Raffaele Poli

University of Lausanne and University of Neuchâtel, Switzerland

raffaele.poli@unil.ch

Understanding globalization through football: The new international division of labour, migratory channels and transnational trade circuits

Over the past 20 years, the problematic of globalization has become a recurring theme in the social sciences. Today it has largely outgrown the economic sphere where it first came into being [4]. The multiplicity of possible approaches to globalization obliges the researcher to have a prior clarification of the concept, in order to avoid the risk of making it 'a buzzword for journalists and the latest generation of vacuous futurolo-gists' [32].

Regarding sport, Bale and Maguire [3] have elaborated an analytical distinction between five different 'scapes' [2] of globalization, each of which comprises the movement of a particular entity: materials for 'technoscapes', money for 'financescapes', images and information for 'mediascapes', ideas for 'ideoscapes' andpeople for 'ethnoscapes'. The scope of this article will focus on the exclusive development of the thematic of the international mobility of sporting talents, especially that of footballers. This case study is particularly interesting for two reasons: the existence of professional championships in a large number of countries and the importance of the international flow of players that football generates.

The article is structured in three parts. First, it presents three perspectives on globalization, clarifying them using references from researchers having worked in the domain of football player migrations. In the second part, it examines the geography of the international flows of footballers and it analyses the quantitative evolution of the presence of expatriate players in the five principal European leagues (English Premier League, Spanish Primera Liga, Italian Serie A, French Ligue 1, German Erste Bundesliga) [8]. The aim is to verify if the general increase in international flows reflects a spatial diversification of migratory routes, or if privileged recruitment channels between the departure and arrival areas still exist. In the third part, we analyse the empirical functioning of transfer networks. The example of three ideal-typical upward career paths of African players having attained the English Premier League is used to illustrate the functional integration of leagues beyond national borders.

From a theoretical point of view, the article is based on relationnism. Contrary to substantialism, which 'takes its point of departure the notion that it is substances of various kinds (things, beings, essences) that constitute the fundamental units of all inquiry', the paradigm of relationnism promotes a view of the society which highlights 'the primacy of contextuality and process in sociological analysis' [13]. From this perspective, the purpose of the article is to show that the general tendency of increase in the international flow of athletes does not occur by itself, as a natural feature of the contemporary world, but concretely depends on the actions of a plurality of actors who, by the relations they build on a daily basis, are responsible for the interconnection between specific zones of departure and arrival. Generally speaking, globalization is not seen as an outcome that actors cannot influence, but as a structural process directly linked to human agency.

Perspectives on globalization and relational approach of footballers' mobility

Depending on the manner in which globalization is envisaged, Held et al. [19] have classed researchers into three categories: the 'sceptics', the 'hyperglobalists' and the 'transformationnists'. The sceptics hold that the concept in itself is not justified insofar as it refers to a relatively old process of internationalization linked to the worldwide diffusion of capitalism. According to their point of view, 'recent growth of international flows represents rising in interactions between well-defined national economies, rather than the emergence of global economic activity' [32]. For Brenner, globalization proceeds by rounds and the current phase is only 'the most recent historical expression of a longue durée dynamic of continual deterritorialization and reterritorial-ization that has underpinned the production of capitalist spatiality since the first industrial revolution of the early nineteenth century' [4]. In the case of the migration of footballers the perspective of Lanfranchi and Taylor [20] can be put into this category. According to Taylor, 'football migration is nothing new, but has a long and complicated history' [46]. For the same author, 'much of the writing on football migration has tended to employ ''globalization'' uncritically, as if it were an established fact rather than a contested concept'. In the sceptics' vision, historical continuity is preferable to a view that emphasizes rupture, such as that held, contrarily, by the hyperglobalist perspective.

The hyperglobalists consider that the concept of globalization is not only justified, but is the only proper conceptual framework with which to analyse the contemporary world. From their perspective 'globalization has created a single economy transcending and integrating the world's major economic regions' [32]. From now on, we live in a world where states have lost a large part of their power. As they are no longer able to limit the international circulation of goods, services and merchandises, states have to be content with encouraging them by the setting up of regulatory frameworks that are less restrictive than in the past [45]. With the development of the NTIC and the process of the annihilation of space by time [18], the erosion of state power is the key aspect on which hyperglobalist researchers insist.

Applied to football, this perspective has been amply used to account for the juridical changes that have come about from 1995 onwards, when the Bosman ruling by the Court of Justice of the European Community liberalized circulation within the EU for players holding a communitarian passport [11]. After having focused their attention on the general increase in international flows of athletes - a hitherto little-studied topic suddenly became worthy of scientific attention - several researchers distanced themselves from the hyperglobalist vision, by notably highlighting the geographical selectivity of the flows [26, 29, 38].

The third perspective on globalization is defined as transformationnist. Contrary to the sceptics, the transformationnists consider that globalization is a new process comprising a spatial interdependence on a transnational scale that has not been seen before. A distinction is made between internationalization and globalization. According to Peter Dicken, while internationalizing processes involve the 'simple geographical spread of economic activity across national borders with low levels of functional integration', the globalizing ones involve 'both extensive geographical spread and also a high degree of functional integration' [9]. In the case of internationalization, the change is more quantitative in nature, while in case of globalization it is more qualitative. The key feature of globalization is the development of economic circuits functionally integrated beyond national borders, to which authors usually refer by the terms of 'global commodity chains' [15] and 'global production networks' [9].

The desire to understand the manner in which a functional integration process intervenes within the framework of the football players' labour market necessitates taking into account the transfer networks of players as an analytical unit, rather than the players individually or the macro-economic structures in which they are integrated. As Dicken et al. have duly remarked, we think that in order to understand the functioning of the global economy, it is necessary 'to transcend ''atomistic description'' of activities of individual actors or meta-individual imaginations of ''deep'' structures' [10]. In the manner of Granovetter, we believe that 'the analysis of processes in interpersonal networks provides the most fruitful micro-macro bridge. In one way or another, it is through these networks that small-scale interaction becomes translated into large-scale patterns, and that these, in turn, feed back into small groups' [16]. Within the paradigm of relationnism, the analysis of interactions between actors in networks allow us to understand how the latter 'globalize' the world, from the point of view of strategies deployed to create or take advantage of opportunities, as well as the constraints that limit or influence their actions.

In taking into account criticisms made by numerous authors regarding the a-spatial vision of globalization [27, 28, 44, 47], according to which all frontiers disappear, we have built our analysis with the help of approaches developed in the context of studies of migration which are based on the notions of networks [30] and channels [14]. Starting from the figurational view of the society promoted by Norbert Elias [12], networks are considered as sets of functionally interdependent individuals. In interlinking with each other, networks form the base of society. They are seen as dynamic social configurations constructed by actors, that both allow and set the boundaries of their actions. While not all individuals have the same room for manoeuvre within networks, and do not have the same possibilities to elaborate new ones, an initial starting position can progress positively or negatively according to the astuteness of the strategic choices made.

In the case of the footballers' transfer market, networks are made up of a plurality actors playing distinct and complementary roles. From a relational perspective, each flow is a concrete, empirical and synthetic output of networks involving, among others, club officials, managers, agents, talent scouts, investors and, last but not least, players themselves and quite often also their relatives. These actors collaborate to make transfers possible and compete to appropriate the financial added value generated by the latter. As a consequence of this reasoning, we consider that no flows occur without the participation of multiple stakeholders who are directly or indirectly linked each other, and whose decision-making power is greater or lesser according to circumstances and opportunities.

While the importance of networks and intermediaries in the mobility of athletes is widely recognized [36, 42], the emphasis in literature on the migrations of athletes has been placed more often on their individual motives. From this perspective, Maguire [23] has identified different types of athletes, grouped according to their migration experience: 'pioneer' type migrants, 'mercenaries', 'nomadic cosmopolitans', 'settlers' and 'returnees'. In a similar fashion, Magee and Sugden [22] have added to the previous typology the categories of 'ambitionist', 'exile' and 'expelled'. As Maguire and Elliott highlight, 'when examining research located in the area of athletic labor migration, it is evident that the majority of the work has focused on the motivations and experiences of migrant athletes' [24]. Nevertheless, the fact remains that the focalization on individual motives and experiences of athletes runs the risk of atomizing the research and tends to lose sight of structural constraints imposed on any player wishing to migrate.

Other studies on the mobility of athletes have taken the structural inequalities between countries as an analytical entity. Principally carried out to account for migrations of African players in Europe, these analyses emphasize the political mechanisms of economic and cultural domination [6]. From a neo-Marxist perspective, Darby et al. hold, for example, that the recruitment of footballers carried out by European clubs in Africa 'may clearly be interpreted as an extension of broader neo-imperialist exploitation of the developing world by the developed world' [7].

The focalization on the political stakes in migrations from a macro-economic perspective entails the risk of forgetting the importance of the actions of individuals who, by their competencies and strategic choices, can influence positively the structures in which they are embedded and of which they are not merely passive actors. While it allows us to understand economic logics and the underlying power games at stake in the migrations of sportsmen, the macro-structural perspective does not furnish the analytical and conceptual tools to understand more precisely the socio-spatial logics at work.

Contrary to what is usually undertaken by researchers following world-system theory or the neo-classic approaches based on the individual rational choice theory, network theories on migration insist on the fact that migratory channels must be carefully traced and described 'instead of being left to external and elusive macro-determinations'. In a relational framework, migration appears indeed 'as being less the residual factor of a confrontation between supply and demand on the international labour market than the expression of global dynamics generated by human interactions' [30].

Having recourse to a relational approach in the study of globalization gives us the possibility to concretely understand the manner in which actors take into account the constraints and opportunities linked to economic, cultural and power differentials existing between territories. In the case of the trade of footballers, such as in other domains of economic life, the relational framework aims at understanding how human actions are reflected spatially and organize in the final instance the dynamics of flows.

Volume and spatiality of international flows

The concept of globalization is generally associated with the process of the new international division of labour. Contrary to the pre-global international division of labour, within which the industrialized countries (the centre) sought from non-industrialized ones (the periphery) raw materials and agricultural products, the new international division of labour manifests itself by a relocalization of part of the industrial production in certain countries of the former periphery. This process can be defined as 'an emergent form of worldwide division of labour associated with the internationalization of production and the spread of industrialization' [31]. Controlled from a few global cities [43], the major transnational companies no longer look only for markets or raw materials from Southern countries, but also for a labour force that is considered to be 'cheap to buy, abundant, and well disciplined' [48].

The analytical framework of the new international division of labour is also applicable in the context of professional football. While, in the industrial sphere, the relocalization of production is reflected in a strong increase of products made in countries having progressively acquired an advantage regarding human capital and factors of production (notably in South-East Asia), in professional football such a process is born out of an increase of the number of players imported from South America and Africa.

Figures 1 represent the number of expatriate players taking part in the 'big-5' European leagues during the seasons 1995/96 and 2005/06, according to their country of origin. The statistical data used for the purpose of this article have been gathered together by the Professional Football Players Observatory (PFPO), a French-Swiss research group co-founded by the author. The origin of players is defined here independently from the nationalities they may have, as the country where they grew up and from which they departed following recruitment by a club overseas. Only flows directly linked to football are thus taken into account. The choice of seasons for the comparison was made with respect to legal changes regarding international flows of players (Table 1). The 1995/96 season is indeed the last prior to the application of the aforementioned Bosman law.

Table 1. Evolution of the composition of the labour market in the 'big-5' European leagues according to players' origin

a

b

Figure 1. Number of expatriate players in the 'big-5' European leagues according to their country of origin: a) 1995/96 season; b) 2005/06 season

Henceforth, in the five major European leagues, almost four footballers out of ten have come to the country where they work following a professional international migration. The increase in the number and the percentage of expatriate players has continued over the last four seasons. In 2008/09, there were 1107 expatriates in the 'big-5' leagues, representing 42.6 percent of the total number of footballers [36].

From 1995 to 2005, in absolute terms, all geographical origins have participated in the increase of expatriate footballers. However, the relative values indicate today a higher proportion of Latin American and African players among expatriates, to the detriment of footballers from Western Europe, and, even more so, from Eastern Europe (Table 2).

Table 2. Evolution of the proportion of expatriates according to zone of origin

From 1995 to 2005, clubs of the best European leagues have recruited players from Africa and Latin America even more than in the other countries of the continent. Consequently, the proportion of non-European expatriate players has increased in two ways, in comparison with the global pool of players, as well as relative to expatriate footballers. In absolute terms, while the number of expatriates from Western and Eastern European countries has increased from 317 to 502 (+58.3%), the number of non-Europeans has augmented from 146 to 496 (+240%). While on the increase in all the leagues, the proportion of the latter still changes greatly according to country.

Regarding international recruitment, Spanish, French and Italian clubs are clearly more orientated towards other continents than English or German ones (Table 3). This result is a first indication of the spatial selectivity of the international flows of footballers. This phenomenon can be fully understood by having recourse to the 'migratory channels' analytical framework [14].

Table 3. Evolution of the proportion of non-European players among expatriates according to country of employment

In the study of the migration of highly qualified personnel, different researchers have shown the importance of networks to enhance and organize international flows. Assuming that 'most international migrants depend on intermediaries to help them to achieve employment and housing in another country', Findlay and Li consider that the 'intermediary agencies, by channelling information and resources, have an influence in moulding the process of international migration'. This forms the basis for the 'migratory channels' framework, so that 'channels not only reflect the existence of a migration system, but are also responsible for the structuring of the system' [14].

According to Meyer, the emergence of persons and enterprises who play an intermediary role in a professional capacity, 'confirms the fact that globalization of the highly skilled labour market does not occur without massive network investment' [30]. The same logic can be applied to the migration of footballers, even though the latter, in particular Africans, due to their early age of migration¹ and a departure abroad most often without a signed work contract, can rather be considered as 'migrants to be qualified' [35]. The intersection of the geographical origin of expatriates and the countries in which they exercise their profession confirms the existence of privileged relations between territories (Table 4).

Table 4. Spatial distribution of expatriate players by zone of origin according to league (2005/06 season)

The strongest concentration pertains to African footballers: 57.2 percent of footballers of this origin play for French clubs. Western European expatriates are highly concentrated in England, while Eastern European ones are over-represented in Germany. Finally, Latin Americans are very present in Spain as well as in Italy, and are the least concentrated players in a single country. They are, however, clearly over-represented in these two Latin countries.

The inequality of the distribution of expatriate footballers according to their origin is confirmed by significant Chi-square tests for both the 1995/96 and 2005/06 seasons. The intensity of the relation has only diminished slightly in 10 years: Cramer's V value has gone from 0.44 to 0.4. The general increase in the international flow of players during the first 10 years after the Bosman law has thus not been reflected in a significant spatial diversification of transfer networks. On the contrary, it has led to a quantitative reinforcement of older channels. The operation of the latter continues to depend on criteria such as geographical proximity (Germany-Eastern Europe, England-Scandinavia, Ireland and other UK nations) or historical links (Spain-Latin America, Italy-Latin America, France-old African colonies, England-USA and Australia). The evolution observed is not, however, the same for all the zones of origin studied (Table 5).

Table 5. Contribution to the Chi according to zone of the origin

The evolution of the contributions to the formation of the relation expressed by the Chi-square value according to the zone of origin of expatriate footballers shows that, in comparison with Africans and Latin Americans, Europeans tend towards a more homogenous distribution between the leagues taken into account. From the point of view of geography, the introduction of free movement for EU players has above all resulted in a more equal spatial redistribution in the different championships. Indeed, during the 1995/96 season, European expatriate footballers were even more concentrated in Germany and in England than they were 10 seasons later. On the other hand, for non-Europeans, the growth took place in a selective manner, from leagues where players from these origins were already over-represented before the Bosman law was decreed. This process has led to higher levels of concentration of Africans in France, such as of Latin Americans in Italy and in Spain.

The different geographies of international recruitment carried out by European clubs in relation to their home country clearly indicate that economic factors (differentials in riches) and legal ones (quota systems) cannot by themselves explain the totality of dynamics at work in the international trade of footballers. Like Maguire and Pearton, we believe that 'although economics play a crucial part in determining the patterns of football migration, they are by no means the only factor involved. Rather, a set of interdependences contour and shape the global sports migration' [25]. To fully understand what is at stake, it is indispensable to analyse in detail the manner in which places are put into relation by different types of actors playing the role of intermediaries in the midst of transfer networks.

Transfer networks and functional integration of spaces

Up until now, we have studied the presence of expatriate players in Europe from the perspective of stocks. In a context of high professional mobility, it is also necessary to analyse the career path of players from a longitudinal point of view. It thus becomes possible to grasp the complexity of the underlying spatial stakes of the trade and migrations of footballers, especially when more than two countries are concerned.

If we speak jointly of trade and migrations of football players, it is because the latter, in the economic context particular to this activity, are not just simply workers under contract with a club. They also have to take on the status of commodities. Indeed, by their transfer, different types of actors, from club managers to players' agents, aim to set up value added chains. In order to achieve this, they seek to gain profit from the economic differentials existing between clubs and leagues by 'buying' or 'placing' a player in a club and then transferring him for money to a team with more means at its disposal.

In order to clarify this idea, we present three examples of upward career paths that have characterized African footballers playing or having played for Premier League clubs in England. The choice of examining only upward career paths was made so as to be fully aware of the speculative logic (creation of value added chains) that structures the transfer market for players. The decision to study the career paths of African players in England is valid for two reasons. The first is the high mobility of Africans: they change club every 2.4 seasons, whereas the average for players as a whole is 2.9 [36]. The second is that the English championship, with a turnover of 2.4 billion euro in 2007/2008, is today the richest football competition in the world [8]. English clubs are capable of attracting a large number of expatriate stars. During the 2008/09 season, 59.3 percent of players in the Premier League were expatriates.

The first example of an upward career path is that of Emmanuel Eboué. This player was born on 4 June 1983 in Yopougon, one of the 10 communes that make up the Abidjan agglomeration. At a very young age he was part of a local organization called Cooper Ecole de Football, named after its founder, Innocent Anzan, nicknamed 'Cooper' [33]. He was discovered during a friendly match by the Académie MimoSifcom of Abidjan, a training centre founded in 1994 by the French trainer Jean-Marc Guillou and the club ASEC Mimosas Abidjan. He became a full-time member of the organization on 10 October 2001. Less than one year later, he was transferred to KSK Beveren, in Belgium, a club of which Jean-Marc Guillou had become a shareholder. After two and a half seasons in Flanders and marriage to a Belgian woman, he was recruited by Arsenal in January 2005. Thanks to the very good relations between Arsène Wenger, the Arsenal manager, and Jean-Marc Guillou, Eboué had already been on a training camp for three weeks with Arsenal during the summer of 2004. Emmanuel Eboué's career path is particularly interesting because it reflects the existence of a migratory channel set up by club officials. This channel linked the Ivory Coast to England, with a stage via an intermediate country, Belgium, where the conditions to obtain a work permit for non-EU players are less restrictive than in England.

The second example of an ideal type of upward trajectory is that of Michael Essien. This midfielder was born in the neighbourhood of Awutu Senya, in Accra, on 8 December 1982. Thanks to a scholarship awarded for his football talents, he spent three years at Saint Augustine's Senior Secondary School (Cape Coast). On his return to Accra, Micheal Essien became part of the Liberty Professionals FC team. He was then picked for youth teams for Ghana. In 1999, he took part in and won the under 20s African Cup of Nations. The same year he played in the under 20s World Cup, where Ghana took third place. Essien was spotted by recruiters from Manchester United, who invited him to England for a training camp. However, the player was already under contract to a French players' agent, the former goalkeeper Fabien Piveteau, an associate of Sly Tetteh, the president and co-founder of Liberty Professionals FC. Manchester United suggested to Micheal Essien that he play for the Belgian partner club of Royal Antwerp in order to have the time to acclimatize himself to European football and to fulfil conditions necessary to obtain a work permit for across the Channel². The player refused and made his way to his agent's house in Monaco.

Essien took part in a first trial at Lille and a second at Bastia, a club for which Fabien Piveteau has played from 1996 to 1998 during the last stage of his professional career. In August 2000, Michael Essien signed his first professional contract for the Corsican club. Three years later, after some excellent performances, he was transferred to Olympique Lyonnais in exchange for 11.75 million euro³. After two years at Lyon, Essien requested and obtained a transfer to Chelsea in London, who paid 38 million euro for him. This figure is the highest so far that has been paid for the recruitment of an African player. Michael Essien's trajectory is particularly revelatory of the role of players' agents in the management of the flow of footballers. Indeed, in most cases, more than the players or the managers of clubs, it is the latter, thanks to their connections, who contribute to the setting up of migratory channels that the footballers end up by following (Poli, 2004a).

A third ideal-type upward career path is that of Henri Camara, a forward born in Dakar on 10 May 1977 to parents of Guinean origin. At an early age, he became part of one of the two flagship clubs of the Senegalese capital: ASC Jaraaf Dakar. In 1999, he was recruited by Racing Club of Strasbourg via the intermediary of the trainer Claude Leroy and the Swiss agent Nicolas Geiger. After having spent many years as a coach in Africa, including Senegal, Claude Leroy was at the time the sporting director of the Alsace club. Nicolas Geiger was himself active in Africa and had tipsters in Senegal and in Cameroon. Camara's transfer was settled in Dakar on 7 July 1999 and the player arrived in France on 20 July of the same year. The managers of RC Strasbourg paid a commission fee of 1.7 million euro to the Team Consult company based in Vevey, in Switzerland, and belonging to Nicolas Geiger⁴.

On 6 September 1999, Camara was loaned to Neuchatel Xamax club, then trained by Alain Geiger, brother of the agent. After a year and a half in Neuchatel, Camara was transferred to Grasshoppers of Zurich, where he stayed for six months winning the Swiss championship. In the summer of 2001, free from all commitments following legal proceedings⁵, Henri Camara signed for CS Sedan through the intermediary of his new and powerful agent, the Franco-Senegalese Pape Diouf, a former journalist who later became the president of Olympique de Marseille. After two seasons, Camara was transferred to Wolverhampton Wanderers for a sum of around 3 million euro, via the intermediary of the associate of Pape Diouf for the English market, the Scottish resident of Monaco, William McKay. In 2004, the player refused to stay at Wolverhampton Wanderers, which, in the meantime, had been relegated. He was subsequently loaned for six months to Celtic in Glasgow, and for an additional six months to Southampton. The following year Wolverhampton Wanderers accepted a bid to sell the player to Wigan Athletic for the sum of 4.6 million euro. Henri Camara moved then to West Ham United, went back to Wigan Athletic and was loaned out again to Stoke City. He is now under contract with Sheffield United. Aged 33, his financial worth is henceforth close to zero.

Generally speaking, the three ideal-typical examples presented show to what extent the migration of African footballers to Europe is carried out through privileged relations between different types of actors (club officials and managers, agents, players, private investors) who interact within the framework of transfer networks. The setting up and development of the latter is part of a logic of creation of value added chains in which players are supposed to acquire worth through movement. Spatially, these circuits very often involve more than two countries. They are indeed usually constructed so as to take advantage of the economic differentials existing between leagues. Within the structure of the international trade of footballers, the different spaces and clubs through which the players transit take on complementary statuses. This allows us to propose a typology of spaces.

The 'platform' space defines the first country to which the player comes from his federation of origin (for example, France for Essien or Belgium for Eboue). The 'stepping stone' space is a country from which a players gains access to another country where the sporting and economic levels of the championship are higher. For example, Switzerland and France have played this role for Henri Camara. The 'transit' space is defined as the country where the player passes through and leaves and where the level of competition to which he is used to remains unchanged (no example for the trajectories cited). The 'relay' space is defined as the country to which the player was loaned before returning to the country from where he came (for example, Scotland for Camara). The 'destination' space is that hosting the wealthiest leagues and clubs of the world (England in this case).

The constant coming into contact of these different types of spaces within the framework of players trade provokes a functional integration of football leagues on a transnational level and, according to the transformationnist point of view, justifies the recourse to the concept of globalization.

Conclusion

It is time to invert the research perspective showing how the analysis of commerce and international migration of footballers serves to better understand the process of globalization, notably by underlining the central role that human intermediation plays in the economic construction of competitive advantages [40] and in the dynamics of spatial inequalities. Throughout this article we have shown that the analytical and conceptual tools developed within the framework of the study of globalization can also be applied to the case of trade and migrations of footballers. It is also true that the analysis of this specific case allows us to better understand the mechanisms at work in the wider context of economic globalization.

The example of football permits us in the first place to show the critical role played by intermediaries in the formation and development of migratory channels. This argument can also be applied to the functioning of the global economy as a whole. While the study of the social relations of production remains important, the example of football demonstrates that it is just as important to examine the 'social relations of circulation' [37] of both men and commodities. The necessity to adopt a biographical approach to fully understand 'the social life of things' has been also underlined by Appadurai, who highlights the necessity in 'breaking significantly with the production-dominated Marxian view of the commodity and focusing on its total trajectory from production, trough exchange/distribution, to consumption' [1]. This realization is more pertinent than ever at an epoch where the growing fragmentation of production provokes a very strong increase in the flow of goods and makes itself felt at the level of the increasing amount of power held by intermediaries [5]. The latter play a crucial role in the management of these flows and largely determine their spatiality. The footballers' transfer market is a perfect case to shed light on this process.

The example of football also shows that the creation of economic opportunities is intrinsically linked to the characteristics of the actors involved: their biographies, their linguistic skills, their trust relationships, etc. Put together, these individual characteristics largely determine the formation of migratory channels, which in the football industry take the form of transfer networks. As a consequence, the advantages regarding the recruitment of players abroad cannot be seen as uniquely stemming from factors of a 'purely' financial nature, such as, for example, the differences in the means at the disposal of clubs according to the territories concerned by the flows of players. If financial gaps matter, these advantages are also socially constructed according to the profiles of actors who are at the root of the setting up and development of transfer networks.

This realization can also be applied to the wider context of economic action and of globalization from an economic perspective. As underlined by Mark Granovetter with the notion of 'social embeddedness' [17], within the framework of production and commerce of goods, the social and political stakes that guide the actors' actions are very often as important, or more so, as the stakes linked to a 'pure' economic rationality, supposedly objective and equal for all. The example of the international trade of footballers clearly shows that the functional integration of spaces on a transnational level is brought about firstly in a relational manner, according to the social capital [41] that the actors involved in transfer networks have at their disposal or are able to mobilize.

In bringing to light the importance of human intermediation, the study of footballers' transfer market also allows us to highlight the fact that globalization cannot be considered as a 'simple' structural outcome, which is external to actors in networks and networks of actors. Though actors seek to take advantage of existing economic differentials, by their daily activities, they also contribute to create them. Their strategic choices have indeed an impact in the constant production or reproduction of competitive advantages. This always occurs in a selective manner, through the mobilization of relational resources whose spatial projection never covers the world in a homogenous fashion.

Generally speaking, the study of the case of football players' transfer market reveals the necessity, as underlined by David Ley, to 'bring the issue of human agency to a globalization discourse that has frequently been satisfied with speaking of a space of networks and flows devoid of knowledgeable human agents' [21]. The example of the migrations and commerce of footballers fully demonstrates the imperative of replacing the agency of human actors and the social embeddedness of their actions at the heart of the analysis of this crucial process of the contemporary world. Notes

References

1.     Appadurai A. (ed.) The social life of things: Commodities in cultural perspective. – Cambridge: Cambridge University Press, 1986.

2.     Appadurai A. Global ethnoscapes: Notes and queries for a transnational anthropology / In: Fox R.G. (ed.) Interventions: Anthropologies of the present. – Santa Fe, NM: School of American Research, 1991. – P. 191–210.

3.     Bale J. and Maguire J. (eds) The global sports arena: Athletic talent migration in an interdependent world. – London: Frank Cass, 1994.

4.     Brenner N. Beyond state-centrism? Space, territoriality, and geographical scale in globalization studies // Theory and Society 1999. – Vol.28, N 1. – P. 39–78.

5.     Burt R. Structural holes: The social structure of competition. – Cambridge, MA: Harvard University Press, 1992.

6.     Darby P. The new scramble for Africa: African football labour migration to Europe. In: Mangan JA (ed.) Europe, Sport, World: Shaping Global Societies. – London: Frank Cass, 2001. – P. 217–244.

7.     Darby P., Akindes G., Kirwin M. Football academies and the migration of African football labour to Europe // Journal of Sport and Social Issues. – 2007. – Vol. 31, N 2. – P. 143–161.

8.     Deloitte Annual Review of Football Finance. – Manchester: Deloitte, 2009.

9.     Dicken P. Global Shift. Reshaping the Global Economy Map in the 21st Century. – London: SAGE, 2007.

10.     Dicken P., Kelly P., Olds K., Yeung H.W.C. Chains and networks, territories and scales: Towards a relational framework for analysing the global economy // Global Networks. – 2001. – Vol.1, N2. – P. 89–112.

11.     Dubey J-P. La libre circulation des sportifs en Europe. – Bern and Bruxelles: Staempfli, Bruylant, 2000.

12.     Elias N. La societe des individus. – Paris: Fayard, 1991.

13.     Emirbayer M. Manifesto for a relational sociology // The American Journal of Sociology. – 1997. – Vol.103, N 2. – P. 281–317.

14.     Findlay A., Li F. A migration channels approach to study of professionals moving to and from Hong Kong // International Migration Review. – 1998. – Vol. 32, N3. – P. 682–703.

15.     Gereffi G., Korzeniewicz M. (eds) Commodity chains and global capitalism. – Westport, CT: Praeger, 1994.

16.     Granovetter M. The strength of weak ties // American Journal of Sociology. – 1973. – Vol. 78, N 6. – P. 1360–1380.

17.     Granovetter M. Economic action and social structure: The problem of embeddedness // American Journal of Sociology. – 1985. – Vol.91, N 3. – P. 481–510.

18.     Harvey D. The condition of postmodernity. – Oxford: Blackwell, 1989.

19.     Held D., McGrew A., Goldblatt D., Perraton J. Global transformations: Politics, economics and culture. – Cambridge: Polity Press, 1999.

20.     Lanfranchi P., Taylor M. Moving with the ball: The migration of professional footballers. – Oxford: Berg, 2001.

21.     Ley D. Transnational spaces and everyday lives // Transactions of the Institute of British Geographers. – 2004. – 29. – P. 151–164.

22.     Magee J., Sugden J. The world at their feet: Professional football and international labor migration // Journal of Sport & Social Issues. – 2002. – Vol. 26, N 4. – P. 421–437.

23.     Maguire J. Global sport: Identities, societies, civilizations. – Cambridge: Polity, 1999.

24.     Maguire J., Elliott R. Thinking outside the box: Exploring a conceptual synthesis for research in the area of athletic labor migration // Sociology of Sport Journal. – 2008. – 25. – P. 482–497.

25.     Maguire J., Pearton R. Global sport and the migration patterns of France '98 World Cup finals players: Some preliminary observations // Soccer & Society. – 2000. – Vol. 1, N 1. – P. 175–189.

26.     Maguire J., Stead D. Border crossings: Soccer labour migration and the European Union // International Review for the Sociology of Sport. – 1998. – Vol. 33, N1. – P. 59–73.

27.     Massey D. For Space. – London: SAGE, 2005.

28.     Mattelart A. Diversite culturelle et mondialisation. – Paris: La Decouverte, 2005.

29.     McGovern P. Globalisation or internationalization? Foreign footballers in the English League, 1946-95 // Sociology. – 2002. – Vol.36, N1. – P. 23–42.

30.     Meyer J-B. Network approach versus brain drain: Lessons from the diaspora // International Migration. – 2001. – Vol. 39, N 5. – P. 91–110.

31.     Murray W. Geographies of globalisation. – London: Routledge, 2006.

32.     Perraton J. The scope and implications of globalisation / In: Michie J. (ed.) The handbook of globalisation. – Cheltenham: Elgar, 2003 – P. 37–60.

33.     Poli R. Le football en Cote d'Ivoire: organisation spatiale et pratiques urbaines. – Neuchatel: CIES, 2002.

34.     Poli R. Les migrations internationales des footballeurs. Trajectoires de joueurs camerounais en Suisse. – Neuchatel: CIES, 2004a.

35.     Poli R. Des migrants a qualifier. Les footballeurs africains dans quatre pays europeens / In: Nedelcu M. (ed.) La mobilite internationale des competences. Situations recentes, approches nouvelles. – Paris: L'Harmattan, 2004b. – P. 143–164.

36.     Poli R. Agents and intermediaries. In: Chadwick S., Hamil S. (eds) Managing Football: An International Perspective. – Oxford: Elsevier, 2009. – P.201–216.

37.     Poli R. Le marche des footballeurs. Reseaux et circuits dans l'economie globale. – Bern: Peter Lang, 2010.

38.     Poli R., Ravenel L. Les frontieres de la libre circulation dans le football europeen. Vers une mondialisation des flux de joueurs? Espace Population Societe. – 2005. – 3. – P. 293–303.

39.     Poli R., Ravenel L., Besson R. Annual review of the European football players labour market. – Neuchatel: CIES, 2009.

40.     Porter M. Competitive advantage: Creating and sustaining superior performance. – New York: Free Press, 1998.

41.     Putnam R. Bowling alone: The collapse and revival of American community. – New York: Simon & Schuster, 2001.

42.     Roderick M. The work of professional football. A labour of love? – London: Routledge, 2006.

43.     Sassen S. The global city: New York, London, Tokyo. – Princeton, NJ: Princeton University Press, 1991.

44.     Savage M., Bagnall G., Longhurst B. Globalisation and belonging. – London: SAGE, 2005.

45.     Urry J. Sociologie des mobilites. Une nouvelle frontiere pour la sociologie? – Paris: Armand Colin, 2005.

46.     Taylor M. Global players? Football, migration and globalisation, c. 1930-2000 // Historical Social Research. – 2006. – Vol. 31, N1. – P. 7–30.

47.     Thrift N. A hyperactive world / In: Johnston R., Taylor P., Watts M. (eds) Geographies of global change: Remapping the world. – Malden, MA: Blackwell, 2002. – P.18–35.

48.     Wright R. Transnational corporations and global divisions of labor / In: Johnston R., Taylor P., Watts M. (eds) Geographies of global change: Remapping the world. – Malden, MA: Blackwell, 2002. – P. 68–78.