (eds.), Sports economics after fifty years: Essays in honour of Simon Rottenberg. (pp. 211-226). Oviedo: Ediciones de la Universidad de Oviedo.

Soriano, F. (2009). La pelota no entra por azar. Barcelona, Spain: El lector universal.

Szymanski, S. (2001). Income inequality, competitive balance and the attractiveness of team sports: Some evidence and a natural experiment from English soccer. The Economic Journal, 111(469), 69-84.

Szymanski, S. & Kesenne, S. (2004). Competitive balance and gate-revenue sharing in team sports. Journal of Industrial Economics, 52(1), 165-177

Szymanski, S., & Kuypers, T. (1999). Winners and losers: The business strategy of football. Middlesex, Penguin Books Limited.


(1) See Sloane (2006) for a discussion of this literature on the invariance principle and the different views on what he refers to as "unambiguous text-book prediction."

(2) See Humphreys (2002) for an alternative measure of competitive balance, which captures variation across seasons.

(3) See Groot (2008) and Pawlowski et al. (2010) for analyses of the evolution of competitive balance for EPL and Montes, Sala-Garrido and Usai (2014) for LaLiga.

(4) The increasing pattern seems to be recovered in the last season 2016-2017, for which, for instance, the coefficient of variation is 0.392.

(5) See, for instance, Pedersen and Satchell (1998) for a survey of the financial-risk measures, including the semivariance and the parallel literature on inequality measurement in economics to which some traditional measures of competitive balance belong.


(7) The remaining 10% is distributed among the clubs in the second division. All the clubs have to contribute 7% of their TV revenue to a fund which is distributed as follows: an amount to compensate those teams that are relegated (3.5%), the Liga de Futbol Profesional (1%), and private and public institutions linked with Spanish football (2.5%).

(8) Szymanski (2001) uses the coefficient of variation in order to describe the financial inequality of English professional football. Samagaio et al. (2009) use a structural equation model where the variables are standardized.

(9) Another alternative is to use the ordered logit model (Forrest et al., 2005; Garcia, Perez, & Rodriguez, 2017), which is a particular case of the multinomial logit model. The econometric analysis was replicated using this model and the results were very similar, but the fit was worse.


(11) The estimated model is available on request.

(12) The effects on the probability of both a home and an away win are consistent with the quadratic profile, since in both cases the maximum is outside the range of variation of the financial imbalance variable.

(13) Notice that these changes are associated with an increase of 5.7% in the mean of absolute differences in revenue in the case of EPL, and of 2.2% for LaLiga, as mentioned in third section.


TV Revenues, Sharing System, and the Absolute and Relative Financial Differences Among Clubs

Consider that club 1 is an entity with greater financial power than club 2 and [a.sub.1] and [a.sub.2] (with [a.sub.1] > [a.sub.2]) are the respective revenues, excluding those coming from TV rights (M). These TV revenues are distributed in two parts as follows: i) proportion a is divided in an equal way between all clubs (N) and ii) the remaining proportion (1 - [alpha]) follows a nonproportional criterion (for instance, merit or attendance), so [n.sub.1] and [n.sub.2] are respectively the share of these nonproportional revenues corresponding to clubs 1 and 2.

We define [ADB.sub.1,2] and [RDB.sub.1,2] as the revenue differences in absolute and relative terms between both clubs before the TV revenue distribution, respectively:

[ADB.sub.1,2] = [a.sub.1]--[a.sub.2] and [RDB.sub.1,2] = [a.sub.1]/[a.sub.2] (A.1)

In a similar way, we define [ADA.sub.12] and [RDA.sub.12] as the revenue differences in absolute and relative between both clubs after the TV revenue distribution, respectively:

[ADA.sub.1,2] = [a.sub.1] - [a.sub.2] + (1- [alpha]) M([n.sub.1] - [n.sub.2]) (A.2)

[mathematical expression not reproducible] (A.3)

It is immediate to show that [ADA.sub.1,2] is an increasing lineal function on M revenues (M) if [n.sub.1] > [n.sub.2].

By considering the derivative of [RDA.sub.1,2] with respect to M in expression (A.3), we can see that [RDA.sub.1,2] is a decreasing function on M if [n.sub.1]/[n.sub.2] [less than or equal to] [a.sup.1]/[a.sub.2] and [a.sub1] [greater than or equal to] [a.sub.2].

[mathematical expression not reproducible] (A.4)

This means that the nonproportional TV revenue portion is distributed in a more balanced way than the incomes from alternative sources to the TV rights, which is a plausible general assumption.

In the case of LaLiga, the change is not only associated to an increase in TV revenues but also in the system of distribution from individual bargaining to collective bargaining, with precise rules to be applied as mentioned above. In this case, and defining [t.sub.1] and [t.sub.2] as the TV revenues previous to the change in the system, the change in absolute differences ([CADA.sub.1,2]) is given by:

[CADA.sub.1,2] = (1 - [alpha])M([n.sub.1] - [n.sub.2]) - ([t.sub.1] - [t.sub.2]) (A.5)

which can be either positive or negative. For a change not implying an increase in the total amount of TV rights (M), the change is going to be positive whenever the difference in TV revenues prior to the change is smaller than a proportion 1 - [alpha] of the difference in TV revenues, which would be obtained according to a merit criterion [Mn.sub.1] - [Mn.sub.2]. If the total amount of TV rights also increases by a proportion 5, this condition is more easily satisfied since now the proportion would be not 1 - [alpha] but (1 - [alpha])(1 + [delta]).

The change in relative differences ([CRDA.sub.1,2]) is given by:

[mathematical expression not reproducible] (A.6)

where [mathematical expression not reproducible] is the TV revenues of the team i after the change in the TV rights distribution, [mathematical expression not reproducible].

This change in relative differences is negative whenever

[mathematical expression not reproducible] (A.7)

i.e., whenever the relative increase in the total revenue of the richest team--when there is a change in the TV rights distribution system--is smaller than that corresponding to the poorest team, which is expected to be the usual situation.

Authors' Note

We wish to thank the associate editor, two anonymous referees, and participants of the XI Gijon Conference on Sports Economics for very helpful comments and suggestions. The first author wishes to express his gratitude for financial support from grant MPCUdG2016. The second author wishes to express his gratitude for financial support from grant MINECO-ECO2014-52238-R.

Miquel Carreras (1) and Jaume Garcia (2)

(1) Universitat de Girona

(2) Universitat Pompeu Fabra

Miquel Carreras-Simo, PhD, is full professor of business economics in the Department of Economics at Universitat de Girona. His interest are industrial organization, financial and economic analysis of firms, and sport economics.

Jaume Garcia is professor of applied economics in the Department of Economics and Business Universitat Pompeu Fabra. He was formerly president of the Spanish Statistics Office. His interests are sports economics, applied microeconometrics, labor economics, health economics, housing economics, and individual behavior.
Table 1. EPL. Evolution of Competitive Balance (CB) and Financial
Inequality (FI) (2002-2016)

             HHI                   [CR.sub.5]            CV
             FI      CB     RCI    FI      CB     RCI    FI

2002-03       0.067  0.054  0.220   0.462  0.344  0.475   0.593
2003-04       0.065  0.054  0.223   0.459  0.349  0.501   0.558
2004-05       0.062  0.055  0.284   0.447  0.347  0.489   0.504
2005-06       0.063  0.056  0.316   0.451  0.358  0.549   0.519
2006-07       0.069  0.054  0.239   0.498  0.353  0.523   0.631
2007-08       0.065  0.057  0.371   0.467  0.363  0.570   0.558
2008-09       0.070  0.056  0.315   0.497  0.366  0.589   0.653
2009-10       0.083  0.056  0.312   0.556  0.360  0.558   0.838
2010-11       0.073  0.053  0.159   0.508  0.329  0.403   0.693
2011-12       0.073  0.055  0.286   0.521  0.348  0.495   0.694
2012-13       0.075  0.056  0.307   0.540  0.364  0.579   0.731
2013-14       0.070  0.056  0.340   0.509  0.372  0.618   0.654
2014-15       0.069  0.055  0.251   0.506  0.356  0.538   0.632
2015-16       0.072  0.054  0.230   0.520  0.303  0.269   0.687
Correlation  -0.016                -0.041                -0.013
CB and FI

             CB     RCI

2002-03      0.292  0.457
2003-04      0.294  0.460
2004-05      0.332  0.519
2005-06      0.350  0.548
2006-07      0.304  0.476
2007-08      0.380  0.594
2008-09      0.350  0.547
2009-10      0.348  0.544
2010-11      0.248  0.389
2011-12      0.333  0.521
2012-13      0.345  0.540
2013-14      0.363  0.568
2014-15      0.312  0.489
2015-16      0.299  0.468
CB and FI

Table 2. LaLiga. Evolution of Competitive Balance (CB) and Financial
Inequality (FI) (2002-2016)

             HHI                  [CR.sub.5]                  CV
             FI     CB     RCI    FI     CB     RCI    FI     CB

2002-03      0.107  0.053  0.157  0.610  0.330  0.408  1.095  0.246
2003-04      0.119  0.053  0.148  0.621  0.322  0.362  1.207  0.239
2004-05      0.138  0.054  0.197  0.663  0.311  0.307  1.364  0.276
2005-06      0.135  0.054  0.207  0.646  0.319  0.349  1.340  0.284
2006-07      0.138  0.053  0.172  0.664  0.335  0.430  1.364  0.258
2007-08      0.129  0.053  0.189  0.647  0.314  0.326  1.292  0.271
2008-09      0.143  0.054  0.195  0.669  0.340  0.454  1.400  0.275
2009-10      0.147  0.056  0.327  0.701  0.360  0.556  1.432  0.356
2010-11      0.163  0.055  0.258  0.716  0.353  0.524  1.543  0.316
2011-12      0.169  0.055  0.264  0.714  0.341  0.463  1.580  0.320
2012-13      0.161  0.055  0.291  0.703  0.363  0.571  1.526  0.336
2013-14      0.160  0.056  0.311  0.707  0.357  0.541  1.518  0.347
2014-15      0.170  0.058  0.415  0.729  0.378  0.647  1.591  0.401
2015-16      0.144  0.056  0.308  0.684  0.365  0.585  1.404  0.345
Correlation  0.785                0.777                       0.801
CB and FI


2002-03      0.386
2003-04      0.374
2004-05      0.432
2005-06      0.444
2006-07      0.405
2007-08      0.424
2008-09      0.430
2009-10      0.557
2010-11      0.495
2011-12      0.501
2012-13      0.526
2013-14      0.543
2014-15      0.628
2015-16      0.541
CB and FI

Table 3. Evolution of Absolute Positive Financial Difference (APFD)
and Relative Positive Financial Difference (RPFD) of EPL and LaLiga

         EPL                                              LaLiga
         APFD                      RPFD                   APFD
         Mean    St. Dev.  Max.    Mean  St. Dev.  Max.   Mean

2002-03   34.46   36.30    144.60  1.86  0.95       6.09   40.25
2003-04   35.00   36.20    131.10  1.78  0.84       4.45   47.52
2004-05   33.37   28.85     88.90  1.80  0.74       3.64   55.34
2005-06   34.62   30.98     97.30  1.80  0.77       3.80   60.70
2006-07   45.56   41.36    138.40  2.11  1.12       6.14   72.30
2007-08   52.09   46.90    147.30  1.90  0.88       4.44   78.12
2008-09   63.07   63.05    210.80  2.03  1.10       5.55   86.71
2009-10   86.73   80.38    302.10  3.53  3.77      27.97   96.83
2010-11   80.49   78.53    280.00  2.17  1.24       6.49  103.76
2011-12   82.32   81.31    267.00  2.17  1.24       6.04  109.99
2012-13   93.79   93.72    307.00  2.27  1.36       6.48  110.40
2013-14  107.35  106.67    350.10  2.05  1.11       5.21  114.10
2014-15  108.70  104.26    316.00  2.05  1.07       4.99  127.90
2015-16  124.66  125.91    427.40  2.12  1.20       5.86  140.30

         APFD              RPFD
         St. Dev.  Max.    Mean  St. Dev.  Max.

2002-03   49.78    181.00  3.11  2.86      16.08
2003-04   64.38    222.40  3.26  3.33      17.35
2004-05   79.89    265.50  3.69  4.26      26.29
2005-06   89.81    275.00  3.42  3.81      17.18
2006-07  105.25    333.80  3.58  4.04      20.41
2007-08  107.95    350.00  3.60  3.96      22.88
2008-09  125.42    385.70  3.91  4.65      26.21
2009-10  137.36    420.00  4.17  4.92      24.33
2010-11  155.90    462.00  4.42  5.60      26.67
2011-12  168.57    494.00  4.53  5.92      27.85
2012-13  169.14    501.00  4.16  5.26      26.05
2013-14  168.21    497.60  4.43  5.55      25.04
2014-15  193.13    557.90  4.71  6.14      30.52
2015-16  203.73    588.10  3.79  4.34      19.38

Table 4. EPL. Differences in APFD and RPFD between Actual Data (season
2014-2015) and the New Scenario (TV Rights Season 2016-2017)

Club            ManU  ManC   Ars.   Chel.  Liverp.  Totten.  Newc.

Manchester U          -0.02  -0.02  -0.04  -0.04    -0.18    -0.42
Manchester C    -1.3          0.00  -0.02  -0.02    -0.13    -0.34
Arsenal         -0.3   1.0          -0.02  -0.01    -0.13    -0.33
Chelsea         -1.8  -0.5   -1.5           0.00    -0.09    -0.27
Liverpool        2.0   3.3    2.3    3.8            -0.09    -0.26
Tottenham        3.0   4.3    3.3    4.8    1.0              -0.08
Newcastle       10.0  11.3   10.3   11.8    8.0      7.0
Everton          7.8   9.1    8.1    9.6    5.8      4.8     -2.2
West Ham         9.5  10.8    9.8   11.3    7.5      6.5     -0.5
Aston Villa      9.0  10.3    9.3   10.8    7.0      6.0     -1.0
Southampton      8.6   9.9    8.9   10.4    6.6      5.6     -1.4
Leicester        7.3   8.6    7.6    9.1    5.3      4.3     -2.7
Swansea         10.0  11.3   10.3   11.8    8.0      7.0      0.0
Crystal Palace   9.8  11.1   10.1   11.6    7.8      6.8     -0.2
Sunderland       9.5  10.8    9.8   11.3    7.5      6.5     -0.5
Stoke City      13.6  14.9   13.9   15.4   11.6     10.6      3.6
W. Bromwich     11.1  12.4   11.4   12.9    9.1      8.1      1.1
QPR             16.1  17.4   16.4   17.9   14.1     13.1      6.1
Hull City       14.8  16.1   15.1   16.6   12.8     11.8      4.8
Burnley         15.6  16.9   15.9   17.4   13.6     12.6      5.6

Club            Evert.  W.H.   A.V.   South  Leices.  Swan.  C.P.

Manchester U    -0.49   -0.49  -0.55  -0.60  -0.75    -0.70  -0.73
Manchester C    -0.39   -0.39  -0.45  -0.49  -0.62    -0.58  -0.60
Arsenal         -0.38   -0.38  -0.44  -0.47  -0.60    -0.56  -0.58
Chelsea         -0.31   -0.31  -0.36  -0.40  -0.51    -0.47  -0.49
Liverpool       -0.31   -0.31  -0.35  -0.38  -0.49    -0.46  -0.48
Tottenham       -0.11   -0.11  -0.14  -0.16  -0.22    -0.19  -0.21
Newcastle       -0.02   -0.02  -0.03  -0.04  -0.09    -0.07  -0.07
Everton                  0.00  -0.01  -0.02  -0.06    -0.04  -0.05
West Ham         1.7           -0.02  -0.03  -0.06    -0.05  -0.05
Aston Villa      1.2    -0.5           0.0   -0.05    -0.03  -0.03
Southampton      0.8    -0.9   -0.4          -0.03    -0.02  -0.02
Leicester       -0.5    -2.2   -1.7   -1.3             0.02   0.01
Swansea          2.2     0.5    1.0    1.4    2.7            -0.01
Crystal Palace   2.0     0.3    0.8    1.2    2.5     -0.2
Sunderland       1.7     0.0    0.5    0.9    2.2     -0.5   -0.3
Stoke City       5.8     4.1    4.6    5.0    6.3      3.6    3.8
W. Bromwich      3.3     1.6    2.1    2.5    3.8      1.1    1.3
QPR              8.3     6.6    7.1    7.5    8.8      6.1    6.3
Hull City        7.0     5.3    5.8    6.2    7.5      4.8    5.0
Burnley          7.8     6.1    6.6    7.0    8.3      5.6    5.8

Club            Sund.  Stoke  W.B.   QPR    Hull   Burnl.

Manchester U    -0.75  -0.68  -0.80  -0.87  -0.94  -1.05
Manchester C    -0.62  -0.56  -0.66  -0.72  -0.78  -0.88
Arsenal         -0.61  -0.54  -0.64  -0.70  -0.76  -0.85
Chelsea         -0.51  -0.45  -0.55  -0.59  -0.65  -0.73
Liverpool       -0.50  -0.44  -0.53  -0.57  -0.62  -0.70
Tottenham       -0.22  -0.18  -0.24  -0.25  -0.28  -0.33
Newcastle       -0.08  -0.05  -0.09  -0.09  -0.11  -0.14
Everton         -0.06  -0.03  -0.06  -0.06  -0.08  -0.11
West Ham        -0.06  -0.03  -0.07  -0.06  -0.08  -0.11
Aston Villa     -0.04  -0.01  -0.05  -0.04  -0.06  -0.08
Southampton     -0.03   0.00  -0.03  -0.03  -0.05  -0.07
Leicester        0.01   0.03   0.00   0.01   0.0   -0.02
Swansea         -0.01   0.02  -0.01  -0.01  -0.03  -0.04
Crystal Palace  -0.01   0.02  -0.01   0.0   -0.02  -0.04
Sunderland              0.0    0.0    0.0    0.0   -0.03
Stoke City       4.1           0.0    0.0    0.0   -0.06
W. Bromwich      1.6   -2.5           0.0    0.0   -0.02
QPR              6.6    2.5    5.0           0.0   -0.03
Hull City        5.3    1.2    3.7   -1.3          -0.01
Burnley          6.1    2.0    4.5   -0.5    0.8

Table 5. LaLiga. Differences in APFD and RPFD between Actual Data
(Season 2014-2015) and the New Scenario (TV Rights Season 2016-2017)

Club          R. M.  FCB     At.M.  At.B.  Sevila  Val.   Villar.

Real Madrid           -0.01  -0.77  -1.68  -1.87   -2.41  -3.29
FC Barcelona   -4.9          -0.73  -1.60  -1.77   -2.30  -3.14
Atl. Madrid   -47.9  -43.0          -0.11  -0.14   -0.26  -0.37
Athl. Bilbao  -31.1  -26.2   16.8          -0.01   -0.07  -0.10
Sevilla       -28.3  -23.4   19.6    2.8           -0.06  -0.09
Valencia      -29.8  -24.8   18.2    1.4   -1.4           -0.01
Villarreal    -20.0  -15.1   27.9   11.1    8.3     9.7
Espanyol      -17.5  -12.5   30.4   13.6   10.8    12.3    2.5
R. Sociedad   -24.4  -19.4   23.5    6.7    4.0     5.4   -4.3
Malaga        -25.1  -20.2   22.8    6.0    3.2     4.6   -5.1
Levante       -17.5  -12.6   30.4   13.6   10.8    12.3    2.5
Celta         -13.6   -8.7   34.3   17.5   14.7    16.1    6.4
Deportivo     -15.6  -10.7   32.3   15.5   12.7    14.2    4.4
Granada       -15.3  -10.4   32.6   15.8   13.0    14.4    4.7
R. Vallecano  -17.9  -12.9   30.1   13.3   10.5    11.9    2.2
Cordoba       -13.6   -8.7   34.3   17.5   14.7    16.2    6.4
Getafe        -18.4  -13.5   29.5   12.7    9.9    11.3    1.6
Elche         -16.3  -11.4   31.6   14.8   12.0    13.4    3.7
Almeria       -15.8  -10.9   32.1   15.3   12.5    13.9    4.2
Eibar         -15.4  -10.5   32.5   15.7   12.9    14.4    4.6

Club          Esp.   R.Soc.  Mal    Lev.   Celta  Depor  Gran.  R.V.

Real Madrid   -3.65  -6.02   -6.45  -5.67  -6.83  -7.54  -8.33  -9.13
FC Barcelona  -3.48  -5.78   -6.20  -5.43  -6.55  -7.24  -8.01  -8.78
Atl. Madrid   -0.42  -1.11   -1.23  -0.90  -1.11  -1.34  -1.54  -1.81
Athl. Bilbao  -0.12  -0.50   -0.56  -0.37  -0.46  -0.59  -0.69  -0.85
Sevilla       -0.10  -0.44   -0.51  -0.33  -0.41  -0.53  -0.62  -0.76
Valencia      -0.39  -0.71   -0.76  -0.65  -0.78  -0.88  -0.98  -1.09
Villarreal     0.0   -0.22   -0.26  -0.14  -0.18  -0.25  -0.31  -0.41
Espanyol             -0.20   -0.23  -0.12  -0.16  -0.23  -0.28  -0.37
R. Sociedad   -6.9           -0.03   0.08   0.09   0.03   0.00  -0.08
Malaga        -7.6   -0.7            0.11   0.12   0.06   0.03  -0.04
Levante        0.0    6.9     7.6          -0.01  -0.07  -0.10  -0.17
Celta          3.9   10.8    11.5    3.9          -0.05  -0.08  -0.14
Deportivo      1.9    8.8     9.5    1.9   -2.0          -0.03  -0.08
Granada        2.2    9.0     9.8    2.2   -1.7    0.3          -0.06
R. Vallecano  -0.4    6.5     7.3   -0.3   -4.2   -2.2   -2.5
Cordoba        3.9   10.8    11.5    3.9    0.0    2.0    1.7    4.3
Getafe        -0.9    5.9     6.7   -0.9   -4.8   -2.8   -3.1   -0.6
Elche          1.1    8.0     8.8    1.2   -2.7   -0.7   -1.0    1.5
Almeria        1.7    8.6     9.3    1.7   -2.2   -0.2   -0.5    2.0
Eibar          2.1    9.0     9.7    2.1   -1.8    0.2   -0.1    2.5

Club          Cord.  Getaf.  Elche   Alm.    Eibar

Real Madrid   -9.57  -10.62  -10.49  -12.85  -17.02
FC Barcelona  -9.20  -10.23  -10.09  -12.39  -16.42
Atl. Madrid   -1.81   -2.23  -2.14    -2.77   -3.92
Athl. Bilbao  -0.83   -1.07  -1.02    -1.35   -1.96
Sevilla       -0.74   -0.97  -0.92    -1.22   -1.78
Valencia      -1.13   -1.29  -1.26    -1.57   -2.11
Villarreal    -0.38   -0.54  -0.50    -0.69   -1.04
Espanyol      -0.34   -0.48  -0.45    -0.62   -0.94
R. Sociedad   -0.03   -0.17  -0.13    -0.25   -0.48
Malaga         0.01   -0.13  -0.09    -0.20   -0.42
Levante       -0.14   -0.26  -0.22    -0.34   -0.58
Celta         -0.11   -0.21  -0.18    -0.28   -0.47
Deportivo     -0.05   -0.15  -0.12    -0.22   -0.40
Granada       -0.02   -0.12  -0.09    -0.18   -0.34
R. Vallecano   0.04   -0.06  -0.03    -0.11   -0.26
Cordoba               -0.09  -0.06    -0.14   -0.28
Getafe        -4.8            0.03    -0.04   -0.17
Elche         -2.7     2.1            -0.07   -0.21
Almeria       -2.2     2.6    0.5             -0.11
Eibar         -1.8     3.0    0.9      0.43

Table 6. Fitted Values of the Log Likelihood Functions

        Absolute Difference  Relative Difference

EPL     -1884.58             -1890.63
LaLiga  -1805.63             -1866.40

Table 7. Maximum Likelihood Estimates of the Multinomial Logit Model
Dependent Variable: Result of a Match

                           EPL                     LaLiga
                           Coef.          p-value  Coef.      p-value

Away win
Difference of revenues
  Linear                   -0.0069        0.000    -0.0062    0.000
  Quadratic                -4.55E-06      0.104    -2.58E-06  0.047
Home team promoted         -0.0068        0.967     0.3250    0.044
Away team promoted         -0.4285        0.013     0.0038    0.982
Constant                   -0.3008        0.000    -0.5826    0.000
Difference of revenues
  Linear                   -0.0038        0.000    -0.0034    0.000
  Quadratic                -7.96E-06      0.004    -2.64E-06  0.010
Home team promoted         -0.0065        0.969     0.2763    0.101
Away team promoted         -0.4189        0.013    -0.1950    0.256
Constant                   -0.3039        0.000    -0.5623    0.000
Log L                   -1884.58      -1805.63
N                        1900          1900

Note: Base outcome is "home win"

Table 8. Simulation Results of the Values of the CB Measures for
Season 2014-2015 when Considering the New TV Rights System for Season

                   Real    Simulation  Difference

Arsenal FC         75      82           7
Aston Villa FC     38      34          -4
Burnley FC         33      29          -4
Chelsea FC         87      92           5
Crystal Palace FC  48      43          -5
Everton FC         47      44          -3
Hull City AFC      35      32          -3
Leicester FC       41      37          -4
Liverpool FC       62      69           7
Manch. City FC     79      88           9
Manch. Utd FC      70      85          15
Newcastle Utd.     39      31          -8
Queens PR FC       30      27          -3
Southampton FC     60      51          -9
Stoke City FC      54      57           3
Sunderland AFC     38      46           8
Swansea City FC    56      52          -4
Tottenham H. FC    64      65           1
West Bromwich      44      48           4
Albion FC
West Ham Utd.      47      47           0
HHI                 0.055   0.057
[CR.sub.5]          0.356   0.393
CV                  0.312   0.356

                Real    Simulation  Difference

UD Almeria      29      31           2
Ath. Bilbao     55      59           4
Atl. Madrid     78      83           5
FC Barcelona    94      97           3
R. Celta de V.  51      47          -4
Cordoba CF      20      19          -1
RCD Coruna      35      34          -1
SD Eibar        35      34          -1
Elche CF        41      40          -1
RCD Espanyol    49      48          -1
Getafe CF       37      37           0
Granada CF      35      33          -2
UD Levante      37      37           0
Malaga CF       50      50           0
R. Sociedad     46      47           1
Rayo Vallecano  49      49           0
Real Madrid     92      93           1
Sevilla FC      76      77           1
Valencia CF     77      77           0
Villarreal CF   60      58          -2
                 0.058   0.058
                 0.378   0.389
                 0.401   0.421

Note: The values in the "Simulation" column are the averages across
1000 replications.
COPYRIGHT 2018 Fitness Information Technology Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Carreras, Miquel; Garcia, Jaume
Publication:International Journal of Sport Finance
Article Type:Report
Geographic Code:4EUSP
Date:Aug 1, 2018
Previous Article:Profiting from the English Premier League: Predictive Elicitation, the Kelly Criterion, and Black Swans.
Next Article:Estimating the Implied Probabilities in the Tennis Betting Market: A New Normalization Procedure.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters