Can advertising make free-to-air broadcasting more profitable than pay-TV?

Kesenne, Stefan

Introduction

The general belief is that pay-TV is more profitable than free-TV for a private broadcaster to air sports events, because the former provides two major revenue sources: subscription fees and advertising revenue, whereas the latter can only rely on advertising revenue (see also Hoehn & Lancefield, 2003).

On the one hand, FIFA and UEFA, the International and the European Football Federation, are protesting against the restrictions imposed by the European Commission on selling football broadcast rights. The European Commission does not allow some major national and international sports events, considered as crown jewels, to be hidden behind decoders and high subscription fees. A national pastime should be open to the public, free of charge.

On the other hand, the organizers of the Primavera, the classic cycling race in Italy, Milan-San Remo have always sold the broadcast rights to open-air TV channels with the exception of only one year in the mid-1990s when the rights were sold to a paychannel, but this turned out to be a big mistake.

Back in the early '90s, the Dutch public boycotted the private TV channel of the Dutch Football Union (KNVB), called 'Sport 7,' which was claiming the exclusive right to broadcast Dutch professional football, charging high fees (see Van der Burg, 1996). The Dutch simply refused to turn to the 'Sport 7' channel to watch the championship games, and 'Sport 7' lived a very short life.

In the US major leagues--NBA, MLB, and NFL--all games are broadcast free-to-air or on (low-fee) cable TV, while the majority of European networks do not show the national football matches free-to-air, but rather on subscription or pay-per-view TV.

So, the question rises if free-TV can be more profitable than pay-TV, and if free-TV channels are willing to pay a higher price for the broadcast rights than pay-TV channels.

Holden (1993) argued that technological progress, which is usually associated with the introduction of new or better products, may also mean new methods of paying for old products. So is also pay-per view, it does not change the production function; the TV programs that can be made are no different from what could be made before; the innovation is that the new technology makes it possible to let only the people who pay to view a specific program such as a sports game receive access. Holden (1993) compares pay-per-view with advertising-supported networks and shows pay-per-view has reduced welfare.

Anderson and Coate (2005) present a theory of the market provision of broadcasting and use it to address the nature of market failure in the industry. Market provision may allocate too few or too many resources to programming. Monopoly ownership may produce higher social surplus than competitive ownership. Armstrong (2005), however, argues that the advent of subscription television overcomes many of the market failures which once existed. He discusses the merits of public intervention in the provision of television broadcasting services and argues that intervention was justified in the past when there were just a few channels and when advertising was the only source of commercial funds.

Anderson and Gabszewicz (2006) observe that many advertising expenditures are incurred for television, which is also mainly supported by advertising revenue. In the past, this might have caused market failures in program duplication and a lack of cultural diversity and quality, due to the competition for viewers of the demographics most desired by advertisers, implying that programming choices were biased towards the tastes of those with such demographics. The ability to use subscription pricing can help improve performance by catering to the tastes of those otherwise under-represented, though higher full prices tend to favor broadcasters at the expense of viewers and advertisers.

Peitz and Valletti (2008) compare the advertising intensity and content of programming in a market with competing media platforms. With pay-TV, media platforms have two sources of revenues, advertising revenues and revenues from viewers. With free-to-air, media platforms receive all revenues from advertising. They show that if viewers strongly dislike advertising, the advertising intensity is greater under free-to-air television. They also show that free-to-air television tends to provide more similar content whereas pay-TV stations differentiate their content. They also discuss the welfare properties of the two different schemes.

The model in this paper is closely related to the analytical model of Dietl and Hasan (2007). They show that the probability of free-TV airing of major sports increases, among other parameters, with a higher sensitivity of sponsoring fees to viewer ratings and a higher price sensitivity of consumer demand. In our model the price sensitivity does not affect the broadcaster's choice between pay-TV and free-TV, but the level of consumer demand does, as also indicated by Szymanski (2003). Furthermore, in our model, it is not the absolute level of consumer demand, nor the absolute value of the sponsoring fees per viewer, that are affecting the choice, but it is the ratio between the level of demand and the sponsors' sensitivity that is crucial for the decision of a profit-maximizing sport broadcaster. In our opinion, there are reasons to believe that this ratio tends to be lower in the US than in Europe, which might explain why in the US, free-TV, rather than pay-TV, dominates.

In analyzing sports and the media, it is important to realize that two markets are at play, which, though interdependent, should be clearly distinguished. The first one is the market of TV rights, with the sports clubs and the league on the supply side and the broadcasters on the demand side. The second market is the down-stream market of televised sports, with the broadcasters on the supply side and the TV spectators on the demand side. If the TV rights are monopolized and sold to just one (highest bidding) broadcaster, another monopoly is created in the market of televised sports. If pay-TV is an option, the price charged to the TV spectators will then be too high and the number of games that are broadcast will be too low. Moreover, in deciding about either free-TV or pay-TV, the broadcaster also has to take into account the interests of the sports organizer, which prefers more TV viewers to less TV viewers.

In this contribution, we try to analyze under what conditions free-TV is more profitable than pay-TV, based on a simple theoretical model of a profit-maximizing TV company that holds a monopoly position in the market of televised games.

A Simplified Model

Assume that the demand for TV sports, supplied by a monopolist broadcaster is given by

p = [alpha] - [[beta].sub.q] (1)

where p is the price per view (or the subscription fee) and q is the number of spectators or the number of subscribers.

The smaller [beta], the more price-elastic is the demand for TV sports. However, pay-TV is not the broadcaster's only source of revenue. Besides the payments by TV spectators, a broadcaster can also receive income from TV advertising before, during, or after the games. Assume that advertisers are willing to pay an amount as large as [lambda] per TV viewer. It follows that the broadcaster's total revenue can be written as:

R = pq + [[lambda].sub.q] = ([alpha] + [lambda])q-[beta][q.sup.2] (2)

The average revenue per spectator AR is then: R/q = p + [lambda] = ([alpha] + [lambda])-[[beta].sub.q], which is parallel to the demand curve D. The marginal revenue MR is: [partial derivative]R / [partial derivative]q = ([alpha] + [lambda] - 2[[beta].sub.q]

On the cost side, we consider two cost categories: the broadcast rights the TV company has to pay, and the operational cost to broadcast the games, including transportation, equipment, and personnel. These costs are all independent of the number of spectators that watch the games on TV. So the total cost C=[C.sub.0] is fixed and the marginal cost is zero (see Kesenne, 2007). In Figure 1, this model is graphically presented.

On the horizontal axis, the number of spectators is measured and, on the vertical axis, the pay-per-view price or the subscription fee. The demand for TV sport is presented by the linear curve D. Given the constant average revenue ([lambda]) from TV advertising, the average revenue and marginal revenue curves of the broadcaster are given by the AR and MR. Because the total cost is constant, the marginal cost is zero and the average cost is the downward sloping hyperbolic curve AC.

Pay-TV versus Free-TV

If the pay-TV broadcaster is a profit maximizer, he will set a price that maximizes profits [[pi].sub.v] = R - C = ([alpha] + [lambda])q - [beta][q.sup.2] - [C.sub.0]. From the first-order condition ([alpha] + [lambda]) - 2[beta]q = 0 the optimal number of spectators can be found as:

[FIGURE 1 OMITTED]

q* = [alpha] + [lambda] / 2[beta] = (3)

The optimal price is then:

p* = [alpha] - [lambda] / 2 (4)

From this result, it can be seen that the optimal price can be negative if [alpha] < [lambda], although this outcome is not very realistic. It would imply that broadcasters are willing to pay their TV viewers, because sponsors are very sensitive to viewer ratings. Substituting the optimal values in the profit function, the broadcaster's profit in the pay-TV scenario is then:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (5)

In the free-TV scenario with zero price, p' = 0, the demand for games or the number of spectators would be q1 = [alpha]/[beta]. So, profit would then be equal to:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (6)

The difference between the profits in both cases can then be calculated as:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (7)

Because the difference is always positive, we can conclude that, for whatever values of the parameters [alpha], [beta], and [lambda], pay-TV is more profitable, which confirms the general belief.

In Figure 2, the profit maximizing number of spectators is found at the point of intersection of the MR and the zero MC, which is q*, and the optimal price is p*. Profits can be found as the product of the number of spectators (q*) and the difference between the average revenue (ar) and the average cost (ac).

[FIGURE 2 OMITTED]

In the free-to-air scenario with the price equal to zero, the number of spectators can be seen to equal q' The average revenue (ar') which now only consists of advertising receipts, is equal to [lambda]. Profits are now indicated by the product of the number of spectators (q') and the difference between [lambda] and the average cost (ac'). The difference in profits will be larger, the higher the demand level ([alpha]) and the more elastic the demand curve (1/[beta]). The more advertisers are willing to pay for TV adds ([lambda]), the lower will be the difference in profits.

The Impact of Board and Shirt Advertising

However, the result in the previous section is not the end of story. If the games are broadcast free-to-air, more spectators are watching, which makes it also more interesting for businesses to advertise on player shirts and field boards, which is an additional revenue to the organizer, the league, or the club. Under these circumstances, the organizer would be willing to sell the broadcast rights at a lower price. The price reduction the organizer is willing to grant to a free-to-air broadcaster will depend on the extra TV spectators that can be expected on free-TV compared with pay-TV, which is in our model:

[DELTA]q = [alpha] / [beta] - [alpha] + [lambda] / 2 [beta] = [alpha] - [lambda] / 2[beta] (8)

If these shirt and board advertisers are willing to pay the same amount of [lambda] per viewer as done by the TV advertiser, the reduced cost of the broadcaster will be:

C = [C.sub.0] - [lambda] [alpha] - [lambda] / 2[beta] (9)

With this new cost function, with again a zero marginal cost, the difference between profits in the pay-TV and the free-TV case becomes:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (10)

As can be seen now: [[pi].sub.p] < [[pi].sub.f] if [alpha]/[lambda]<3. So, it is not generally true that pay-TV is more profitable than free-TV. Free-to-air broadcasting can be more profitable than pay-TV, depending on the ratio [alpha]/[lambda], that is: depending on the level of demand and the price that advertisers are willing to pay per TV viewer. The lower the demand for televised sport and the more advertisers are willing to pay, the more profitable free-TV will be. As distinct from the result in Dietl and Hasan (2007), the elasticty of the spectators' demand does not affect the choice for free-TV.

In Figures 1 and 2, the impact of shirt and board advertising can be introduced by a downward shift of the average cost curve in the case of free-TV, so that it is possible that free-TV profits can be higher than pay-TV profits.

What happened to the broadcasting of the classic bike race Milano-San Remo offers a good illustration of this effect. When the TV rights were sold to a pay-TV network, the number of TV spectators dropped dramatically due to the pay-per-view price. So, the sponsors of the racing teams and the board advertisers threatened to reduce their sponsorship money. The next year and ever since, the organizer decided to sell the rights of the Primavera again to a free-to-air network, although pay-TV channels were willing to pay a higher price for the broadcast rights.

One can take the analysis one step further by assuming that the marketing effect of a direct TV add, which can show a video with a story or a clear message, is more effective than the effect of just a brand name on a board or a shirt. So, the money that advertisers are willing to pay per viewer is lower for board and shirt advertising. Assume that the price which advertisers are willing to pay per spectator for board and shirt advertising is y with y<1. In that case on can derive for the difference in profits:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (11)

This implies that free-TV is more profitable than pay-TV if [alpha] - [lambda] / [gamma] <2 which is less likely than the previous condition [alpha]/[lambda]<3. One might also consider including in the analysis the fact that many spectators do not like interruptions for TV adds. If TV adds reduce the demand for televised sports, this implies a lower value of a, which makes it more likely that free-TV is more profitable than pay-TV.

Discussion and Conclusion

In this contribution we have tried to show, using a simple theoretical model, that payTV is not always more profitable than free-TV for a profit-maximizing broadcaster. The main reason is that one also has to take into account the interests of the board and shirt advertisers and the spectators' aversion for TV adds.

Can this result also shed some light as to why in the US major league sports, free-TV prevails, while in European soccer, pay-TV is more dominant? In other words, are there reasons to believe that the ratio between demand level and sponsor's sensitivity ([alpha]/[lambda]) tend to be lower in the US and higher in Europe? Apart from historical, structural, and institutional differences on both side of the Atlantic (see Hoehn & Lancefield, 2003; Szymanski, 2003), an important factor is the scale effect in the US, with its large market compared to the smaller national markets in European soccer. However, in the US, four major leagues, with partly overlapping seasons, have to compete for the TV spectator's interest, whereas in most European countries, soccer is so dominant that there is hardly any competition from other televised sports. Moreover, as argued by Dietl and Hasan (2007), the average season appeal for games in Europe might be higher because of the promotion and relegation system, forcing the weaker teams to fight in order not to be relegated to a lower division, and because of the fight to qualify for the European competitions (UEFA Champions League and Europa League), which provide an additional contest in the national championships, possibly leading to a high demand for televised soccer in Europe. In our model above, the demand level is given by the size of the parameter a. We can only think of one reason why the value of [lambda], that is the price advertisers are willing to pay per TV viewer, should be higher in the US than in Europe. Soccer in Europe goes on uninterrupted for two times 45 minutes. The major league sports in the US, such as basketball, football, and baseball are interrupted more often, by time-outs and by their segmentation into thirds, quarters, or ninths, which allows networks to run more commercials. So, the relative size of these parameters might be part of the explanation why the major league sports in the US are broadcasted free-to-air and Europe has to pay to watch soccer on TV.

References

Anderson, S., & Coate, S. (2005). Market provision of broadcasting: A welfare analysis. Review of Economic Studies, 72, 947-972.

Anderson, S., & Gabszewicz, J. (2006). The media and advertising: A tale of two-sided markets. Handbook on the Economics of Art and Culture, 1, 567-614.

Armstrong, M. (2005). Public service broadcasting. Fiscal Studies, 26(3), 281-299.

Dietl, H., & Hasan, T. (2007). Pay-TV versus free-TV: A model of sports broadcasting rights sales. Eastern Economic Journal, 33(3), 1-24.

Hoehn, T., & Lancefield, D. (2003). Broadcasting in sport. Oxford Review of Economic Policy, 19(4), 552-568.

Holden, S. (1993). Network or pay-per-view? A welfare analysis. Economics Letters, 43, 59-64.

Kesenne S. (2007). The economic theory of professional team sports, an analytical treatment. Northampton, MA: Edward Elgar.

Peitz, M., & Valletti, T. M. (2008). Content and advertising in the media: Pay-tv versus free-to air. International Journal of Industrial Organization, 26, 949-965.

Szymanski, S. (2003). The economic design of sporting contests. Journal of Economic Literature, 41(4), 1137-1187.

Van der Burg, T. (1996). Het voetbal-monopolie. ESB, 81, 710-711.

Stefan Kesenne

University of Antwerp

Stefan Kesenne, PhD, is a professor in the Economics Department. His research interests include sport economics and labor economics.