Integrated women's football teams can attract larger stadium crowds

ABSTRACT

Research Question

Women's sports have seen a trend of diversification recently, with women's and men's teams merging to form integrated football clubs. While such integrated clubs (e.g. Paris Saint-Germain), in particular, have determined the media headlines with record attendances, governing bodies have begun drafting policies to accelerate such strategic integrations. However, empirical evidence on whether integrated women's football teams (which may benefit from brand spillover effects, among other things) attract larger crowds to their stadiums is limited at best. Naturally, this raises the question of whether being an integrated football club matters for women's football stadium attendance demand.

Research Method

Using ordinary least squares regression with season-level fixed effects with a dataset of 1,506 games played over 11 seasons before and after the COVID pandemic in Germany, France, and Sweden, we are the first to present robust evidence on the association between being an integrated football clubs and stadium attendance demand for European women’s football across multiple markets.

Results and Findings

We find that integrated women's football teams can attract larger crowds. Intriguingly, such demand synergies, however, are not a given emerging from a strong brand effect, per se. Instead, women's football club executives must proactively unlock these synergies to avoid adverse effects.

Implications

Our findings suggest that women’s football may achieve a higher potential through integration, but only if the women’s team is managed as more than an ancillary second team. As such, despite being formulated with the best intentions, forced integration policies might backfire.

KEYWORDS:

• Attendance
• brand management
• demand
• diversification
• ticketing

Introduction

While women’s football is still often neglected academically (e.g. Thomson et al., 2023), two trends are evident and necessitate further investigations to improve our understanding of the developments in women’s football. First, following the 2022 Union of European Football Associations (UEFA) European Women’s Football Championship stadium attendance demand for women’s football has increased significantly in Europe, with record attendances breaking the news almost routinely (e.g. Guinness World Records, 2023). Second, women's football has recently seen a trend of diversification, with women’s and men’s teams merging to form integrated football clubs (FC; European Club Association, 2021). Intriguingly, while anecdotal evidence suggests a strong and positive association between both phenomena, empirical evidence on whether such integrated women's football teams (which may benefit from brand spillover effects, among other things) attract larger crowds to their stadiums is limited at best, with those few existing findings being inconclusive.

The advent of integrated FCs is driven by strategic managerial decisions and top-down policy provisions (Valenti et al., 2021). In theory, integrated FCs could leverage an established brand to raise awareness for women’s football and, thus, attract new, previously uninterested stadium visitors, among others. As anecdotal evidence suggests, this strategy appears to have worked well in Europe recently. In Spain, for instance, FC Barcelona Femení sold out the men’s stadium (i.e. the Camp Nou), setting a new world record of 91,533 spectators (Burhan, 2022). Similarly, in France and Germany, Paris Saint-Germain and Werder Bremen only recently hosted matches in front of 43,255 (Europe1, 2023) and more than 20,000 spectators (Lorenzen, 2022) respectively. Despite these one-time events, data generated from the German football market before the recent COVID pandemic, however, also raises some doubt about the robustness of the association's link, with two of the three most-visited teams back then being non-integrated clubs and relatively little stadium attendance demand for some integrated clubs (c.f., soccerdonna.de, 2023), despite featuring well-established men’s football brands. Naturally, this raises the question of whether integrated clubs systematically realize stadium demand synergies beyond the relocated women’s football games we highlighted earlier, often accompanied by significant (local) media interest. Put differently, we ask whether women’s football teams can profit from significant brand spillover effects, in the sense that the stadium attendance demand increases when integrating with a men’s football brand. Or are they better off independently?

While growing, the otherwise already rich extant literature on the determinants that may shape stadium attendance demand is still surprisingly shy on women's sports (c.f., Schreyer & Ansari, 2022), despite some more recent efforts (e.g. Agha & Berri, 2023; Baecker et al., 2023; Stephenson, 2023). More specifically, to our knowledge, only a handful of studies currently exist modeling the demand for women's football in only three different markets. While three of these studies model the stadium attendance demand for women's football in Germany (Meier et al., 2016; Meier, 2020) and, across Europe, the UEFA women’s Champions League (Valenti et al., 2020) relatively broadly, three studies are more focused. LeFeuvre et al. (2013) explore the potential effect of U.S. national team success on stadium attendance demand for the U.S. domestic league. More recently, also covering the U.S. market, Chahardovali et al. (2023) explore whether a women's team location affects the demand for women's football, while Stephenson (2023) is (again) interested in spillover effects from international tournaments and potential superstar externalities, mainly mirroring what's been done in men's football previously (e.g. Schreyer & Singleton, 2023; Shapiro et al., 2017; van Ours, 2022). Given that the determinants shaping the demand are not necessarily robust across men’s and women’s sports (e.g. Agha & Berri, 2023; Konjer et al., 2017; Meier & Leinwather, 2012), this sparse literature on women’s football is surprising.

Intriguingly, as indicated earlier, regarding potential brand spillover effects, the literature is not only sparse but also ambiguous, with Valenti et al. (2020) and Meier et al. (2016) presenting contradicting conclusions and exploring one competition each (rather than multiple markets). That is, while Valenti et al. (2020) find a positive effect if the visiting team uses an integrated brand, Meier et al. (2016) note that there are no such effects in the long run. Still, despite their evident merits, both studies offer no definitive answer to the question at hand. For instance, while one study does not differentiate brand effects of home and away teams, the other one presents empirical evidence based on data generated before the more recent – and substantial – wave of women club's integration. In general, these ambiguous brand spillover effects, however, perfectly align with the broader diversification literature, which points towards both potential synergies and dyssynergies (c.f., Ahuja & Novelli, 2017).

Adding to the still only slowly emerging discussion around diversification strategies in sports (e.g. Holzmayer & Schmidt, 2020), in this manuscript, we thus attempt to investigate the diversification dys-/synergies by integrating men’s and women’s FCs, while taking the women’s team perspective.¹ As such, we attempt to advance the discussion around women’s football demand and diversification in three significant ways. First, by focusing on brand and stadium relocation effects from integrated clubs, we intend to provide a better understanding of the demand synergy potential and success factors of integrating women’s and men’s teams. Second, by including eleven seasons from three domestic women’s football leagues, we assess the temporal and cultural robustness of demand determinants of women’s football stadium attendance, adding to the burgeoning women’s football research stream (see Thomson et al., 2023; Valenti et al., 2018). More precisely, using ordinary least (OLS) squares regression to explore the association between integrated women's football teams and stadium attendance demand for 1,506 games played over a cumulated eleven seasons and in three countries, we are the first to present robust evidence across multiple football markets: Germany (9 seasons), France (1 season), and Sweden (1 season), as well as different time periods, namely pre- and post-COVID (i.e. between the seasons 2011–2012 and 2018–2019, in the season 2021–2022).² Third, by analyzing a micro-mechanism of diversification theory in a related diversification case (i.e. demand synergies via brand spillover from integrating men’s and women’s clubs), we investigate potential demand synergies from the resource-based view of diversification. As such, we contribute to a better understanding of the diversification-performance link in strategic (sport) management (Ahuja & Novelli, 2017).

Below, we proceed as follows. First, we review related literature from the resource-based view of diversification theory and live-sport stadium attendance demand to establish the theoretical foundation for our research. Second, we provide an overview of recent developments in the women’s football industry with a specific focus on the trend of integrated football teams to ensure a common understanding of the context of this paper. Third, we introduce our data and the demand model to provide transparency on the empirical functioning of this study. Fourth, we discuss the model results and derive findings to summarize the key takeaways of our work. Fifth, we conclude with a synthesis of the practical implications of our study for policy- and decision-makers to translate our findings into action, as well as potential future research endeavors.

Related literature

Evidently, our research question relates to two different literature streams, one on diversification and one on stadium attendance demand. Below, we will review these two streams to establish a common understanding of the theoretical framework. First, we will synthesize diversification theory focusing on diversification in sports and potential demand synergies. Second, we will summarize relevant findings from the literature on stadium attendance demand.

Diversification theory in professional sports

Research on diversification in sports is still relatively nascent, and a notable gap exists for empirical studies and synergy investigations. Theoretically, questions related to diversification in sports are grounded in the resource-based view of the firm (Penrose, 1959), which hypothesizes that a firm, in our case a sporting club, is incentivized to diversify its portfolio to leverage excess capacity of unsaleable resources in other business areas to increase firm performance (e.g. Montgomery & Wernerfelt, 1988; Peteraf, 1993; Wernerfelt, 1984). Only recently, Schmidt and Holzmayer (2019) conceptualized diversification efforts of professional sports along a geographic (e.g. owning clubs in multiple countries) and a business dimension (e.g. offering new products). Within both dimensions, the existing literature focuses predominantly on qualitative descriptions of arguments for internationalization (e.g. Fleischmann & Fleischmann, 2019) and business diversification (e.g. Andrews & Harrington, 2016; Pizzo et al., 2022; Storm, 2009). The few empirical studies on the diversification of sports clubs focus on high-level performance indicators, establishing a positive performance outcome (Fühner et al., 2021; Holzmayer & Schmidt, 2020). For sports organizations, Holzmayer and Schmidt (2020) observed the economic superiority (i.e. revenue and profitability) of related business diversification (i.e. investments that are relatively close to the organization’s core business; e.g. cultivating and owning a farm team) over unrelated business diversification (e.g. investments in start-ups), which finds support in general diversification research (e.g. Palich et al., 2000; Rumelt, 1982).

Surprisingly, as yet, no empirical study on diversification in sports has addressed the effectiveness of the most natural, and thus related, form of business diversification – integrating women’s and men’s teams within the same sport. As Ahuja and Novelli (2017) argue, most studies in the field of management, which investigate this diversification effectiveness, tend to focus on high-level diversification-to-outcome relationships, while failing to establish a more informative mechanism on the synergy level. In other words, they search for high-level correlations between diversification and performance levels, without decomposing the underlying levers, which eventually affect the overall outcome. In this regard, integrating men’s and women’s teams in football presents an ideal phenomenon to investigate such underlying levers from related diversification. This is particularly true given the recent trend to integrate women’s teams in various sports (e.g. Lusted & Fielding-Lloyd, 2017; Valenti et al., 2021). In theory, such synergies can be realized on either the supply (e.g. Teece, 1980) or demand side (e.g. Ye et al., 2012). Supply-side synergies may reduce the cost of production based on economies of scale and scope (e.g. Teece, 1980, 1982), creating, for example, club overhead efficiencies. Additionally, efficient know-how transfer (e.g. Johanson & Vahlne, 1977; Lu & Beamish, 2004) may foster more effective team management. Demand synergy research often acknowledges superior customer value and subsequently customer willingness to pay when offering a more diversified portfolio (e.g. Mawdsley & Somaya, 2018; Priem, 2007; Priem et al., 2012; Schmidt et al., 2016). These demand synergies could potentially lead to higher stadium attendance.

Related diversification may create stadium attendance demand synergies

Based on diversification theory, integrating men's and women's teams may create stadium attendance demand synergies through three competitive advantages: (1) Increased consumption utility, (2) increased customer productivity, and (3) reduced search (or shopping) costs. Below, we sketch out each advantage.

First, given their awareness and attitude toward the club’s brand & product portfolio (Cottrell & Nault, 2004; Nayyar, 1990, 1993), both extant spectators of the men's matches and previously uninvolved third parties may enjoy increased utility from also watching women’s sports at integrated clubs, positively shaping stadium attendance demand. As Capron and Hulland (1999) show, using a shared brand after integration generally increases demand synergies. Especially in related diversification cases, brand strength positively correlates with favorable customer attitude towards diversification (Aaker & Keller, 1990; Boush & Loken, 1991). This positive relationship between brand strength and demand is especially pronounced by the emotionally charged nature of the sports business (e.g. Bauer et al., 2008; Blumrodt, 2014; Couvelaere & Richelieu, 2005; Underwood et al., 2001).

Second, integrated clubs may offer spectators increased customer productivity (Hinterhuber, 2002) if the women’s team plays in the men’s stadium. Meier et al. (2016) show that stadium-quality plays no significant role in explaining stadium attendance demand in women's football in Germany. Still, sharing a stadium with the men's team may streamline game day logistics, as extant club fans are familiar with the travel journey and as men’s team stadium may offer more efficient infrastructure. In fact, Chahardovali et al. (2023) found that an efficient stadium journey is likely to increase stadium attendance demand for women's football in the U.S. market. On a different note, it may increase spectator productivity if fans initially affiliated with the men's team can reuse existing merchandise (e.g. flags and scarves) for women's games and vice versa.

Third, spectators may reduce shopping or search costs (Klemperer & Padilla, 1997), given their habitual consumption of their favorite team’s news or communication. As such, integrated clubs may be able to leverage this for special promotions like relocated top games. Previous research confirms an association between special promotions (Bruggink & Eaton, 1996) or homecoming events (DeSchriver & Jensen, 2002) and stadium attendance demand in baseball and college football, respectively. Naturally, due to the historically stronger demand for men's matches, only integrated clubs have the infrastructure resources and promotional reach to promote such special relocations effectively.³

Previous research on whether related diversification may create stadium attendance demand synergies are inconclusive

As a rule of thumb, the economic foundation of stadium demand is based on the standard consumer theory model, which postulates how consumers aim to maximize utility from consumption within a budget constraint (Borland & Macdonald, 2003). Accordingly, as Borland and Macdonald (2003) synthesize, the determinants of stadium attendance can be clustered into five categories, namely (1) characteristics of the sporting event (e.g. game outcome uncertainty, quality of the participating teams, and relevance of the game), (2) the quality of viewing (e.g. day of the week, kick-off time and the weather), (3) economic factors (e.g. the market size), (4) consumer preference (e.g. habit persistence), and (5) capacity supply. Still, whether these determinants are robust across men's and women's sports is in doubt (e.g. Agha & Berri, 2023; Konjer et al., 2017; Meier & Leinwather, 2012), necessitating an independent research stream on women's sports in general and women's football in particular.

With only a few studies modeling stadium attendance demand for women's football (e.g. Schreyer & Ansari, 2022), it may not surprise that evidence on whether related diversification may create stadium attendance demand synergies is both limited and inconclusive. More specifically, to the best of our knowledge, and as indicated earlier, only two such studies exist, both drawing contradicting conclusions. On the one side, Meier et al. (2016) and an updated version of the article in Meier (2020) measured the impact of integration by the home team on demand for German Frauen-Bundesliga with panel data. In the short term, these authors document a significant and negative effect (i.e. the year of the integration). However, no considerable brand spillover effects seem to exist in the long term. On the other side, Valenti et al. (2020), modeling stadium attendance demand for the UEFA Women’s Champions League, observe that the integration status is positively associated with demand if the visiting team is an integrated club. However, no previous study has accounted for the brand equity (i.e. brand awareness and brand attitude) of the integrated brand, even though brand equity differs strongly among professional sports clubs (e.g. Woisetschläger et al., 2019). Moreover, no study has compared demand determinants and brand effects across different countries, cultures, and seasons for women’s football.

Background on women’s football

Women’s football has recently made headlines around record-breaking stadium attendance demand and increased integration with men’s FCs. Below, we shed some light on these developments to provide the relevant background information for our study. More specifically, in the following paragraphs, we focus on (1) the development of women’s football stadium attendance demand, (2) recent integration policies by governing bodies, (4) trends in club integration, and (4) actions by club decision-makers.

While some highlight games have attracted fan and media attention, the underlying fan demand for first-tier women’s football has plateaued mostly since 2011. In the summer of 2022, 87,192 stadium visitors watched the Women’s EURO 2022 final game. Back then, the event set a new stadium attendance demand record for a European Championship final, including the men’s tournaments (UEFA, 2022). It has showcased how far professional women’s football has come to transcend the historic inequities in the male-dominated football world (Caudwell, 2011). At the same time, women’s football has seen similar, ultimately unsustainable, spillover effects before (e.g. LeFeuvre et al., 2013; Meier, 2020). For instance, according to our data set, in Germany, stadium attendance demand for the Frauen-Bundesliga has increased considerably from below 300 spectators per game in the season 1999–1900 to 1,121 in 2011–2012 after the home world cup. Since then, however, the league did not achieve further growth in this regard. As such, it might be unsurprising that matches in France, Germany, and Sweden tended to attract less than 1,000 spectators per game in 2021–2022.

To further grow the demand for women’s sports, over the last few years, football’s governing bodies have started prioritizing the development of women’s football with dedicated agendas on all association levels – global, national, and club. Globally, Fédération Internationale de Football Association (FIFA) published its women’s football strategy in 2018, aiming to ensure that all its member associations formulate a dedicated women’s football strategy by 2022 (FIFA, 2018). Nationally, many national football associations followed suit or had already drafted roadmaps to promote women’s football. One common feature of the policies of the national associations of Brazil, China, and Italy is to mandate the integration of women’s teams within existing professional men’s teams (Valenti et al., 2021). Similarly, on club level, the European Club Association agrees to promote this integration in its most recent women’s football strategy (European Club Association, 2021).

Consequently, over the last decade, the share of integrated teams in top-tier women’s football has increased considerably. For our research purposes, we classify all clubs as integrated if they have a women’s team in the first division and a professional men’s team in the first or second division.⁴ More specifically, here, we analyze three out of the seven most prestigious domestic women’s football leagues (i.e. German Frauen Bundesliga, French Division 1 Féminine, and Swedish Damallsvenskan).⁵ As we show in Figure 1, in 2010, 33% (12 out of 36) of the analyzed first-division women’s teams met our criteria for being an integrated club. Intriguingly, by 2022, this share and the number of integrated clubs has nearly doubled to 61% (23 out of 38).

Figure 1. Share of integrated teams within first divisions of Women’s football in Germany, France, Sweden + combined Total. Source: weltfußbll.de, own depiction.

Independent of recent strategic considerations, the increased integration of women’s teams in the first division is driven by organic builds or executing inorganic acquisition strategies by professional men’s football teams. Organically, existing women’s teams within integrated clubs thrived and were promoted to the first division (e.g. FCO Dijon). Inorganically, various first-division women’s teams were acquired or merged with professional men’s teams and rebranded under an eponymous brand (e.g. 1. FFC Frankfurt and Eintracht Frankfurt in 2020, Kopparbergs/Göteborg FC and BK Häcken in 2021). Interestingly, academics and club representatives alike claim mutual benefits, including but not limited to access to infrastructure, investment, professionalization, and publicity (e.g. Eintracht Frankfurt, 2020; Leslie-Walker & Mulvenna, 2022; Smith, 2021; Valenti et al., 2021; Welford, 2018). However, as we have shown earlier, the verdict of whether it is beneficial for the women’s team to adopt the brand of a professional men’s team is still out. As such, by exploring whether an integrated brand attracts more crowds to women’s games, we attempt to offer one answer to this crucial question.

Data and empirical study

To explore the potential association between integrated women's football teams and stadium attendance demand, we model the stadium attendance demand for 1,506 games played over 11 seasons in three countries; nine seasons in Germany and one each from Sweden and France. Below, we describe our model, including the different specifications, our dependent variable, the independent variables proxying the various aspects of being an integrated club, and the controls, all of which emerge from the already rich stadium attendance demand literature.

In line with most previous stadium attendance demand research (e.g. Agha & Rhoads, 2018; Reade et al., 2021; Sung & Pyun, 2023), we investigate the determinants of women’s football stadium attendance using an ordinary least squares (OLS) regression approach. Here, we control for season-fixed effects in our multi-season data set for Germany and country-fixed effects with separate models for each country.⁶ We use clustered standard errors on home-team-season level for our multi-season data set as the integration status within a season and thus for home matches generally does not vary.⁷ Given the right-skewed distribution of our dependent variable attendance, we model the logarithmic stadium attendance for football matches i, (1) $\ln (\mathrm{attendanc}{\mathrm{e}}_i)=\alpha +\beta {\mathrm{I}}_i+\gamma {\mathrm{X}}_i+{\epsilon }_i$(1) where our dependent variable, attendance, is the total number of distributed tickets, often set equal with the number of stadium visitors (c.f., Schreyer & Ansari, 2022),⁸ I_i is a vector of variables of interest, describing the integration status of participating women’s teams in match i, and X_i is a vector of all other control variables, meaning determinants that may vary from match to match. The associated estimation coefficients of interest are row vectors β and γ, while ϵ is the error term. Further, for team-related variables, we include the variable for both the home and away teams. Given the insignificance of economic factors (e.g. population, income) in previous women’s football research (Meier et al., 2016; Valenti et al., 2020),⁹ and the irrelevance of capacity constraints given typically underutilized stadiums,¹⁰ we have excluded these variables in our models. Moreover, unlike previous Women’s football research (e.g. Meier et al., 2016), we refrain from including past attendance as a proxy for habit persistence, despite understanding the allure capturing season-ticket holder behavior, among other things. However, by doing so, we understand we would create a dynamic model, distorting the interpretation of our main variables of interest because previous attendance is, effectively, a function of all previous seasons’ variables and coefficients, including the past integration status.

In total, we present five model specifications of our baseline model, which assesses variations within a season. To begin with, we analyze data from nine seasons from the German Frauen Bundesliga before checking the robustness of our findings across French Division I Féminine and Swedish Damallsvenskan. More specifically, to leverage the comparably good data availability for the Frauen Bundesliga, in our first model, we include eight seasons between the seasons 2011–2012 and 2018–2019, adding season-fixed effects.¹¹ To allow for better comparability between the different leagues, in our second model, we then only show estimations from the league's last season (i.e. the season 2018–2019) which we then compare to the estimates of the French Division 1 Féminine in our third model. While we are unaware of reliable post-COVID data for the French league, such data exists for the Swedish league. As such, in our fourth model, somewhat mirroring our previous approach, we only show estimations from the Frauen Bundesliga's more recent season, 2021–2022, which is a post-COVID season, which we then compare to the Swedish season 2022 in our fifth model.¹² Unfortunately, for Sweden, no reliable pre-COVID data is available.

Later, because we note a surprising association between integrated women's football teams and stadium attendance demand, we then extend our first model, also investigating the specific role of brand strength in shaping stadium demand synergies in more detail (c.f., Table 3). More specifically, we include detailed information on the integrated brands’ awareness and attitude toward professional football brands in Germany (e.g. Woisetschläger et al., 2019),¹³ primarily to scrutinize the role of the strength of the integrated brand. Naturally, these brand ratings are only available for integrated teams. In sum, we run three model variations, one featuring matches by integrated home teams, one featuring games by integrated away teams, and one featuring games between integrated home teams versus integrated away teams.¹⁴ This allows us to leverage the exogenous variation of match-ups to assess whether an integrated club attracts more stadium visitors either on average or under the condition that the opposing team is also integrated.

In Table 1, we present the summary statistics of our dependent, independent, and control variables, which we hold constant across all models, except our first model. Notably, testing our set of explanatory variables against multicollinearity, all variance inflation factors (VIFs) were below the threshold of 10, which indicates little concern for multicollinearity (Baum, 2008).

Table 1. Summary statistics.

Variable	Description	m	std	min	max	source
Dependent variable
Attendance	Stadium attendance demand^a	950	1,113	80	25,907	(a)
Team integration
Home integrated^b	Home team shares brand with men’s team	0.5	0.5	0	1	(b)
Home brand awareness	Brand awareness (home team; %)	0.9	0.1	0.8	1	(c)
Home brand attitude	Brand attitude score (home team)	0.5	0.0	0.4	0.6	(c)
Away integrated^b	Away team shares brand with men’s team	0.5	0.5	0	1	(b)
Away brand awareness	Brand awareness (away team; %)	0.9	0.1	0.8	1	(c)
Away brand attitude	Brand attitude score (away team)	0.5	0.0	0.4	0.6	(c)
Shared stadium^b	Home team shares stadium with men’s team	0.1	0.3	0	1	(a)
Special relocation^b	Match relocated to men’s team stadium	0.0	0.1	0	1	(a)
Sporting event characteristics
Outcome uncertainty	Absolute difference between home/away team win	7.7	8.5	0	60.7	(d)
Game quality	Sum of league rankings of home/away team	13.2	5.0	3	27	(a)
UCS Home	Game relevance to clinch the championship title for the home team …	1.4	5.4	0	100	(a)
UCS Away	… and away team, respectively	1.4	4.9	0	100	(a)
Viewing quality
Weekend^b	Weekend game	0.8	0.4	0	1	(a)
Morning^b	Kick-off between midnight and noon, …	0.2	0.4	0	1	(a)
Afternoon^b	… noon and 5 pm, …	0.7	0.5	0	1	(a)
Evening^b	… and 5 pm and midnight	0.1	0.3	0	1	(a)
Distance	Distance between the stadia (in km)	327.9	172.8	0	1169	(e)
Temperature	Average game day temperature (host city, in °C)	9.9	5.4	−4.3	27.2	(f)
Precipitation	Game day precipitation (host city, in mm)	1.9	3.9	0	38.1	(f)

Notes and sources: ^aRounded; ^bDummy variable (yes = 1; no = 0); (a) weltfussball.de, (b) transfermarkt.de, (c) Woisetschläger et al. (2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019), (d) oddportal.com, (e) maps.google.com, (f) visualcrossing.com.

We investigate potential demand synergies from brand spillover by integrating the women's team with an eponymous men's team. We use eight independent variables to capture potential integration effects, differentiating between integrated home and away teams in fixtures. Accordingly, all models include two dummy variables that take the value of 1 if the home or away team shares the brand with a professional men's team in the first or second division. Further, we also include two explanatory variables capturing whether, first, the game was hosted in a shared stadium (i.e. a stadium where both the men's and the women's teams held their home games) or, second, the women's team had to move to the men's stadium as part of a special relocation. As indicated earlier, these relocated games have added publicity from club promotions and a considerable media echo, which should increase fan interest. In both cases, playing in a women's only stadium is the reference category. Finally, in our extended model (c.f., Table 3), we also include four variables capturing the home and the away team’s perceived brand awareness and attitude towards the brand (c.f., Keller, 1993), respectively, thus leveraging the work of Woisetschläger et al. (2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019), who, for the period between 2012 and 2019, surveyed more than 4,000 individuals each year to assess the brand strength of professional FCs in Germany.

Beyond our primary focus, we also add numerous controls, thus adding some first light on whether the determinants that might shape stadium attendance demand in one market are robust across different markets, a mostly ignored facet of stadium attendance demand research. More specifically, mirroring previous research (c.f., Borland & Macdonald, 2003), we add controls approximating the characteristics of the sporting event, and the quality of viewing.

Regarding the sporting event’s characteristics, our model captures three aspects; game outcome uncertainty, quality, and, for both teams, the sporting relevance. To proxy outcome uncertainty, we exploit betting data, thus following Valenti et al. (2020), among many others. More precisely, we capture outcome uncertainty by calculating the absolute difference between the betting odds for the home team winning and the away team winning. While we use betting odds to capture uncertainty, we rely on less dynamic table rankings, or, more precisely, the sum of the opponent’s ranking position before the game, to capture game quality. Further, supplementing short-term game outcome uncertainty with mid-term uncertainty, we refer to a well-known measure approximating the relevance of the game for the championship outcome (UCS; Janssens & Késenne, 1987). Frequently employed in the field (e.g. Pawlowski & Anders, 2012), UCS incorporates the number of points needed to win the championship (w), the maximum number of points possible throughout the season (m), the number of points the team has already gained (g), and the maximum number of points it could have gained (p), and increases the closer the team is (g) to clinching the championship (w) and turns zero as soon as the team is eliminated from contention. The calculation for UCS follows equation 2(2) $\mathrm{UCS}=\{\begin{array}{c}{\displaystyle \frac{100}{w-g}\mathrm{if}w-g\le m-p}\\ 0\mathrm{if}w-g>m-p\end{array}$(2) : (2) $\mathrm{UCS}=\{\begin{array}{c}{\displaystyle \frac{100}{w-g}\mathrm{if}w-g\le m-p}\\ 0\mathrm{if}w-g>m-p\end{array}$(2) Further, we also introduce five control variables related to viewing quality. More precisely, we account for scheduling effects by capturing the day of the week, the corresponding kick-off time, the travel distance, and two variables capturing weather conditions (i.e. precipitation and temperature). Regarding scheduling effects, we differentiate between matches played on weekdays, which are likely to attract fewer spectators, and the weekend. Similarly, we capture kickoff time with a categorial variable, capturing some expected non-linearity, by distinguishing between matches kickoff in the morning (until noon), afternoon (12:01 pm to 5 pm), or evening (5:01 pm until midnight). Further, we measure the travel distance between the stadia, a well-known proxy for geographical rivalries and, naturally, also potential away team support, in kilometers. Finally, the favorableness of the weather conditions is approximated by the temperature, measured by the daily average temperature in degrees Celsius on the gameday in the host city, and precipitation, as the daily amount of rainfall in mm on gameday in the host city.

Results and discussion

Below, we structure our results around five key findings. As a rule of thumb, we find that integrated women's football teams can potentially attract larger crowds within a season. Intriguingly, such demand synergies are, however, not a given emerging from a strong brand effect, but beyond that, integrated clubs need to proactively unlock the synergy opportunities arising from integration to avoid potentially even adverse effects (i.e. strengthening the opponent financially).

Our first finding relates to the question of whether matches hosted by an integrated women's football team attract larger crowds per default. Perhaps surprisingly, our results do not support such a claim in any of the three football markets (c.f., Table 2), instead pointing to the opposite. More precisely, in Germany (c.f., Table 2, specification 01, 03, and 04), we note a significant negative association between an integrated home team and stadium attendance demand within a season, robust for pre- and post-COVID seasons. As such, our findings mirror extant anecdotal evidence on the stadium attendance demand for men's reserve teams. Intriguingly, these teams regularly rank towards the bottom of attendance rankings, despite playing with a prestigious crest on their jersey. In both cases, the more established first men’s team seems to overshadow the integrated team.

Table 2. Results.

	Dependent variable: Log attendance
Variables	(01) Germany (full)	(02) Germany (2018–2019)	(03) France (2018–2019)	(04) Germany (2021–2022)	(05) Sweden (2022)
Team integration
Home integrated	−0.48*** (0.08)	−0.60*** (0.10)	0.18 (0.18)	−0.41** (0.16)	−0.06 (0.10)
Away integrated	0.02 (0.03)	0.14 (0.11)	0.18 (0.17)	−0.03 (0.14)	0.40*** (0.10)
Shared stadium	−0.47*** (0.1)		−0.10 (0.47)	−0.55* (0.24)	−0.24° (0.14)
Special relocation	1.10*** (0.25)		1.46** (0.54)		2.56*** (0.59)
Sporting event characteristics
Outcome uncertainty	−0.01 (0.00)	−0.01 (0.00)	−0.00 (0.01)	−0.01 (0.01)	−0.01 (0.01)
Game quality	−0.09*** (0.01)	−0.07*** (0.01)	−0.05** (0.02)	−0.06*** (0.16)	−0.05*** (0.01)
UCS Home	0.02*** (0.00)	0.04** (0.01)	0.05** (0.02)	0.04* (0.02)	−0.01 (0.03)
UCS Away	0.00 (0.00)	0.02 (0.03)	0.06 (0.04)	−0.01 (0.03)	−0.01 (0.02)
Viewing quality
Weekend	0.11 (0.07)	−0.05 (0.26)	−0.33 (0.41)	0.4 (0.29)	0.23 (0.27)
Morning	−0.05 (0.05)	−0.33* (0.13)			−0.27 (0.24)
Evening	−0.14° (0.08)	−0.54° (0.29)	0.33 (0.30)	−0.11 (0.27)	0.52 (0.27)
Distance	−0.00** (0.00)	−0.00 (0.00)	−0.00 (0.00)	0.00 (0.00)	0.00 (0.00)
Temperature	0.02*** (0.00)	0.02 (0.01)	−0.1 (0.01)	0.01 (0.01)	−0.00 (0.01)
Precipitation	−0.01° (0.00)	−0.01 (0.01)	−0.05° (0.03)	−0.00 (0.02)	0.01 (0.02)
Season effects (German national team success before season)
2012–2013	−0.28° (0.17)
2013–2014 (UEFA, Winner)	−0.01 (0.15)
2014–2015	−0.05 (0.11)
2015–2016 (FIFA, Semi-finalist)	0.25° (0.13)
2016–2017 (Olympia, Winner)	0.01 (0.12)
2017–2018 (UEFA, Quarter finalist)	−0.06 (0.11)
2018–2019	−0.07 (0.13)
N	996	119	99	122	170
Mean VIF	1.42	1.73	1.56	1.72	2.32
R²	0.52	0.60	0.54	0.32	0.38
Adj. R²		0.55	0.47	0.25	0.32

Notes and sources: ° p < .1. * p < .05. ** p < .01. *** p < .001. Reference categories for all models are weekday, afternoon kick-off, and women’s only stadium. For (01), season 2011–2012 (home world cup) is reference category.

Interestingly, this does, however, not mean that women’s teams may not benefit from integration. Quite contrary, we, second, note that special relocations of the women's team to the men's team's stadium, often accompanied by promotional activities, are robustly nurturing spectator interest in all three markets within a season. More importantly, even, as sharing a stadium seems to be disadvantageous, per se, in Sweden and Germany (c.f., Table 2, 01; Meier et al., 2016), the relatively strong effect size suggests that using special relocations and probably related communication, thereby reducing a potentially interested spectator’s search costs, and also creating unique events and experiences, might be essential to nurturing spectator interest in women’s sports.

While an integrated women’s football team might not benefit from attracting more spectators, per se, our results, third, nevertheless suggest that spectators are likely to prefer matches featuring an integrated women’s team or, more specifically, an integrated visiting team that is both well-known and perceived as being reasonably controversial according to its brand awareness and attitude scores. That is, while we do not observe such a significant brand effect in two markets, except for the Swedish market (c.f., Table 2, 05), our extended German model (c.f., Table 3), including more nuanced brand information, reveals a significant positive association between the integrated away team’s brand awareness and stadium attendance demand (c.f., Table 3, specification 02). The extended models further reveal that this positive relationship is not conditional on the home team being integrated or not. As such, our findings, to some degree mirroring earlier results from empirical studies modeling the stadium attendance demand for the men's equivalent (e.g. Pawlowski & Anders, 2012) or spectator no-show behavior (e.g. Schreyer, 2019), depict a puzzling scenario, in which the management of a women's football team about to decide whether to become an integrated team is likely to only cater to the financial needs of the competitions, at least when it comes to generating stadium attendance demand.¹⁵ Naturally, this encourages a conflict of interest. On the one hand, a women's football league's administration will likely encourage integration efforts, thus raising the overall level of attendance within a season. At the same time, an individual club's management might be disincentivized to integrate. As can be seen in the business practice already, one approach to resolving such an emerging conflict could be a dual branding approach, as it is apparently successfully applied in the U.S. in the NWSL. According to our data, NWSL attracts the most visitors of any women’s football league without any team using an integrated brand, despite many of them sharing an owner with a men’s football team.

Table 3. Results, extended German model.

	Dependent variable: Log attendance
Variables	Extended Model Home	Extended Model Away	Extended Model Both
Home team brand
Home integrated	omitted	−0.46*** (0.08)	omitted
Home brand awareness	−1.14* (0.50)		−0.91 (0.64)
Home brand attitude	−2.87** (1.04)		−2.77** (1.15)
Away team brand
Away integrated	0.04 (0.04)	omitted	omitted
Away brand awareness		1.78*** (0.34)	2.25*** (0.53)
Away brand attitude		−0.34 (0.4)	−1.03° (0.55)
Stadium integration
Shared stadium		−0.54*** (0.11)
Special relocation	1.39*** (0.15)
Controls
Sporting event characteristics	yes	yes	yes
Viewing quality	yes	yes	yes
Season effects	yes	yes	yes
N	487	487	225
Mean VIF	1.56	1.58	1.79
R²	0.57	0.57	0.62

Notes and source: Standard errors clustered on home-team-season level; Reference category for all models is women’s only stadium; ° p < .1. * p < .05. ** p < .01. *** p < .001.

As a rule of thumb, fourth, these previous results on whether integrated women's football can attract larger crowds are mostly robust across the different markets and periods we explore, even though not every effect is significant throughout. In particular, this robustness applies to the pronounced effect of special relocations, which we observe in all specifications, if applicable. Further, except for the French market, which has the most significant number of integrated teams, we note a robust negative association between a home team's integration and stadium attendance demand. While we obviously cannot make any causal claims, one explanation for this might point to the relatively long-standing tradition of integrated teams in France, where all Division I Féminine champions since 2007 were integrated clubs and where mergers between women's and men's teams have not been observed lately, which, over time, may have helped to ease a host's potential brand dyssynergia. The opposite, of course, may apply to the German and the Swedish markets, both of which have a long tradition of successful independent women’s clubs (Melkersson, 2013; Kjær & Agergaard, 2013) but where multiple mergers have occurred in recent years, thus routinely shaking previously well-known fan preferences.

Finally, leaving branding effects aside, our control variables mainly align with the results documented in previous women’s football stadium attendance demand research (e.g. Valenti et al., 2020). As such, while we note no support for Rottenberg’s (1956) much-debated uncertainty of outcome hypothesis, spectators seem to prefer fixtures featuring two quality opponents (c.f., Meier et al., 2016), and, in Germany and France, the relevance of the game for championship outcome for the home team. Further, while not robust across countries, we note that scheduling matters in women’s football, with lower turnout for rather early and late kick-offs, respectively in Germany.

Conclusion

Summary

Throughout this empirical study, we attempt to provide a robust perspective on the potential determinants of stadium attendance demand for women’s football games in Germany, France, and Sweden, specifically exploring potential demand synergies from integrating the women’s team with a men’s team. By investigating 1,506 games played across three countries and over 11 seasons, we are the first to provide a multi-country analysis of women’s football stadium attendance, addressing an evident, previously unaddressed research question in times when women’s football has seen a surge in public awareness. Intriguingly, this emergence has led to both increased consumer interest and, perhaps as a result, enormous interest in diversifying men’s clubs by either building or integrating existing women's teams. While our results indicate that such integrated women's football teams can, in fact, attract larger crowds within a season, we also note that the expected synergies are, however, not a given emerging from a strong brand effect. In contrast, integrated clubs need to proactively unlock synergy opportunities arising from integration to avoid potentially even adverse effects on game attendance. One method is a temporary relocation to the men's home, which, if not permanent, is likely to nurture stadium attendance demand within a season. Further, brand effects are likely to emerge from attractive visiting teams. Naturally, as we discuss, this encourages a conflict of interest, with the individual club's management potentially being disincentivized to integrate, but the league eager to promote collective integration efforts to raise the overall level of attendance.

Practical implications

In sum, given the discussed potential disadvantages for previously independent women's teams, our findings support a more liberal policy stance while advancing the literature on women's sports attendance and diversification theory. More specifically, while policymakers at national associations are well-advised to incentivize the integration of women's football teams in an attempt to boost total stadium attendance demand through recognizable and polarizing brands, we also recommend taking a more skeptical perspective on mandatory integration policies. This is, in particular, true if the extant management fails to prioritize women's football within the integrating club (e.g. by not hosting the women’s team in the men’s stadium for special games). In this case, the observed dyssynergia may outweigh the potential benefits, hurting the development in countries with an otherwise already well-established women's football ecosystem.. From a branding perspective, a dual branding strategy may be as effective as an integrated brand, even though the empirical verdict on this matter is still out. According to our findings, the most promising lever, though, is leveraging the hype potential (e.g. through special relocations). Thereby, integration may contribute to women's football realizing its full potential. In a broader context, our work thus uncovers a micro-mechanism of how diversification can and cannot create demand synergies from brand spillover in the case of related diversification (i.e. from men's to women's football).

Limitations and suggestions for future studies

While we acknowledge several limitations, we believe that addressing these limitations in future research may offer an exciting road for better understanding the determinants of stadium attendance demand for women's sports and beyond. In particular, we can see three further research avenues relating to women's sports demand research, diversification synergies, and integrated brand activation. First, while we are first to present a relatively comprehensive study on the potential determinants of women's football demand within a season across different leagues, our data, evidently, lacks reliable historical information for two of the three countries (i.e. France and Sweden). As such, while better data sets could help increase the robustness of our results, not only by extending our sample's period but also across more markets, we believe that we, as a field, could only benefit from partnering with the practice in an attempt to model more nuanced (e.g. no-show behavior) or alternative (e.g. tv audiences demand) consumer decisions, also from different sports. In fact, though scarce, previous research has long begun to exploit disaggregated stadium attendance data sets (e.g. Popp et al., 2023; Schreyer, 2019; Sung & Pyun, 2023), modeling season ticket holder decision-making (e.g. Amberger et al., 2023; Karg et al., 2021; Schreyer et al., 2016), among others. Second, though consciously, our study is limited in scope, focusing on stadium demand synergies rather than painting a complete picture. As such, we believe that further investigations could scrutinize other synergy sources. More precisely, following diversification theory, it may be worth exploring additional benefits to integrated women’s teams, most notable in the form of growth in other revenue streams (e.g. sponsoring, merchandizing), cost synergies (e.g. overhead reduction, infrastructure investments, knowledge transfer) or the outcome of the diversification-performance-link (i.e. financial or sporting performance). Similarly, it will be worthwhile to investigate the objective function from the perspective of the men’s club, which acquires a women’s team. Third, given that the expected increases in stadium attendance demand from relocating the women's team are temporary and may wear off over time, we believe it is time for a more nuanced perspective on successful activation practices, specifically to better understand how such integrations can be fully leveraged effectively.

Disclosure statement

The authors report there are no competing interests to declare.

Data availability statement

The data supporting this study’s findings are available on reasonable request from the first author.

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Notes

1 On a somewhat more philosophical note, several colleagues have voiced concerns regarding integration to prevent managing women’s football as the men’s team’s little sister (e.g., Welford, 2018). Despite these concerns and inconclusive evidence on potential demand synergies, several football associations around the globe tend to force the integration of men’s and women’s FCs to increase women’s football visibility (e.g., Valenti et al., 2021). Therefore, uncovering a more nuanced perspective on the demand synergies of integrated FCs has far-reaching implications for policy- and decision-makers.

2 Generally, such cross-market attendance demand studies are surprisingly rare (c.f., Reade & Singleton, 2021, for a notable exception).

3 Put differently, integrating clubs may benefit from reduced customer acquisition costs by leveraging existing club fan databases, including data (Akçura & Srinivasan, 2005; Zander & Zander, 2005). In this regard, bundling tickets for women's and men's games and cross-selling them to the same (existing) customer pool may create a superior product integration (Stremersch & Tellis, 2002).

4 These league levels can be considered as the lowest common denominator to assume a minimum professional level in all three countries.

5 According to soccerdonna.de, the women's football pendant to transfermarkt.de, the seven leagues with the highest accumulated market values also include the U.S. National Women's Soccer League (NWSL) and the UK Women's Super League (WSL), as well as the first leagues from Italy and Spain. Ultimately, we decided to exclude NWSL and WSL because both leagues show too little variation between integrated and Women's only clubs to enable a statistical investigation. More specifically, in the NWSL, no team shares a brand with a professional men's team, while the opposite is true in the UK. Further, we had to exclude both the Italian and the Spanish leagues as, to our knowledge, no reliable stadium attendance data was available.

6 As one reviewer has rightly pointed out, statistically, our methodological approach is not ideal. Quite the contrary, when establishing causal claims, the ideal approach would be running a randomized trial, in our case, a field experiment, with everything else held constant except whether a club is integrated, which would be randomized. Obviously, this is an impossible ask, and, unfortunately, to the best of our knowledge, there's also no natural experiment to exploit, which would undoubtedly be the second-best approach. For an event study on the league level or a difference-in-difference approach, according to the excellent reviewer the third-best option, we certainly miss enough historical data. As such, we would consider our approach the best viable empirical strategy.

7 Our multi-season data set yields 97 clusters, which seems sufficient (c.f., Angrist & Pischke, 2009). Importantly, due to the limited data availability and the resulting degrees of freedom constraints, obviously, we do not cluster standard errors on the home-team-season level for single-season models.

8 As such, again, in line with most research, we do not differentiate between attendance and physical attendance, sometimes called admissions (e.g., Schreyer et al., 2016), excluding no-shows (c.f., Amberger & Schreyer, 2024).

9 One potential explanation might point to the relatively short observation periods in attendance demand research exploiting match-level data, typically only a few years, in which economic trends are often stable and, therefore, unlikely to explain much variance.

10 In our dataset, less than two percent of all games exceeded 90% of the stadium utilization, which makes alternative robustness checks (e.g., introducing an additional TOBIT model; c.f., Schreyer & Ansari, 2022) obsolete.

11 We include season-level fixed effects are included in the error term, i.e., ϵ_i = s_T(i), where T(i) = t is just an indicator that football match i took place in season t.

12 Sweden's Damallsvenskan plays its season in a single calendar year. We have refrained from parallelizing the timespans with the German 2021–2022 season to avoid introducing a systematic error from varying league composition and unequal home-away-team pairings.

13 For clarification, here we rely on (public) information collected and published annually during our period of observation by Woisetschläger and his co-authors.

14 To further increase the robustness of our results, we also simulated a model with a dummy value of zero for brand dimensions of independent women teams for lack of a better proxy. Naturally, this assumes that independent women’s football teams are both entirely unknown (brand awareness) and unfavored (brand attitude). Intriguingly, our results remain essentially robust in direction, effect size, and significance. However, as we cannot theoretically support the assumption behind the value of zero, we opted not to include these additional model variations.

15 An alternative, more nuanced explanation would be that these clubs are supported by more away fans. In this regard, the integrated home team could even note a home disadvantage, somewhat affecting the team’s sporting success. Unfortunately, though, we cannot differentiate between home and away attendances.