Abstract
The present study aims to explore the social conditions and underlying mechanism of match-fixing, especially when it is not facilitated in a quid pro quo relationship. Analyzing scandals in Taiwan’s Chinese Professional Baseball League, our research investigates how both individual agendas and organizational interests are served in match-fixing. Practically, the prerequisites and actors’ rationales for their involvement in match-fixing are explained. Theoretically, actors’ motivations and the exchange of intangible resources (not in the form of money) are examined. Our findings speak to theories of structural, cognitive and relational social capital by elaborating on the dark side of social capital. The relational attributes possessed by actors embedded in their social networks are involved in the cognitive process of actors’ decision making. Alternative policy implications are proposed to reduce the potential for match-fixing.
Introduction
This paper examines match-fixing, which exemplifies the corrupt side of sports, from the perspective of social capital, which has its own dark side (Numerato and Baglioni, 2012). As ‘a modality of sports related crimes’ (Haberfeld and Abbott, 2013a: 3), match-fixing has become a global problem (Anderson, 2019). Given the long history of corruption in sports (Rosen, 2008), attempting to disassemble the linkage between sports and gambling may be an unfeasible and outdated focus because:
Sports benefit from betting revenues (Forrest and Simmons, 2003).
Fans are attracted by both sports and gambling, which are both marked by uncertainty (Villeneuve and Pasquier, 2019).
Instead, what matters more is to ensure sports’ immunity from betting-related match-fixing because the manipulation of sports results is harmful to the sports industry as a whole (European Commission, 2012).
Scholars have attempted to better conceptualize match-fixing, including its definition (e.g. Brasseur, 2012, as cited in Villeneuve and Pasquier, 2019: 2), taxonomy (e.g. Boeri and Severgnini, 2011; Maennig, 2005) and techniques (e.g. Cheloukhine, 2013; Park, 2018; Rebeggiani and Rebeggiani, 2013; Vodde, 2013). With regard to its causality, economists have offered productive insights and have even established their own schools of thought (see Wolfers, 2006). Scholars of law and criminology (e.g. Blackshaw, 2012; Rodenberg et al., 2016; Van Rompuy, 2015; Yilmaz et al., 2018).) as well as scholars of history and culture (e.g. Huggins, 2018) have also joined the discussion. However, compared to the prosperous development of micro (individual) and macro (societal) analyses (Kihl, 2018b), little attention has been paid to the meso level (Huberts, 2010). Because match-fixing is linked to the involvement of criminal organizations (Villeneuve and Pasquier, 2019) and ‘there is often more to play for off than on the field of play’ (Blackshaw, 2012: 6), we propose an integrated micro-meso approach to study match-fixing. This approach is in line with the argument of Gorsira et al. (2018), who proposed investigating corruption from both individual and group perspectives. Moreover, while scholars such as Costa (2017, 2018), Hill (2008, 2009, 2013, 2016), Numerato (2008, 2016), Numerato and Baglioni (2012), Tak (2018) and Tak et al. (2018) have studied match-fixing based on empirical findings, constant academic effort is still required. By venturing out of the theoretical realm and engaging in the real world, we focus on the interaction among match-fixers and their collaborators, which is where negotiation and operation occur (Blackshaw, 2012).
According to Haberfeld and Abbott (2013b), a comprehensive framework for research on match-fixing should incorporate at least three dimensions: cognitive, structural and relational. First, the intrinsic and extrinsic utility value of sports cheating is a common factor in economists’ analysis (Rebeggiani and Rebeggiani, 2013), and these two types of utility value are strongly associated with actors’ cognition (Kasser and Ryan, 1993). Second, the non-monetary costs of match-fixing (see Bag and Saha, 2011; Forrest and Simmons, 2003) are related to actors’ social norms (Kohlberg, 1975), and this issue should be examined from a relational perspective (Jancsics, 2014). Third, the social network dimension cannot be ignored because:
Decisions are made based on various interpretations and situations in which reference points are recognized in social networks (Emirbayer and Mische, 1998).
Normative conditioning and cultural influence are circulated and adopted as pipelines via social networks (Mische, 2011).
A more systematic analytical framework composed of these three dimensions can therefore be an alternative to compensate for economists’ rational choice approach, which cannot thoroughly explain the more nuanced social interaction in the setting of match-fixing. Certainly, economistic approaches can ‘single out different factors influencing the behavior of the involved actors and . . . evaluate their respective contribution to the cheating decision’ (Rebeggiani and Rebeggiani, 2013: 165). Though this ability has both its strength and weakness, a cost-benefit evaluation is indeed strongly associated with actors’ cognitive processes (Kasser and Ryan, 1993). Moreover, because match-fixing is facilitated by the ‘fixing ring’ (Hill, 2016: 233), the way that actors interact with each other should also be examined. This includes actors’ embedded structure as well as their relational attributes with significant others.
Under what social conditions and through what underlying mechanism can match-fixing be activated without quid pro quo? More specifically, what prerequisites are involved, and how do actors rationalize their engagement? What motivates actors, and what do they anticipate? What is exchanged, if not in the form of money or financial interests? Given the repeated institutional governance failure of the Chinese Professional Baseball League (CPBL), should we infer that there are invisible hands behind the scenes? To unveil the complex dynamics, we highlight in this study the interaction among syndicates (as organized criminal groups), coaches and players (as partners mobilized by syndicates) (Mutschke, 2013). This aspect is rarely examined by academics, and its impact is thus under-evaluated (Vodde, 2013). Our examination is based on social capital constructs that have two levels of analysis and three dimensions. We believe that Kihl (2018b: 205) is correct in arguing that ‘the sport landscape offers fertile opportunities to examine organizational corruption through a multilevel approach that could build theories that combine these levels in an interconnected framework’.
Contextualizing CPBL scandals
Sports corruption is becoming increasingly globalized, and match-fixing, one form of sports corruption, is no exception. This issue has received extensive attention internationally (Opie and Steele, 2018). A review of match-fixing studies identifies several key points:
Match-fixing often accompanies betting or gambling (Menz and Skene, 2018; Spapens and Olfers, 2015).
Police, judicial efforts or public prosecutors are considered important in fighting match-fixing (Murnane and Moller, 2018).
Critical approaches to prevention endeavors are suggested (Moriconi and Almeida, 2019; Tak et al., 2018) at both the individual (player) and organizational levels (Chung et al., 2018; Forrest, 2013).
Nevertheless, some unexplored issues exist in the scholarship. First, although cost-benefit analysis has been conducted from the perspective of fixers, empirical research on the costs and how actors perceive these costs is relatively rare (Huggins, 2018). Second, compared to the rich knowledge generated by studies examining the betting aspect of the market, namely, the ecosystem of gambling operators (Spapens, 2017; Spapens and Olfers, 2015), less is known about the fixing aspect of the market, especially methods and strategies to realize match-fixing proposals (Costa, 2018). Third, the existing literature mainly addresses external factors, leaving discussions of the internal mechanisms among actors less examined.
In the context of globalization, Taiwan is not isolated from the global match-fixing trend. Established in 1990, Taiwan’s CPBL witnessed five notorious large-scale betting scandals from 1997 to 2010: the 1997 China Times Eagles scandal, the 2005 La New Bears scandal, the 2007 Chinatrust Whales scandal, the 2008–2009 Dmedia T-Rex scandal and the 2009–2010 Brother Elephants scandal. Court transcripts revealed that some players were either threatened or bribed to fix matches; some were kidnapped (in the 1997 case) or forced by club owners who were themselves the mafia (in the 2008–2009 case). As for the other three cases, it was reported that the mafia contacted some tier-1 players 1 to arrange collusions, and these tier-1 players reached out to other tier-2 or even tier-3 players within their social networks.
According to Arangure (2015), match-fixing has threatened baseball in Taiwan for years, but it is difficult to uncover given the mafia’s predominant intervention. Arangure (2015) further explained that the CPBL is susceptible to match-fixing because:
The Asian gambling market is large.
Taiwanese players are among the lowest paid players in professional baseball.
Match-fixing is not considered an illicit activity.
Good education for players is lacking.
A good example can illustrate the structural difficulty – with the aim to ‘regulate’ forces in the sports-gambling market, the legalization of sports lottery in 2009 2 was not as effective as expected. This is because the gamblers and syndicates, compared to the teams and clubs, were able to offer more to the players.
Thus, the history of the CPBL provides a rich set of material to examine the nature of match-fixing. A repeated pattern of illicit practices can be summarized as follows: coaches and players are mobilized and manipulated by the mafia to fix games so that the mafia can make a fortune from their gambling businesses (Lee, 2017; Lin and Chen, 2015). As a matter of fact, the mafia deployed a long-term investment strategy before these issues came to light. The player-mafia interaction has been in place for years in the form of: (a) financial support, such as sponsorships, scholarships or daily allowances; and (b) social bonds strengthened by organizing ‘extra-curricular’ activities that penetrate players’ personal and social lives (Lee, 2017). In Taiwan, the mafia may involve politicians or businessmen who have a passion for sports and supporting players. Their hidden agenda is well concealed, and it is legal and beneficial to their reputation to offer financial support to sports talent.
Conceptualizing social capital in sports, sports corruption and match-fixing
This section demonstrates how the notion of social capital can help answer the research questions. It is beyond the scope of this section to review all aspects of the interplay among social capital, sports, sports corruption and match-fixing. Instead, it is important to discuss the existing discourses and issues that have not been adequately examined. In this way, we can advance knowledge about match-fixing as a social phenomenon (Numerato, 2016).
The dark side of three-dimensional social capital
There are three major streams of social capital theories: the individual level (e.g. Bourdieu, 1986; 3 Coleman, 1990), the organizational and inter-organizational level (e.g. Burt, 1992; Nahapiet and Ghoshal, 1998) and the societal level (e.g. Putnam, 1993a, 1993b, 1995) With regard to sports and social capital, the academic works of Putnam and his followers are the mainstream (Jarvie, 2003; Spaaij, 2012; Widdop et al., 2016). However, Putnam’s approach, as noted by Morrow (1999: 761), serves to explain ‘the notion “civicness”, trust in institutions, and interest in public affairs’. This is not our topic and is therefore not applicable to this study. Notably, Numerato and Baglioni (2012: 597) warn that applying Putnam’s to examine sports phenomena of various natures can ‘run the risk of conceptual slippage’. On the other hand, the lack of meso-level discussions is one of the real problems facing the scholarship (Widdop et al., 2016). What is more interesting is that most analytic frameworks fall into the three dimensions proposed by Nahapiet and Ghoshal (1998), as shown by a further review of studies considering sports and social capital at the micro-meso level (Ivy et al., 2015; Pieters et al., 2012).
The convergence of the micro-meso level is applicable in this study, and the social capital constructs developed by Nahapiet and Ghoshal (1998) are given prominence in our inquiry. Articulating social capital at both the organizational and individual levels, they define social capital as ‘the sum of the actual and potential resources embedded within, available through, and derived from the network of relationships possessed by an individual or social unit’ (Nahapiet and Ghoshal, 1998: 234). The three dimensions of social capital are summarized in Table 1.
The measurement of organizational social capital.
Source: Nahapiet and Ghoshal (1998: 251).
The three dimensions of social capital can have both positive and negative effects, though the negative effects are less investigated (Banerjee, 2016). For the structural dimension, close interorganizational social networks can negatively influence organizations’ performances (Adler and Kwon, 2002); social capital facilitates corrupted public-private partnership when social ties exist between civil servants and firm executives (Collins et al., 2009). For the cognitive dimension, social capital consolidates the taken-for-grantedness of corruption (Nyamori and Gekara, 2016), which becomes accepted and even institutionalized within organizations (Anderson and Jap, 2005). Moreover, shared systems of meaning shape decision makers’ individual cognition, which then influences organizations’ responses to their embedded institutional environment (Collins et al., 2016). For the relational dimension, prevailing norms modify organizational behaviors – firm corruption spreads when decision makers’ entrepreneurial orientation surpasses their innovative orientation (Karmann et al., 2016). Finally, cross-dimensional effects are found in some cases – in light of institutional failures and the lack of trust in many African economies, it is essential for African entrepreneurs to efficiently develop their business networks (Lux et al., 2016).
Overall, social capital is a double-edged sword, and scholarly discussion about its dark side is mainly at the individual and organizational levels. Now that the association between social capital and corruption is presented, the next step is to focus on links between social capital constructs and match-fixing.
Sports, sports corruption, match-fixing and social capital
There is a longstanding debate about whether corruption is due to micro, meso or macro attributes (Kish-Gephart et al., 2010). For instance, De Graaf (2007) argues that corruption cannot be thoroughly explained via only the micro or meso approach. Given the nature of match-fixing and social interactions among the actors involved, 6 an integrated micro-meso perspective is suggested in this study. In particular, the social capital of our concern is at the micro-meso level. Another issue is that topics concerning social capital and sports corruption are not adequately explored by sports sociologists, and studies addressing both social capital and match-fixing are even rarer. We acknowledge the insights of some pioneers, such as Numerato and Costa, which significantly inspire our analysis. However, we also recognize the methodological and theoretical issues to be addressed.
Methodological innovation awaits us in this new domain, and it is necessary to incorporate other qualitative data for the following two reasons. First, actors’ network configuration provides the background of the broader picture, leaving the underlying mechanism unexplained (Stolle, 2001). Compared to social network analysis applied by Costa (2017, 2018), an ethnographic approach is adopted by Numerato (2008, 2016) and Numerato and Baglioni (2012). Clearly, these two sets of methods have their respective strengths, but an integrated design can be even more convincing. In this vein, Numerato’s approaches prove effective, and the subjective bias of interviewees (Flick, 1992) could also be resolved provided that secondary data are verified. Second, rather than the meso level, the discussion about the dark side of social capital in sports scholarship is usually at the macro or micro level (Numerato and Baglioni, 2012). Nonetheless, issues at the organizational level should not be ignored in regard to corruption. To mend this gap and to bring meso-level analysis into the scene, a qualitative approach is desired because data at the meso level, especially those of a sensitive nature, are less present in the extant literature.
Besides, unresolved theoretical puzzles also require further clarification. For instance, Costa (2017, 2018) considers the concept of social capital as a single entity without defining what social capital is or examining how it functions. In other words, he takes social capital at face value and treats it as a collective concept. This is understandable because his interest is in social network analysis. Indeed, recognizing who links with whom provides an outline of the phenomenon. But investigating the rationales of the connections can also generate other sociological implications that quantitative approaches cannot (Lin, 2001). Compared to Costa, Numerato’s approaches prove more convincing, although his theoretical explorations could have been further refined by a detailed distinction of social capital constructs.
First, although Numerato identifies how the cognitive dimension influences the function of relational and structural dimensions, he could have further distinguished the absence and negativeness of social capital constructs. While his focus is ‘the lack of social trust’, ‘lack of identification’ and ‘deconstruction of social ties’ (Numerato, 2008: 31), one should bear in mind that all of these do not equal the negative impacts produced by social ties and other attributes. Second, epistemologically, social capital is considered a ‘heuristic device’ by Numerato and Baglioni (2012: 597). However, they pay little attention to the correlations between each dimension and corrupt sports governance. In addition, while they argue that abusing social ties and trust explains the dark side of social capital, other social capital constructs could have been made more evident in their discussion, as recognized in their cited examples. 7 Third, Numerato (2016) concludes the normality and complicity of match-fixing as well as three ideal types of match-fixing (direct, mediated and meta). He argues that public secrecy about match-fixing affects social relations in the context of sports, especially in terms of distrust. However, the leverage between trust (or distrust) and match-fixing could have been further examined by a cross-level analysis.
In brief, our justification of applying social capital constructs to understand match-fixing is twofold:
Match-fixing is frequently related to sports governance (Numerato, 2016: 705), in which social capital plays a role (Numerato and Baglioni, 2012).
Less academic attention is paid to sports governance than to sports practice in examinations of the dark side of social capital (Numerato and Baglioni, 2012).
Similar to these scholars, we consider social capital as a whole. Furthermore, we believe it is necessary to comprehensively reflect on each construct of social capital when examining match-fixing (Hedström, 2005). This approach will avoid the ‘black-box correlations’ of social capital constructs, as Seippel (2006: 171) warns.
Methodology
To address the methodological gap, a two-mode approach was adopted to keep the conversation open and to encourage informants to speak. The importance of referring to reported cases cannot be overemphasized. To avoid situations in which an informant might refrain from discussing relevant topics, we began with the analysis of published court transcripts and press commentaries. This strategy served two functions. The first was the generation of supporting evidence that helped to formulate the interview materials. The second was the setting of an environment where informants were less defensive or hesitant to discuss sensitive issues, since the talking materials did not directly involve them personally. In brief, thanks to this strategy we were better equipped to trigger discussions in our fieldwork.
In-depth interviews also generated valuable output. The mafia 8 that organized the match-fixing, the coaches who were the go-betweens, and the players who were manipulated were interviewed. The mafia can mobilize other two parties through violence, sexual favors and other material incentives. However, they also practice soft coercion via social capital – that is exactly the core concern of this current study. Nine in-depth interviews and two focus groups were conducted with four coaches (A, B, C, D), five players (E, F, G, H, I) and five syndicates (J, K, L, M, N). In general, interviewees were invited to briefly draw a network graph of their perceived ‘fixing ring’. Based on the visible network architecture, more questions about the cognitive process and relational attributes were raised. Moreover, field notes were taken when participant observations were conducted during informants’ off-work time (with C, D, I). Rich data and valuable insights were collected in these informal conversations.
Because the issues were sensitive, we invited the interviewees to talk about the general facts of which they were aware while excluding themselves from the scene. As stated earlier, the key was to keep the conversation open and to encourage informants to provide their comments. We therefore did not insist that informants specify whether they were telling their personal experience. Additionally, it is important to emphasize that even though the informant(s) used the expression ‘I’ or ‘we’, this does not necessarily mean that the informant(s) conducted the (mis)behavior mentioned. Under the codes of language (Mandarin and Taiwanese), the expression ‘I’ or ‘we’ denotes generality without specifying any specific individual(s).
Overall, the conversations were streamlined thanks to this two-mode approach featuring in both the analyses of archival data and face-to-face interactions.
Intertwined dynamics of social capital behind match-fixing
We present our findings along two lines: (a) from network configuration to actors’ perceptions and interpretations; and (b) from the mafia’s perspective to that of players while keeping coaches present in the collective system. We elaborate on relevant dynamics when: (a) a network’s appropriability is manipulated; (b) shared narratives are the reference frame; and (c) a match-fixing proposal is executed in relation to norms and trust.
The grooming mafia and the groomed players
The focus of our study is on match-fixing that is not facilitated in a quid pro quo relationship; that is, when innocent players are mobilized to fix matches. It is thus necessary to first distinguish corrupted and innocent players 9 in our context. In addition, the position and role of the mafia need further elaboration. Occupying the central node of the network, the mafia manages to construct ties with diverse actors – from clubs and teams to coaches and players – to different degrees and for various purposes. As the funding source, the mafia sets the scene and intentionally constructs networks which coaches and players are co-opted into. These syndicates mobilize players directly or indirectly, and they can even intervene in the operations of teams. 10
In general, the mafia can access innocent players through three paths of influence: through corrupted coaches, through corrupted players 11 and directly to innocent players. In other words, the link between the mafia and innocent players can be dyadic or configured via a third party. This configuration also implies that the long-term relationship can be transformable and appropriable for other purposes. Nahapiet and Ghoshal (1998) use the term ‘appropriable organizations’ to refer to the fact that a set of networks, despite being initially created for one purpose, can be transformed to serve other functions. In our case, a variety of ‘appropriable organizations’ deserve further distinction as follows.
First, the coach–player relationship is similar to a father–son or parent–offspring relationship, and the relationships among players are similar to sibling–sibling and friend–friend relationships (Lee, 2017). When there are unwanted variations in actors’ deep relationships through these two sets of network configurations, a typical ‘appropriable organization’ manifests. Second, the mafia–player relationship is more complicated. On the whole, generational differences and the mafia’s return on investment (ROI) are the two factors. In the old days, the older generation of players grew up in Taiwan when professional baseball was not yet established. Leaders of these mafia groups groomed players either out of their passion for baseball or as part of their human resource development. 12 This fits the concept of ‘appropriable organizations’ because the mafia groomed players not for match-fixing. In contrast, when Taiwan later commenced its professional baseball system, the mafia intentionally groomed players and considered it an investment from which future returns might be secured through match-fixing. 13
Networking among players, coaches and the mafia lays the preliminary foundation for a match-fixing proposal. A well-cultivated relationship is indispensable for the mafia to mobilize players, either directly or via a third party: ‘I must protect myself. So when someone approaches me, I first check our common links. If there are none, then it’s more difficult to say yes and to get the deal done’ (A, coach).
Compared to Philippe and Seiler’s (2006) finding that a good coach–player relationship correlates to players’ performance, we note that a good coach–player relationship opens the door for match-fixing. The role of a coach is that of a ‘project manager’, as Hill (2009) describes with regard to football scandals. But how can the mafia mobilize players through appropriable networks? The symbiotic system deliberately constructed by the mafia is the key: the mafia invests in building relations with players from their childhood, with returns expected when the players become adults. The players, on the contrary, are not aware of the mafia’s hidden agenda. Because each ROI takes approximately 10 years and the ROI of every player varies, the mafia finds it indispensable to build ties with as many young players as possible. This arrangement is a hybrid between Hill’s (2009: 21) dichotomy of quantity-oriented ‘fast bucks’ and the quality-based ‘counterfeit intimacy’ methods by which players are approached. Syndicate H verifies the mafia’s ‘investment’ in players:
Of course, we want them to pay back. They’ve received our support for a long time, so it’s normal (that we want something from them). And you know what? Compared to the amount we’ve spent on them, our return on investment is much more. (K, syndicate)
Our findings are in line with those of Numerato and Baglioni (2012), who believe that players’ networks with sponsors can lead to match-fixing. However, in our case, sponsors are syndicates as criminal groups that, in the back stage, manipulate players. These facts suggest that the mafia manipulates operations and abuses the network configuration. However, it is players who make the ultimate decision whether to refuse or accept a match-fixing request, demonstrating that a static network configuration is not the only significant variable to explain the dynamics of various negotiations contained therein. To understand how actors interpret their interactions and why they are willing to be involved in illicit behaviors, we need to consider other variables.
The shared frame of reference
The mafia efficiently mobilizes players by manipulating tacit understanding and fostered comradeship among players. These two factors, reinforced by shared antecedents, help to accomplish match-fixing deals. First, when players ask for their fellows’ help to fix the game, quite often it is without excessive effort of verbal communication or explanation. This is because a good rapport enables their tacit understanding. This also explains why the mafia can efficiently mobilize a constellation of players:
Among players, they’ve already got a good relation. So if one day player A (tier-1) has difficulties and the mafia offers to help A in exchange for his cooperation to throw a match, A will find even more fellows (tier-2 and tier-3) to join the match-fixing. (M, syndicate)
Second, tacit understanding further involves the fostered comradeship incarnated in and strengthened by common languages and narratives:
Compared to referees, coaches are much more often targeted by the mafia. You know why it’s more efficient to mobilize players via coaches (than via referees)? One reason is that coaches and players share much more similarities in their life experiences and career path. Before becoming a coach, we used to be a player. We know they have strong comradeship. We know how to take advantage of that. (C, coach)
In Jowett’s (2003) terms, shared languages and narratives lead to co-orientation in the coach–player relationship. In our study, this shared frame of reference and co-orientation also exist in player–player interactions. Tacit understanding and fostered comradeship develop in long-term relationships, and they have spillover effects on players’ evaluation of match-fixing proposals – in players’ cognition, what is highlighted in ‘helping match-fixing’ is the helping, rather than the fixing: ‘BUT ATTENTION: I help my buddy (to throw a match) doesn’t mean that I fix a match. I don’t care if it’s a fixing but I do care that I’ve helped my buddies’ (K, syndicate).
Additionally, two other facts have similar spillover effects and will be elaborated below.
Players’ mutual support as the norm
Supporting each other and helping those in need without asking for money are considered obligations by players. More specifically, there is a transformation from obligations to taken-for-granted subgroup norms when players identify each other as being in the same community. Since helping fellows is an obligation for players, failing to do so thus makes them feel bad even if the process is illicit or wrong and might lead to negative consequences. This can be further analyzed from the following two perspectives.
First, obligations are transformed into norms through identification, and the fact that members identify each other as in the community of a team sport strengthens the conditioning power of subgroup norms:
You can have a dirty player on a clean team and vice versa. One rotten apple spoils the barrel. The tier-1 player can seduce his fellows, not necessarily through money. Because they see each other as buddies and mutual help is an obligation, this tier-1 worries less when asking a favor. When more and more players are involved, you just can’t easily change such unspoken rules. (B, coach)
Second, these norms further transform inappropriate behavior in society into something that is acceptable in their subgroup. A player might be excluded if he/she fails to comply with the subgroup norms that rationalize collective actions: ‘In Taiwan we do have honest players. But it’s hard for them to stay innocent, especially when they have more and more fellows that are stained. They can be quite isolated as well’ (C, coach).
As Coleman (1990) notes, obligations occur at the personal level, while norms are more generalized at the collective level. More specifically, Goa et al. (1996) note the importance of maintaining harmony in Chinese culture. Therefore, as long as a set of subgroup norms exists, one can convince and comfort oneself that what is done is acceptable regardless of whether it verges on the illegal. In Numerato’s (2016) terms, this marks the normality of match-fixing. Evidently, the norms of the subgroup surpass those of the external group, and this makes it possible to negotiate match-fixing.
The social circle in Taiwan’s professional baseball is quite small, making it easier to maintain subgroup norms and more difficult for players to reject a match-fixing proposal in a collective system. The obligation to help each other through thick and thin, together with the subgroup norms of ignoring potentially detrimental issues, contribute to the maintenance of a long-term relationship. This in itself is both a means and an end. In addition, it also differentiates actors’ trust in this study from other ideal types of trust.
Players cannot help but to trust
Mutual trust enables the possibility of match-fixing because it guarantees actors protective arrangements so that things are under control. Once actors are reassured, they can concentrate on operational techniques of match-fixing. As for the possibility to incubate trust, the existing pan-family relationships among actors stimulate trust to a certain level where temptations or interventions are more likely to occur.
Hard coercion through violence, threats and extortion does exist. However, soft coercion is the highlight of this study. Players are compelled to trust and help their fellows while ignoring potentially detrimental issues to a certain degree. But why are they willing to run the risks? First, the obligation of mutual help, together with identification in the same community, lays the foundation for players’ belief that the helped will not get the helping into trouble. As a former CPBL player mentioned: ‘We have known each other since childhood. I find no reason why my buddies would do something against me’ (E, player). Second, if players reject help, they may run other risks that they cannot afford, such as destroying their relationships with fellows. In reality, ‘many players find it almost impossible to reject. Quite often, the requests are from someone who used to help you, and you owe him an enduring debt of gratitude’ (I, player). Finally, the motivation to trust increases when protective measures to guarantee less risk are expected, provided that subgroup norms value mutual help without considering the detrimental consequences. In other words, identification, obligation and norms in the collectivist system jointly lower the costs of trust.
Players consider two contrasting sets of criteria when evaluating the costs and risks of match-fixing proposals. On the front stage, they think about their reputation in society and their career development should their illicit activities be disclosed. A coach who was formerly a player described the situation in the worst years: ‘We even dared not go out for shopping or whatever because if people recognize you’re a player, you just feel shameful’ (D, coach). On the back stage, they must pay attention to their reputation among their fellows and follow their coaches’ instructions. That way, ‘(w)e’re not marginalized and can avoid being put on the bench’ (H, player). Players have no choice but to trust when subgroup norms matter more than those outside the subgroup and when preventive measures are believed to be taken. Consequently, players are forced to trust, and the costs of trust decrease.
Briefly stated, the mechanism functions with various interlocked elements. It requires identification embedded in a relation in which obligations are expected. Conformity to subgroup norms rationalizes and normalizes practices that are not entirely legally correct. Actors have no choice but to trust, with reassurance expected.
Discussion
Our case differs from other works in different contexts in the following three ways. First, the players in our case are usually asked to lose a game rather than to win a game. This is different from other match-fixing cases (e.g. Brooks et al., 2013). Second, our case contrasts with Hill’s (2009) observation because the mafia can 14 access players directly without intermediaries. Finally, compared to other corrupt sports contexts observed by Nowy and Breuer (2017), it is not always the quid pro quo relationship that motivates players in our case. Material incentives are not always involved in the process, especially on the players’ side. Instead, players are willing to fix a game for free upon the request of their fellow players, coaches or even the mafia.
Our empirical findings can also contribute to scholarship on both sports corruption and social capital in the following ways. First, to the best of our knowledge, the current study is the first to apply micro-meso social capital constructs to examine the social conditions and underlying mechanism of match-fixing. Following Ashforth et al. (2008: 674), we aim to ‘understand why entire industries harbor corruption’. In particular, what concerns us is why and how match-fixing without quid pro quo is facilitated and what theoretical implications can be concluded. Little is known from existing scholarship (Hill, 2008, 2013, 2016), and the approach of Nahapiet and Ghoshal (1998) proves efficient in understanding the context of match-fixing, especially match-fixing for free. It is not easy to resolve macro-meso-micro issues (Huberts, 2010), but Kihl (2018b: 205) is correct when claiming that ‘to better understand sport corruption’s complexity, it is also critical to look beyond individual organizations’. The micro-meso approach of our study has merits because:
The mafia, as an organized criminal group, drives the collective actions of coaches and players in the system.
The mafia’s mobilization occurs at both the individual and organizational levels.
Second, social capital plays a role in match-fixing. The relational attributes possessed by actors embedded in their social networks are involved in the cognitive process of decision making. Through the lens of the structural, relational and cognitive dimensions, we can better understand the interaction of the actors involved, their activities and the resources that are exchanged. Thus, a framework that includes the networks, relations and cognitive aspects is desirable. Kihl (2018b: 202) argues that ‘corruption (in sports) is both a state and process’. Nevertheless, further distinction between the two is lacking in her analysis. Our evidence suggests that actors’ networks function as the state, while their cognitive and relational social capital are embodied in the process. More elaborations are presented as follows to explain why social capital constructs help understand the fabrics of match-fixing that is deeply ingrained in actors’ reciprocity.
The significance of three-dimensional social capital in explaining match-fixing
According to Kihl (2018a), social network analysis is necessary to better understand sports corruption. However, she does not distinguish the nuances of social networks and relationships. We argue that social networks per se cannot thoroughly explain match-fixing. While Numerato and Baglioni (2012) recognize that social networks serve negative purposes, our findings affirm that cognitive and relational social capital also lead to negative consequences. Although they do not specify their theoretical perspective, their perspective corresponds to the three dimensions of social capital. The definition of the dark side of social capital, according to Numerato and Baglioni (2012: 596), is ‘situations in which trust, social ties and shared beliefs and norms that may be beneficial to some persons are detrimental to other individuals, but also to sport movements, as well as to the society at large’.
The function of each dimension
Each of the three dimensions, with its respective function, ironically provides solid foundations for the realization of match-fixing. First, the structural dimension describes the network configuration among actors. It also manifests the appropriable organization that creates more space for alternatives. Second, the cognitive dimension denotes the reference frame. It connects the antecedents and the present that features in match-fixing proposals and evaluation. It also ‘enables the creation and transfer of new interpretations’ (Nahapiet and Ghoshal, 1998: 254), even though it is linked to match-fixing. Third, the overall dynamics are executed in the relational dimension. The match-fixing context ironically highlights norms and trust. According to Nahapiet and Ghoshal (1998: 255), norms ensure ‘the motivation to engage in such exchange’. Trust, in contrast, is the willingness to believe in a partner’s good intentions, competence and reliability (Mishra, 1996).
Appropriable organizations of social networks
Introducing nuances concerning appropriable organizations of social networks, our findings speak to arguments of Nahapiet and Ghoshal (1998). Currently, grooming is intentionally implemented in the mafia’s long-term relationship with players. It appears on the face of it not to be applicable to the appropriable organizations. However, some variables can influence the utility or otherwise of social networks. For instance, the asymmetric structure between two parties of a social tie matters. If one party builds the tie with its own hidden agenda while keeping the other party in the dark, then the initial purpose is not shared by all parties involved. From the perspective of the innocent party, who is therein abused, this scenario still represents an appropriable organization. Furthermore, one should bear in mind that the quality of social networks changes over time and the possibility of using or even abusing that set of social networks floats.
Cognitive and relational social capital
Kihl (2018b: 205) argues that stakeholders’ actions are influenced ‘through structures, systems, and culture that guide their values, priorities, and practices’. Based on our findings, tacit understanding and fostered comradeship strengthen players’ belief that helping should be weighted much more than fixing in ‘helping match-fixing’. This is their shared codes. Furthermore, the fact that players have no choice but to trust can be considered the joint effect of identification, obligations and norms. Players believe that the stakeholders whom they identify as being in their same community will not get them into trouble; they deem that their own interests will not be endangered because protective measures are taken to hide illicit agendas. In match-fixing, these factors jointly decrease the level of risks perceived by players because protective measures are expected. In brief, the functions of shared norms, obligations, trust and identification are consolidated to sustain the sense of a shared community. A hotbed of crime is thus created.
Social capital: positive attributes serving negative consequences
Our results show how constructs that are generally believed to have positive attributes lead to negative consequences. As exemplified, individuals of a given group or organization are not necessarily passive victims of corruption at the meso level. Instead, they can be actively involved in the reproduction of corruption. Through interpersonal interaction, each member develops his/her personal positioning within the collectivity and cultivates his/her own social capital. This is notably crystalized in terms of trust, shared norms, identification with the group, and obligations to support each member and the group as a whole. Unless members are strongly aware of the corruption problem and highly motivated to fight against it, they tend to let it be and thus become part of the accomplice structure (Labianca, 2014; Pecujlija et al., 2015).
Conclusion
Considering match-fixing without quid pro quo as a given fact, we probe its complex operative facets. We argue that negative effects of structural, cognitive and relational social capital can explain: (a) essential prerequisites and actors’ rationales for their involvement; and (b) what motivates actors, what is anticipated and what is exchanged.
Truly, both bright and dark sides of social capital coexist. A side effect of players helping their fellows in the name of social capital might be the threat to the sustainability of sports. Nevertheless, one should not be too pessimistic because the three-dimensional social capital proposed by Nahapiet and Ghoshal (1998: 244) also has anti-corruption functions. Based on our empirical findings and reflections on theories, three approaches are proposed. First, the diversification of players’ social lives should be encouraged in their education and professional trainings. This is because the structural connection is about ‘who you reach’, and relationships can influence actors’ behavior. Next, a systemic governance framework for the organizational field is desirable, given that the cognitive dimension is about ‘systems of meaning among parties’. Finally, a joint effort of the aforementioned two is feasible if we recognize the importance of a clean sports environment for the interests of all actors involved.
Future methodological and theoretical endeavors are proposed. Methodologically, a more complete picture can be presented by obtaining viewpoints of other stakeholders (e.g. professional sports team officials, referees, and family members or acquaintances of players). It is hoped that mediated and meta match-fixing can be analyzed accordingly. Theoretically, incorporating other perspectives such as economic sociology and sociological institutionalism can shed light on how corruption is sustained or even institutionalized. Furthermore, these match-fixing operations occur in the Taiwanese society that is culturally rooted in the Chinese tradition. Future research should therefore incorporate emic Chinese constructs, such as the code of brotherhood among actors.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article: The authors acknowledge the financial support from the Ministry of Science and Technology, Taiwan, Funding number: MOST 108-2410-H-142-013-MY2.
