The behavioral economics of competitive balance: Theories, findings, and implications.

Budzinski, Oliver ; Pawlowski, Tim

Abstract

Despite the prominence and relevance of the uncertainty-of-outcome hypothesis (UOH) for professional sports worldwide, decades of empirical research have not been successful in establishing clear evidence for the importance of outcome uncertainty for stadium attendance and TV audience. In this regard, some recent papers were developed drawing upon a body of behavioral economic thoughts that might help to better understand the divergence between the UOH, competitive balance, and consumer choices. Since this literature has so far focused on different facets of behavioral economics, it is the objective of this paper to summarize and review the existing literature, discuss possible policy implications that can be derived from behavioral thinking in this context, and point out further research avenues.

Keywords: uncertainty of outcome, competitive balance, behavioral economics, antitrust, competition policy

Introduction

The concepts of competitive balance (CB) and the uncertainty-of-outcome hypothesis (UOH) have been highlighted as distinctive features of the economics of sport from the beginning of sport economics research, pioneered by Rottenberg (1956) and Neale (1964). Overall, the hypothesis suggests that an increasing outcome uncertainty of games (short-term), sub-competitions (mid-term), or leagues over time (long-term) increases (marginal) utility of fans and therefore the demand for sports (for tickets/stadium attendance or broadcasts). (1) As such, competitors in sports markets are in need of each other in order to provide an economically viable product. Moreover, they not only need to allow their competitors to survive, they also have an individual as well as a common incentive to preserve the competitiveness of their competitors (Rottenberg, 1956). This CB argument serves as a main pro-competitive justification in the US to defend agreements otherwise prohibited by antitrust laws (Mehra & Zuercher, 2006). Also in Europe, involved parties have brought forward the CB defense in virtually all competition policy cases on a community level (for a case overview, see Budzinski, 2012). Surprisingly, however, despite the prominence and relevance of the UOH for professional sports worldwide, decades of empirical research have not been successful in establishing clear evidence for the importance of outcome uncertainty for stadium attendance and TV audiences. (2)

Most studies analyzing the potential impact of short-term uncertainty on stadium attendance found either no significant effect (e.g., Benz, Brandes, & Franck, 2009; Forrest & Simmons, 2002, 2006; Szymanski, 2001) or an effect not supporting the UOH (e.g., Buraimo & Simmons, 2008; Czarnitzki & Stadtmann, 2002; Feddersen, Borcherding, & Maenning, 2006; Pawlowski & Anders, 2012; Pawlowski &Nalbantis, 2015; Peel & Thomas, 1992) while only few papers found support for the UOH (e.g., Knowles, Sherony, & Haupert, 1992; Rascher, 1999). Moreover, while studies focusing on TV demand provide comparably more support for the UOH, even those studies struggle in providing clear evidence for the importance of short-term uncertainty across settings. (3) Although there are fewer empirical studies explicitly examining the impact of long-term uncertainty on stadium attendance or TV viewership, the available evidence contradicts the UOH since a (slightly) increasing imbalance, like in the top European football leagues (Pawlowski, Breuer, & Hovemann, 2010), has been accompanied by increasing attendance in the stadiums (Flores, Forrest, & Tena, 2010). Also with respect to mid-term uncertainty, evidence is mixed. While a significant positive effect was often detected if a team still had a chance to contend for the championship (Jennett, 1984; Pawlowski & Anders, 2012; Pawlowski &Nalbantis, 2015; Scelles & Andreff 2015; Scelles, Durand, Bonnal, Goyeau, & Andreff, 2013) or to earn promotion (Forrest & Simmons, 2002; Scelles et al., 2013), neither the chance to avoid relegation (Jennett, 1984) nor to qualify for the UEFA Champions League (Pawlowski & Anders, 2012) was found to be of any apparent importance to consumers.

In summary, the existing empirical literature displays "a lack of certainty about outcome uncertainty" (Leach, 2006, p. 117) in professional sports and rather suggests a rejection of the UOH in many settings (Coates, Humphreys, & Zhou, 2014). What remains unclear is whether this lack of evidence is a result of the UOH just being wrong (or less relevant as suggested by Szymanski, 2006) or the fact that traditional economic models regularly rely on strict and (in reality) hardly observable assumptions, in particular regarding the behavior of the agents. In this regard, to better understand the beneficial effects of CB and uncertainty of outcome for fan utility and fans' behavior, we might need to move away from simplistic notions of perfect information, perfectly rational behavior, and textbook-level microeconomics. In contrast to standard assumptions, cognitive resources are scarce, and individuals economize on these resources by limiting information gathering (Simon, 1955; Stigler, 1961), relying on mental models as interpretative forces of the brain (Denzau & North, 1994; Kahneman, 2003a, 2003b; Kahneman & Tversky, 1979), and focusing their scarce cognitive resources on those problems where their employment promises to yield extraordinary revenues and reverting to heuristics-following behavior in more routine situations (Budzinski, 2003; Vanberg, 2004).

By drawing on an empirically well-supported image of the rational-behaving economic subject that includes imperfect information, subjective and constructive perception, as well as rule-following, heuristic behavior, branches such as institutional economics, experimental economics, or behavioral economics have emerged and contributed to enrich our understanding of economic behavior. In this regard, rational behavior does not so much describe "right" behavior in terms of statistical facts, but instead--less ambitiously--the consistency of mind and action. Among the core phenomena of the widespread and quite diverse field of behavioral economics are framing effects and reference-dependent preferences. Individuals perceive and value a situation not in an isolated vacuum but within a given context that has pre-shaped expectations (reference points) and transports pre-evaluation (framing)--both more often subconsciously to the individual than not. An individual will derive different utility from a given consumption, for instance, depending on her tacit expectations (reference-dependence) and on the framework within which the consumption (decision) took place (framing effects). Both concepts derive from the scarcity of cognitive resources and lead to several further phenomena and ostensible behavioral anomalies. (4)

During the past three decades, researchers started to transfer such concepts into the context of sports to better understand behavioral "anomalies," such as the hot hand phenomenon (5) (or fallacy) in basketball (e.g., Gilovich, Vallone & Tversky, 1985). An early application of reference-dependent preferences represents the phenomenon that often silver medalists feel worse than bronze medalists directly after the competition because the silver medalist reflects on the "loss" of the gold medal (reference point), whereas the bronze medalist refers to the non-medal fourth rank as a reference point (e.g., Medvec, Madey, & Gilovich, 1995).

More recently, researchers also started to transfer such concepts into the context of sports demand. Particularly, papers were developed drawing upon a body of behavioral economic thoughts that might help to better understand the divergence between the UOH, CB, and consumer choices. Since this literature has so far focused on different facets of behavioral economics, it is the objective of this paper to summarize and review the existing literature, discuss possible policy implications that can be derived from behavioral thinking in this context, and point out further research avenues.

Reference-Dependent Preferences and Loss Aversion

A decade ago, Koszegi and Rabin (2006) put Kahneman and Tversky's (1979) concept of reference-dependent preferences into an axiomatic framework. Based on this model, Coates, Humphreys, and Zhou (2014) developed a theoretical model that considers fan preferences for home wins, upsets, as well as game uncertainty by distinguishing between two types of utility that a consumer receives from attending a sporting event (i.e., consumption utility and gain-loss utility). While consumption utility corresponds to utility from standard consumer theory, gain-loss utility reflects utility derived from differences between expected (reference point) and actual game outcome. In their model, the UOH emerges when the marginal utility of a home win (when a home loss is expected) exceeds the marginal utility of a home loss (when a home win is expected). When, however, the marginal utility of an unexpected loss is larger than the marginal utility of an unexpected win, a consumer exhibits loss aversion known from prospect theory (Kahneman & Tversky, 1979), and uncertainty of outcome does not create any net utility.

Humphreys and Zhou (2015) added utility from league standing changes (6) as well as utility from the (absolute) quality of the game. Like in the Coates, Humphreys, and Zhou (2014) model, it allows for different relations between home win probability and attendance to play out. If the utility from uncertainty of outcome dominates all other effects, then an inverted u-shaped relation surfaces; that is, attendance is high with medium probabilities and (comparatively) lower with small and high home win probabilities. However, strong home win preferences drives up attendance for games with high home win probabilities. Furthermore, two complementary effects (loss aversion and the quality of the game) lead to (comparatively) lower attendance if home win probabilities are in the middle ranges and to (comparatively) higher attendance if home win probabilities are small. In the case of loss aversion, the behavioral phenomenon of fans' disliking negative upsets more than liking positive upsets surfaces. Similarly, if fans derive strong utility from watching high-quality games, then the attendance should rise when top teams visit. In extension to Humphreys and Zhou (2015), brand effects and superstar effects can be introduced and should display similar effects (i.e., an increase of attendance when consumers can witness teams with top brands or with superstar players). The economic theory of superstars and stardom emphasizes the preference of consumers to witness the best; second-best talent can be a poor substitute for first-best talent (Rosen, 1981), known talent qualities may be preferred to unknown new talent (MacDonald, 1988), and positive network externalities including boulevard externalities may increase the consumption utility if superstar teams or athletes are present (Adler, 1985; Franck & Nuesch, 2007). Thus, witnessing a superstar (team or athlete) dominating a sport may be exciting for consumers. (7) As such, brand and superstar effects partly go somewhat beyond objective quality of a teams' roster. Altogether, if home win preferences, loss aversion, quality and/or brand, and superstar effects dominate the preference for outcome uncertainty, then a u-shaped relation should display with higher attendance for games with small probabilities and big home win probabilities and (comparatively) lower attendance for games with middle-sized probabilities.

Most of the empirical studies using home win probabilities and home win probabilities squared as predictors of stadium attendance provide evidence towards a u-shaped rather than an inverted u-shaped relationship between home win probabilities and attendance. (8) All these findings support the reference-dependent preferences model with loss aversion; that is, a plausible explanation, grounded in a well-established behavioral economic theory, for the empirical phenomenon that fans in many settings seem to favor uneven games with a favorite (home or away) team. Notwithstanding, as the theoretical reasoning described shows, favorite away teams may frequently attract many fans due to their strong brand and the opportunity to see star players. Taking this further, the empirical findings of increasing attendance with decreasing home win probability might be explained by either reference-dependence preferences for upsets (Coates, Humphreys, & Zhou, 2014) or by preferences for absolute talent combined with network-externalities-styled superstar effects as well as with brand effects (Pawlowski & Anders, 2012). Importantly, however, the u-shaped relation also remains when quality adjusted home win probabilities are used (Humphreys & Zhou, 2015). Moreover, in contrast to many other studies conducted before, Coates, Humphreys, and Zhou (2014) as well as Humphreys and Zhou (2015) control for visiting team fixed effects in their structural model and therefore capture (at least parts of) such superstar and brand effects while the u-shaped relation remains. Therefore, these two explanations seem to be complements rather than substitutes, suggesting that increasing attendance with decreasing home win probability might be explained by both reference-dependence preferences for upsets and (in addition) superstar and brand effects.

Furthermore, Humphreys and Zhou (2015) noted an important (under-)identification since the common empirical specification that only provides two estimates (i.e., one for home win probabilities and one home win probabilities squared) does not allow identifying all three preference parameters (i.e., fan preferences for home wins, for game uncertainty, and loss aversion). While the parameter capturing home win preferences can be separately identified, a u-shaped relationship between home win probabilities and attendance does not allow for ruling out the existence of fan preferences for game uncertainty. Rather, it only allows inferring that fan preferences for game uncertainty are dominated by fan loss aversion.

In summary, by introducing the concepts of reference-dependent-preferences and loss aversion into the analysis of sports demand, researchers generated a plausible theoretical explanation for the common empirical finding that fan interest is maximized when either the home or the away team is favored. This finding has remarkable policy implications, since it may be more attractive for individual teams to play against much weaker or much stronger teams, while CB may still represent the superior situation for the league as a whole, creating a conflict of interests (Humphreys & Zhou, 2015). However, though the u-shaped relation between home win probabilities and attendance or TV viewing is the most often empirical finding in the literature, several studies found an inverted u-shape, suggesting that fan preferences for game uncertainty dominates fan loss aversion. Therefore, there is a need for future research to explore the moderating role of different settings (such as kind of sports, countries, and time period) and reveal the conditions under which loss aversion dominates the preferences for uncertain (game) outcomes, and vice versa. Nevertheless, the framework of reference-dependent-preferences and loss aversion offers for the first time a behavioral-theoretical rationale for the common empirical finding that the relation between home win probabilities and demand is u-shaped (i.e., fans do not seem to care much about short-term uncertainty). Moreover, the extended model by Humphreys and Zhou (2015) sensitizes researchers that even a u-shaped relationship between home win probabilities and attendance does not allow for ruling out the existence of fan preferences for game uncertainty.

Threshold Effects and Satisficing Utility

Threshold effects correspond to an important behavioral economics-qualification of the concept of optimality; instead of aspiring to an optimal level of satisfaction in regard to the consumption of any good, individuals are typically less ambitious and settle for a "satisficing" (combining the verbs to satisfy and to suffice) level as a cognitive heuristic (Bolton & Faure-Grimaud, 2010; Caplin, Dean, & Martin, 2011; Guth, 2010; Guth, Levati, & Ploner, 2012; Harstad & Selten, 2013; Simon, 1955). Once a certain level of satisfaction (aspiration level) is reached, no further cognitive resources are spent on optimizing the consumption in question. Instead, the scarce cognitive resources are focused on consumption areas where no satisfying level has been reached yet. In other words, (small) variations above the satisficing level do not matter. However, if the satisficing threshold is undercut, then a strong (demand) reaction is triggered (discontinuity effect).

The existence of such satisficing thresholds might generally explain why fans do not have an (explicit or implicit) notion of an optimum regarding CB. Moreover, CB changes above the satisficing level would not be perceived to be relevant and, thus, actually would not be relevant for consumption behavior. Then, only a drop of CB below the satisficing level is likely to cause (discontinuous, perhaps even extreme) consumption reactions. In this regard, previous findings on the irrelevance of short-, mid-, and long-term uncertainty for stadium attendance and TV viewership might be explained by variations in uncertainty just taking place above the satisficing threshold. It is part of the behavioral anomaly compared to the traditional model that such (subconscious) thresholds can exist even without a (conscious) notion of optimality or where the optimal value (of CB) lies.

Since fans would not target CB levels beyond the satisficing level, the existence of such satisficing thresholds is generally hard to verify empirically. To test this "in the field," a continuum of different levels of (short-, mid-, and/or long-term) uncertainty as well as a corresponding set of repeated observations of fan behavior is required. Since such data is unfortunately not available at present, we have to rely on studies using survey data with information about individually weighted evaluations of CB (i.e., perceived competitive balance [PCB], as well as intention-to-consume or willingness-to-pay measures). While earlier studies settle with a description of the empirics, the behavioral line of reasoning offered here provides a possible explanation for some peculiar observation in this regard.

The concept of PCB was developed and introduced by Pawlowski (2013a, 2013b) and Pawlowski and Budzinski (2013, 2014), and relies on subjective evaluation by fans. The author(s) try to capture the subjectively perceived level of CB within a league by asking fans in Germany, Denmark, and the Netherlands to rate the level of suspense in the first football divisions on a scale of 0-10 (0=not at all suspenseful to 10=very suspenseful). (9) The developed studies offer some first evidence on the existence of a satisficing threshold. While Pawlowski (2013a, 2013b) finds that the league PCB conditional demand curves are s-shaped (indicating an area of inelastic response for both very high and very low values of PCB), findings by Pawlowski and Budzinski (2013) suggest that changes in fan willingness-to-pay for improvements of CB are triggered by CB falling below a crucial threshold (i.e., WTP "jumps" to a higher level as a reaction to this).

Next to some critiques related to the stated preference methods in general (see, inter alia, Zou & Hobbs, 2006) these studies come along with two study-specific shortcomings. First, it remains unclear what the measure of PCB as employed in these studies really means as it relates to the league in general. (10) Second, the employed intention-to-consume measures used by Pawlowski (2013a, 2013b) are quite vague and hard to evaluate by fans. (11) Nalbantis, Pawlowski, and Coates (2015) tryto tackle these issues in a follow-up study by collecting statements of fans of a German 1st Bundesliga club a few days prior to a game about their willingness to pay for a ticket to the upcoming game, as well as about their perception of suspensefulness of the game under consideration. This is an improvement compared to the aforementioned studies, since their measure relates to a single game and the choice scenario (i.e., willingness-to-pay for stadium admission for a specific game a few days before kick-off) is much more concrete and easier to evaluate by fans. Regression results suggest that willingness-to-pay is higher the more suspenseful the game is perceived to be. This is true, however, only until a certain level, after which increases in the degree of suspensefulness have no additional effect. In this regard, the findings by Nalbantis, Pawlowski, and Coates (2015) provide some further evidence towards the existence of such discontinuity effects.

In summary, there is some supportive evidence based on survey data that the relationship between uncertainty of outcome and fan consumption includes a discontinuity in terms of some kind of a "tipping point" or threshold above which changes in CB are not very relevant for fans, whereas fan consumption behavior does change significantly once CB falls below that crucial threshold. Such threshold effects would imply important policy and management conclusions. Situations of unexpected drops of CB levels below certain satisficing thresholds that would trigger discontinuous and severe decreases of demand must be of concern for league management (by sports associations, league organizations, or promoting companies). In such situations--and perhaps only in such situations--the goal of protecting and stabilizing revenues would require CB-preserving or CB-improving interventions. However, while some modifications with regard to the measurement of variables and the scenario used could already improve the validity of results offered by previous PCB studies, it remains unclear so far how the measures of PCB relate to objective measures of CB (OCB). Moreover, it is important for league organizers to have concrete figures for satisficing thresholds with regard to short-, mid-, and/or long-term uncertainty. Exploring this is subject to further research in this area.

Framing Effects

Framing effects (Gachter, Orzen, Renner, & Stamer, 2009; Tversky & Kahneman, 1981) imply that the context of a perception or a decision situation matters for interpretation and action; individuals are framed by past experiences as well as the environment of a situation, and this influences how they perceive and interpret a given phenomenon. Consequently, a phenomenon like CB is not just a number, it gains meaning for fan behavior from the context in which a given CB-situation surfaces; that is, how has CB been in this league before, what CB are fans expecting to see (various types of reference points; Sagi, 2006), etc. Theoretically, this implies that changes in CB values become more important than cardinal levels because the past levels of CB represent reference points for fan perception (Budzinski & Pawlowski, 2015; Pawlowski & Budzinski, 2014). If CB has been very low in a given league, then fans have been accustomed to low CB levels and take the past imbalance as a reference point. Any improvement from that low level may consequently be perceived as being relatively high CB then. On the other hand, if CB levels used to be high, any small deterioration of CB may already be perceived as low CB because fans have adjusted their reference point to the high level. The subjective assessment of CB is then driven by a mismatch of CB expectation and actual CB--conceptually very similar to the gain-loss utility previously described in this manuscript. In this regard, fan behavior deviates from a traditional rational action model in the sense that the same OCB may be valued differently depending on the framing. Therefore, framing effects might lead to differences between OCB and PCB.

Based on standard OCB measures capturing long-term uncertainty Pawlowski and Budzinski (2013) show that the Danish Superligaen appears to be at least as balanced (or unbalanced) as the Dutch Eredivisie and the German Bundesliga. However, fans in Denmark assess their league to be considerably less balanced (suspenseful) than fans in Germany and the Netherlands do regarding their leagues. For instance, they are willing to pay (per stadium ticket per game) around 2[euro] more than fans in Germany and the Netherlands to increase the current level of CB within their league (Pawlowski & Budzinski, 2013), suggesting a comparably lower perceived level of CB within the Danish Superligaen. In this regard, the existence of framing effects might offer a possible explanation for this peculiar observation. While both the Dutch Eredivisie and the German Bundesliga experienced similarly low OCB levels in the years before, OCB of the Danish Superligaen displayed a sharp decline over time. This decrease in OCB seems to influence perceptions in a stronger way than the level of OCB in the case of the Danish Superligaen. Thus, the difference between OCB levels and PCB levels may be explained by changes of CB (as a framing factor) being a stronger influence on fans perception than absolute CB levels.

If this finding would hold in general, important policy implications would follow. First and foremost, league organizers should try to avoid massive declines in the level of CB over time by implementing appropriate measures. So far, however, it remains unclear what "massive" means in this context. In addition, the only empirical study available suggesting the existence of framing effects (Pawlowski & Budzinski, 2013) is based on a cross-country comparison rather than a longitudinal perspective of respondents from a single country. Moreover, the general reservations with regard to stated preference approaches, the PCB measure employed, and the choice scenario used in this study apply here as well. Nevertheless, while much more sophisticated empirical research is necessary to validate the existence of framing effects, first insights point towards their potential relevance in the context of CB and uncertainty of leagues and competitions.

Attention Level Effects

In general, individual valuations of goods depend on the degree of an individual's attention that is drawn to a specific phenomenon (Bernheim & Rangel, 2009; Sexton, 2015; Tversky & Kahneman, 1979). Since attention represents a scarce cognitive source, attention levels depend--next to individual preferences--on salience-related aspects like media intensity (presence in broadcasting, newspapers, internet, boulevard media, etc.) and relative importance of specific subparts of the overall phenomenon. In regard to the valuation of goods, a typical consequence is that valuations of those products that receive high attention levels outshine those of products with low attention levels in the perception of the consumer.

With respect to the CB of top-level football leagues, the effects of diverging attention levels maybe particularly relevant since the relative importance of competition among teams differs significantly, depending on the positions within the league's ranking that these teams are fighting for (Pawlowski & Budzinski, 2014). Put drastically, competition for top positions is considerably more important than competition in the "dull" midfield. The relative importance of the championship race is higher than the race for a midfield position and, correspondingly, media intensity (regarding all dimensions) will be much higher for the former than for the latter. Looking at typical top-level European football leagues, one can identify several sub-competitions that exceed the "ordinary" fight for position in the league ranking, in terms of relative importance and media intensity. These include, for instance, the championship race, the race for the qualification positions for the European-level competitions (UEFA Champions League; UEFA Europe League), the race against relegation, and qualification positions for play-off rounds (depending on the league's championship structure). Consequently, perceptions of balance and uncertainty by fans are more driven by the closeness of these comparatively important sub-competitions than by the CB of the overall league. In other words, mid-term measures that focus on important sub-competition might be more relevant for consumer behavior than long-term measures that usually reflect the situation with regard to uncertainty and balance within a whole league.

Some supportive evidence for the existence of attention level effects is provided by Pawlowski and Budzinski (2014) based on a comparison of further OCB and PCB measures that could be derived from the three-country study previously referred to in this manuscript. Both objective measures and subjective perceptions indicate that the fight for the championship in the Danish Superligaen was extremely unbalanced, and compared to the German Bundesliga and the Dutch Eredivisie, significantly less balanced in the time period under consideration. The relative differences between these leagues are similar for other sub-competitions, such as the fight for qualifying places for the UEFA Champions and Europa Leagues. This supports the hypothesis that league-wide PCB is more driven by the closeness of comparatively important sub-competitions (i.e., by the CB among the contenders for the relevant positions in the league ranking) than by the CB of the overall league. The imbalance of the championship race in the Danish Superligaen dominates the balance of the overall league in regard to the assessment and behavior of the fans.

The finding on the relevance of certain sub-competitions implies that improving overall league balance is less important than improving balance among the limited number of teams "in the hunt" for important decisions (championship race, etc.). In other words, it might be enough to have a sufficient number of teams strong enough to provide a close fight for the championship (ideally not decided before the last match) and it would not matter much whether the rest of the teams fall significantly behind them. Quite in contrast, it could be particularly beneficial to have an oligopoly of title contenders and an oligopoly of teams battling relegation--and merely few teams in between. Such a league structure may very well promote PCB--and fan utility--more than an overall balanced league (given that the distance between these two groups of teams is not too much). In that regard, the Spanish football league with its persistent duopoly of Real Madrid CF and FC Barcelona, as well as Club Atletico de Madrid competing for the championship may be more attractive for fans than a more (overall) balanced league in France, which, however, lacks a close championship race due to the overall dominance of a single club, such as Paris Saint-Germain FC in the last years. The potentially provocative and radical character of these conclusions becomes clear when considering the implications for re-allocation mechanisms and budget-equalizing regulations that breathe the spirit of overall league competitiveness. If some degree of inequality is inevitable (which is theoretically sound if teams are heterogeneous), then attempts to strengthen the weakest ones at the expense of the strongest ones may just reduce the (oligopolistic) number of teams that are able to fight the leader team. Instead of aiming to reduce the gap between the first and the last, it may be more efficient to aim to reduce the gap between the first and the fourth or fifth--because the gap between the top teams and the weakest teams may not matter much for consumer utility.

However, much more advanced empirical analysis is required to corroborate this reasoning. As previously mentioned, with regard to stated preference approaches, the PCB measure employed and the choice scenario used in this study apply here as well. Moreover, it is important to understand whether a championship battle among six, seven, or eight rather equally strong teams would really be superior to a fight among three or four teams, for instance. The latter might be sufficient for a satisficing level of PCB. Finally, it is important to better understand the relevance of permeability between the different performance segments within a league. If, however, further evidence supports these preliminary findings, far-fetching modifications should follow for both empirical research on the relation between uncertainty and the demand for sports as well as the design of re-allocation mechanisms or budget-equalizing regulations as currently in practice in many leagues around the globe.

Conclusion

The notion that more CB increases fan utility--and therefore their willingness-to-pay--and the revenues of commercial and professional sports events has represented one of the core elements of sports economics. As such, it has served sports associations and authorities as a justification for a wide variety of interventions into the inextricable network of sports competition and economic competition. Among the more popular interventions, regulations that display (otherwise) restrictive and anticompetitive effects always played an important role and were often justified by the CB defense. However, empirical support for this theory has always been ambiguous and ambivalent at best. What remains unclear is whether this lack of evidence is a result of the UOH just being wrong, less relevant, or the fact that traditional economic models regularly rely on strict and (in reality) hardly observable assumptions, in particular regarding the behavior of the agents. In this regard, several recent papers were developed, drawing upon a body of behavioral economic thoughts that might help to better understand the divergence between the UOH, CB, and consumer choices. Since this literature has so far focused on different facets of behavioral economics, it was the objective of this paper to summarize and review the existing literature, discuss possible policy implications that can be derived from behavioral thinking in this context, and point out further research avenues.

The discussion in the previous sections highlights some highly relevant implications for sport policy and league management that might be derived from preliminary findings of the already existing studies in this context. First, the framework of reference-dependent preferences combined with loss aversion offers for the first time, a theoretical rationale for the common empirical finding that the relationship between home win probabilities and demand is u-shaped (i.e., fans do not seem to care much about short-term uncertainty). It may be more attractive for individual teams to play against much weaker or much stronger teams, while CB may still represent the superior situation for the league as a whole, creating a conflict of interests. Second, the existence of threshold effects would imply that situations of unexpected drops of CB levels below certain satisficing thresholds would trigger discontinuous and severe decreases of demand and must be of concern for league management (by sports associations, league organizations, or promoting companies). In such situations--and perhaps only in such situations--the goal of protecting and stabilizing revenues would require CB-preserving or CB-improving interventions. Third, framing effects suggest that league organizers should try to avoid massive declines in the level of CB over time by implementing appropriate measures. Fourth, attention level effects would imply far-fetching modifications for both empirical research on the relationship between uncertainty and the demand for sports, as well as the design of re-allocation mechanisms or budget-equalizing regulations, as currently in practice in many leagues around the globe.

However, behavioral sports economics is still a comparably young branch of sports economics research. Moreover, as previously discussed, many studies in this area suffer from considerable shortcomings. Therefore, more research is required to further validate the preliminary findings along the following lines. First, there is a need for future research to explore the moderating role of different settings (such as kind of sports, countries, and time period) and reveal the conditions under which loss aversion dominates the preferences for uncertain (game) outcomes and vice versa. Second, it seems to be highly relevant to test whether this relationship unveils also in settings, where balance and suspense of a game is measured by the subjective evaluation of fans (i.e., PCB). Third, while some improvements with regard to the measurement of variables and the scenario used could already improve the validity of results offered by previous PCB studies, it remains unclear so far how the measures of PCB relate to objective measures of CB. Fourth, it is important for league organizers to have concrete figures for satisficing thresholds with regard to short-, mid-, and/or long-term uncertainty. Exploring this is subject to further research in this area. Fifth, defining concrete values about what "massive" means in the context of "framing effects" appears to be of great importance for competition managers. Sixth, with regard to attention level effects, much more research is necessary to define the appropriate size of contenders within each performance cluster as well as the relevance of permeability between the different performance segments within a league.

References

Adler, M. (1985). Stardom and talent. American Economic Review, 75, 208-212.

Andreff, W., & Scelles, N. (2015). Walter C. Neale 50 years after: Beyond competitive balance, the league standing effect tested with French football data. Journal of Sports Economics, 16, 819-834.

Benz, M. A., Brandes, L., & Franck, E. (2009). Do soccer associations really spend on a good thing? Empirical evidence on heterogeneity in the consumer response to match uncertainty of outcome. Contemporary Economic Policy, 27, 216-235.

Bernheim, B. D., & Rangel, A. (2009). Beyond revealed preference: Choice-theoretic foundations for behavioral welfare economics. Quarterly Journal of Economics, 124, 51-104.

Bolton, P., & Faure-Grimaud, A. (2010). Satisficing contracts. The Review of Economic Studies, 77, 937-971.

Budzinski, O. (2003). Cognitive rules, institutions, and competition. Constitutional Political Economy, 14, 215-215.

Budzinski, O. (2012). The institutional framework for doing sports business: Principles of EU competition policy in sports markets. International Journal of Sport Management and Marketing, 11, 44-72.

Budzinski, O., & Pawlowski, T. (2015). Sportpolitik und Verhaltensokonomik: Sollten FuBballverbande den Ligawettbewerb regulieren? [Sports policy and behavioral economics: Should sport associations regulate competitive balance in a league?]. In C. Muller & N. Otter (Eds.), Behavioral economics und Wirtschaftspolitik [Behavioral economics and economic policy] (pp. 281-308). Stuttgart, Germany: Lucius & Lucius.

Buraimo, B., & Simmons, R. (2008). Do sports fans really value uncertainty of outcome? Evidence from the English Premier League. Internationaljournal of Sport Finance, 3, 146-155.

Buraimo, B., & Simmons, R. (2009). A tale of two audiences: Spectators, television viewers and outcome uncertainty in Spanish football. Journal of Economics and Business, 61, 326-338.

Cairns, J., Jennett, N., & Sloane, P. J. (1986). The economics of professional team sports: A survey of theory and evidence. Journal of Economic Studies, 13, 3-80.

Caplin, A., Dean, M., & Martin, D. (2011). Search and satisficing. American Economic Review, 101,2899-2922.

Cartwright, E. (2011). Behavioral economics. London, UK: Routledge.

Coates, D., Humphreys, B. R., & Zhou, L. (2014). Outcome uncertainty, reference-dependent preferences and live game attendance. Economic Inquiry, 52, 959-973.

Csapo, P., Avugos, S., Raab, M., & Bar-Eli, M. (2015). The effect of perceived streakiness on the shot-taking behaviour of basketball players. European Journal of Sport Science, 15, 647-654.

Czarnitzki D., & Stadtmann, G. (2002). Uncertainty of outcome versus reputation: Empirical evidence for the First German Football Division. Empirical Economics, 27, 101-112.

Denzau, A. T., & North, D. C. (1994). Shared mental models--Ideologies and institutions. Kyklos, 47, 3-31.

Feddersen, A., Borcherding, M., & Maennig, W. (2006). The novelty effect of the new football stadia: The case of Germany. International Journal of Sport Finance, 1, 174-188.

Flores, R., Forrest, D., & Tena, J. D. (2010). Impact on competitive balance from allowing foreign players in a sports league: Evidence from European soccer. Kyklos, 63, 546-557.

Forrest, D., & Simmons, R. (2002). Outcome uncertainty and attendance demand: The case of English soccer. Journal of the Royal Statistical Society: Series D (The Statistician), 51, 229-241.

Forrest, D., & Simmons, R. (2006). New issues in attendance demand: The case of the English Football League. Journal of Sports Economics, 7, 247-266.

Franck, E., & Nuesch, S. (2007). Avoiding 'Star Wars': Celebrity creation as media strategy. Kyklos, 60,211-230.

Gachter, S., Orzen, H., Renner, E., & Stamer, C. (2009). Are experimental economists prone to framing effects? A natural field experiment. Journal of Economic Behavior & Organization, 70, 443-446.

Gilovich, T., Vallone, R., & Tversky, A. (1985). The hot hand in basketball: On the misperception of random sequences. Cognitive Psychology, 17, 295-314.

Guth, W. (2010). Satisficing and (un)bounded rationality: A formal definition and its experimental validity. Journal of Economic Behavior and Organization, 73, 308-316.

Guth, W., Levati, M. V., & Ploner, M. (2012). Satisficing and prior-free optimality in price competition. Economic Inquiry, 50, 470-483.

Harstad, R. M., & Selten, R. (2013). Bounded-rationality models. Journal of Economic literature, 51, 496-511.

Humphreys, B. R., & Zhou, L. (2015). The Louis-Schmeling paradox and the league standing effect reconsidered. Journal of Sports Economics, 16, 835-852.

Jennett, N. (1984). Attendances, uncertainty of outcome and policy in Scottish league football. Scottish Journal of Political Economy, 31, 176-198.

Kahneman, D. (2003a). A psychological perspective on economics. American Economic Review, 93, 162-168.

Kahneman, D. (2003b). Maps of bounded rationality: Psychology for behavioral economics. American Economic Review, 93, 1449-1475.

Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47, 263-292.

Knowles, G., Sherony, K., & Haupert, M. (1992). The demand for Major League Baseball: A test of the uncertainty of outcome hypothesis. The American Economist, 36, 72-80.

Koszegi, B., & Rabin, M. (2006). A model of reference-dependent preferences. Quarterly Journal of Economics, 121, 1133-1165.

Leach, S. A. (2006). Financial viability and competitive balance in English football. (Doctoral dissertation), Imperial College, London, UK.

MacDonald, G. (1988). The economics of rising stars. American Economic Review, 78, 155-166.

Medvec, V. H., Madey, S. F., & Gilovich, T. (1995). When less is more: Counterfactual thinking and satisfaction among Olympic medalists. Journal of Personality and Social Psychology, 69, 603-610.

Mehra, S. K., & Zuercher, T. J. (2006). Striking out competitive balance in sports, antitrust, and intellectual property. Berkeley Technology Law Journal, 21, 1499-1545.

Meier, H. R., & Leinwather, M. (2012). Women as 'armchair audience'? Rvidence from German national team football. Sociology of Sport Journal, 29, 365-384.

Nalbantis, G., & Pawlowski, T. (2016). The demand for international football telecasts in the United States. Houndmills, UK: Palgrave.

Nalbantis, G., Pawlowski, T., & Coates, D. (2015). The fans' perception of competitive balance and its impact on willingness-to-pay for a single game. Journal of Sports Economics, doi: 10.1177/1527002515588137

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, 1-14.

Pawlowski, T. (2013a). Wettbewerbsintensitat im Profifu[beta]ball - Eine empirische Untersuchung zur Bedeutung fur die Zuschauer [Competitive balance in professional football--An empirical analysis of its relevance for fans]. Wiesbaden, Germany:_Springer-Gabler.

Pawlowski, T. (2013b). Testing the uncertainty of outcome hypothesis in Ruropean professional football: A stated preference approach. Journal of Sports Economics, 14, 341-367.

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

Pawlowski, T., Breuer, C, & Hovemann, A. (2010). Top clubs' performance and the competitive situation in Ruropean domestic football competitions. Journal of Sports Economics, 11, 186-202.

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

Pawlowski, T., & Budzinski, O. (2014). Competitive balance and attention level effects: Theoretical considerations and empirical evidence. In O. Budzinski & A. Feddersen (Rds.), Contemporary research in sports economics: Proceedings of the 5th ESEA Conference (pp. 149-162). Frankfurt, Germany: Lang.

Pawlowski, T., & Nalbantis, G (2015). Competition format, championship uncertainty and stadium attendance in Ruropean football: A small league perspective. Applied Economics, 47, 4128-4139.

Peel, D. A., & Thomas, D. A. (1992). The demand for football: Some evidence on outcome uncertainty. Empirical Economics, 17, 323-331.

Rascher, D. A. (1999). A test of the optimal positive production network externality in Major League Baseball. In J. Fizel, E. Gustafson, & L. Hadley (Eds.), Sports economics: Current research (pp. 27-45). Westport, CT: Greenwood.

Rosen, S. (1981). The economics of superstars. American Economic Review, 71, 845-858.

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

Sagi, J. S. (2006). Anchored preference relations. Journal of Economic Theory, 130, 283-295.

Samson, A. (Ed.). (2016). The behavioral economics guide 2016 (with an introduction by Gerd Gigerenzer). Retrieved from http://www.behavioraleconomics.com

Scelles, N., Durand, C, Bonnal, L., Goyeau, D., & Andreff, W. (2013). Competitive balance versus competitive intensity before a match: Is one of these two concepts more relevant in explaining attendance? Applied Economics, 45, 4184-4192.

Schreyer, D., Schmidt, S. L. & Torgier, B. (2016a). Game outcome uncertainty in the English Premier League: Do German fans care? Journal of Sports Economics, doi: 10.1177/1527002516673406

Schreyer, D., Schmidt, S. L. & Torgier, B. (2016b). Game outcome uncertainty and television audience demand: New evidence from German football. German Economic Review, doi: 10.1111/geer.l2120

Schreyer, D., Schmidt, S. L. & Torgier, B. (2017). Game outcome uncertainty and the demand for international football games: Evidence from the German TV market. Journal of Media Economics, doi: 10.1080/08997764.2017.1281818

Sexton, S. (2015). Automatic bill payments and salience effects. The Review of Economics and Statistics, 97, 229-241.

Simon, H.A. (1955). A behavioral model of rational choice. Quarterly Journal of Economics, 69, 99-118.

Stigler, G. J. (1961). The economics of information. Journal of Political Economy, 69, 213-225.

Szymanski, S. (2001). Income equality, competitive balance and the attractiveness of team sports. The Economic Journal, 111(469), 69-84.

Szymanski, S. (2006). Uncertainty of outcome, competitive balance and the theory of team sports. In W. Andreff & S. Szymanski (Eds.), Handbook of sports economics (pp. 597-600). London, UK: Edward Elgar.

Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, 211 (4481), 453-458.

Vanberg, V J. (2004). The rationality postulate in economics: Its ambiguity, its deficiency and its evolutionary alternative. Journal of Economic Methodology, 11, 1-29.

Zou, N. N. H., & Hobbs, J. E. (2006). Modelling functional food choice and health care impacts: A literature review (No. 91556). University of Alberta, Department of Resource Economics and Environmental Sociology.

Endnotes

Oliver Budzinski (1) and Tim Pawlowski (2)

(1) Ilmenau University of Technology, Germany

(2) University of Tubingen

Oliver Budzinski, PhD, is a professor of economic theory and the head of the Institute of Economics, as well as board member of the Institute of Media and Mobile Communication. His research interests include microeconmic theory, industrial economics, media economics, and sport economics.

Tim Pawlowski, PhD, is a professor of sport economics and chair of the Department of Sport Economics, Sport Management & Media Research. His research interests include the economics of league competitions, the econometric analysis of sports demand, as well as the financing of sport systems.

(1) Following the general systematization by Cairns, Jennett, and Sloane (1986), uncertainty-of-outcome refers to a specific match (short-term), to in-season sub-competitions (mid-term) and to the domination (or not) of a league by some teams over time (long-term).

(2) For a detailed analysis of the literature on European professional football see Pawlowski (2013b). For a recent overview on game level evidence on the correlation between expected game outcome and attendance see Coates, Humphreys, and Zhou (2014). For a recent overview on game level evidence on the correlation between expected game outcome and TV demand see Nalbantis and Pawlowski (2016).

(3) For instance, only few studies on TV demand in European professional football found either clear (Buraimo & Simmons, 2009; Meier & Leinwather, 2012; Schreyer, Schmidt, & Torgler, 2016a) or at least some support for the relevance of close games (Schreyer, Schmidt, & Torgler, 2016b, 2017).

(4) For overviews on the state of behavioral economics (which is still a very dynamic research field), inter alia, see Cartwright (2011) or Samson (2016).

(5) According to the hot hand phenomenon "a basketball player is more likely to hit his next shot after having made his previous shots" (variant of confidence or over-confidence effects) (Csapo, Avugos, Raab, & Bar-Eli, 2015, p. 647).

(6) The league standing effect was introduced by Neale (1964) and suggests that frequent changes in the league table (which increasingly occur when teams are close in the standings) increase utility and consequently consumer demand.

(7) Without the imbalance-creating successes of stars like Michael Schumacher or Tiger Woods, their respective sports would hardly have experienced such an upward swing of sports consumer interest in their home country. However, this does not negate the possibility that continuing dominance of the national hero beyond some yet-to-be defined point may significantly decrease spectators' value and interest in this country.

(8) A rough overview on this literature is provided in the introduction of this paper. More detailed overviews on the correlation between expected game outcome and attendance and TV demand are provided by Coates, Humphreys, and Zhou (2014) as well as Nalbantis and Pawlowski (2016). In line with this literature, the most recent empirical models by Coates, Humphreys, and Zhou (2014) as well as Humphreys and Zhou (2015) based on Major League Baseball (MLB) regular season games point towards a u-shape.

(9) For more detailed descriptions of the survey and the methods of analysis see Pawlowski (2013a, 2013b) and Pawlowski and Budzinski (2013, 2014).

(10) Some further analysis by Pawlowski (2013a, 2013b) suggests that the PCB measure used in these studies reflects the three dimensions of uncertainty. However, the variance explanatory power of 11 items reflecting the short-, mid-, and/or long-term uncertainty is quite low, suggesting that the PCB measure consists of more than just these three dimensions.

(11) The information used to estimate the non-parametric Kaplan-Meier survival functions (i.e., conditional stadium and TV demand curves, as well as general interest curves), was derived from the following three questions: "At which level of overall suspense (on a scale of 0-10) would you (1)... start to lose interest in the 1. Bundesliga; (2)... not watch a match in the stadium; (3) ... not watch a match on television" (Pawlowski, 2013b, p. 7).