Esports Governance: Exploring Stakeholder Dynamics

: Esports is a rapidly growing industry. However, the unidentiﬁable governance structure of the industry has contributed to a number of integrity-eroding activities. By exploring esports stakeholder dynamics, this paper answers the question, “Is the esports governance model sustainable?” Data were sourced from documentation, focus groups ( N = 3) and semi-structured interviews ( N = 6). Thematic analysis was conducted using Nvivo. The ﬁndings suggest that (1) the current esports governance framework features some attributes of the “lead organisation-governed network”, with the power residing mainly in game publishers; (2) the rising power of other stakeholders in the network seeking to address integrity issues has caused fragmentation of the esports governance framework; (3) esports governance is evolving towards a network administration organisation (NAO) model. Such evolution has a few challenges—most notably, the compliance of game publishers. Given the social impact of the integrity issues, governments should play a main role in facilitating a NAO model.


Introduction
The growth of digital technologies is providing opportunities for new and emerging industries. These nascent industries require regulation and governance, not only from a business and legal perspective but also to ensure appropriate social impacts. Esports is an example of such an industry. While contested for its definition, esports commonly refers to competitive (pro and amateur) video gaming [1]. In this paper, the focus is on esports governance and the ability of the esports ecosystem to ameliorate integrity-eroding activities. While esports continues to expand exponentially, the industry is facing a multitude of potential challenges. One of the biggest, if not the biggest, question for esports is the lack of a formal and durable governance structure that fits the unique features and circumstances of esports [2]. In other words, the current governance and regulatory landscape surrounding esports is uncertain [3].
Despite the lack of formal governance mechanisms, the esports industry continues to evolve financially. Esports is a fast-growing industry, with USD 957.5 million of revenue generated in 2019, a 23.3 percent increase since 2018 and a projected annual growth rate of 15.5 percent between 2018 and 2023 [4]. Unsurprisingly, the financial rewards on offer have induced an increasing number of integrity-eroding activities-similar to regular sport-such as cheating, doping, match-fixing and illegal gambling [5][6][7]. As this problem has not been addressed through traditional policy or law enforcement mechanisms, the esports industry has initiated its own integrity-promoting activities [8,9]. However, their efforts to do so have been beset with incoherence and crises. Against this backdrop, this paper explores the current esports "governance model" through the lens of network governance and examines the interactions amongst key network members. This is in order to make sense of Sustainability 2020, 12, 8270 2 of 15 the current esports ecosystem and to be in a clearer position to recommend policies to improve its governance going forward. The rationale for this research is the lack of research focusing specifically on esports governance and stakeholder dynamics. In addition, there is a need to better understand and explain the incoherence and crises affecting the esports industry, with a view to providing directions by which to improve the sustainability of the esports governance model. Based on this rationale and research gap, the following research questions were developed: (1) Who are the key and emerging stakeholders in the esports governance system? (2) What is the current esports governance framework and why? (3) How are stakeholders facilitating/constraining the evolution of esports governance?

The Esports Ecosystem as a Network
To tackle the governance of the esports ecosystem, it is important to have an initial understanding of its actors and their relationships. Scholz [10] notes that "The eSports industry, with its various stakeholders, can be seen as an interwoven network, where stakeholders need each other to work and to succeed." Scholz [10] presented an initial categorisation of esports stakeholders and, more generally, the esports ecosystem ( Figure 1). Within the esports ecosystem, esports players, together with game developers, professional players and teams and tournament organisers, are placed as endemic actors. Beyond these actors, non-endemic stakeholders such as media, sponsors, investors, suppliers, providers and shareholders also interact with each other, providing resources for the ecosystem to grow. Despite the ostensible consideration of social and ecological dimensions of the esports ecosystem, Scholz [10] solely focuses on the economic/business aspects. His argument is that "Many eSports organizations are, due to the lack of governmental regulation and without the legal possibility of non-profit strategies, following the business narrative" [10].
Sustainability 2020, 12, x FOR PEER REVIEW 2 of 16 [8,9]. However, their efforts to do so have been beset with incoherence and crises. Against this backdrop, this paper explores the current esports "governance model" through the lens of network governance and examines the interactions amongst key network members. This is in order to make sense of the current esports ecosystem and to be in a clearer position to recommend policies to improve its governance going forward. The rationale for this research is the lack of research focusing specifically on esports governance and stakeholder dynamics. In addition, there is a need to better understand and explain the incoherence and crises affecting the esports industry, with a view to providing directions by which to improve the sustainability of the esports governance model. Based on this rationale and research gap, the following research questions were developed: 1) Who are the key and emerging stakeholders in the esports governance system? 2) What is the current esports governance framework and why? 3) How are stakeholders facilitating/constraining the evolution of esports governance?

The Esports Ecosystem as a Network
To tackle the governance of the esports ecosystem, it is important to have an initial understanding of its actors and their relationships. Scholz [10] notes that "The eSports industry, with its various stakeholders, can be seen as an interwoven network, where stakeholders need each other to work and to succeed." Scholz [10] presented an initial categorisation of esports stakeholders and, more generally, the esports ecosystem ( Figure 1). Within the esports ecosystem, esports players, together with game developers, professional players and teams and tournament organisers, are placed as endemic actors. Beyond these actors, non-endemic stakeholders such as media, sponsors, investors, suppliers, providers and shareholders also interact with each other, providing resources for the ecosystem to grow. Despite the ostensible consideration of social and ecological dimensions of the esports ecosystem, Scholz [10] solely focuses on the economic/business aspects. His argument is that "Many eSports organizations are, due to the lack of governmental regulation and without the legal possibility of non-profit strategies, following the business narrative" [10]. Nevertheless, not all esports organisations follow the business narrative. For example, organisations such as international and national esports associations-missing in Figure 1-are not business-driven but interested instead in the sustainability of the esports ecosystem at the Nevertheless, not all esports organisations follow the business narrative. For example, organisations such as international and national esports associations-missing in Figure 1-are not business-driven but interested instead in the sustainability of the esports ecosystem at the international and national levels. Another limitation of the business perspective adopted by Scholz [10] is that players are treated only as a stakeholder to monetise, without consideration of the health and social issues they may encounter. In the end, if Scholz [10] offers a relevant starting point for the consideration Scholz's (2020) emphasis on stakeholders' interdependence suggests the relevance of shared governance. However, when considering actors not included by Scholz [10], such as international and national esports associations, one may question whether business-driven esports organisations need them and would accept a shared governance with them.
Besides the above, networks are subject to change, because network members adjust constantly to complexities, interdependencies and dynamics [23]. Previous research has identified various factors that drive the evolution of network governance forms. For example, Provan and Kenis [12] pointed out that the evolution of network governance generally follows the logic of searching for greater effectiveness. A shared governance network may evolve to a lead organisation or a NAO governance, as the network grows in size and requires more efficient management of the network. In addition, based on resource dependency theory, networks evolve in forms so that they are better positioned to access valuable and rare resources (e.g., knowledge or technology) and to sustain competitive advantage as a whole [24]. Legitimacy building is another important factor that influences the evolution of a network [25]. To sustain a network, it is important that internal and external groups perceive high status and credibility [26]. Externally, a network may adjust its strategic structure in accordance with the social-political environment, such as addressing government rules, regulations and standards so as to gain legitimacy among a wide range of stakeholders; internally, a network may also change its culture to address individual interests and fairness in treating employees and customers [27].
Consequently, as a result of the various influencers, the three forms of network governance introduced above can evolve from the very basic form (i.e., shared governance) to the other two forms (i.e., lead organisation or NAO) [12]. A lead organisation governance network emerges when one organisation in the original network grows in power and vitality and seizes the most significant and irreplaceable resources (e.g., core technology or intellectual property). A lead organisation form may become the NAO when a powerful organisation no longer sustains legitimacy internally and is not trusted by members; therefore, an independent entity needs to be created to represent the network's interest [12]. The transition from a shared governance or a lead organisation form to a NAO form involves a change in contingency factors, such as the diffused trust among members or the lack of a power organisation within the network to take the lead to govern other members. Externally speaking, it can be a coercive force, such as the government, imposed on the network and mandating a NAO form to take shape due to its concerns for the network.
Provan and Kenis [12] also mentioned that it is unlikely for a lead organisation or a NAO form to evolve into the form of shared governance once institutionalised governing leadership is established. A NAO is arguably the most stable and formalised form of network governance amongst the three, so when this type of governance is established, it is unlikely for a NAO to change to other forms.

Data Collection
Because of the limited research conducted on the esports governance model, this study employed an exploratory case study research design [28]. Yin [29] suggested that when it comes to answering the "what", "why" and "how" questions, a case study design has a distinct advantage, because it allows investigators to focus on a case and retain a holistic and real-world perspective.
To answer the first research question (i.e., who are the key and emerging stakeholders in the esports governance system?), we relied on secondary data. By searching keywords such as esports/gaming governance, esports/gaming regulation and esports integrity in Google, we identified online articles generated by esports media platforms such as the Esports Observer, Esports Insiders, Esports News UK (ENUK), Esports Marketing Blog, Twitch, the NUEL and Esports Integrity Commission (ESIC); policies and documents published by various national (e.g., the British Esports Association and the Association for UK Interactive Entertainment (UKIE)) and international (e.g., the International Esports Federation, the Global Esports Federation, and the World Esports Association) esports-related associations were also reviewed. Documents were selected based on the criteria of authenticity and credibility [30].
Focus groups and semi-structured interviews were conducted to gain a deeper understanding of the dynamic relationships between esports stakeholders. Focus groups provide the opportunity for the researcher to see how participants respond to each other's views and build up a view out of the interaction that take places within the groups [30]. Three focus groups were conducted focusing on the theme of "esports governance", with representatives of the majority of the stakeholders identified in the first step (e.g., esports players at both professional and amateur levels, tournament and league organisers, team managers, etc.). The participants were recruited through social media (e.g., Facebook) and emails. The participant selection criteria required participants to be (1) above 18 years old; (2) involved in esports for at least 2 years. Participants could be either gender and involved as a professional or amateur player, enthusiastic viewer or tournament organiser. The three focus groups were arranged in accordance with the availability of participants. The first focus group consisted of ten participants (3 females and 7 males) and lasted two hours. The second and third focus groups consisted of five participants (1 female and 4 males in each group) and lasted approximately one hour each. All focus groups started with some open questions such as "how many years have you been involved in esports and how?" and "what are the experiences like being involved in the esports industry?". Then, more specific questions were asked around the governance of esports.
Examples include "what issues are existing in esports?", "who is or should be responsible for esports regulation/governance?" and "how can esports governance be improved?". A more detailed list of questions is presented in Appendix A. Participants were encouraged to contribute freely based on their individual experience and not to be influenced by others' viewpoints. Focus groups were assigned pseudonyms of FG1, FG2 and FG3.
Six semi-structured interviews (all males) were conducted with other stakeholders via Skype and telephone calls. Questions were similar in style to those used in the focus groups (see example questions in Appendix B). These questions were asked in a flexible manner, responding to the inquiring process and informants' behaviour. This enabled a natural flow of conversation between the researcher and interviewees. These interviewees were selected through purposive sampling, which is to recruit participants because they are relevant to the proposed research questions [30]. All six participants had rich experience in the esports industry and possessed good knowledge of the current esports governance, which made them "ideal" candidates to discuss this topic. Interviews were digitally recorded and ranged from one to two hours. For confidentiality reasons, respondents were assigned pseudonyms (i.e., R1, R2, R3, . . . R6). Table 1 presents the profiles of the participants in both the focus groups and interviews. Table 1. Participants in focus groups and interviews.

Data Analysis
The first step of data analysis was to map an exhaustive list of esports stakeholders, identifying whether they were key (i.e., core to the ecosystem) or emerging (i.e., appearing in the ecosystem to help address the issues that key stakeholders are not able to tackle or interested in tackling). This was conducted through reviewing the documents collected. The second step relied on focus group and interview data that were subject to thematic analysis [31]. The data analysis was initially framed around Provan and Kenis' [12] three forms of network governance and Sørensen and Torfing's [13] four rationalities to answer the second research question (i.e., what is the current esports governance framework and why?). As such, four initial themes were identified: publishers as leading organisation, profit-driven, legitimacy to govern and tensions/conflicts. We then allowed new themes to emerge to inform us about the relationships amongst different stakeholders and the change in such relationships in time. After a few rounds of grouping the themes, five main themes remained: tensions/conflicts, power, interest, legitimacy and fragmentation. These themes were used to seek answers to the third research question (i.e., how are stakeholders facilitating/constraining the evolution of esports governance?). This strategy is consistent with Ryan and Bernard's [32] suggestion to use the elements of research questions and theoretical framework/underpinning in establishing themes. Finally, the trustworthiness of the data was maximised by a triangulation technique [33]. By selecting three different data collection methods, we were able to compare and contrast the data, and no contradictions between focus group and semi-structured interview data were evident. Member-checking was also used throughout the process to ensure data interpretation accuracy.

Findings
This section, setting out from the list of stakeholders of esports governance identified in the first phase of data collection (see Table 2), will explore the dynamics amongst the key and emerging stakeholders in the esports governance network. It will also examine how each of these stakeholders facilitates, constrains or is constrained in the governance mechanism, in relation to integrity-eroding activities and their regulation. The two main findings derived from the data analysis are: (1) game publishers are the dominant powers in their own network; and (2) emerging stakeholders, with their diversified interests, create fragmentation of esports governance. The section is organised around these two main findings.

Key Stakeholders: Game Publishers Are the Dominant Powers
Game publishers refer to companies publishing video games that are produced either internally or externally (e.g., Tencent Games or Sony Interactive Entertainment). Publishers finance the design, development, marketing and distribution of video games; therefore, they are the ultimate owners of video games' intellectual property [2,34]. When it comes to governance, each esports game is governed by its publisher and different publishers may have their own governance approaches [35]. This means that the esports governance network is fragmented by nature. Some publishers have more advanced governing mechanisms than others. For instance, Riot Games tends to be more proactive in regulating its games than Blizzard (both Riot Games and Blizzard Entertainment are two game publishers based in the United States), which can be evidenced by its recently established Riot Scholastic Association of America, as a governing body for its College League of Legends. Despite the discrepancies in their governing approaches, the role of publishers is essentially similar across various esports and was perceived to be at the centre of the esports governance network, as respondents in focus groups depicted when questions were asked about their perceptions of publishers: They (publishers) can do what they like with their game; this is not like FIFA stopping you and me playing football on my front lawn, because they cannot do that, they do not own football. But Riot [Games] does own League of Legends, they can do whatever they like with it. They can give it to you, they can take it away, and they can change the meta, which they do all the time (FG1).
It was evident from the interviews that publishers have the absolute power in the esports industry because they are the owners of the games. Publishers wish to control esports as an extension of their marketing activities. A respondent commented: Esports is the 400 guys at the top. The 50 million is what the business is about, Riot is a business, EA is a business. Esports exists at their whim, for their pleasure and for their advancement. The minute it is no longer of interest to them, look at Heroes of the Storm, right? The minute it was of no interest any longer to Blizzard, they just drop it. There you go; we are no longer an esports (FG1).
Esports is a less stable industry compared to others due to its constant updates, changes and elimination of video games by publishers. Esports governance is dominated by the publishers. Therefore, publishers tend to show interest in governing esports games to ensure that their popularity and reputation are maintained. For example, Blizzard requires tournament organisers to obtain a license and meet specific criteria in terms of chat moderation and exclusion of certain sponsors [36]: If you broadcast your competition online, you are responsible for ensuring that any accompanied chat is adequately moderated to prevent vulgar, abusive or an otherwise mean-spirited environment.
The competition cannot be sponsored by any companies that sell or promote any of the following products or services: Multinational consumer brands or products or services related to the video game industry without approval from Blizzard; pornography (or extremely mature materials); alcohol and marijuana; tobacco or cigarettes.
By using licensing agreements to create a governance framework, publishers ensure that stakeholders such as tournament organisers comply with their interests. As such, a network emerges with publishers as the most influential group of organisations, as represented in Figure 2. This representation contrasts with Scholz's [10] perspective by putting the game publishers rather than the players at the centre of the network.
With a cascading effect, tournament or league organisers provide another layer of regulations. For example, the ESL (formerly known as Electronic Sports League, is an esports organiser and production company that produces video game competitions worldwide) created a rulebook for players, teams and sponsors associated with their competitions, including players or teams' bans and penalties for harassment or hate speech: Participants must not engage in harassment or hate speech in any form. This includes, but is not limited to: hate speech, offensive behavior, or verbal abuse related to sex, gender identity and expression, sexual orientation, race, ethnicity, disability, physical appearance, body size, age, or religion [37]. By using licensing agreements to create a governance framework, publishers ensure that stakeholders such as tournament organisers comply with their interests. As such, a network emerges with publishers as the most influential group of organisations, as represented in Figure 2. This representation contrasts with Scholz's [10] perspective by putting the game publishers rather than the players at the centre of the network. With a cascading effect, tournament or league organisers provide another layer of regulations. For example, the ESL(formerly known as Electronic Sports League, is an esports organiser and production company that produces video game competitions worldwide) created a rulebook for players, teams and sponsors associated with their competitions, including players or teams' bans and penalties for harassment or hate speech: Participants must not engage in harassment or hate speech in any form. This includes, but is not limited to: hate speech, offensive behavior, or verbal abuse related to sex, gender identity and expression, sexual orientation, race, ethnicity, disability, physical appearance, body size, age, or religion [37].
For those who violate the code of conduct set by the tournament or league organisers such as the ESL, punishments (e.g., penalty points, barrages or immediate and permanent disqualifications) are applied to discourage undesired behaviours. Media like Twitch also have rules in place such as prohibiting both streamers and viewers from posting sexually explicit or suggestive content, although Twitch's definition of sexual content has been identified as vague, subjective and contradictory and enabling discrimination [38].
Esports players and fans co-exist on a virtual platform which is commonly referred to as a "community". Players can join each other in teams to train themselves at different skill levels. Fans could watch other people playing to learn specific gaming skills [1]. When not associated with organised competitions such as tournaments or leagues, these groups are often not governed by publishers directly. Therefore, as respondents pointed out, "the low level is where there's toxicity in the esports community" (FG 3). A respondent in FG 2 provided an example: There was a 15-year-old who beat one of the best Smash players in the world at a tournament and knocked him out and his followers then ousted her out on social media and just gave her a ton of abuse […] unfortunately a lot of amateur esports players are like that and communities will do that […] but no one is there to protect their mental health.
Participants in FG 1 emphasised the exploitation of esports players who transform from amateur to semi-professional or professional levels: For those who violate the code of conduct set by the tournament or league organisers such as the ESL, punishments (e.g., penalty points, barrages or immediate and permanent disqualifications) are applied to discourage undesired behaviours. Media like Twitch also have rules in place such as prohibiting both streamers and viewers from posting sexually explicit or suggestive content, although Twitch's definition of sexual content has been identified as vague, subjective and contradictory and enabling discrimination [38].
Esports players and fans co-exist on a virtual platform which is commonly referred to as a "community". Players can join each other in teams to train themselves at different skill levels. Fans could watch other people playing to learn specific gaming skills [1]. When not associated with organised competitions such as tournaments or leagues, these groups are often not governed by publishers directly. Therefore, as respondents pointed out, "the low level is where there's toxicity in the esports community" (FG 3). A respondent in FG 2 provided an example: There was a 15-year-old who beat one of the best Smash players in the world at a tournament and knocked him out and his followers then ousted her out on social media and just gave her a ton of abuse [ . . . ] unfortunately a lot of amateur esports players are like that and communities will do that [ . . . ] but no one is there to protect their mental health.
Participants in FG 1 emphasised the exploitation of esports players who transform from amateur to semi-professional or professional levels: I see it literally every week. It's a terrible situation, the exploitation of those players who are trying to move from the amateur to the professional, through that kind of really murky semi-pro scene. And the lower down the tier of game you go, the worse it gets. So, there are massive problems. But they're currently beyond the scope of any kind of central regulation, unless the game publisher decided to do something community by community. And honestly, I see no will, I talk to every publisher regularly. I have seen zero will from them to organise in any meaningful way the pathway to pro.
It appears that although game publishers take the lead in governing the esports industry at the elite level, there is a lack of interest in governing the entire industry. In particular, game publishers do not coordinate network-level activities that can be considered major, such as the exploitation of non-elite players. Since all major network-level activities and key decisions are supposed to be coordinated through and by the lead organisation according to [12], game publishers cannot be considered as the Sustainability 2020, 12, 8270 9 of 15 lead organisation of the entire esports governance network. The consequence is that the activities not coordinated by game publishers require or open the door to a regulation by other actors. As a result, a number of stakeholders start to emerge in the industry to facilitate the network governance.

Emerging Stakeholders: Diversified Interests Lead to Fragmented Governance
Despite game publishers possessing an authoritative, perhaps monopolistic position in the esports governance network, some influential stakeholders within the network are emerging. Based on their attributes, we classify these endeavours into two types: (1) national and international esports (governing) bodies; (2) self-proclaimed industry guardian organisations. The rise of these esports stakeholders causes a fragmentation of governance in the industry.

National and International Esports (Governing) Entities: A Legitimacy Issue
To represent and protect their best interests, some esports league/tournament organisers, teams and players have cooperated to create national or international esports governing organisations, separate from game publishers. These attempts have resulted in more than one esports governing organisation.
At present, three organisations have a claim to being the international governing organisation for esports. The International Esports Federation (IeSF), with 63 national esports organisations as its members, aims to legitimise esports as a sport [39]. The World Esports Association (WESA) was the result of joint efforts between professional esports teams and tournament organisers (e.g., ESL) [40]. The Global Esports Federation (GEF), with a somewhat similar mission as IeSF, partnered with one of the biggest game publishers, Tencent, to create industry guidelines [41]. Although tensions exist, none of these organisations seems to have acquired the legitimacy needed to govern esports worldwide. A respondent commented: None of them are significant in any way. They are small organisations with very little clout, because [ . . . ] in top end esports, tier one esports, or games like League of Legends, Counter-Strike, etc. nobody cares about country versus country. What they care about is team versus team, and they have no power to impose regulations to act like a governing body, because nobody recognises them as legitimate (R4). This is understandable because they do not possess any legitimacy to govern a game without the permission of game publishers, and this permission is unlikely to be forthcoming any time soon. When asked the reasons behind the lack of legitimacy of these international bodies, respondents pointed out: None of them wishes to give up autonomy over any aspect of their business. That is what it boils down to. They do not wish to cede authority on any single issue, to anybody outside their business (FG 1).
At the national level, the issue of having multiple esports governing bodies in one country is rare. However, in line with the international esports governance status quo, equivalent national esports governing displays mixed loyalties. For example, as a member of the IeSF, the Australian Esports Association (AESA) would focus on their vision of esports being formally recognised as a sport in Australia [42]. As a member of the Global Esports Federation, the British Esports Association (BESA) focuses on educating and promoting esports at the grassroots level. What is worth mentioning is that, instead of claiming to be the governing body of esports in a specific country, national esports associations like AESA and BESA have both acknowledged that they are not governing bodies but national (advocacy) bodies of esports in their respective countries.
A respondent pointed out the challenge for national esports federations to have an effective governance approach in their countries, i.e., imposing regulations on the esports industry: Bear in mind that you're talking about a global industry here, so if you create barriers in one country, so you decide that you're going to implement a bunch of regulations in France, you've got to bear in mind that the esports industry has a choice as they'll just go somewhere else. You won't solve the problem. You'll just ruin the system within your country because everybody will just decide that France is not the place for them and they'll go to Germany or Italy or Spain (R1).
International esports organisations use a number of strategies to acquire legitimacy. This includes establishing a governance structure with statutes and associating with an already legitimate organisation. For example, the IeSF imitated the governance structure of traditional sport governing bodies by establishing a board representing the interests of members (e.g., national esports associations) and a general meeting where board members could meet regularly and make decisions collectively [39]. Written statutes were also created by IeSF specifying the roles, rules and obligations of the organisation and its members [39]. In addition, both IeSF and GEF associate themselves with traditional sports organisations such as the WADA or national Olympic committees to legitimise their governance [39,41].
In some countries, the government has also involved themselves. For example, in South Korea, the government authorised the establishment of the Korean e-Sports Association (KeSPA) to govern esports in Korea. As such, the KeSPA have jurisdiction rights over the leagues on which they operate. This includes the capacity to charge players or teams with criminal offenses under the country's criminal law [43]. In France, there is a law to protect and regulate professional esports players by setting limits on player contracts to a five-year maximum and 12-month minimum term. In addition, players under 12 years of age are prohibited from competing in tournaments offering prize money [44]. In China, esports was not only officially recognised as a sport by the government, but esport players are also considered athletes [45]. In this sense, the esports ecosystem in China is entirely different from that in other countries, as the Chinese government has the power to directly govern the esports industry in the country, including game publishers and players Through the organisation of State Administration of Press, Publication, Radio, Film and Television.

Self-Proclaimed Industry Guardian Organisations
Apart from esports players, teams, tournament organisers and game publishers, there are also what we term "self-proclaimed industry guardian organisations" emerging. One example is the Esports Integrity Commission (ESIC), which is a non-profit organisation that partners with esports stakeholders to protect the integrity of esports. To tackle the threats of match manipulations, betting fraud and other integrity challenges posed to esports, ESIC created voluntary codes of conduct for players, teams, tournament and league organisers and even national and international esports (governing) bodies [46]. An ESIC representative commented: We've created a set of standards and a set of regulations that various stakeholders can voluntarily adhere to and cooperate in and that's what we're doing, literally company by company by company, getting them to voluntarily accept in a very narrow area of regulatory, anti-corruption and integrity (R4).
As ESIC's regulatory framework has been deemed effective in combating esports corruption, several major esports tournament operators such as ESL and DreamHack have embraced ESIC's regulations. However, there are challenges encountered by this voluntary organisation when reaching the wider esports community with their regulatory framework: Getting them to accept our jurisdiction and help is difficult [ . . . ] because all companies [publishers] are competitors so it's not like they're all desperate to work together. It's my job, in a way, to persuade them that it's in their commercial interest to work together and that's a very difficult job (R4).
Another example includes gaming industry international trade bodies such as the Entertainment Software Association (ESA), the Association for UK Interactive Entertainment (UKIE), the Interactive Games and Entertainment Association in Australia and New Zealand (IGEA) and the Entertainment Software Association of Canada (ESAC). In a joint announcement, these organisations unveiled four principles-safety and wellbeing; integrity and fair play; respect and diversity; positive and enriching game play-which are expected to be adhered to by the members of their organisations, including the game publishers [47]. Adherence to these principles may have reputational and image benefits for game publishers and be used in their corporate social responsibility strategies. These dimensions may be the drivers so that game publishers accept external regulation and, indirectly, work together.
Although the emergence of these governance alliances alongside the publishers has created a chance for the nascent industry to address governance issues which the publishers may have neglected, their diversified roles and interests have made the current esports governance a rather fragmented picture.

Discussion
The findings provide answers to the three research questions. Indeed, they show that: (1) There are (at least) seven key stakeholders and three emerging stakeholders of esports governance (see Table 2).
(2) The current esports governance framework features some attributes of the "lead organisation-governed network", with the power residing mainly in game publishers.
(3) The rising power of other (emerging) stakeholders in the network seeking to address integrity issues has caused fragmentation of the esports governance framework.
This section provides further discussion of the dynamics amongst publishers and emerging stakeholders (5.1), before suggesting an evolution of esports governance from fragmentation to a network administration organisation (NAO) model (5.2).

The Dynamics amongst Publishers and Emerging Stakeholders
Macey et al. [7] noted that "the rapid growth of esports, combined with its grassroots nature, does not exist within the same cohesive governance that is present for many sports." This was reflected in our findings, which demonstrated that the current esports industry features some aspects of a lead organisation governance model [12]. More specifically, game publishers act as the lead organisation at the elite level in their respective game's network. Since game publishers own the intellectual properties of video games, they are the main players in the network. By regulating esports players, teams, sponsors, broadcasters and tournament organisers, game publishers coordinate the network to achieve network goals that align closely with their own goals, i.e., maximising profits [48]. As a market-oriented governance network, which is derived from a procedural rationality [13], the esports industry opts for less state intervention and more market procedures and regulations in its governance mechanism.
However, there is a downside with a procedural rationality-based governance, which is its inefficiency to prevent instability, externality and inequality [13]. In the case of the current esports governance model, game publishers tend to show more interest in governing elite esports and less incentive to govern where there are less profit-making opportunities, such as at the grassroots level of esports. Such inefficiency has led to the neglect of rising unethical behaviours in the esports community and therefore caused severe consequences for players' mental health and esports integrity.
Catalysed by the rising concerns surrounding the industry, alliances started to emerge within and outside the network to facilitate the governance of esports. Alliances between esports players, teams, tournament organisers and, in some cases, governments and third-party organisations (e.g., ESIC) attempt to offer a solution to the governance issues in the esports community. However, such coalitions are in conflict with publishers in terms of the ownership of intellectual properties. The exclusivity of publishers has prevented these new alliances from gaining legitimacy over the governance of the industry. Third-party voluntary organisations such as the ESIC and gaming industry international trade bodies, similarly, exist as advocacy rather than governing roles in the industry due to such tension. In this sense, publishers have constrained the establishment of a comprehensive governance mechanism that would take into consideration the protection and representation of all stakeholders, particularly those in a weaker position.

Evolution of Esports Governance: From Fragmentation to Network Administration Organisation?
Although struggling with legitimacy issues, new esports governance alliances are following a trend of moving away from fragmentation to a network administration organisation (NAO) model. Provan and Kenis [12] pointed out that a lead organisation governance model may be best suited when network members are less able to solve issues on their own, but not conducive to long-term sustainability. In the esports industry, it is evident that there are governing "blind spots" that game publishers are not able or even willing to attend to as the sole governing body, and for a more sustainable and healthier ecosystem, it is necessary to take all stakeholders into consideration. The NAO governance model offers a potential solution, as members of the NAO are typically committed to the network-level goals and involved strategically as a whole [12]. Thus, goal consensus and the creation of the conditions to achieve such goals are favoured in the NAO form.
However, building a NAO governance model in esports has a few challenges. First and foremost is the compliance of game publishers. Without the engagement of publishers, a NAO will have limited power over the governance of the industry. In the case of current NAO "governing" bodies, such as national or international governing bodies, very few of them have involved publishers in their governance approach; therefore, their existence is as essentially mere tournament organisers that partner with publishers or as educators with the aim of training and educating esports players. They do not have the real power to govern the industry and this is detrimental to the sustainability of such networks. Human and Provan [25] pointed out that the failure to build legitimacy can lead to network collapse. Nevertheless, the cases of China and France, where governments are involved in esports governance, potentially provide a solution to this challenge. With the increase in social issues encountered by the esports industry (e.g., verbal abuse, impact on mental health) and given their legislative and jurisdictive power, governments are increasingly expected to play a role in the governance of the esports industry. However, this solution probably works better in places where political relationships are vital, even to game publishers.
The other challenge is the fragmented reality of the esports industry. Game publishers have a different approach towards governing their games, and in different countries, there are various policies and regulations in place, e.g., some recognise esports as a sport and others do not. The current fragmented governing entities (e.g., multiple international esports governing bodies) have already caused confusion to people both inside and outside the esports industry. Esports being an international phenomenon makes it challenging to regulate the industry with a standardised approach. Although NAOs are probably best suited to networks that demand a wide variety of competencies [12], such as the esports industry, it is still early for the mix between lead organisation-governed model at the elite level and fragmentation overall to fully evolve into the NAO model that caters to such variances, as the esports industry is still at its incipiency. This by no means suggests that an effective NAO model will not take place at some point in the future, but it may need some time before stakeholders converge and agree that such a NAO model should be in place.

Conclusions
This study examined the sustainability of the esports governance model. The latter is currently fragmented, which may undermine its sustainability. However, a recent trend towards moving away from fragmentation to a network administration organisation (NAO) model may improve the sustainability of the esports governance model. As such, this paper provides a basis for the development of a more sustainable and healthier esports ecosystem, one that caters to the interests and rights of all stakeholders, not simply the publishers.
This study has a number of practical implications and theoretical contributions. On the practical side, three implications are identified. First, through an overall investigation of the governance structure of the esports industry, the paper presents a holistic picture for stakeholders within the esports industry to understand the accountability capacity of the existing governing entities. Second, this paper provides some explanations of the current fragmented governance mechanism, which is caused by the divergence of interests and the dynamics amongst game publishers and its stakeholders. This is of particular importance to endemic stakeholders such as sports organisations, particularly post-COVID-19, as an influx of sports businesses such as the National Basketball League (NBA), Major League Baseball (MLB), NASCAR and Formula 1 turn to esports to replace their lost sporting