Beyond performance? The importance of subjective and objective physical appearance in award nominations and receptions in football

Abstract

Despite an extensive body of research indicating multifaceted advantages for employees deemed physically attractive, factors that limit or even negate the attractiveness premium have not been sufficiently investigated. In this paper, we are interested in whether a rich set of physical appearance factors matter when performance information is transparently available, rather than – as is mostly the case in the labor force – imperfect or costly to obtain. To this end, we explore the labor market of professional football players in one of the world's most prestigious football leagues, namely the German Bundesliga. We investigate whether a rich set of physical appearance factors (including subjective attractiveness, facial symmetry, facial width-to-height ratio (fWHR), and body gestalt) affect players’ chances of being nominated or receiving the prestigious Player of the Month award. This study offers a unique opportunity to explore selection heuristics among team captains who are responsible for the nomination of the players, and the public who votes on the three top selections by captains. Our results indicate that individual performance tends to be a stronger driver of award success than physical appearance. Interestingly, performance can moderate some physical appearance factors (e.g., beauty matters more for lower performing players) and some positive assortative effects between captains and players along physical appearance are observed (captains with higher fWHR scores are more likely to nominate players who also have higher fWHR scores). In general, heuristics for captains are easier to identify empirically than for the public. We also find some differences; for example, only the public demonstrates a bias towards domestic players.

Introduction

The finding that attractive individuals are perceived to have more skills and competencies, as well as superior character traits (Dion et al., 1972), has initiated an abundance of research as to the effects of physical attractiveness on the assessment of human abilities. Hamermesh (2011) stresses that humans are obsessed with beauty, and research on the topic tends to conclude that “beauty matters” (Rosar et al., 2008: 64). These findings appear to hold across a wide spectrum of areas studying the attractiveness stereotype (e.g., Hamermesh and Parker, 2005; Süssmuth, 2006). Attractive individuals are deemed to be predestined for, and favored by, success as a result of others’ assessments of their physical appearance.

Apparently, this effect is present not only in private interpersonal relationships but also in professional situations where the attractiveness stereotype has been proven to hold in a myriad of contexts. Research has found that attractive individuals enjoy better treatment in the labor market (e.g., Hamermesh and Biddle, 1994; Averett and Korenman, 1996; López Bóo et al., 2013); for example, beauty premiums in employee salaries (e.g., Frieze et al., 1991; Biddle and Hamermesh, 1998; Kanazawa and Still, 2018) and better career opportunities both in the military (Mazur at al., 1984) and in the private sector (e.g., Li et al., 2019) as well as in academia (e.g., Hamermesh and Parker, 2005), where attractive university professors are perceived to offer better lectures. Moreover, individuals’ physical appearance may also influence the investment decisions of investors in hedge funds (Pareek and Zuckerman, 2014). Thus, the attractiveness stereotype may even affect presumably free markets.

Recent studies have found evidence for the presence of the attractiveness stereotype in the evaluation of leadership personalities, indicating increased confidence in attractive leaders’ competencies. For instance, attractive CEOs have been found to boost share prices immediately after taking office simply by virtue of their appearance (Halford and Hsu, 2013). Moreover, facial appearances have implications for how people select individuals, such as when choosing leaders (e.g., Little et al., 2007; Little, 2014; Van Vugt and Grabo, 2015). Similarly, more attractive constituency candidates are argued to have better prospects of obtaining a seat in the parliament (e.g., Efran and Patterson, 1974; Marwick, 1988; Klein and Rosar, 2005) only because of their appealing physical appearance.

However, within the beauty premium literature, there are also studies which question the findings that good looks pay off. Deryugina and Shurchkov (2015a) observe evidence of beauty premium in a bargaining setting which is driven by statistical discrimination (biased beliefs about the performance). However, subjects learned to adjust quickly as they receive performance signals meaning that already in the second round, beauty was no longer correlated with wage bids. Dilmaghani (2020)’s results show no evidence of a beauty premium for men in the Canadian labor market while younger women and women with a perfect education-job match faced a beauty penalty. However, attractiveness was positively correlated with the number of fringe benefits for both sexes. Ruffle and Shtudiner (2015) explored in a field experiment callback in job openings. They observed that beauty discrimination (positive and negative) happens at the earliest stage of job search. They find a discrimination against attractive women while males benefitted from a beauty premium. Thus, attractive females suffer from negative returns in the labor market. Based on follow-up survey they suggest that female jealousy and envy are part of the explanation. Agthe et al. (2011) showed that when the person being evaluated is of the same sex as the evaluator, it leads to a disadvantage. Bi et al. (2020) looked at public speaking fees among scientists find that social scientists benefit from a beauty premium while natural scientists benefit from an unattractiveness premium.

Surprisingly, the central tenet as to whether physical attractiveness or physical appearance – as opposed to real performance – leads to such success has not yet been explored in much detail. More specifically, besides evidence by Mobius and Rosenblat (2006), Andreoni and Petrie (2008), Deryugina and Shurchkov (2015b), and a theoretical modeling approach by Altman et al. (2021), little effort has been invested in examining the magnitude of the attractiveness stereotype on success when the actual performance of the individuals under consideration is overt and transparent. We therefore contribute to the literature by focusing on sports as an environment where performance information is transparently available. Thus, in contrast to the performance rating approach of Ross and Ferris (1981), in this study, we exploit an opportunity to measure individual working performance combined with objectively and subjectively gathered employee appearance data by using the sports environment as a setting in which performance variables are widely available to the public. Labor economics studies often struggle to find settings in which objective performance assessments are available for a large group of people under natural working conditions. Moreover, it is too often a very precarious task to adequately compare performance among various heterogeneous individuals, taking into account varied job descriptions, distinct unobserved individual strengths and weaknesses or different incentive structures and mechanisms. On the other hand, the sports environment offers a real-world laboratory (e.g., Goff and Tollison, 1990; Torgler, 2009; Chan et al., 2021a; Kahn, 2000) to explore the relevance of individual appearances in more detail. This allows us to test questions in a relatively controlled high-stakes environment, where incentives are clearly defined, and information is transparently available for outsiders to examine (cf., Chan et al., 2021a).

In our study, we are particularly concerned with examining the question of whether attractive and more masculine male individuals are deemed superior and granted more occupational success compared to their less attractive or less masculine players. To be more specific, our analysis explores players’ facial attractiveness, facial symmetry, the facial width-to-height ratio (fWHR), and their body gestalt. Thus, we are able to explore whether and how physiological cues such as facial appearances matter for their success when controlling for their performance. Evidence indicates that human beings are able to evaluate faces on a range of different trait dimensions (Todorov et al., 2005). In addition, we also explore subjective physical attractiveness based on a large number of rating assessments from more than 2200 online raters. Exploring both subjective and objective attractiveness ratings is a major contribution of the study.

Furthermore, to measure success, we explore whether those characteristics explain nominations and public votes for the “Fußballer des Monats” (Player of the Month; POM) award in the German Bundesliga, which is one of the most prestigious football leagues in the world. When focusing on the POM award, we are able to separately explore two stages in the award bestowal process. First, we explore how those characteristics affect the award nomination. For a few years, each month, team captains were requested to nominate the league's POM. Based on those nominations, three players were selected who then went through the public electoral process to finally receive the POM award. Thus, while captains had a large pool from which to select their candidates, the public could only choose between three options. Thus, we can provide insights into internal (or insider) voting (captains’ choices of success) and external voting behaviors (public's choices of success).

From a theoretical point of view, we explore whether facial characteristics act as a heuristic among bounded rational captains, providing a decision based on factors that go beyond players’ performance. In addition, we complement previous studies on facial appearances focusing on the political leader selection process via electoral votes (e.g., Sigelman et al., 1990; Berggren et al., 2010). We also contribute to the growing empirical literature on the economics of awards (e.g., Frey and Gallus, 2017a, 2017b; Chan et al., 2014). As discussed earlier, the sports environment provides a comparable controlled environment in which to explore awards (e.g., Bünning et al., 2021) similar to research on awards in academia (e.g., Azoulay et al., 2014; Chan et al., 2014).

Although performances alone cannot explain the award and nomination outcome, they matter a lot, being a more vital driver of award success than physical appearance. Thus, our results show that physical appearances play less of a role in an environment where individuals’ performances are transparently visible. Physical appearance may therefore be taken as a proxy-indicator of (expected performance) in situations where an assessment of actual performance is very time-consuming, extremely costly, or just impossible. Moreover, we find that positive performance changes or more attention in the form of play time positively influence the award selection. However, we also observe that performance acts as a moderator for appearance, i.e., a beauty premium is more likely to be visible among lower-performing players. Physical appearances also matter more when we match captains’ and players’ characteristics finding evidence for positive as well as negative assortment. Our results also indicate that it is easier to identify the choice heuristics of captains than the public ones. Beyond that, we were also able to identify some choice biases among captains and the public.

The remainder of this article is structured as follows: In the next section, i.e., Section 2, we provide the theoretical considerations underpinning our investigation. In Section 3, we provide a summary of the application setting around the POM award in the Bundesliga. Section 4 presents the data and methodology, including the construction of physical appearance measures and information of players. In Section 5, we report and discuss the results of the two award stages. Finally, Section 6 concludes with relevant managerial implications and the limitations of our study.