Artistic work intermediaries as industrial relations institutions: The case of musicians

In artistic and cultural work, where unions and formal industrial relations institutions have weak influence, market intermediaries often shape the terms of exchange for workers. Focusing on musicians and drawing mainly on examples from France, this article shows how these intermediaries shape employment relations, in particular the distribution of risks and surpluses. While intermediaries and musicians have many shared interests, there is also a structural antagonism between them, which could be exacerbated by digitalization.


Introduction
Intermediaries in cultural fields have been subject to a steady stream of studies, starting with Bourdieu's discussion of cultural intermediaries in Distinction (1984). He defines them as the professionals engaged in the production and circulation of symbolic goods and services. These intermediaries determine how cultural products are produced and consumed (Hesmondhalgh, 2006). Despite progress among students of cultural work in understanding community building (Cornfield, 2015), industrial citizenship (Legault and D'Amours, 2011) and collective action (Greer et al., 2018), cultural work is under-studied in the employment relations literature (Barré et al., 2018). Little is known about how the relations between cultural workers and their employers are structured by these intermediaries.
This article examines artistic work intermediaries as industrial relations actors, using the music industry as a case. Situated between producers of artistic products and their clients or employers, they intervene in market transactions by helping artists succeed in the market, influencing their output to make it more marketable, and in other ways influencing the diffusion and circulation of artists' works. Artistic work intermediaries may work for artists, such as managers or union hiring halls; for employers or customers, such as talent buyers (Foster et al., 2011) or comedy scouts (Friedman, 2014); and for both or either, such as agents and online platforms (Azzellini et al., 2019). Together with other cultural intermediaries 1 such as radio hosts, critics, publishers and record labels, artistic work intermediaries shape the wider cultural landscape and help to determine which artists succeed on the market and which ones are excluded from it. Their role in labourmanagement relations, usually overlooked, became headline news in April 2019, when Hollywood writers fired their agents en masse in a contract dispute (Koblin, 2019).
Some authors have argued that technological change makes cultural intermediaries superfluous by displacing them with direct transactions between producers and consumers over online platforms (Hirsch and Gruber, 2015). However, empirical studies have found a continued proliferation of intermediaries and a diversification of their functions (Azzellini et al., 2019;Caves, 2000;Jeanpierre and Roueff, 2014;Karpik, 2007;Lizé et al., 2011). Below we review secondary literature and documentary evidence on artistic work intermediaries in music to show how these intermediaries shape working lives. Our focus is primarily (but not exclusively) on France because of the rich literature on intermediaries in music and because the French labour code clearly spells them out as three métiers: agents, managers and tourneurs. 2 Inspired by John Commons's (1906) discussion of hiring halls created by American Federation of Musicians local unions, we argue that these intermediaries are centrally important industrial relations actors in the music industry. They shape the terms on which musicians engage with potential customers by determining the distribution of (1) risks faced in navigating the market (putting the musician or intermediary in a precarious position) and (2) surpluses generated through work (claimed by the musician or intermediary). We will show that many artistic work intermediaries occupy precarious situations, much like the artists they represent; that intermediaries depend for their livelihoods on the artist's success on the market; and that both artists and intermediaries are often victims of wider inequalities in music industries transformed by the 'star system' and digitalization. We also show that, among 'ordinary musicians' (Perrenoud, 2007), attempts to digitalize artistic intermediation can intensify the structured antagonism between intermediaries and artists themselves as intermediaries become more 'customer-facing' (Azzellini et al., 2019;Umney, 2017). Recent contributions have advocated more cooperative forms of intermediation adapted to digitalized labour markets (e.g. Scholz, 2016), but to date there appears little evidence of these innovations in arts and culture.
We begin by spelling out how cultural work intermediaries shape transactions in these markets, noting the prevalence of highly individualized forms of intermediation over their collective and public counterparts. Then, we look at the how intermediaries' role in managing the transaction translates into a regulatory function in musicians' work. Third, we examine the antagonism between intermediary and producer in music that determines the distribution of risk and surpluses and how it is altered by digitalization. Finally, we draw out implications for other kinds of work where intermediaries play a centrally important role, including the broader arts and culture sector, staffing agencies and the digitalized platform economy; as well as considering the possibility for more collective models of intermediation.

Intermediaries and the institutional regulation of work
Industrial relations literature typically examines collective bargaining and labour law as the main institutions that structure labour-management relations. This kind of institutional approach is difficult to use in sectors like music, where it is not straightforward to identify a single employer; unions are weak to nonexistent; many written rules are dead letter; and where workers depend on individual reserves of social capital to gain access to work. In markets outside the arts where similar conditions apply, intermediaries also play a central role in how formal rules and norms are enacted (Wagner, 2015) and how cultural capital is recognized and valued (Samaluk, 2016). What, then, are the relevant rules and procedures, formal and informal, that govern work in this kind of a market? In this section, we show how the forms taken by intermediation in music tend to be highly individualized, often small-scale, and, in many respects, highly precarious.
It is beyond the scope of this article to retrace the history of cultural professions and their hazy professional boundaries (see Lizé et al., 2011;Naudier, 2013;Rosselli, 1983;Weber, 2002). One driver of intermediaries' development in music has been a rapid increase in the number of working musicians: in France, from 11,820 in 1982France, from 11,820 in to 32,213 in 2007. Over this period, musical life has become increasingly internationalized, digitalized, and dominated by large multinational firms. These have led to growth in musical activity, a growing gap between the incomes of ordinary working musicians and elite 'stars', and an increasingly detailed division of labour in production, diffusion and promotion. In this context, intermediaries have become increasingly important.
Artistic work intermediaries vary widely in their form and function. Unions may establish intermediaries to shape the labour market in workers' collective interest. A classic example is the US union hiring hall, in which musicians organized transactions in the same physical space where they set wage minima and other standards -and organized boycotts of venues that used non-union musicians (Commons, 1906). Unions no longer control musical labour markets to this degree, but collective bargaining is common in established cultural institutions like orchestras or theatres and they do set guideline rates freelancers can follow voluntarily (Umney, 2016). Unions also campaign over issues of political relevance to musicians, or try to raise awareness of persistent problems such as low or unpaid work (Greer et al., 2018).
Alternatively, some countries have public intermediaries in music, as with the Pôle Emploi Spectacle (PES) in France and the Künstlervermittlung of Germany's Bundesagentur für Arbeit. These are specialized organs of national public employment services focused on arts workers. In France, artists usually join in order to assert their rights to unemployment services associated with the intermittence du spectacle regime. However, even here, this kind of intermediation has little significant role as a means of matching the supply and demand of cultural labour, because its interventions tend to provide only low-value jobs in which performers are easily interchangeable (Renard, 2005;Sigalo Santos, 2013).
In the French musical field, most intermediaries are more individualized and less concerned with collective standard setting. Agents and managers are representatives acting on behalf of artists, paid in the form of a commission on their income (generally between 10% and 20%), and whose job involves prospecting and negotiating with employers. The agent's main task is to seek jobs for artists and negotiate on their behalf, while a manager intervenes more directly in artists' career development. In this sense there appears to be relatively little structured antagonism in the relationship between artists and intermediaries, since both have a strong interest in securing more frequent and better-quality work for the artist. Agents dominate classical music, while managers dominate popular music; in jazz and certain traditional music genres, they play an equal role. Tourneurs, by contrast, organize tours and productions and usually employ artists directly. The informal norms and aspirations of musicians themselves are one important reason why public and collective intermediaries are marginal: many workers enter the field of music in part because they want to escape formal rules, standards and bureaucracy which they associate with other types of working life (Umney, 2016). Until 2010, the legal framework tended to confine agents to an artist placement role, while the tourneur was also concerned with finding concerts for musicians.
The functions of intermediaries change in part due to jurisdictional contests (Abbott, 1988) between different categories, but also due to qualitative shifts in their role. France, for example, has seen a 'managerialisation' of intermediary roles in music (Lizé et al., 2011) and cultural activity in general (Chiapello, 1998;Dubois, 1999). As a generational contrast between the impresario and the manager shows, over time the latter professional habitus has come to marginalize the former in the musical field (Lizé et al., 2011). Intermediaries have become more managerial as their representation role has increasingly taken the form of 'career development' and the entrepreneurial rationalization of placement strategies and artist promotion.
In France, this change in intermediary roles was enshrined in a 2010 law that removed the legal basis for the agent/manager distinction. It liberalized access to licences for artistic intermediary work and widened agent prerogatives to include a wider range of functions related to career advice and administration. It also liberalized rules around the accumulation of different activities: an artistic intermediary could now also be a venue manager, producer, or publisher, allowing in an extreme case all of artists' activity falling under the mandate of one enterprise.
Artistic intermediary work is becoming more porous also because of wider changes in the music industry. Due to digitalization and its consequences for the recorded music industry, the importance of live music at the heart of the music economy has grown (Allen, 2015;Guibert and Sagot-Duvauroux, 2013). Previously, tours would have been used to promote record sales; the decline in revenue from recordings has reversed this relationship. As record labels have disengaged from artists' careers, managers have become increasingly involved in the organization of concerts as part of their promotion role. Managers' roles have expanded further with the diversification of administrative and entrepreneurial activities in music, in which musicians are generally unskilled (Hracs, 2015;Thomson, 2013).
Much of this intermediary activity, however, exists on a small scale under highly precarious conditions. The trends described above, as well as tending towards an expansion of managerial activity, has also involved a transference of economic risk downwards, from record labels onto artists and their intermediaries. In the new music ecosystem, the manager, the agent and the tourneur face precarity and therefore pressure to act according to a risk-laden and entrepreneurial logic.
Here, music should be differentiated from other creative sectors, such as film and television, where the market for intermediaries is more concentrated (Bielby and Roussel, 2015). In France, five powerful agencies represent all of the most famous actors, and many smaller agencies represent less famous actors; the vast majority of actors have an agent (Naudier, 2015). However, many musicians do not have agents, and the intermediaries that do exist in most cases represent only one. This means that music intermediaries are generally smaller, weaker and more precarious than in film and television.
The database of Centre d'information et de ressources pour les musiques actuelles (IRMA) provides information on almost all popular music artists and groups (8996 in 2008), and their intermediaries (2510 in 2008) (Lizé et al., 2011). These populations of artists and intermediaries are composed of some well-integrated professionals, alongside a majority of semi-professionals, and some volunteers. The 2510 intermediaries include: Beyond the clear numerical superiority of managers, note that 527 (21%) intermediaries have numerous functions. This indicates the porosity of boundaries among these functions and the pluri-activity inside the music world. Other data also indicate there is a strong pluri-activity of intermediaries with jobs outside the music world (Lizé et al., 2014). Pluri-activity therefore appears to be a necessity to try to build a decent revenue, and to invest in the career development of the group or musician. It also protects against the risk of leaving the job: as with other labour markets for cultural intermediation, those in popular music experience strong turnover. This should be seen alongside the fact that the large majority of agents and managers represent only one artist. It is thus a diffuse and atomistic market, as well as a heavily stratified one. Most intermediaries are semiprofessional and coexist with voluntary intermediaries acting merely for one artist or group -possibly a parent, friend or spouse -and who generally have other professional activities sometimes unrelated to music. Then, there are more established professionals who act as gatekeepers to success and who are frequently approached by musicians looking for intermediaries (Lizé et al., 2014). The more an artist ascends the hierarchy, the more intermediaries surrounding them are differentiated and professionalized, and the more extended the network that participates in the promotion of the artist (Lizé et al., 2011). For international stars, there is usually further differentiation between a personal manager, a business manager and several regional managers -for example one each for Europe, Asia and North America (Lizé, 2016).
Therefore, although the overall demand for professionalized intermediaries is larger than supply, many artistic intermediaries occupy precarious situations. Smaller-scale intermediaries have limited capacity. A manager can, with difficulty, represent more than three musicians or groups at the same time, an agent can represent 10 if she/he only does placements and the work of a tourneur, more occasionally. To extract a net revenue from taxes, costs and investments of 30,000 euros per year, a manager with three artists taking a commission of 15% of revenue must manage musicians who earn on average almost 100,000 euros each per year. That is a small minority.
Although musicians have a strong incentive to work with an intermediary, most cannot access one. Among the 8996 artists and groups in French popular music identified in the 2008 IRMA census, 46% declared neither an agent, nor tourneur, nor manager. The high percentage indicates the weak integration of many musicians in the market (Perrenoud, 2007), but it also demonstrates the possibility of accessing some work without the aid of intermediaries. Many artists evidently practise do-it-yourself intermediation: the solo search for work opportunities, the most accessible of which are concerts. Others work with an 'artist developer' that carries out diverse cultural intermediation tasks in a particular region, often with public funding to work with little-known acts. 3 It is likely that the situation for artistic work intermediaries is being reshaped by new digital tools, much as the music industry in general has already been. Jeanpierre (2012) suggests that digitalization forces cultural intermediaries to re-evaluate the way they present themselves as evaluators of cultural value. Azzellini et al. (2019), in a study of Germany and the United Kingdom, highlight the growing role of digitalized intermediaries, which assemble large rosters of artists in an online repository through a relatively simple 'sign-up' procedure (often, artists will be asked simply to submit video or audio clips for approval). These artists can then be browsed and, often, sorted according to price or some form of 'quality' or popularity indicator (see also Umney, 2017). In such cases, there appears more scope for artists to access agencies. However, the role of the agency also shifts: it functions less as an individualized representative of a small number of bands, and more as a venue for buyers to compare a wider range of potential acts. Examples in the United Kingdom include Last Minute Musicians and Alive Network. As yet, however, Azzellini et al. (2019) find that there is a weak likelihood of these kinds of digitalized agencies displacing established professional agencies representing elite performers. Instead, they provide a seemingly more open way of accessing intermediaries for musicians outside of the elite, with potential consequences for the forms and distribution of risk, thus intensifying the structured antagonism between musician and intermediary. To understand these developments further, it is necessary to consider the way in which artistic work intermediaries intervene to shape transactions.

The management of transactions and the terms of market exchange
Artistic work intermediaries shape the workplace by virtue of the way they manage transactions between buyer and seller. Like digital platforms or staffing agencies, these intermediaries may have a profound impact on industrial relations. In this section, we identify three specific ways in which they do so. They facilitate the matching of supply and demand by accumulating and providing information for market participants; they manage relationships between musicians and clients, sometimes acting as an artist's representative in negotiations over the terms of work; and they influence the market value of musicians and their work.

Accumulating and providing information
Like many other labour intermediaries (Biglaiser, 1993), agents, managers and tourneurs accumulate information on job vacancies that may not be available directly to musicians themselves. In providing this information, intermediaries transform it to increase the attractiveness of the opportunity for buyers and sellers (Stigler, 1962). This tends to reduce the uncertainty that surrounds quality. In artistic markets, where the imperative of uniqueness, the weakness of institutional skills certification and changes in taste prevent any consensus over the evaluation of quality, overcoming this uncertainty is crucially important. Because engagements in live music are normally brief (often lasting just a few hours) and transactions frequent and one-off, intermediaries' information is continually in demand.
The coordination of buyers and sellers is fundamental to the translation of the artistic value of their clients into economic value. Without an explicit price list for each musician, knowledge of practices and prices on the market enables them to negotiate prices up to the benefit of their client. Intermediaries observe fluctuations in values, build up benchmarks of price growth, and identify better ways of gaining profits and optimizing the market value of musicians.

Managing relations between musicians and clients
Artists not only seek out intermediaries for their ability to find relevant information, but also for their individual social capital (Bourdieu, 1984). Personal relationships with other professionals in the field give the intermediary skills, experience and past collaborations valued by the artist. This social capital provides access to new opportunities and may also facilitate the task of negotiating contracts.
The address book is the intermediary's primary resource for prospecting for work from diverse employers, producers, financers and labels. This expands the artist's social network, thus reducing uncertainty surrounding the selection of work partners and access to employment. This in turn facilitates the realization of projects, the accumulation of symbolic capital and, thus, the careers of musicians (Dowd and Pinheiro, 2013;Faulkner, 1983;Lizé, 2016;Scott, 2012).
Taken together, networks of the artistic work intermediaries constitute the artistic labour market and its various channels, especially for well-established professional musicians. As William and Denise Bielby (1999) have shown with regard to 'talent' agents in the USA, these are markets that individual intermediaries shape by building and developing their network, as shown, for example, by the extension of the market through intermediaries who prospect at international level. Intermediaries' networks in music enable them to multiply and diversify artists' revenue sources by grasping diverse opportunities: legal (with intellectual property, for example 4 ), socio-economic (enlargement of the basis for calculating rights) and technological (internet, digital media 5 ).

Increasing the market value of artists and their work
Because they are paid on a commission basis, agents and managers have an economic interest in working with well-known artists: the greater the commercial and symbolic value of the artist, the greater their revenues. The same goes for tourneurs when they sell or produce concerts. This is why a large part of these intermediaries' activities and strategies aim, in Bourdieusian terms, to grow artists' symbolic capital and convert it into economic capital (Bourdieu, 1996). At stake is income and the intermediary's reputation, on which it depends to attract new partners and renowned artists.
Artistic work intermediaries have an influence on the value of an artist's work, first, as gatekeepers to the market. With the growth in the number of musicians, artists compete for intermediation on a seller's market. In France, where the number of working musicians has grown rapidly since the early 1980s, over half of all musicians have no intermediary, as noted above. Agents and managers are in a position to be highly selective in admitting artists. Representation by an intermediary not only helps a musician to access jobs, it is also a mark of professionalism that is in itself a form of recognition.
Music work intermediaries also influence the qualities of artists. We see this in the advice function that they exercise around artists in order that they adjust their creative product and their professional expectations to fit possible work. This action on the professional disposition (Bourdieu, 1990) of musicians is all the more visible when managers develop the posture of 'coach' to influence the artist's way of life: private behaviour and its relationship to work are changed so that aesthetic entrepreneurs succeed on the market. Managers view artists as 'human capital', whose ensemble of personal qualities must be invested to improve 'employability' (Lizé et al., 2011). Artists have to 'be professional' at work and in their private life: this extends from personal conflicts in an orchestra or a recording studio to the interests, ways of talking, dressing and conducting a home and family life. Artistic work intermediaries drive musicians to comply with the entrepreneurial regime of artistic activity Agents, tourneurs and managers develop artists' careers in diverse ways, which vary by type of artist and musical genre. As noted above, it often happens through intermediaries' interventions in artists' professional habitus and in the creation. It requires bringing out the uniqueness of artists and their works, knowhow and image, in order to produce symbolic capital. This means promoting the artist, in ways that vary depending on whether they are targeting the public or professional partners. If they act in collaboration with the latter (record labels, notably), managers, agents and tourneurs play a decisive role in promotion.
Finally, intermediaries produce value and develop artists' careers through network strategies which are at the root of social capital, a critical resource for artists (Down and Pinheiro, 2013;Faulkner, 1983) and intermediaries themselves (Bielby and Bielby, 1999;Lizé et al., 2011). As noted earlier, one of the main reasons artists seek out managers, agents and tourneurs is to increase their social capital: in other words, the number and quality of ties they have with professional partners in the music world. Thus, acting in various respects as gatekeepers, which can reinforce inequalities within cultural labour markets, intermediaries shape the hierarchies of perceived talent that determine the distribution of success in cultural scenes (Jeanpierre, 2012) or act as lynchpins who can control and limit others' participation in the scene (Woo, 2012). In this respect, intermediaries in cultural work have been identified as mechanisms through which existing inequalities, such as the privileging of whiteness and homophily, can be reinforced (Doane, 2009). Intermediaries thus shape the increasing inequalities within the arts labour force (O'Brien et al., 2016), particularly in music (Umney and Kretsos, 2015).

The unequal distribution of risks and surpluses
Artistic work intermediaries and musicians have a common interest in increasing the artist's symbolic capital and maximizing their income from engagements, and both suffer from the winner-takes-all star system (Krueger, 2019). Indeed, with the severe decline in record sales, major labels have shifted a part of the economic risks associated with uncertainty about the success of music productions to artists and intermediaries, for example, by outsourcing artistic management and reducing contract duration and sales advances. So far, then, the structured antagonism between musicians and intermediaries is not immediately apparent, since they have significant common objectives and face significant shared threats. However, intermediaries are often suspected of ripping off artists, and these suspicions are nourished by well-known cases of abuse and weak formal regulation. For example, although using written contracts is widely considered best practice, and in France is required by law, oral contracts are frequently used, even by well-known artists, and artists may suffer from intermediaries doing little to ensure prompt payment, or else extracting higher commission rates in untransparent ways (Allen, 2015;Umney, 2017). This suggests potential for more antagonistic relations between artists and intermediaries which are poorly understood. How do intermediaries cope with their own, often precarious, situation, and in what ways may the interests of intermediaries and the artists they represent diverge or conflict? In this section we show that, particularly in a context of increasing digitalization, there may be more scope for conflict to emerge.
Artistic work intermediation is itself a highly risky business, employment within intermediaries is highly insecure, and intermediaries have to compensate for the uncertainty of revenues. Parent (2012) finds severe economic precarity in a survey of French 'artist developers'. Among the 32 small intermediaries studied, only 20% of working time was remunerated according to fixed-term or permanent employment contracts, and that part-time and unpaid work were extremely common. In a separate survey of popular music intermediaries, Lizé (2014) reports that 38% of them declared that they had earned nothing from intermediation activities in 2009, and half reported income of between 0 and 5000 euros. Only in the higher regions of the popular music field, can the activity of the manager, agent or tourneur earn manageable, even comfortable, incomes.
For the music work intermediaries, one solution is to diversify income streams and functions. In France, they have been very active in prospecting and even inventing new streams of income for artists. For example, based on the 1985 law on intellectual property related authorial rights, on new media and technologies and on transformations in the law, artistic work intermediaries have derived new income streams from digital streaming, endorsements, crowdfunding, and business angels. In addition, pluri-activity (the assumption of additional music-related functions), well-known among artists (Bureau et al., 2009), applies equally to intermediaries. In France, after the July 2010 reform, an artistic agent, a manager or other intermediary representing a performing artist can also manage a venue, be a radio or television programmer or producer, artistic director within a record label, or music publisher. Henceforth, all these activities can be entrusted by the artist to one person, which can allow the intermediary to present to the artist a diverse range of opportunities for their works. In any case, as with the 360 degree contracts proposed by major labels (Marshall, 2012;Stahl and Meier, 2012), the concentration of different intermediation activities within one person is not without risk for the artist's autonomy, however. Indeed, it may be easier for an artist to find room for manoeuvre when working with several partners than by depending on a single partner for all of his/her activity.
More generally, the majority of agents, managers, tourneurs and artist developers practise at least one other remunerated activity in music (Lizé, 2014;Parent, 2012). Polyactivity (the addition of functions not related to music) is also very common: although it only concerns a third of tourneurs, it also touches on half of agents and managers investigated. The vast majority of intermediaries have diversified in one or both of these ways: mono-activity accounts for 27% of tourneurs, a third of agents and only 10% of managers (Lizé, 2014).
Aside from diversification, agents and managers seek surpluses and shift risks in a very different way from tourneurs. In France their relationship with the artist is defined in law as a mandat to look for work, represent the artist in the negotiation and execution of contracts with other partners, and to advise the artist in career development. The artist generally accepts an exclusive relationship with one manager. Most agents and managers receive a commission with an upper limit of 10% of the artist's revenues or 15% when organizational and career development tasks are conferred to them, as with the majority of managers, plus costs. This is fixed in article D.7121-8 of the Code du Travail and is similar to levels in the US (Allen, 2015;Thomson, 2012). On both sides of the Atlantic a booking agent will have a lower fee because it limits itself to the traditional role of placing artists in engagements; managers and agents who take on a broader marketing and career development role have a higher fee. Certain managers fix commission at 20% or even 25% of all the artist's revenues, if their role extends further to handling authorial rights. In interviews agents and managers justify this increase in the commission in terms of the resources that they make available to the artist, including social capital, financial investment, economic management and production structures (Lizé, 2016).
Tourneurs operate under a different legal framework, resulting in different licences and contracts. They combine functions of concert promoters (who organize shows) and agents, but for the duration of the show the tourneur usually is the employer of all the personnel, artistic or not, involved in the production of the shows (the 'plateau artistique'). Tourneurs find dates and sometimes sell concerts in ceding rights. Certain among them also concern themselves with advice and development of artists' careers, indeed functioning also as agent and/or manager. Hybrid contracts also exist between labour contract and mandat (Guibert and Sagot-Duvauroux, 2013) which cross, in their way, the porosity between these professions and their intermediation functions.
For tourneurs, the main sources of profit come from their methods of bringing a live show to market. In direct production the tourneur is paid through ticket sales but assumes the risk entailed in paying the plateau artistique, hiring a venue, and managing contracts for other services. This is most common where the performer is a star and the risk low. Where ticket sales numbers are more difficult to predict, the tourneur may remain employer of the artists but sell the show to a venue as a cession du spectacle. This transfer of risk usually takes place with subsidized venues whose mission is to present artists of sufficient fame to guarantee a full house. The tourneur's optimization of revenues thus depends either on the star status of the artists or a shift of risk onto the cultural institutions that support venues (Guibert and Sagot-Duvauroux, 2014).
The relationship between agents and managers and their artists, while ostensibly one of representative and represented, should be examined critically because of potential antagonisms embedded within it. In some cases, examples of antagonism between artists and agents may reflect an inherent 'art versus commerce' tension, where the intermediary seeks to commodify the artist's activity and thus exert control over their decisions (Morrow, 2013). Moreover, there can also be conflicts over division of revenue -for instance whether agent commissions should be calculated on net or gross profit -which have led to discussions among artists of 'horror stories' about agents, and calls for greater professional regulation (Morrow, 2013). Where agents have a comparatively large roster of acts, it may be difficult or intimidating for artists to challenge them over issues like delayed payments (Umney, 2016). Despite this scope for conflict, there is not a large literature on the tensions between artists and their agents.
Emerging research suggests that technologically-driven changes may alter this relationship further. Specifically, the distribution of surpluses and risks between artists and agents should be reconsidered in the light of new forms of digitalization which may sharpen the structured antagonism between musicians and intermediaries. As Azzellini et al. (2019) show, while more accessible digitalized agencies may potentially mitigate problems of access facing musicians, they also raise other problems in terms of the distribution of risk. In particular, by shifting focus away from a negotiation/representative role and towards acting as a venue for price comparison, they may normalize more intense price competition between musicians. In these authors' data, intermediaries with large online rosters of musicians were disinclined to negotiate higher fees for artists, despite retaining a commission-based model, because their priority had shifted towards quickly arranging a large volume of transactions, making 'customer service' more important. Moreover, Azzellini et al. (2019) also show that many digitalized agencies tend to be beyond the scope of collective bargaining arrangements with relevant unions that more 'traditional' agencies have in place to regulate the division of surplus (for instance by setting guidelines around commissions). From an industrial relations perspective, this shift reveals changes in the relationship between client, musician and intermediary.
Where the intermediary appears more as a customer-facing venue for comparison between potentially thousands of artists, poor fees and enhanced risk imposed on musicians may be less bad practice than a feature inherent to the model. In these cases, the agent begins to market itself more explicitly to the buyer, and because of the high volume model, the shared interest in raising individual artists' profile and income is greatly diluted or even removed.

Conclusion
Cultural workers have often been perceived as the 'canary in the coalmine' for labour markets more generally under conditions of neoliberalism (Barré et al., 2018;Perrenoud and Bataille, 2018). They pursue a highly individualized career path, in which value is often difficult to quantify and based on artists' largely intangible characteristics. Artists and intermediaries share some common interests: intermediaries may themselves be precarious and have an interest in bolstering the artist's profile. But there is also an underresearched structural antagonism between artist and intermediary that determines the distribution of risks and surpluses between market participants.
One contribution of this research is to highlight the importance of market intermediaries in the regulation of work in cultural occupations. Like Commons (1906), we see industrial relations in cultural work as strongly influenced by intermediaries: in particular, they influence pay by advancing the market position of those who do access them (and implicitly therefore also potentially disadvantaging the relative position of those that do not) and by acting as a gatekeeper to work. Unlike Commons's hiring halls, however, the main intermediaries are not collective (nor are they public): in part because of the nature of musical work and the motivations of those doing it these intermediaries are highly individualized. Another distinctive factor is the small-scale nature of much this activity, and the high rates of 'pluri-activity': many intermediaries are individuals who represent only one artist and rely on various other sources of income. We draw attention to the heightened precarity and 'DIY' nature of the intermediary landscape and hope to open up new lines of enquiry for industrial relations researchers interested in cultural sectors beyond music. Digitalization is not eliminating the need for intermediaries, but they may be forcing open cultural labour markets through more extensive but also more customer-facing intermediary models with a weaker interest in individual artists' development, and hence a potentially more conflictual relationship between these actors.
Cultural work intermediaries are so important in music because the employment relationship is not a common way of organizing transactions and the standard toolbox of institutional analysis is inadequate. In this sense, our insights regarding intermediaries as industrial relations actors are highly applicable to other labour markets in arts and cultural sectors -a growing topic of interest in industrial relations scholarship (Barré et al., 2018;Friedman, 2014;Umney and Symon, 2019). However, there are also important common themes between our research and that in other sectors where labour law and collective bargaining are similarly weak, and where cultural and social capital play an important role in shaping market conditions. For example, Samaluk (2016) finds that labour-supply agencies placing workers from Slovenia in jobs in England similarly muster social capital and contact networks to facilitate exchange, and there are similar concerns about the distribution of surpluses. Wagner (2015), furthermore, shows that the enactment of industrial relations rules in Germany is prevented by the transnational network of subcontractors that govern worker posting in construction.
In the emerging literature on the platform economy (Forde et al., 2017;Moazed and Johnson, 2016), we also find certain parallels. Here, the use of algorithmic techniques and automation of transactions appears quite different from the kinds of intermediation discussed in this article. However, our comments on the digitalization of intermediation suggest the possibility of overlap, since they indicate the downgrading of individual cultural and social capital in favour of a higher-volume way of working that bears some similarities with 'platformization' (Azzellini et al., 2019). Ultimately we suggest that digitalization may sharpen the antagonism between intermediaries and those they represent, not just in arts and entertainment but more generally.
These findings have a clear relevance for arts workers and their intermediaries. First, we have drawn attention to trends in cultural intermediation that may have worrying implications. We have suggested that in some highly digitalized contexts, intermediaries continue to play an important industrial relations role, but in a form where conflict between artists and their ostensible representatives is more likely; and where they may intensify, rather than mitigate, competitive pressures on arts workers. Trade unions, in particular, seeking to represent arts and cultural workers need to be paying close attention to these developments, though at this stage we cannot offer a ready-made solution. However, we argue that cultural workers would benefit from a re-emergence of collective forms of intermediation in cultural work. There are already attempts to reformulate collective and cooperative intermediation in highly digitalized contexts, but unfortunately these have rarely been applied to the kinds of labour markets we have discussed here (e.g. Scholz, 2016). In response to digitalized intermediation, we ask whether trade unions and musicians' collectives could pursue cooperative models, such as online labour brokerages with stronger minimum fee rules. Such forms, potentially, could push back against those elements of digitalized intermediation that intensify competition and thus undermine working conditions. We also stress the need for more collective mobilization by artists themselves in defence of working conditions. Greer et al. (2018) document collective action among musicians which extends from more traditional trade union membership to campaigning collectives. These actors have introduced new intermediation functions to fulfil urgent needs, such as lobbying against legislation which is detrimental to musicians and seeking out new resources and spaces for performances. Through these innovative practices, new forms of cultural intermediation are created which can intervene collectively on behalf of workers in artistic labour markets. The question of how these collective models could extend into market intermediation and embed social obligations is one for future empirical enquiry and one to which actors in the field will need to find answers.