TY - JOUR
AU - Grosse Ruse-Khan, Henning
AB - Abstract This article explores the challenges associated with creative work in the digital economy at both a conceptual and practical level, through the conjoined lenses of labour law and copyright law. We begin by developing a conception of the capitalist work relation and the distinct struggles intrinsic to it. This allows us to better understand the functions of creative work in contemporary ‘digital’ capitalism and the various regulatory challenges to which these different functions give rise. We then use this analysis to explore some of the conceptual and practical challenges that arise in both labour and copyright law when it comes to regulating creative work in an age of ‘digital platforms’, with a particular focus on the question of how to secure creators a fair remuneration, and potentially, a living, for their work. The concluding section discusses the potential and limits of new European Union rules on mandatory protections for authors and users to deal with these challenges. “It’s what we saw with Chuck Berry getting a Cadillac instead of royalties. It doesn’t really matter if it’s vinyl or an app, every time there’s a new way of doing music, the creators always get screwed.”1 INTRODUCTION Creative work, the productive activities that generate the content of various forms of media,2 has become an important source of profit for firms in the contemporary, digital economy. Often presented as an anti-thesis of capitalist work—as a form of social, or collaborative, work that allows for genuine self-expression and autonomy—creative work is increasingly being harnessed by capital to serve the ends of capital accumulation.3 While creative work has long been a key source of revenue for firms, the potential for firms to profit from creative work has increased exponentially with the emergence of the digital network environment.4 On the one hand, those responsible for the reproduction and dissemination of creative works (‘publishers’) can now exploit creative works not only in physical form (as published books, CDs, DVDs, etc) but also in digital formats, making them available on their websites or allowing third parties to exploit those works, for example, in databases and on platforms.5 On the other hand, as digital platforms such as YouTube, Facebook and Instagram have come to monopolize an ever-greater share of public attention, individuals have become increasingly reliant on those platforms to find an outlet for their creative work, and potentially, to earn an income. Not only does this place the platform in a powerful bargaining position when it comes to collecting the necessary rights in that work, but so too does it enable them to hive off data generated by the users attracted by those works so as to produce ‘advertising’ commodities that can be sold for profit. What this means is that, while the ‘digital network revolution’ creates new outlets for creativity, for creative workers to find new audiences for their works, as digital platforms become increasingly structured so as to serve the ends of capital accumulation, the emancipatory potential of such platforms as mediums for the expression of creativity—and as potential sources of subsistence income for individuals—is diminished. The traditional legal mechanism to address issues of economic dependence in the context of capitalist work relations has primarily been through labour and social security law. In this article, we aim to show that these mechanisms are insufficient, and often, inappropriate, to respond to the distinct challenges that economic dependence creates in the context of the creative industries today. These legal regimes operate on the basis of narrow conception of the capitalist work relation that fails to embrace the full diversity of forms which capitalist work relations assume, and this is particularly so in the context of the creative industries where direct employment is often inappropriate. In turn, the structure of these regimes, and of the rights they provide, is often ill-equipped to deal with the complex, and fragmented, relations that exist between producers and the firms that commercially exploit their works in the digital economy today.6 The issue is not simply, however, that much of the creative work performed for and on behalf of capital today currently falls outside labour law’s protective scope. But, in turn, many of the creative activities from which capital benefits today actually take place outside of the market, and thus, outside of a capitalist work relation, entirely.7 In this context, the problem is not so much the creator’s dependence on her creative activities for subsistence and the struggle, by capital, to lower the value of that subsistence (exploitation). Rather, the issue is that the intellectual property (IP) regime, in combination with a usually generous deference to freedom of contract, effectively enables firms to enclose and appropriate the products of those activities—either ‘for free’ or with little remuneration. In each case—both the case of the creative worker dependent on his/her creative work for subsistence and the creative individual engaging in autonomous acts of self-expression—the interests of the creator, and the public, are being subordinated to the demands of capital accumulation—and it is not entirely clear to what extent labour law might have the capacity, or might be the best mechanism, to address these problems. While labour lawyers and IP lawyers have both devoted considerable time and attention to exploring some of the regulatory implications of these issues, at present, and particularly with the growth of digital platforms, there exists little communication between the disciplines.8 Given that both labour and copyright law are concerned, in one way or another, with the issue of ‘work’, and the allocation of rights in, and over, ‘creative work’ and ‘works’, however, we believe that much can be gained from integrating the two perspectives, exploring the conceptual and practical challenges that creative work—in its various forms—poses in the digital economy today. Copyright lawyers have for a long time now focused on how the digital network environment changes the modes of, and liabilities for, exploiting copyrighted works—for example, in debates how to deal with hyperlinks (to protected works) as essential communication tools, or with the role of Internet service providers and other ‘intermediaries’ that incentivize, facilitate or at least offer technology for users to share copyrighted content online.9 While this has included significant discussion on more effective mechanisms to protect users and to deal with potentially copyright-infringing user-generated content, most of this discussion concerns user’s ability to generate, share and receive such content as a matter of ‘freedom of expression’.10 Current discourses invoke ‘user rights’ as tools to counter contractual and technological overreach by copyright owners and platforms in the form of blocking access to user-generated content—somewhat overlooking that on platforms (such as YouTube or Instagram) that employ automated content management systems (like Content ID); blocking nowadays is the exception and monetizing is the rule.11 Less attention therefore, in the digital context, has been paid to those that depend on creative work for their subsistence, or to the means of monetizing user expression online, in particular by platforms such as YouTube, Facebook, Instagram or TikTok. Labour law, on the other hand, has devoted considerable attention to the effects of digital surveillance and algorithmic control on traditional modes of work (the problem of the so-called ‘gig-economy’), but less attention has been paid to the effects of the digital network environment on those working in industries, such as the cultural industries, in which self-employment has long been the norm. Because creators of media content are owners of their creative products, their position is often compared favourably to that of the traditional employee, alienated from the products of her work, such that the significance of capital–labour relations in these industries for individuals’ subsistence has often been overlooked. As digital platforms transform more and more people into potential ‘creators’, the effects of the increased competition among creators on those dependent on their creative work for subsistence can no longer be ignored. In such a context, and given the limitations of traditional mechanisms of labour protection (unfair dismissal law, minimum wages, working time limits), it may be time to start thinking more creatively about how alternative legal regimes, such as copyright law, might help perform, in relation to those persons, some of the functions which, in other industries, have long been performed by labour law. In this article, we explore the challenges associated with creative work in the digital economy at both a conceptual and practical level, through the conjoined lenses of labour law and copyright law. We begin by developing a conception of the capitalist work relation, and the distinct struggles, or antagonisms, intrinsic to it. This allows us to better understand the role, or functions, of creative work in contemporary ‘digital’ capitalism, and the various regulatory challenges to which these different functions give rise. We then use this analysis to explore some of the conceptual and practical challenges that arise in both labour and copyright law when it comes to regulating creative work in an age of ‘digital platforms’, with a particular focus on the question of how to secure creators a fair remuneration, and potentially, a living, for their work. The concluding section discusses the potential, as well as the limits, of the new European Union (EU) rules on mandatory protections for authors and users to deal with these challenges. We find that while these rules do provide a foundation on which to build, they do not go far enough to qualify freedom of contract—and are likely to be hampered in their implementation environment by the wider labour law and industrial relations environment in which they must operate. CAPITALIST WORK: A CONCEPTUAL FRAMEWORK The concept ‘capitalist work’ is used in this article to refer to the historically specific form that working relations assume in the context of capitalism.12 It identifies as the distinguishing feature of capitalist work relations the struggle over surplus and subsistence,13 the capitalist’s attempts to increase the ratio of unpaid to paid labour time and the worker’s attempt to maximize the value of the wage, relative to the labour time expended.14 This struggle is a function not of the terms of a particular contract, but the wider class structure in which that contract is concluded. That is, of workers’ dependence on capital for access to the means of production and subsistence, their need to work ‘for’ capital, and on capital’s terms, to live. It is this dependence that provides the basis both for surplus value production, and for firms to profit from the unpaid labour of their workers, for it provides capital with the power to decide how workers’ labour power will be used, and/or the terms on which its product will be sold, and thus, to extract more labour from them than is represented in the wage. It follows from this that capitalist work relations, relations to which the struggle over surplus and subsistence is endemic, include all those working relations whereby the worker, by reason of their exclusion from the means of production, has no choice but to work for capital15 in some capacity in order to gain access to subsistence: whether that be by producing commodities (goods or services) for sale (productive workers) and so, producing surplus value; whether it be by contributing to and facilitating the production/realization process (workers hired in retail, or in managerial roles); and/or working on behalf of the state with a view to securing the overall functioning, and reproduction, of the capitalist system (unproductive workers). In all these cases, even where surplus value is not produced, workers’ dependence on capital is such that the firms, or organizations, for which they work, are in a position to extract more labour from them than they pay for in the wage, and thus, to profit from their (unpaid) work.16 Because, moreover, there is no guarantee that individual firms will actually pay their own workers a ‘subsistence wage’ in practice, regulatory mechanisms—such as labour law—are required to keep the struggle over surplus and subsistence in check, and in turn, to keep the exploitation of labour power within socially and economically sustainable limits. Capitalist work relations might, however, take any number of legal forms: workers might be hired as employees, or might offer their labour services as freelancers. Workers may also produce some ‘creative’ commodity (such as a copyrighted work in the literary, musical, dramatic or artistic domain, or a patentable invention), and sell their IP rights in that commodity (if those rights are initially attributed, in law, to the creator) in the hope/expectation of future payment. To the extent that they depend on capital to find an outlet for that work, capital still enjoys a power over that worker in terms of influencing what is produced, and how, even if a particular firm can only benefit from that work once a contract has actually been concluded.17 The precise legal or contractual form by which rights in the labour product are obtained (through a contract of employment, or a contract for services etc) is, then, irrelevant: what distinguishes the relationship as a ‘capitalistic’ work relation is not a particular form of contract, or a particular form of contractually underpinned subordination and control, but the underlying compulsion, rooted in social property relations, which forces workers to adapt their work processes to meet the needs or demands of capital (such as the, usually market-driven, tastes and preferences of publishers) and which empowers capital to extract from them a given proportion of unpaid labour. In the context of the ‘self-employed’ freelancer, for example, unpaid labour is extracted not by intensifying the work process and extending the working day through centralized management, as with traditional employees, but through the payment of piece-wages, paying per word, per article, per song or per minute of film etc—or even, as is the case on digital platforms, per hits, clicks, or view time. This leaves unpaid all that labour that individuals must engage in, in order to compete in a highly competitive market for creative work and thus find a firm, or platform, willing to exploit their work—all that time spent, developing and researching ideas, creating personal profiles, revising and improving works and/or promoting oneself. Identifying these distinguishing features of capitalist work is extremely important: capitalistic work relations are profoundly affected by the drive, by each firm, to maximize private profit in a way that productive activities performed outside that relationship are not (such as the work of genuinely independent producers, and/or the private and communal acts that give rise to the features of the cultural domain). This drive manifests in attempts, by firms, to increase the ratio of unpaid to paid labour time; to lengthen the working day (to increase the labour time expended); to increase the intensity of work and/or the rate at which commodities are produced. All these strategies have a profound effect on the organization of the labour process and on worker’s basic terms and conditions of work.18 The drive might even manifest in attempts by firms to depress the wages of their own workers below subsistence level or to shift the costs of reproducing labour onto other firms and the state.19 While it has historically been labour law that has placed constraints on these strategies—for example, through legally imposed limits on the working day, minimum wages and/or the regulation of working conditions—not only can many labour law norms effectively be excluded, or modified, by agreement, but many capitalist work relations fall outside of labour law’s protective scope entirely. 20 In such contexts, firms are thus relatively free to make use of workers’ labour power without having regard to the conditions required for its reproduction. CAPITALIST WORK RELATIONS: PLACING THE SPOTLIGHT ON LABOUR LAW Modern European systems of social and labour protection developed in the mid-20th century, in a context where most large manufacturing and service firms pursued a strategy of vertical integration, and mass manufacture, where collective bargaining was widespread, and full-time, stable employment for a single employer was the norm.21 As a result, the labour rights established under protective labour and social legislation came to be associated with just one legal form of capitalistic work relation: employment (or a ‘contract of service’). Broadly speaking, then, it remains the case today that to come within the scope of labour law, an individual must be working under a contract of employment with a firm, a contract whereby the employee (or ‘worker’) agrees to provide their personal services in exchange for a wage/remuneration; are under the direction and control of their employer; and/or be integrated into the employer’s enterprise.22 The direct, centralized, employment model of capitalist work was particularly conducive to the dominant competitive strategies of the era, namely, increasing surplus value production and firm competitiveness via the rationalization, and re-organization/standardization, of production. In such a context, and allowing for differences between industries, direct control over the production process, and a highly differentiated, or de-compartmentalized, division of labour, was integral to a firm’s profitability, and all of these things were supported by the legal incidents associated with the contract of employment.23 In an age in which firm’s competitive strategies have shifted more and more towards ‘modern’ market demands—such as competing via product differentiation and improvement (that is, towards dynamic rather than static efficiencies), as well as advertising and branding—and where the gains to be had from productivity increases are more limited, such a model appears increasing out-moded and under-inclusive.24 This shift towards competition driven by innovation also allows firms to perpetuate consumption patterns—as it is easier to create consumer demand for something that is new, innovative or otherwise an improvement to existing commodities than to facilitate demand on the basis of price alone. Today, in a context of vertical disintegration and a shift to ‘lean production’, many workers are contracted on an ad hoc or intermittent basis, and the contract of employment is no longer the dominant legal form by which labour is ‘hired’. Nor is life-long employment for a single employer any longer the norm.25 In such a context, labour law faces a real challenge. As we have seen, capitalistic work relations take their form from the broader structures in which they are embedded, from the exclusion of the direct producers from access to the means of production, and from the resulting dependence of those producers on capital to realize their labour capacity—in other words, the direct producer’s capacity to access their subsistence ‘through’ their work. The functions that have historically been performed by labour law, the need for some mechanism capable of regulating the struggle over surplus and subsistence, are no less pressing in relation to capitalist work relations that arise from sub-contracting or freelance arrangements, but at present, these functions are only performed (however imperfectly) in relation to those capitalist work relations that assume the form of, or can be conceptualized as an example of or analgous to, direct employment.26 The problem here is that, in a context of a structural inequality of bargaining power, it is capital that decides how labour will be contracted.27 As such, if it is in the interests of capital to contract labour through a series of commercial arrangements, rather than a contract of employment, then this is the legal form that will be adopted. While the use of this legal form necessarily influences the dynamics of the relationship, providing the ‘self-employed’ persons with various advantages of flexibility and autonomy; a change in legal form does of course not modify the respective socio-economic positions: the role of the formally ‘self-employed’ in capital’s circuit. As such, self-employed persons providing work or services to a capitalist firm are, very often, wage-dependent persons: dependent on that self-employed relation for subsistence, and this means that capital is in a position of considerable power when it comes to the extraction of unpaid labour time, and in influencing the terms and conditions of work. In order to respond to these problems, some legal systems have introduced a third form of work relation, conceptualized as ‘semi’, ‘para’ or ‘quasi’ dependent labour, located between employment and self-employment. The idea here is to extend some labour protections to those relations that resemble, in many respects, employees—the so-called dependent contractor—but lack one or more of the criteria commonly associated with the contract of employment.28 In the UK, this is known as the limb (b) worker,29 a concept introduced in the 1970s with a view to extending some but not all, labour rights, beyond employment. The problem, however, is that these extensions simply do not go far enough, because they ‘still’ presuppose an agreement to work in exchange for remuneration, and this excludes many of the ‘genuinely self-employed’ persons who are, nonetheless still dependent on capital for access to subsistence, and as such, over whom capital still exerts considerable power when it comes to their overall terms and conditions of work. This is precisely the case, as we shall see, with many people working in the creative industries, as well as the growing numbers of persons attempting to earn an income by creating media content for digital platforms. These issues of scope aside, it is not sufficient, or practical, to simply extend the scope of existing systems of protection to forms of working relationship which the legislator did not have in mind when those protections were enacted. Indeed, it is one thing to recognize that a freelance journalist, dependent on selling articles to multiple newspapers, websites and journals to earn a living, is in a comparable socio-economic position to an employee; and quite another to design a set of regulations capable of providing that individual with labour rights equivalent to unfair dismissal protection, or a guaranteed minimum (living) wage. It is not clear, for example, that unfair dismissal legislation really makes sense in a context in which what the individual depends on is not being provided with regular work by a firm, but the opportunity to access the market which that firm dominates. Similarly, a right to be paid a minimum sum per unit of work (as with minimum wage legislation) is relatively unhelpful in a context in which labour must be expended prior to the conclusion of a contract, and where the amount and nature of that labour is inherently speculative and uncertain. In such a context, what the individual needs is not so much a guaranteed minimum sum per unit of work, but a right to receive a reasonable proportion of the profits made from the exploitation of the finished work, and, potentially, a right to be paid a ‘subsistence payment’ to cover living costs throughout the duration of the creative process. Neither of these things is, however, rights currently envisaged under, or even compatible with the conceptual architecture of, labour law. There are thus real practical, as well as conceptual, difficulties involved when it comes to extending labour law protections to certain forms of capitalist work: not only the question of how minimum wage legislation could be structured and applied to the individual uploading content on YouTube, who is dependent on the advertising revenue that YouTube shares with content providers for subsistence, but also, questions such as: how should those individuals be protected against—or at least be given a say in—tweaks in YouTube’s, TikTok’s or Instagram’s algorithms that determine the feed of content to users,30 hence impacting significantly whose content is pushed, and hence may be able to earn from one’s own content on these platforms? Should an Uber driver be paid when the app is switched on, or only when a trip has actually been accepted?31 How might a right to strike be operationalized in a context in which content creators are under no legal obligation to work and creators are so numerous, and dispersed, that the pressure on platforms likely to result from any withdrawal of ‘labour’ is likely to be minimal? But the problem is not solely one of how to adapt and extend existing systems of protection to cater for some of the unresolved challenges at the heart of particular types of capitalist work relation; part of the problem today is that, with the growth of the digital network environment, the boundaries between work and non-work have become increasingly blurred. Capital is increasingly benefitting from a number of non-work activities (such as the activities and interactions of ‘users’ of social media platforms), activities that have the capacity to indirectly affect workers’ standards of living, and which, in turn, generate distinct regulatory challenges of their own. It is in this context that the broader notion of ‘work’ in copyright law might be instructive. Whereas the concept of capitalistic work relation derives its meaning of the term ‘work’ from an analysis of the relationship through which a particular productive activity is performed,32 and the labour law concept of employment focuses on just one of the many possible legal forms of such relationship, in copyright law, the concept of work, as a ‘creation of the mind’,33 is much broader. Here, ‘work’ is theorized as an object—one that does not necessarily have to manifest itself in tangible form—by reference to the act of human creation emanating from the human mind, and the nature and quality of ‘intellectual labour’ that produces it. It thus encompasses the products of a wide spectrum of ‘creative’ activities, including those performed ‘through’ a capitalist work relation, ‘and’ those which are not. By this latter, we mean any ‘free’ or ‘spontaneous’ acts of creation—ie those outside the sphere of compulsion described above—that generate the cultural goods which in turn contribute to the ‘creative’ or ‘knowledge’ commons. For our purposes, this would include those that create, often on the basis of existing copyrighted materials, user-generated content and upload it onto platforms where it is being viewed, shared and further modified or commented on by other users. Capital, here in the form of those claiming rights in that content and those operating the platforms on which such content is shared, can simply rely on the technological infrastructure and/or on claiming rights over that content in order to extract value from the creative acts users engage in, acts on which users do not depend for subsistence, even if their ability to engage in such acts presupposes an alternative source of income. While these creative ‘works’ are not produced ‘in’ or ‘through’ a capitalist work relation, and so, are not subject to the same struggle (or not in the same way)34 over surplus and subsistence, they are still works from which capital can, and often does, benefit. Such benefit might be obtained, for example, by gathering, usually by means of contracts, the copyrights in works produced outside capitalist work relations and in that way ‘appropriating’, or enclosing, those works from the ability of others, including the creator, to access and utilize them. Or it might involve capital drawing on products of culture (myths, legends, traditional songs or religious symbols) in the design of ‘original’ brands or trademarks appropriating and manipulating the meaning of those cultural products for commercial purposes—as with Disney’s trademarking of the Swahili phrase ‘Hakuna Matata’, or Kim Kardashan’s trademarking of her shapewear ‘Kimono’. In these ways, public works or other creations of the mind can provide a source of profit for capitalist firms, without adding to the (surplus) value in circulation, and without paying respect to the origins of those works or their embeddedness in cultural identities.35 The products of these free acts of creation outside capitalist work relations therefore may, to the same extent as works produced within such relations, become objects of private property, pressed into the services of profit-making. Their enclosure has little regard for the interests of those producing these works, or the wider public in their enjoyment. At the same time, the firms that rely on these ‘free gifts’ are able to obtain a competitive advantage, or monopoly, over certain goods, which can profoundly affect questions of distribution. Of course, the tendency for capital to make use of various ‘free gifts’, or natural and social resources outside the market, is not unique to the creative industries.36 Of the various public goods appropriated by capital, we might include not only inputs from the cultural and scientific domain,37 but also ‘free gifts’ of nature (soil fertility, wind power, etc), the talents and skills that individuals and communities cultivate outside the workplace, as well as the unpaid labour expended, often by women, in the household.38 The issue in this context is not that the individuals responsible for producing these ‘social’ goods are exploited by capital39—that is, that they are dependent on these activities for subsistence, creating a distinct imbalance of power. Rather, the common denominator is that aspects of the social, cultural or natural domain are been enclosed, and privatized, for the purposes of making profit—instead of remaining outside the reach of a market, and often available for use, and enjoyment, by individuals or communities. Not only are the communities from which these goods are appropriated excluded from sharing in the resulting financial gains, but they also lose control over the meaning, and sense of identity, that certain social or cultural goods often connote. When exploring the distinct challenges that labour law faces when it comes to regulating the creative industries, then, we need to bear in mind not only the difficulties that labour law confronts when it comes to extending and adapting labour law protections to non-employment, capitalist work relations; we also must be conscious of the quite distinct challenges that arise in relation to the various forms of non-work activities from which digital platforms in particular benefit today, and which have the capacity to affect, albeit indirectly, questions of subsistence, given their impact on distribution and quality of life. LABOUR LAW IN THE CREATIVE INDUSTRIES: A CONCEPTUAL AND PRACTICAL CHALLENGE As systems of labour and social protection have become increasingly complex, it has been increasingly difficult to tailor labour protections to the concerns of specific industries.40 With the decline of sector and industry-specific collective bargaining, tailoring employment rights to the conditions of particular groups, and with the growth of individual employment rights, the tendency in most European jurisdictions has thus been towards a general set of labour rights which, inevitably, fail to cater for the ‘distinct’ challenges faced by different workers in different industries, such as the creative industries, a problem which is simply exacerbated by the various ‘digital’ transformations taking place. One particular challenge here is that, in the context of ‘creative’ work, capital must concede a degree of autonomy to creative workers if the use value of their ‘creativity’ is to be realized in full. As something arguably more intrinsically than extrinsically motivated,41 creativity (and to that extent, creative work) cannot be ‘ordered’. That is, if the final product is to bear the stamp of creativity—or, as the European Court of Justice puts it: ‘a personal touch’42—that will usually enhance its attractivity to consumers, then autonomy in the act of creation is essential.43 Providing space and freedom in regard to the conduct of creative work is thus critical if workers—or to use copyright’s terminology: authors or creators—are to come up with innovative ideas, and/or to produce the new, and distinctive products that consumers will want to buy. In fact, the ability to make ‘free and creative choices’ is the raison d'être for copyright to vest in a particular work.44 And capital can readily concede this control to creative workers precisely because (and to the extent that) these authors largely lack the capital that would be required for them to reproduce, disseminate or otherwise exploit, and so profit, from their own work. As such, while the creative worker/author often can produce creative works independently from capital, dependence on publishers and other firms—such as online platforms in the digital context discussed below—for commercial exploitation affords the latter power and control over the former. What we have very often in the ‘creative industries’, therefore, is a particular legal form of the capitalistic wage relation, albeit one that lacks many of the hallmarks of traditional employment. Against this background, the relation between those who create independently, that is, outside a formal employment relationship (here generally referred to as ‘freelancers’), and those who control the means to exploit this creative output serves as the principal example for our discussion of the creative industry. Freelancers are rarely paid for the time expended on producing a particular good, but are paid by output: per article, per song, per film, etc.45 And yet, they are not in a position to charge high prices for their creative products because, with the market for freelance work over-stocked, they tend to bid-down the price of their own works. Capital knows that these freelancers (often) have no control over the means to commercially exploit their works, nor access to an alternative outlet for their works, and can take advantage of that position accordingly. While, originally, this dependence took the form of a dependence on publishing houses and/or printers, record labels and/or film production companies, with payment often conditional on sales of physical media (books, CDs, DVDs etc); today, increasingly, this dependence takes effect also, or alternatively, as a dependence on the owners of digital platforms. Platforms such as YouTube, Instagram, Facebook or TikTok afford creators with virtual access to a public audience for their creative works, providing the technical infrastructure, the software code and—most crucially—an opportunity to reach out to a global network of users through which they can be ‘published’. And given that these platforms depend for profitability not merely on the exploitation of creative content itself, but also on the communication devices (the hardware and software) that this content helps to sell and the opportunities for advertising revenue it creates, they are less interested in increasing the profit from any given work, and more in increasing the overall quantity of content available so as to expand the choice available to ‘users’. This removes bargaining power from creative workers as they effectively become substitutable and, as such, reduces the pressure on platforms to ensure fair remuneration to producers/creators to continue to attract creative content. In short, by controlling ‘digital’ or ‘virtual’ capital, platforms can extract value from the creative works of their users. It follows that platforms rarely require that the audience pays, in monetary terms, the platform or the creator for access to creative works. Instead, platforms typically ask, in their terms of reference and ‘policies’, their users to accept the extraction, use, and sale of user data in exchange for access to the platform.46 That data in turn attracts advertisers to the platform which pay the platform for access to those users they wish to target with their political or commercial messages. Depending on the platform’s business model, creators receive a share of the advertisement revenue, promote brands or have found other ways to benefit (or even earn a living) from their platform presence.47 Focusing on those platforms that allow creators to monetize content, the income of the creative worker thus becomes conditional not (or not necessarily) on the sale of the creative work by a ‘publisher’, but on the advertising revenue generated by the platform: the money flows not from publisher to creator, but from advertiser to creator, via the platform. As such, these creators would struggle to fit themselves within even the more expansive definitions of worker (or equivalent) in labour law, for even these presuppose some form of bilateral, wage-work exchange. Given the limitations (discussed above) of many statutory employment protections in this context, it is perhaps no surprise that creative workers have long tried to rely on forms of collective organization (as opposed to individual statutory labour and social protections) to secure more favourable terms and conditions of work. The problem, however, is that, today, it is likely that many agreements between unions of creators and publishers/platforms will fall foul of EU competition law—at least to the extent that they touch core contract terms such as remuneration.48 While collective agreements between employers and employees are generally exempted from the scope of EU competition law, this will not be the case for agreements between ‘undertakings’—for example, between publishers or platforms and the genuinely self-employed. While there may be scope for arguing that the individuals are ‘false’ self-employed, dependent on a particular organization for work,49 such an argument is unlikely to succeed in the case of YouTubers and other content creators, given the absence of any contractual obligation between them and the platform to provide work.50 Aside from the legal and (practical) challenges of organizing among self-employed creators, and of getting those unions protected and recognized by publishers/platforms, gaining enough support from creators whose absence would be noticeable enough to make a difference to platforms such as YouTube is a major challenge. The sheer numbers of potential content creators, and the fact that currently, only a minority of those content creators actually depend on these platforms for subsistence (as opposed to reputation and self-expression), and so, will have something to gain from joining a union, means that the pressure such unions can exert on platforms will always be limited. Indeed, this seems to be exactly what the (relatively new) YouTubers Union (based in Germany) is experiencing; YouTube has little incentive to respond to the union’s demands so long as most major creators are not members, and are not prepared and ready to take action. The Union’s demands—for, inter alia greater transparency, for parity of treatment between creators, a more equitable approach to monetization, and a seat at the table where decisions are taken that are likely to affect creator’s livelihood—are very much in line with the campaigns pursued by unions in other industries for decades.51 History tells us, however, that, however important unionization and collective action might be in theory, in practice, they are highly dependent on the existence of a legal framework that places positive obligations on capital—publishers and platforms—to recognize and bargain with unions, and a state-backed procedure for extending agreements to non-members. Not only is such a framework absent in most countries52—particularly in relation to the self-employed, but it would also likely be frustrated today by EU competition rules.53 While the digital network revolution initially appeared to have ‘democratized’ access to information, communication channels and the ability to reach (in theory) a global audience, an increasing concentration and marketization of social media has meant that this ‘revolution’ has simply shifted control over the relevant means of production from traditional publishers to online platforms. Because the creative industries have always appeared comparatively attractive to many, the effect has simply been to intensify competition among creators. In the digital environment, widespread access to IT devices (such as smartphones) with sophisticated recording and editing technologies and immediate options for communicating and sharing results online with a global audience has further intensified competition among those who dream of becoming a ‘star’. This same process of competition forces creators to produce ‘more, and arguably, more “popular” works, known to be in high demand’, in the hope of increasing their chances of finding an outlet for their work. The result, then, is that firms benefit not only from an ever broader supply base, but also from a vast amount of extra labour time spent developing, researching and refining creative output—without having to pay for it. Despite these problems, the economic position of creators as freelancers is often compared ‘favourably’ to that of creator-employees, because usually the former, but often not the latter, hold copyrights in their works. This is equally true of those creators producing content for sharing on a digital platform in the hope of obtaining a fraction of advertising revenue, peer recognition, fame, or all of that. Indeed, legal ownership of a right that allows to exercise control over the reproduction (by means of making copies), initial publication, distribution and communication to the public of the creative output might seem like an essential bargaining chip for freelancers that often will not be available to employees. However, and while there are important differences in terms of allocating initial ownership of copyright in employment scenarios between what can be roughly divided along Common Law and Civil Law country lines,54 we show below that neither these differences nor the distinction between employees and freelancers, eventually matter much ‘in practice’: that is, once bargaining powers, the resulting copyright contracts, and (implied) assignments or licenses are factored in. In effect, copyright law’s relative deference to freedom of contract undermines the critical, or emancipatory, potential of its broad conception of ‘work’. COPYRIGHT PROTECTION FOR CREATORS ON DIGITAL PLATFORMS In order to consider copyright’s potential in providing creative workers with a key bargaining chip vis-à-vis those who exploit their works online, this section reviews common approaches to the allocation of first ownership, as well as on assignment and licensing, taking into account differences amongst major jurisdictions, especially concerning employees and freelancers (as well as other creators). In Common Law jurisdictions—for example on the basis of section 11(2) of the UK CDPA—the initial ownership of a copyright in a work created by an employee in the course of employment is vested not in the (natural) person who created the work (who generally is considered as the author, section 10(1) CDPA)), but in the employer.55 The same approach is taken in, for example, Japan,56 Korea,57 Ecuador,58 Argentina,59 Italy60 and Brazil61—and in essentially all jurisdictions considered here, it can be qualified, or deviated from, by contract, collective agreements or relevant practice. Whether a creator in fact is considered an ‘employee’ and when that person has created ‘in the course of employment’ then, more or less, depends on traditional employment criteria (such as the degree of control exercised, the integration within an organizational structure, and/or the application of tax- and insurance regimes) and the proximity of the creative work to the employment.62 Those employed under a contract of service hence do not, from the outset, own copyright in works created in the course of their employment (which in turn generally depends on the scope of their duties, as set out in the contract). While some Common Law countries (such as the UK, Australia and Canada) offer employee-creators a degree of ‘moral rights’ protection—that is rights which allow to claim authorship and to object derogatory treatment of the work—those protections are often heavily qualified in the context of employment, or can be effectively waived.63 Finally, employee-performers, ie those who perform copyrighted (dramatic, literary or musical) works, are not necessarily subject to a statutory allocation of rights to the employer—but can fully assign those rights, including prospectively.64 As discussed above, the creative industries seldom enter into a formal employment relationship with those who produce creative works for them. And of course does such a relationship not exist between platforms and their users/creators. On the other hand, freelancers, others that create under a contract for services (those traditionally regarded as ‘self-employed’), as well as creators that accept a platform’s terms of service, will tend to be initial copyright owners in their works.65 To the extent that performers are not subject to an initial allocation of their rights to an employer or commissioner, they also fall under this category. The same goes for users that create works for upload onto a platform. However, all those are likely to lose their rights either by means of explicit assignment66 or on the basis of an implied assignment or license, construed to give ‘business efficacy’ to the underlying arrangement with a publisher (or other entity in control of the means of exploiting their work or performance).67 And while freelancers in the creative industries might occasionally win a legal battle over the scope of the exploitation rights granted (explicitly or by implication) to capital,68 this generally cannot undo the underlying economic power relations: history shows that capital firms (in our case, publishers or platforms) are likely to adapt their standard contractual terms for all future licences, and might prefer to simply stop using existing works if freelancers and other creators are not willing to accept retroactive licence extensions.69 In Civil Law countries, natural rights theories about the foundation of author’s rights (droit d’auteur) often mean that the person creating a work will always be its first owner—even in employment situations.70 While here employees retain formal title over their copyright, the law in Germany, for example, allows (and generally operates in favour of) extensive transfers of all relevant ‘exploitation rights’ (Nutzungsrechte) to the employer if that finds support in their underlying contractual relationship.71 In France, the cradle of the droit d’auteur approach, employers will usually have to arrange for an express transfer of all the exploitation rights sought for, as contracts are generally interpreted in favour of the author (in dubio pro auctorem).72 China in principle adopts the Civil Law position, allocating copyright to the employee-creator. However, the employer generally has a ‘priority right’ to use an ‘employment work’ within its scope of business and enjoys all relevant exploitation rights with regard to specific categories of works by virtue of Article 16(2) of the Chinese Copyright Act.73 In sum, while most Civil Law jurisdictions that follow a droit d’auteur approach appear to formally allocate copyright to employee-creators rather than their employers, they often assume an implied transfer, based on the nature of the employment relationship, of all relevant exploitation rights for works created by an employee in the course of her employment. And where no such implied transfer of rights to the employer is assumed, usually little prevents the latter from relying on their bargaining power to seek extensive transfers of all relevant rights to exploit the work. However, this is not always the case. Relying on Germany as an example, we want to show how some mandatory protections for authors are designed to address this imbalance. They have been primarily drafted with freelancers and other creators in mind, but can be applied to employees as well—if the nature and terms of the employment relationship do not suggest otherwise.74 The main question for our purposes then is whether such mandatory protections for authors can be effectively invoked, maybe even via collective bargaining or in conjunction with collecting societies, by those who create and share content on platforms. In a range of provisions, the German Copyright Act (UrhG) deals with transfers of exploitation rights and licensing in a way that generally adopts a pro-author approach. While an extensive treatment is not possible here,75 the following outlines key protections that freelance authors in the creative industries as well as creators on platforms can potentially avail themselves of. Starting point is the ‘specific purpose’ principle (Zweckübertragungsgrundsatz) whereby in cases where the type, nature, (territorial and temporal) scope or limits of a transfer of exploitation rights has not been explicitly agreed to, in principle, rights are transferred only to the extent necessary to give effect to the underlying contractual relationship and its purpose.76 This ‘no more than necessary’ principle can help freelancers (and other non-employee creators),77 in particular in cases where they are producing works in an ad hoc or one-off scenario, for a particular purpose, or in—in the digital context—for a specific platform: any uses outside that respective context will need a separate agreement, with arguably a fresh opportunity for bargaining. Section 37 UrhG follows this with specific legal assumptions in favour of the author concerning the type of rights that remain with her in case, for example, rights to reproduce or make available to the public are being transferred. Similarly, the law assumes a prohibition to adapt the work, to modify its title or to remove attributions to the author.78 Contracts concerning future (yet unknown) forms of exploitation and those concerning rights over future (to be created) works of the author require written form, and can be cancelled once the author has been notified of the new mode of exploiting her work—while a commitment to transfer rights over future works can be terminated after five years.79 Furthermore, onward transfers of rights or sublicenses by the publisher to third parties generally require the explicit permission of the author.80 Of course, one might counter that most of the protections referred to so far are merely dispositive and hence can be overcome by explicit terms in a contract that provide otherwise—for example, transferring rights over all (known) forms of exploitation, without time limits or further restrictions and worldwide, including an express permission to transfer onwards and to sublicense. Again, the underlying economic relations and the resulting power dynamics will make it unlikely that a lot of freelancers in the creative industries can actually avail themselves of these protections effectively. An analysis of the terms of service used by major platforms in the context of digital networks confirms this. Such standard terms usually include specific clauses on the extent to which creators license their content to the platform. At the time of writing, YouTube, for example, asks all its users to ‘grant to YouTube a worldwide, non-exclusive, royalty-free, transferable, sublicensable licence to use [the creator’s] Content (including to reproduce, distribute, modify, display and perform it) for the purpose of operating, promoting, and improving the Service’.81 Creators also have to grant other YouTube users a worldwide, non-exclusive, royalty-free licence to access and use their content, as enabled by YouTube services. In principle, that licence continues until a creator removes her content from YouTube. However, in contrast to the example of TikTok discussed below, this licence is more limited in scope, and, most importantly, does not preclude creators from benefitting financially from the content they produce. TikTok, the App for sharing short videos where creators often ‘perform’ to licensed music and other third-party content, requires all its creators/users to sign up to significantly less favourable terms.82 While it allows creators to incorporate and combine (parts of) videos produced by other users as well as licensed third-party content and then share their resulting video on TikTok and other platforms, creators have to agree to effectively unlimited commercial exploitation of their content by TikTok without receiving any shares in profit, income or other consideration.83 Creators also grant the platform: an unconditional irrevocable, non-exclusive, royalty-free, fully transferable, perpetual worldwide licence to use, modify, adapt, reproduce, make derivative works of, publish and/or transmit, and/or distribute and to authorise other users of the Services and other third-parties to view, access, use, download, modify, adapt, reproduce, make derivative works of, publish and/or transmit your User Content in any format and on any platform, either now known or hereinafter invented.84 From composers and recording artists, TikTok further requires express assurances to clear all relevant rights with their labels or publishers, and finally asks all its creators/users to assign or waive—to the extent possible—any moral rights they might enjoy in their videos.85 Overall, TikTok’s comprehensive and detailed provisions leave little room for the application of dispositive pro-author copyright contract rules along the lines of the German UrhG. Other platforms that are less directly concerned with users sharing their creative works—such as Instagram or Facebook—nevertheless include express licences in their terms of service whereby a user that uploads, posts or shares content on or in connection with Instagram or Facebook, also grants ‘a non-exclusive, royalty-free, transferable, sub-licensable, worldwide license to host, use, distribute, modify, run, copy, publicly perform or display, translate, and create derivative works’ in relation to that content.86 While the rights under such a licence generally terminate once a user deletes content on the platform, Facebook, for example, explains that in case the content continues to be used by others, ‘this licence will continue until the content has been fully deleted’.87 In contrast to Instagram, Facebook however does allow its users to monetise their creative works on its platform, offering them—roughly equivalent to YouTube—a 55 per cent share of the advertisement revenue generated.88 In sum, and while individual contractual terms differ, the explicit licensing rules in the terms of service used by major platforms generally contract out of a range of the potential creator protections under copyright contract law. Returning to the licensing provisions in German copyright law, the perhaps most relevant form of protection of authors, however, cannot be contracted out of or waived—not even by the choice of a foreign law to govern the contract, if the relevant contract otherwise would be governed by German (copyright) law, or if the relevant acts of exploitation take place in Germany.89 This protection comes in the form of two provisions that both aim to secure fair and adequate remuneration of authors, in particular freelancers, in exchange for the transfer of exploitation rights. Its social function is to enable freelancers to make a living from their creative undertakings: that is, by selling the rights to exploit their works—which in turn makes them highly dependent on the remuneration they receive.90 In light of the expansive licence terms employed by platforms vis-à-vis content created by their users, pared with the essential role some of these platforms play for commercializing content online, contract-proof protections for creators offer an appealing solution—if they can be effectively operationalized in practice. In this context, section 32 UrhG generally entitles the author to ‘adequate remuneration’ (angemessene Vergütung)—a fair share in the profit generated by the exploitation of her work.91 While the law defines this primarily in reference to a range of factors to be applied in the individual circumstances (determining what is customary in the relevant trade or profession, taking into account the type, scope, etc of the transfer), it further encourages collective agreements between authors and those exploiting their works.92 Once such an agreement exists, the remuneration set out therein is considered adequate, and an individual author can invoke claims for that amount against any contracting partner involved in negotiating the collective agreement.93 For exploiting uses unknown at the time of the transfer, the author deserves separate, adequate remuneration.94 In addition to these forms of adequate remuneration, the author has a right to obtain ‘additional remuneration’ in case the profit generated by exploiting the work is significantly disproportional to the remuneration initially afforded to the author.95 Finally, several provisions in the German Copyright Act foresee further (and again mandatory) remuneration of the author for specific forms of exploitation (such as cable transmission and auctions of artistic works), usually based on EU rules harmonizing copyright in the EU single market.96 As those forms of exploitation are less relevant in the context of exploiting works on platforms, they are not further discussed here. These various forms of mandatory remuneration rights specifically aim to protect freelancers and certainly offer legal tools which can be invoked to address the unequal bargaining positions that otherwise underpin typical work relations in the creative industries. But how effective are they on the ground, including for global online utilizations? In the context of platforms commercializing creator’s content, it is worth noting that TikTok, for example, has a specific German version of its Terms of Service which—although generally as unfavourable to creators as its counterparts for other jurisdictions—explicitly leaves claims for adequate remuneration under section 32, 32a UrhG intact.97 There is, however, no indication how authors would be able to invoke these claims. TikTok does not pay any (let alone adequate) remuneration in exchange for the broad licence creators have to grant. Instead, TikTok’s express denial, in its terms of service, of any rights for creators to commercialize their works or to receive a fraction of TikTok’s profit generated by exploiting these works suggests that the platform does not seem particularly interested in living up to its obligations.98 Since TikTok’s licence also includes the right for other TikTok users to utilize the creator’s content without compensation in their own videos, a song uploaded onto TikTok can serve as basis for hundreds or thousands of memes or lip-synchs by other users—without creator’s seeing any penny for it.99 Another technical–legal, but important, point is that the German version of TikTok’s terms which includes the reference to section 32, 32a UrhG only applies to creators with a habitual residence in Germany. Under section 32b UrhG, however, providing adequate remuneration is an internationally mandatory norm which cannot be contracted out by choosing a foreign law—and that applies also when works are ‘exploited in Germany’ (that is, when accessing, viewing and sharing the work via TikTok takes place in Germany).100 The platform hence does not act in conformity with its obligations to offer creators remuneration in accordance with the German mandatory author protections. Other platforms do not appear to integrate any of the mandatory protections for authors in their terms of service, including in cases of users with habitual residence in Germany, or where exploitation of the works created by users takes place in Germany. At the same time, YouTube and Facebook do allow creators to commercialize their content, and generally offer a share of around 50 per cent of the advertisement revenues. Given that the cut provided by YouTube seems to serve as global benchmark for what creators can expect online,101 it may well be seen as indicator for adequate remuneration under section 32(2) UrhG—although this eventually will always depend on a case-by-case analysis.102 And while there are calls (especially in Germany) for YouTubers to unionize and assert collective bargaining powers for a ‘more stable and equitable approach to monetization’, those so far seem not having been framed in the context of collective agreements over adequate remuneration under section 36, 36a UrhG.103 In addition, the terms of service used by several platforms more generally refer to the rights of ‘consumers’ under applicable EU consumer protection rules, including the jurisdiction of courts in EU Member States to handle disputes arising with consumers habitually resident in the EU.104 Even under the assumption that such references can be construed to include mandatory protections for creators along the lines of the German Copyright Act, this would only protect a creator that occasionally uploads content onto the platform. On the other hand, a freelancer or other creator aiming to make a living commercializing her creative work online arguably will not be considered a consumer. In its most recent terms of service, YouTube explicitly distinguishes between consumer and business users—explaining in its FAQs that ‘if you're a creator whose main source of income is monetising videos on YouTube (or other platforms) you would also most likely be a Business User’.105 However, not all clauses and determinations made by platforms in their terms of service will necessarily survive litigation in national courts. In France, the Tribunal de Grande Instance de Paris held not only that Twitter users are consumers as they pay for the platform’s service with their data, but also that a non-exclusive, worldwide blanket license (along the lines described above) in Twitter’s general terms of service, allowing Twitter to use all content generated by a user on the platform, including future works, breaches various author protections rules of the French Copyright Code and hence is invalid.106 The same Court subsequently also held equivalent licence clauses in Google’s and Facebook’s terms of service to be void as they grant the platform a right of free use of all content, including copyrighted works, generated by the user without sufficiently specifying the works at stake, the nature of the rights conferred and the type of uses which are allowed.107 While these cases had been brought by a consumer association on behalf of consumers, the relevant author protection rules in the French Intellectual Property Code not only apply to consumers, but authors in general.108 Nevertheless, it appears doubtful that individual court decisions will lead to a significant change of the situation for freelancers and other creators on the ground—who often earn comparably little from their creative work. Studies on the income of professional authors (defined as allocating more than half of their time to writing) in Germany and the UK show that an average earning of 42 per cent and 64 per cent of the national gross median wage in 2004/2005.109 Another study analysing the effect of introducing the above-discussed mandatory creator protection in Germany similarly refers to an average income that is less than half of that earned by self-employed persons generally.110 The latter study focuses on the impact of the obligation to pay an adequate remuneration (section 32 UrhG) and additional remuneration in case the profit generated by exploiting the work is in significantly disproportional to the remuneration initially afforded to the creator (section 32a UrhG). It finds that 10 years after the introduction of these legislative reforms to improve the contractual position of creators, not much has actually changed on the ground. Few collective agreements on adequate remuneration under section 36, 36a UrhG have been concluded; and the German Federal Supreme Court awarded only marginal additional payments in application of sec tion 32 UrhG.111 Somewhat more successful—but primarily for works those exploitation had been highly profitable—had been attempts by creators to claim additional remuneration under section 32a UrhG: German courts have held that any remuneration that is less than 50 per cent of the adequate remuneration owed under section 32 UrhG triggers a claim for additional remuneration112—which importantly can be exercised not only against the contracting partner, but ‘any’ licensee that profited from the work. Nevertheless, from the creator’s perspective, the German law so far does not appear to have achieved its aim to ensure adequate remuneration for creative work.113 There are few reasons to believe that the existence of mandatory protections for creators, including in form of an obligation to provide adequate remuneration, can—without more—ensure fair contracts with platforms, or more generally improve the situation of user/creators on platforms. To begin with, even on platforms like YouTube that do share ad revenues with their creators, only very few among the popular influencers can make a living from the content they upload. While YouTube does not publish data on this, reports suggest that even breaking into the top 3 per cent of most-viewed channels is likely to generate advertising revenue of no more than about US$16,800 a year.114 In research conducted by Bärtl in 2018, these top 3 per cent of video creators of all time attracted more than 1.4 million views per month.115 One of the key observations presented in Bärtl’s study unsurprisingly then is the overwhelming dominance of very few channels over the rest of content on YouTube, and a ‘macro-manifestation of the rich-get-richer phenomenon’.116 As the supply of videos on YouTube continues to increase, and with other platforms like TikTok and Instagram (which don’t share advertisement revenues) competing for user attention, it is unlikely that except the very few actually can make a living from their creative content shared on platforms. And it is unrealistic to assume that this will change simply by introducing mandatory rules on adequate remuneration: if, for the vast majority of creators, there is generally very little to earn (in the case of platforms, by means of advertisement revenue) from an individual piece of content uploaded onto the platform, then the latter has very little to share in terms of remuneration. But this does not mean improving copyright contract law and mandatory remuneration rules has no purpose: it can provide at least the legal tools to ensure that in case the platform does make (significant) profit, an adequate share can go to the creator. In light of the new EU rules on adequate remuneration discussed in Section ‘New Rules for the Digital Market as a Way Forward?’, the principal hurdle to overcome then may not necessarily be the lack of an appropriate legal framework (at least not in the EU), but effective tools for creators to enforce their rights for adequate compensation. While virtually omnipresent, the platform’s physical ‘remoteness’ for users, combined with most user’s dependency on the platform as an essential element of their social existence, will significantly affect the latter’s willingness to actually invoke their rights, while, at the same time, making collective organization of users/creators extremely difficult. This is not to say that the legal framework should not be designed in a way that generally allows claims by users for fairness in how their creative works are exploited. It simply points to the further difficulties about effective enforcement on the ground (or rather: in the virtual world) which need to be tackled, for example, via strengthening collective bargaining frameworks, and/or by reconsidering the role, and position, of collection societies in defending creators’ rights. In the end then, traditional publishers and virtual platforms as the new digital intermediaries normally end up with all relevant rights to commercially exploit the creative work of creators that produce for capital. While most creators (including freelancers) will be the initial legal owner of their work, that is, however, only until they give up those rights to capital. In the traditional context of the creative industries, many freelancers are entirely dependent on one or more firms to reproduce and disseminate their works so that firms frequently make the transfer of copyrights (or—where that is legally not an option—an exclusive, permanent and world-wide license) a condition of the contract. We have shown how equivalent outcomes are produced online when creators wish to share their works on platforms—be it in hope for money, fame or simply social interaction. Either by statute, by contract or by implication, giving away the exclusivity an IP right (usually copyright or related rights) entails is therefore a condition for the individual to not only (possibly) earn an income from their ‘work’, but—especially in the virtual environment—to use one’s work to communicate or express oneself. Where competition between creators is particularly intense, and where the industry is monopolized by just a few capitalist firms, these firms are able to exploit creative work, without actually guaranteeing them any income at all. This is exactly what happens in the context of digital platforms: at best, the creator is provided with a share of any (advertising) revenue generated from the content they provide. As shown above, only in very rare cases will this lead to any meaningful income. Other platforms—such as TikTok—are able to attract creators and collect all relevant rights in their works without offering any share in their profits. And this is not something that is likely to be improved via collective organization: not only is organization legally and politically fraught, but creators are so dispersed, and their interests so diverse, that they are unlikely to form a sufficiently large, homogeneous, group to impose a credible threat to the platform. In such a context, moreover, as our discussion shows, simply introducing mandatory rights for an adequate remuneration will not solve the issue—unless effective enforcement mechanisms are included. This is a lesson that was learned decades ago in the context of the statutory minimum wage: only by providing both a contractual right to be paid a minimum sum, and imposing effective (criminal) penalties on employers who failed to pay such sums, have mandatory rights to minimum remuneration been even remotely effective.117 Even then, where individuals rely on intermittent work for multiple firms, minimum remuneration rights do little to guarantee a stable, and reliable, income for those individuals who do rely on their ‘work’ for subsistence. Before the final section considers the potential of recently introduced EU rules on mandatory protections for creators and users in this context, the following section discusses another important factor that further reduces chances for creators to financially benefit from their works on platforms. CONTENT PROTECTION ON PLATFORMS AND ITS IMPACT ON CREATORS While the previous section has critically reviewed the potential for copyright to empower creators and to secure them an income from their works, we now turn to the use of automated copyright (or rather: content) protection systems on platforms as a tool to ‘demonetize’ creators—that is, to actively prevent them from being remunerated for their works. As platforms like Facebook, YouTube, Instagram or Twitter depend on users ‘freely’ sharing content in exchange for the data that users generate in the process, they have an inherent interest in maximizing the creation, communication and sharing of user-generated content. However, the content that users create and share is often not solely comprised of material independently created by the user. Instead, it is frequently based on, or at least involves elements of, existing, often copyright-protected content. User creativity—for example in the form of memes, mash-ups, play-alongs or supercuts—tends to be expressed through re-purposing, transforming or otherwise modifying or adding to existing content: a form of ‘remix culture’ that somewhat resembles sampling in 1990s Hip Hop. A significant proportion of that remixed content potentially infringes copyright under applicable domestic laws—although much will depend on each legal system’s respective quantitative or qualitative thresholds for infringement and on any defenses or exceptions, such as fair use concepts or specific exceptions for the purpose of parody, pastiche, satire, quotation, news reporting, education, research and so on.118 Hence, platforms that allow such content to be shared run the risk of being held (contributorily or accessorily) liable for the infringing acts of their users—unless of course they can rely on so-called ‘safe harbors’ that exempt liability for online intermediaries under certain conditions. While details differ and not all major jurisdictions offer specific safe harbors, these safeguards from liability converge around a ‘notice and take-down’ model whereby intermediaries have to act expeditiously when they become (or should be) aware of infringing acts of their users (often via a notification).119 In order to avoid liability, platforms have taken steps to respond to infringement notifications, primarily by removing or blocking access to allegedly infringing content.120 Increasingly, however, platforms also rely on filtering systems and other automated content protection tools that check, usually in real-time, any content uploaded or otherwise shared by users against reference files in huge databases for content submitted by those claiming rights in that content.121 As the specific mechanisms and technical tools used by platforms such as Google, YouTube, Facebook or Instagram differ, we focus on one of the most sophisticated and arguably most profitable filtering system: YouTube’s ‘Content ID’.122 Once Content ID’s automated content recognition tool detects a (partial) match with a reference file, the entity (or entities) claiming rights over that content is notified. That entity can then decide (case-by-case or in form of setting general rules) over the further fate of the user content uploaded or shared. It can of course be blocked – but, given the popularity of user-generated content which often gets thousands (and occasionally millions)123 of views, content owners may find it much more economically interesting to monetize user content, primarily by taking the user’s share of the revenue generated by running ads against it. In other words, they can profit from user uploads that incorporate anything matching with content they claim. This is not only the default and most relied on option on YouTube’s Content ID,124 but equally available on Facebook and Instagram; while TikTok does not offer automated copyright enforcement tools to right holders and instead tries to ensure all of its user-generated content is fully licensed.125 Employing automated enforcement tools that allow to monetize user-generated content therefore creates quite a lucrative market for platforms, as they usually will be able to claim a (significant) share in the advertising revenue (or other modes of monetization) generated by sharing the user’s content. In addition, large-scale availability of user content attracts more users and keeps existing users on the platform—generating even more revenues, including via shares in monetization activities by creators and those claiming rights in user content. The principal effect of automated enforcement tools such as Content ID for creators then is not an option to secure earnings for their works shared on the platform, but an ‘exclusion’ from any revenue otherwise generated in relation to works claimed by Content ID. That is because, first, most individual creators will not be eligible to use Content ID as YouTube indicates that the tool is only for those who ‘own exclusive rights to a substantial body of original material that is frequently uploaded by the YouTube user community’.126 More importantly, any upload claimed as allegedly infringing by Content ID can only be monetized by the entity making the Content ID claim, preventing creators faced with such a claim from earning any shares in the revenues generated from ads run against their works on YouTube.127 For several reasons, this basic feature of Content ID significantly reduces the already poor options for creators to earn an income from their works shared on YouTube. First, automated systems in general and Content ID in particular are unable to identify whether the content claimed via a reference file is actually copyright protected and if so, who owns rights to it.128 Secondly, even sophisticated matching algorithms as those employed in Content ID cannot, from comparing the reference file with the user upload, determine whether the content used in an upload actually constitutes copyright infringement under the applicable copyright law, including the important question of whether that upload can benefit from any exceptions or limitations to copyright protection (such as fair use, fair dealing, or specific exceptions for quotation, parody, pastiche or illustration for educational purposes).129 Thirdly, YouTube in principle does not monitor or otherwise check whether the Content ID claimant actually holds any rights and/or whether the claimed match constitutes an infringement.130 Anecdotal evidence of all kinds of false positives in claiming content not copyright-protected, not owned by the entity claiming it, or clearly within the ambit of copyright exceptions (in particular fair use) shows that leaving it to claimants to ‘self-regulate’ and simply warn them not to over-claim has not worked well in the past.131 This is also because, fourthly, the dispute settlement system YouTube offers to challenge a Content ID claim (i) significantly disincentivizes especially monetizing creators to file a dispute (and therefore is hardly ever used);132 (ii) involves no independent third-party review;133 and (iii) requires disputing users, regardless of their domicile, to eventually submit to the jurisdiction of US Courts.134 Overall then, automated content protection on platforms such as YouTube allows the platform and those major copyright owners that enjoy access to its sophisticated matching tools to monetize user content, thereby potentially enclosing creators from opportunities to earn income from their works. In other words, the remix culture of user-generated content is turned into a profit enterprise for platforms and publishers135 that can jointly exploit the user’s ‘digital labour’136 involved in creating and sharing content. As the discussion above suggests, exploiting user’s content through monetization has no legal basis if the use of content falls under a copyright exception, or is otherwise non-infringing. But even if user-generated content does involve copyright-infringing content (eg a one- or two-minute extract of a recorded song that serves as background in a 15-minute video), it is not immediately clear that the user content as a whole (the whole 15-minute video that otherwise does not infringe) should be monetized exclusively by the platform and the copyright owner.137 In sum, creators not only have their rights collected by platforms via unfavourable terms of service, but, as this section has shown, their remaining opportunities for monetization can be prevented by publishers and platforms who rely on automated enforcement tools which—while unavailable for most individual creators—allow for unchecked claims on creator/user-generated content, and make challenges to such claims extremely difficult. NEW RULES FOR THE ‘DIGITAL MARKET’ AS A WAY FORWARD? In this article, we have considered how creative work is treated and conceptualized in labour and copyright law, placing this analysis against the broader context of socio-economic power relations within which individual creators produce their work(s). With a focus on the exploitation of these works by digital platforms, we have shown that neither labour nor copyright law is currently able to adequately deal with creative work(s) in a way that would meaningfully improve the situation of individual creators—which would regulate the struggle over surplus and subsistence that is intrinsic in even creative forms of capitalist work relation. This concluding section offers some thoughts about a possible way forward, in light of recent EU legislation which addresses both the use of copyrighted content on digital platforms and the protection of creators in their contractual relations with publishers and others exploiting their work. As the implementation period for this EU Directive is, at the time of writing, still ongoing, we refrain from a detailed, technical analysis of the relevant provisions—but rather consider whether they, broadly speaking, can be employed as tools to address the problems identified in this article.138 Article 17 of the EU Directive on ‘Copyright in the Digital Single Market’139 aims to correct the so-called ‘value gap’ that some content producers (mainly publishers and other copyright owners from EU Member States) perceive to exist between the valuable (copyrighted) content they publish online and its subsequent ‘appropriation’ by online platforms that allow their users to share such content and benefit from such sharing via advertisement revenues, without those revenues being ‘fairly’ shared with copyright owners. The basic solution adopted in Article 17 is to expand the exclusive right of copyright owners to authorize (and license) the communication of their works to the public so as to include the activities of ‘online content-sharing service providers’ (read: YouTube and other platforms whose principal purpose is to ‘enable users to upload and share large amounts of copyright protected content’ and in that way compete with legitimate, licensed sources).140 Based on that inclusion, platforms covered by Article 17 are under an obligation to enter into a copyright licensing agreement for their services, which in turn is meant to cover also the acts of platform users.141 If platforms fail to negotiate or otherwise obtain a license from all copyright owners whose content is available on the platform, the platform will be liable for copyright infringement (and unable to rely on the ‘safe harbours’ mentioned in the previous section)142 unless it takes a range of measures, including (i) best efforts to obtain a license; (ii) best technological efforts to ensure notified unlicensed works cannot be accessed or shared through its service (arguably by means of upload filters); and (iii) act fast to prevent further (including future) access to such works.143 Since creators who opt to monetize their work on platforms like YouTube will already have entered into a license agreement as part of the terms of service,144 however, Article 17 will not improve their legal position. For creators whose works are used on platforms ‘outside’ any licensing agreements they have with those platforms, Article 17 would provide them with a remedy—but one which primarily allows them to ask for access to their works to be blocked under Article 17(4)(c) Digital Single Market (DSM).145 If platforms cannot show they have done enough in terms of best technological efforts to prevent unauthorized uploads of their works, creators should (in theory) be able to claim damages as well—but unless creators are represented via a collection society (which will be much better placed to negotiate deals with platforms), they would have to fight their case against a platform like YouTube in court (and hence will hardly ever pursue this). For creators that incorporate pre-existing content, such as a copyright-protected video or audio recording, into their works (as very common in user-generated content),146 Article 17 may also improve, in theory, their ability to rely on copyright exceptions that allow, under specific requirements, such incorporation. Article 17(7) requires EU Member States to ensure that non-infringing user content remains ‘available’ on the platform. In particular, users must ‘be able to rely on’ the quotation, criticism and review, as well as the caricature, parody and pastiche exception.147 While a lot will depend on how exactly EU Member States implement Article17(7), it seems to focus on preventing that non-infringing content is being ‘blocked or removed’. In light of our analysis in the previous section, that alone is not enough. As the Content ID example shows, nowadays blocking or removing content is the rare exception and monetization of user generated content the rule. EU members should consider this when implementing the DSM. Since the ability to ‘rely on’ the listed copyright exceptions arguably implies an option to invoke these exceptions also against attempts to monetize user content that amounts to a parody, quotation, criticism or review, one would hope that Article 17(7) will be implemented to cover monetization. If EU Member States take this approach, creators that incorporate existing copyrighted material in their works in a non-infringing way might be able to rely on domestic rules implementing Article 17(7) to act against Content ID and equivalent automated enforcement tools which allow, simply on the basis of a claim, to appropriate all monetization of the work and in that way exclude creators from earning income. It however remains to be seen whether domestic implementation of this provision will be in the form of an actual remedy afforded to creators.148 Finally, Article 17(9) of the DSM Directive might improve the situation of creators as part of the users of platform services because it requires platforms covered by Article 17 to put in place ‘effective and expeditious complaint and redress mechanisms’. Our analysis in the previous section has shown that in light of the flaws and abuse potential exemplified by Content ID and other automated enforcement tools, we should consider how to design more effective and fair remedies against claims about user uploads matching copyrighted content. Users must be able to contest these matters in meaningful dispute settlement proceedings offered by the platform which (i) must involve an independent and impartial review,for example by an ombudsperson for platform users; and (ii) allow users to challenge claims—importantly also those involving monetization—in front of courts with jurisdiction over the protection claimed by a copyright owner. In case a user contests a claim, the onus should be on the claimant to show a prima facie case of copyright or neighbouring right infringement—away from the current default where the ‘process accepts the rightsholder’s claim of infringement as correct until proven otherwise’.149 If automated enforcement allows allegedly infringing use to be stopped just on the basis of a claim, the decision to automatically enforce must be taken with care and be reasoned and comprehensible to users—which for the time being might well mean that there needs to be a ‘human in the loop’. Article 17(9) addresses some of these demands,150 but again focuses on blocking and removing content—hence to a large extent disregarding exploitation of creators’ works through monetization.151 Finally, akin to a duty to indemnify for damages suffered because of a request for preliminary (injunctive) relief or related court orders, one might even consider strict liability of those who decide to automatically enforce, should a court or other independent third party later find that no infringement existed. Overall, the idea is to adopt measures that account for, and re-balance, the fundamental shift which is inherent in algorithmic enforcement where our ability to experience content online is pre-determined by what automation allows us to see and hear. In addition to Article 17 specifically addressing the use of copyrighted works on platforms, Articles 18–22 of the DSM Directive set out basic principles of protecting creators as weaker parties152 in their contractual position vis-à-vis those exploiting their works. Again, we refrain from a technical–legal analysis of these provisions in detail as much will depend on how they are implemented. In light of our analysis in Sections ‘Labour Law In The Creative Industries: A Conceptual and Practical Challenge’ and ‘Copyright Protection for Creators on Digital Platforms’, it is unlikely that mandatory rules in favour of creators as such are sufficient—in particular in the context of works exploited on platforms. We therefore focus on the limits and potential of Articles 18–22, which—perhaps together with our suggestions on a meaningful dispute settlement system under Article 17(9), and wider reforms to labour law—could improve the situation of creators who try to earn income from their work(s). Article 18 sets out the ‘principle of appropriate and proportionate remuneration’ whereby Member States must ensure that authors and performers receive appropriate and proportionate remuneration when they license or transfer their exclusive rights for the exploitation of their works. Apart from suggesting a few factors153 that EU Member States should take into account when putting this principle into practice, the Directive leaves Members much freedom in implementing this provision. It should be noted, however, that Article 18 does not provide a right to adequate or reasonable (or fair) remuneration. Rather, it provides for remuneration that is ‘appropriate’ and ‘proportionate’, where, according to recital 78, this is to be assessed in light of the revenues generated from the exploitation of the work in question, and the degree of contribution by the creator to the work. The problem with this is that, not only is an individual’s specific contribution to a particular work extremely difficult to assess, but by linking remuneration to the commercial success of the work, the effect is to expose dependent creators to the same risk of commercial failure as the publisher or platform. This is further exacerbated by the fact that, according to recital 78, the assessment of what counts as appropriate and proportionate is to be linked with existing practices in specific sectors, and so, is essentially constrained by ‘market practice’. A more favourable formulation of this article would have granted creators a right to receive ‘fair and reasonable remuneration’ for their creative works. Such a formulation has historically proven significant in ‘fair wages’ mechanisms in various countries,154 because the language of what is ‘fair and reasonable’ has long been deemed to separate the right to payment both from the profit of the enterprise and the state of the market (existing practice), linking it instead to what is fair and reasonable from the perspective of the individual’s needs. Historically, such provisions have been rooted in collective bargaining mechanisms, given the importance of collective negotiations when it comes to tailoring remuneration to the needs of particular workers and the demands of different types of work. While Article 18 does provide that the right to adequate and proportionate remuneration can be implemented via ‘collective bargaining and other mechanisms’, those must be compatible with the structures already in place in certain member states (whether these be mechanisms obtained through collective agreements or based on mandatory collective management). As we have seen, in the EU at least, only those individuals recognized in labour law as employees (or workers) enjoy a right to collectively bargain, and only collective agreements agreed between firms and unions of workers and employees are exempt from EU (and very often domestic) competition law. As a result, in many countries, it is unlikely that collective bargaining mechanisms in relation to content creators and platforms will actually be ‘compatible with the structures already in place’. Historically, collection societies have played an important role in the creative industries when it comes to representing the financial interests of creators and protecting them against unauthorized uses of work. However, the rights administered by these societies often primarily concern acts of secondary, further exploitation—such as playing copyrighted songs on the radio or copying literary works in libraries and in educational institutions—while creators are commonly required to transfer principal exploitation rights (for controlling reproduction, publication, distribution and making available to the public) in their works to publishers. In addition, these societies are, like unions of self-employed workers, recognized as undertakings under EU law, and so are also potentially subject to competition rules—rules that do not always operate in the interests of the societies’ members.155 While collection societies traditionally operated as national monopolies—a strategy that was integral to their role in protecting creators’ rights, such a situation was soon recognized as coming into conflict with the EU’s goal of creating a single market. The EU’s response to this has thus been to encourage competition between collection societies: a strategy that often conflicts with the societies’ goals of strengthening the bargaining power of their members.156 More promising appears the idea of facilitating, via collection societies, the multi-territorial licensing of musical works for online uses under Articles 23–32 of the EU’s Collective Rights Management Directive157—which however will only be able to benefit creators as long as they have not already transferred all relevant rights to platforms or music publishers. Given, moreover, that Article 18(2) of the DSM Directive expressly allows Member States to take into consideration the principle of contractual freedom and a ‘fair balance of rights and interests’, there is a real risk that this ‘right’ will be left to the vagaries of contract. To some extent, the limitations of Article 18 might be mitigated by Article 20 which introduces a ‘contract adjustment mechanism’ that applies when remuneration for the creator is disproportionately low compared with the revenues derived from the exploitation of a work. Interestingly, this article provides that authors or performers are entitled to claim additional, appropriate and fair remuneration, when the remuneration turns out to be disproportionately low compared to all the subsequent relevant revenues derived from the exploitation of the work or performances. How broadly Member States interpret this Article is yet to be seen, but the reference to what is ‘fair’ may help to encourage an interpretation that is favourable to the creator. For example, one might envisage a use for this article where the creator has agreed to low remuneration in proportion to the success of the work (such as 2 per cent of total revenues) but new forms of exploitation become possible at some later date such that it can be argued that this royalty would never have been acceptable to a creator had they known about this in advance.158 In many ways, however, the efficacy of both Articles 18 and 20 depends on Article 19, which establishes certain transparency obligations, requiring that authors and performers must receive, on a regular basis (at least once a year), ‘up to date, relevant and comprehensive information on the exploitation of their works and performances’. This information must be provided not only by the parties to whom they have licensed or transferred their rights, but also their successors in title, and covers in particular the ‘modes of exploitation, all revenues generated and remuneration due’. Article 19 works in conjunction with Articles 18 and 20 insofar as it is only if the creator has up-to-date and complete information about exploitation that they will be in a position to assess whether remuneration is ‘adequate or proportionate’ within the meaning of Articles 18 and 20. The problem, however, is that if Article 19 is to be effective, the creator must be able to obtain information about exploitation not only by the licensee, but also sub-licensees down the entire chain of exploitation. And yet, by virtue of Article 19(2), creators do not have an automatic right to transparency but need to make a request to the sub-licensee to be given any information the first licensee does not already hold. Not only this, but member states are free to decide whether the creator can contact the sub-licensee directly, or must do so through the licensee. Because the creator will have no direct contractual relation with subsequent exploiters down the chain, therefore, there is a risk that creators will struggle to obtain the information they need to meaningfully assess the adequacy of their remuneration within the meaning of the Directive. Finally, while Article 19(5) provides that the transparency rules can be negotiated collectively, it is difficult to see how much benefit will be gained from collectively negotiated transparency rules if they cannot then be used to inform collective negotiations about remuneration (see above). At the end of the day, while the Directive certainly provides a framework on which to build, at present, it simply gives too much weight to contractual bargaining and is too willing to defer to ‘market practice’. In effect, the Directive, as with many aspects of labour law and copyright law more generally, remains blind to the structural inequalities that underpin relations between creators and publishers/platforms, and thus, to the origins of the imbalance of power that the Directive purports to address. At a more practical level, the Directive, while concentrating primarily on copyright issues, is nonetheless premised on the existence of effective labour law and industrial relations frameworks which, as we have seen, are likely to be of little assistance to creators because they systematically privilege a particular form of employment relation over others. As long as EU and domestic competition law remains hostile to collective bargaining attempts among ‘self-employed’ persons, while also subjecting collecting societies to the strictures of competition law159 and domestic labour law regimes restricted to those in traditional employment relationships, it will be difficult for creators to take advantage of even the more promising aspects of the Directive. In this respect, extending national frameworks for collective bargaining to ‘all’ those in a ‘capitalist work relation’—and thus, to all those dependent on capital (publishers/platforms) for subsistence—will be a vital preliminary step when it comes to realizing the potential that this particular, and further future, copyright reform has to offer—when it comes to securing not merely a proportionate, but a ‘fair’ remuneration for creative workers and creative work.160 Copyright law and labour law must thus work together if we are to establish the sort of legal framework on which a fairer, more equitable, ‘market’ for creative and digital work, and works, can be built. Otherwise, the situation as bluntly captured in the introductory quote—that ‘every time there’s a new way of doing music, the creators always get screwed’—is likely to remain a reality, including on digital platforms. Footnotes 1 Brett Gurewitz, Band Member of Bad Religion, in an interview about new opportunities for artists on digital platforms, 2019. 2 There are many more forms of creative work necessary to capitalism, such as the creative work spent on designing innovative new products (which as inventions may be subject to patents); on adapting and improving existing ones (which in some jurisdictions can be protected as utility models or petty patents); and the creative work applied in the branding and advertising strategies of firms (which may be subject to trade mark, passing off or unfair competition protection). In this article, we focus only on the creative work that produces ‘creative’ products, or works—the basis for products and services in the media, information and other cultural industries, and frequently subject to copyright protection. 3 Mark Banks, The Politics of Cultural Work (Springer 2007). 4 We use this term to refer to integrated, often global, networks of devices that enable communicating information in electronic formats. Such networks are essential for sharing media content in digital form, which in turn often involves creative work output. 5 Giuseppina D’Agostino, Copyright, Contracts, Creators: New Media, New Rules (Edward Elgar 2010). 6 For a detailed analysis of the distinctive features of creative work that tend against employment, see Nicole S Cohen, ‘Cultural Work as a Site of Struggle: Freelancers and Exploitation’ (2012) 10 tripleC: Communication, Capitalism & Critique. Open Access Journal for a Global Sustainable Information Society 141. 7 This observation sets the analysis in this article apart from the analysis of Fuchs, for example, who conflates these two forms of ‘work’ as all examples of capitalist exploitation, or ‘labour’. See Christian Fuchs and Sebastian Sevignani, ‘What Is Digital Labour? What Is Digital Work? What’s Their Difference? And Why Do These Questions Matter for Understanding Social Media?’ (2013) 11 tripleC: Communication, Capitalism & Critique. Open Access Journal for a Global Sustainable Information Society 237. 8 For an integrated approach, see Michael D Birnhack, ‘Who Owns Bratz: The Integration of Copyright and Employment Law’ (2009) 20 Fordham Intellectual Property, Media & Entertainment Law Journal 95. 9 See, for example, the extensive literature that evolves around the WIPO ‘Internet’ treaties (including the WIPO Copyright Treaty (WCT), Geneva, 20 December 1996, 2186 UNTS 121) and the ‘Copyright in the Information Society Directive’ (EU Directive on the harmonisation of certain aspects of copyright and related rights in the information society (2001/29/EC) OJ L167, 22 June 2001) as well as its subsequent application by courts, including on hyperlinking, (see Christina Angelopoulos, ‘Hyperlinks and Copyright Infringement’ (2017) 71 CLJ, 32) and intermediary liability (see Jani Riordan, The Liability of Internet Intermediaries (OUP 2016)). 10 On various aspects of user-generated content, see, for example, Henning Grosse Ruse-Khan, Nadine Klass and Silke von Lewinski, Nutzergenerierte Inhalte als Gegenstand des Privatrechts - Aktuelle Probleme des Web 2.0 (Springer 2010) and Daniel Gervais, ‘The Tangled Web of UGC: Making Copyright Sense of User-Generated Content’ (2009) 11 Vanderbilt Journal of Entertainment and Technology Law 841. On freedom of expression and other ‘user rights’ in copyright, see the Court of Justice of the European Union (CJEU) judgment in Case C-476/17 Moses Pelham GmbH v Ralf Hütter (2019), ECLI:EU:C:2019:624; as well as the Canadian Supreme Court judgment in CCH Canadian Ltd. et al v Law Society of Upper Canada [2004] 1 SCR 339; and generally Pascale Chapdelaine, Copyright User Rights (OUP 2017) as well as Graham Greenleaf and David Lindsay, Public Rights: Copyright’s Public Domains (CUP 2018). 11 See our discussion in Section ‘Content Protection on Platforms and Its Impact on Creators’. 12 On the nature of ‘social form’ analysis, see Patrick Murray, The Mismeasure of Wealth: Essays on Marx and Social Form (Brill 2016) ch 1. 13 Ursula Huws uses a similar idea to distinguish labour ‘inside the knot’: Ursula Huws, Labor in the Global Digital Economy: The Cybertariat Comes of Age (Monthly Review Press 2015). Sheila Cohen conceptualizes these struggles as the essence of capitalistic exploitation: Sheila Cohen, ‘The Labour Process and Class Consciousness’ <http://oro.open.ac.uk/18868/1/pdf158.pdf> accessed 6 January 2020; Nicole S Cohen, ‘Cultural Work as a Site of Struggle: Freelancers and Exploitation’ (2012) 10 tripleC: Communication, Capitalism & Critique. Open Access Journal for a Global Sustainable Information Society 141 (n 6). 14 This conception of capitalist work draws on Marx’s labour theory of value. This theory operates at the aggregate or societal level, and so is not directly concerned with the relationship between an individual worker and his employer, and the amount of time that worker works relative to the amount he receives as wages. However, even at the individual level we can speak of a struggle over surplus and subsistence in the sense that capital seeks to maximize how much time each worker works relative to how much it pays in wages, for such is integral to the question of profit (conceived as the relationship between revenue and costs). 15 By ‘capital’ we mean an individual, enterprise or other organization (including those controlled by the state), in possession of the means of production, and so, in a position to appropriate the surplus labour of others. 16 By distinguishing ‘capitalist work’ from the narrower category of productive labour (labour that produces surplus value), we re-iterate Marx’s point that, in capitalism, the distinctions between different types of workers is obscured, as all workers are placed into competition with each other in the market. 17 The terms capital and labour are used in this article to refer to the class positions people assume when they participate in various practices. 18 Cohen (n 13). See, more generally: Harry Braverman, Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century (25th anniversary edn, Monthly Review Press 1998). 19 For an analysis of this point and its implications for value production, see: Zoe Adams, ‘Understanding the Minimum Wage: Political Economy and Legal Form’ (2019) 78 The Cambridge Law Journal 42. 20 Key examples include the possibility of securing opt-outs to working time regulations; to the right not to be subject to deductions from wages; and the general freedom the parties’ enjoy to modify, or exclude, many of the terms implied into contract of employment with a view to better balancing the parties’ interests. 21 Hugh Collins, ‘Independent Contractors and the Challenge of Vertical Disintegration to Employment Protection Laws’ (1990) 10 Oxford Journal of Legal Studies 353. 22 Despite subtle differences, the essential markers of an employment relationship are broadly similar throughout different jurisdictions. For an historical and comparative overview, see Bruno Veneziani, ‘The Employment Relationship’ in Bob Hepple and Bruno Veneziani (eds), The Transformation of Labour Law in Europe: A Comparative Study of 15 Countries 1945-2004 (Hart Publishing 2009) and Bernd Waas and Guus Heerma van Voss, Restatement of Labour Law in Europe (Bloomsbury Publishing 2020) vol 2. 23 See in particular the literature on labour process theory: Braverman (n 18). 24 On the nature and significance of such shifts, see Tony Smith, ‘Flexible Production and the Capital/Wage Labour Relation in Manufacturing’ (1994) 18 Capital & Class 39. 25 For an analysis of these trends, see Judy Fudge, ‘Beyond Vulnerable Workers: Towards a New Standard Employment Relationship Special Section on Administering Labour Law - Papers from the Annual UWO Labour Law Conference: Part 1’ [2005] Canadian Labour & Employment Law Journal 151; Judy Fudge, ‘The Future of the Standard Employment Relationship: Labour Law, New Institutional Economics and Old Power Resource Theory’ (2017) 59 Journal of Industrial Relations 374. 26 In some jurisdictions, such as the UK and Italy, an intermediate status between self-employed and employee has been created to extend labour protections to those deemed to be in a similar economic position to employees. However, all such approaches stop short of extending employment protections to the ‘genuinely self-employed’. For an analysis of these relations of ‘para-subordination’, see Ulrike Muehlberger and Silvia Pasqua, ‘Workers on the Border Between Employment and Self-Employment’ 24. 27 Collins (n 21). 28 See in the Italian context, art 409(3) of the Civil Procedure Code (introduced in 1973), which refers to the the para-subordinate worker, or collaborazioni coordinate e continuative; and in the German context, the quasi-salaried person, or ‘arbeitnehmerähnliche Personen’ a definition of which can be found in s 12a of the Act on Collective Bargaining Agreements (Tarifvertragsgesetz). These are effectively economically dependent persons in need of social protection, but not necessarily personally subordinated. These individuals enjoy some basic employment rights. Similarly, while France does not have an intermediate category, some persons are considered to be employees on the grounds of economic dependence, but are not employees per se. 29 This refers to the legislative definition contained in s 230(3)(b) ERA 1996. 30 For an illustrative example that shows the impact (and unintended consequences) of changes to algorithms utilized to suggest further content to users—usually to maximize the amount of time users spent on a platform, see Kevin Roose, ‘The Making of a YouTube Radical’ The New York Times (8 June 2019), New York <https://www.nytimes.com/interactive/2019/06/08/technology/youtube-radical.html>. On the impact of changes to YouTube’s algorithm for creators, see generally Emma Grey Ellis, ‘YouTubers Must Unionize, No Matter What Google Says’ (Wired, 24 October 2019) <https://www.wired.com/story/youtube-union/> accessed 19 February 2020. 31 This issue was discussed extensively by the UK Government’s independent review of working practices (Matthew Taylor, ‘Good Work: The Taylor Review of Modern Working Practices - GOV.UK’ <https://www.gov.uk/government/publications/good-work-the-taylor-review-of-modern-working-practices> accessed 18 March 2020), and was also discussed at length in Aslam v Uber BV [2018] EWCA Civ 2748. One of the key challenges being that if the answer to this question is yes, a worker could theoretically be working for more than one ride-hailing firm (Uber, Lift etc) simultaneously. 32 The term productive is used in the general sense of something that produces a useful effect, rather than in the Marxian sense of something that produces surplus value. . 33 ‘Work’ as the subject matter of copyright is part of what nowadays is commonly referred to as ‘intellectual property’—a range of intangible assets emanating from the human mind for which there is no agreed abstract definition, but what the World Intellectual Property Organization (WIPO 2017) describes loosely as ‘creations of the mind’—see WIPO, ‘What is Intellectual Property?’ <https://www.wipo.int/about-ip/en/> accessed 13 August 2020. 34 As Reveley notes, ‘social activities’ can have an indirect effect on the value of labour power, and in this way, indirectly effect (and in the long term) the struggle over surplus and subsistence, but this is an indirect feedback loop effect: James Reveley, ‘The Exploitative Web: Misuses of Marx in Critical Social Media Studies’ (2013) 77 Science & Society 512. 35 For an excellent discussion of IP and culture, see: Rosemary J Coombe, The Cultural Life of Intellectual Properties: Authorship, Appropriation, and the Law (Duke UP 1998) <https://read.dukeupress.edu/books/book/2240/the-cultural-life-of-intellectual> accessed 16 October 2019. 36 Tony Smith, ‘The “General Intellect” in the Grundrisse and Beyond’ (2013) 21 Historical Materialism 235. See also the notion of ‘appropriation’ in Nancy Fraser’s work: Nancy Fraser, ‘Expropriation and Exploitation in Racialized Capitalism: A Reply to Michael Dawson’ (2016) 3 Critical Historical Studies 163. 37 Apart from the creative works discussed so far, such input would include, on the one hand, ideas, discoveries, myths and legends, traditional knowledge and other elements that form, under current laws, part of the knowledge commons and can be freely incorporated into IP-protected subject matter. On the other hand, it would also include creations of the mind that intellectual property law recognizes as protected subject matter (inventions, industrial designs or signs indicating commercial origin). 38 Nancy Fraser, ‘Behind Marx’s Hidden Abode for an Expanded Conception of Capitalism’ [2014] New Left Review 55. 39 In specific sense of working beyond the point required to reproduce their labour power. On mistaken uses of the term exploitation in this context, see Reveley (n 34). 40 While in Austria, for example, a distinction is still maintained between the rights of blue- and white-collar workers, and special rules apply to special groups of white-collar workers (journalists, actors, employees in agricultural enterprises and pharmaceutical employees), the relevance of these special acts is diminishing and they often only cover some aspects of the employment relationship, such as wage determination or working time. See Restatement of Labour Law in Europe : The Concept of Employee (Hart Publishing 2017) ch 1. 41 See Dan Ariely and others, ‘Large Stakes and Big Mistakes’ (2009) 76 Review of Economic Studies 451 and Daniel H Pink, Drive: The Surprising Truth About What Motivates Us (Riverhead Books 2009). 42 Court of Justice of the European Union, Case C-145/10 Eva-Maria Painer v Standard Verlags GmbH et al (2011) ECLI:EU:C:2011:798, para 92. 43 Bill Ryan, Making Capital from Culture, The Corporate Form of Capitalist Cultural Production (Reprint 2010, De Gruyter 2010). 44 In the EU, this ability is seen as an implication of the general requirement in copyright law for works to be ‘original’ in the sense of involving an ‘own intellectual creation’; see Eva-Maria Painer v Standard Verlags GmbH et al (n 42) para 89. While other jurisdictions employ different understandings of originality (such as the traditional approach in Common Law countries to demand a sufficient degree of ‘skill, judgement or labour’), there has been a degree of convergence towards originality conceived as ‘intellectual creation’ (see art 10 of the WTO Agreement on Trade Related Aspects of Intellectual Property, TRIPS) or involving a ‘spark of creativity’ (see Feist Publications, Inc., v Rural Telephone Service Co. (1991) 499 US 340). 45 Cohen (n 6); D’Agostino (n 5). 46 See for example the data policies of Facebook (https://www.facebook.com/about/privacy/update) and Instagram <https://help.instagram.com/519522125107875/?helpref=hc_fnav&bc[0]=Instagram%20Help&bc[1]=Privacy%20and%20Safety%20Center>. 47 YouTube is perhaps best known for allowing creators to monetize their videos by taking a cut (around 50%) of the advertisement revenue generated from running adds with their videos—as long as they meet basic eligibility requirements in terms of channel popularity (see <https://support.google.com/youtube/answer/72857?hl=en-GB&ref_topic=9257989>) and if all content used in the video has been produced by the creator, is in the public domain or licensed for commercial use on YouTube (see <https://support.google.com/youtube/answer/2490020?hl=en-GB>). On Instagram, users with a significant following (of around 10,000 followers upwards) can monetize their online presence for example by product placements, or directly advertising or selling branded products (see <https://www.uscreen.tv/blog/how-to-monetize-instagram/>). A prominent exception to user monetisation along the lines mentioned above is the platform TikTok, an increasingly popular App that allows users to upload short videos (usually between 15 and 60 seconds) which they can produce, using a range of tools offered by the App, including music, films and sound effects users can select from large databases of licensed content. In 2019, TikTok had been installed more than 738 million times, and is considered the second most downloaded App—right after Whatsapp (see <https://techcrunch.com/2020/01/16/44-of-tiktoks-all-time-downloads-were-in-2019-but-app-hasnt-figured-out-monetization/>). This popularity allows TikTok users to reach an ever-increasing audience where users with a large following (‘influencers’) can earn significant amounts via tips received during live streams or, more importantly, sponsor deals where they promote products to their fan base (see <https://www.distractify.com/p/can-you-make-money-on-tik-tok>). Further, curating a fan base on TikTok allows to divert audience to channels that can be monetized—such as YouTube. The approach of earning money through fan base has also led to the development of TikTok money calculators (see <https://influencermarketinghub.com/tiktok-money-calculator/>). These are not based on TikTok’s own algorithm, but aim to estimate the earning potential of influencers on TikTok, based on, for example, the extent of engagement an influencer has with her fan base. 48 The same is true with regard to agreements providing collecting societies with a monopoly on collecting royalties for particular types of creators is contrary to Article 102 TFEU. While the CJEU has held that such monopolies are lawful under art 102 TFEU, it went on to argue that such societies might abuse their dominant position, thereby contravening Article 102. Previous decisions to challenge the way collection societies operate under Article 101 have failed. See: ‘Copyright Owners, Collecting Societies and the Competition Conundrum’ (Managing Intellectual Property, 31 March 2014) <https://www.managingip.com/article/b1kc25rmj22kw0/copyright-owners-collecting-societies-and-the-competition-conundrum> accessed 18 March 2020. Crucially, because collecting societies are deemed to be undertakings for competition law purposes, they potentially face competition law barriers whenever negotiating collective rights for their members. 49 The CJEU has argued that ‘[O]n a proper construction of EU law, it is only when self-employed service providers who are members of one of the contracting employees’ organisations and perform for an employer, under a works or service contract, the same activity as that employer’s employed workers, are “false self-employed”, in other words, service providers in a situation comparable to that of those workers, that a provision of a collective labour agreement, such as that at issue in the main proceedings, which sets minimum fees for those self-employed service providers, does not fall within the scope of Article 101(1) TFEU. It is for the national court to ascertain whether that is so’. FNV Kunsten Informatie en Media v Staat der Nederlanden C-413/13 EU:C:2014:2411 [42]. 50 See also the decision of the European Committee on Social Rights in ECSR, Irish Congress of Trade Unions v Ireland, Complaint No 123/2016, Merits, 12 December 2018 concerning a pre-2017 ban in Irish law on collective bargaining by freelance journalists, voice-over actors and session musicians, which the committee suggested was in violation of the European Social Charter’s protection for freedom of association. However, the reasoning was linked in part to the fact that the individuals were ‘truly self-employed’ because they weren’t dependent on multiple firms, and the fact that the ban was absolute. 51 Ellis (n 30). 52 Organization among self-employed individuals is much easier than is the organization of collective bargaining. However, in many countries, there exist real obstacles to organization, a position that is only exacerbated by competition authorities. For a detailed study of the position of self-employed persons under collective labour law, see L Fulton, Trade Unions Protecting Self-Employed Workers (ETUC 2018). 53 In the UK, the Independent Workers Union of Great Britain (IWGB) has enjoyed some success organizing in the gig-economy, but most of those in the Union have been classed in law as ‘workers’ and thus, not as ‘genuinely’ self-employed. Moreover, even the IWGB has had to confine itself primarily to protest, getting recognition for the purposes of collective bargaining has proven difficult. See: Case TUR1/985 Independent Workers’ Union of Great Britain (IWGB) v RooFoods Limited T/A Deliveroo (2016) (no right to invoke the statutory recognition procedure because Deliveroo cycle couriers did not meet the legal definition of ‘worker’). 54 See generally Michael Birnhack, ‘Working Authors’ (October 2008) Tel Aviv University Law Faculty Papers, Working Paper 97, 15–23 <http://law.bepress.com/taulwps/art97>, as well as the contributions on many common and civil law countries in Sanna Wolk and Kacper Szkalej, Employees’ Intellectual Property Rights, AIPPI Law Series (Wolters Kluwer 2018). 55 In the USA, the ‘work made for hire doctrine’, as set out in s 101 of the Copyright Act (title 17 of the US Code), allocates first ownership not only to the employer in case the work is prepared by an employee within the scope of her or his employment, but also if the work has been ordered or commissioned for use in a range of situations, such as part of a motion picture or audio-visual work. In Australia, next to works created by employees ‘in pursuance of the terms of his or her employment’ under s 35(6) of the 1968 Copyright Act, also certain types of commissioned artistic works (s 35(5)) and works created by journalists (s 35(4)) will not—or not in full—belong to the creator, but to the employer, commissioner, or publisher; see Ann Monoti, ‘Australia’ in Wolk and Szkalej (n 54) 556–59. s 17 of the Indian Copyright Act takes a very similar approach. In Canada, s 13(3) of the 1985 Copyright Act also allocates ownership to the employer for works created by employees in the course of their employment; see Norman Siebrasse, ‘Canada’ in Wolk and Szkalej (n 54) 596. 56 See art 15 of the 1970 Copyright Act. Interestingly, this allocation in favour of the employer includes moral rights; see Tsugizo Kubo, ‘Japan’ in Wolk and Szkalej (n 54) 474–75. 57 See art 9 of the Korean Copyright Act (as revised in 2009). A qualifying condition for works to be assigned to the employer under art 9 is that the work is published under the employer’s name—which in turn means that employees own copyright in works published under their own name; see Gyooho Lee, Republic of Korea, in Wolk and Szkalej (n 54) 497. 58 See art 16 of the 1998 Ecuadorian Intellectual Property Law. 59 While there seems to be no explicit rule for copyright, an analogue application of Employment Contract Law 20,744 leads to an allocation of copyright in employee works to the employer, see Martin Bensadon, ‘Argentina’ in Wolk and Szkalej (n 54) 623–24. 60 Akin to the approach in Argentina, Italian Copyright law lacks a specific rule for employment relations, but commentators suggest an analogue application of art 64 of the Italian Industrial Property Act—with the effect of ‘derivative and direct’ assignment of copyright to the employer; see Andrea Ottolia, ‘Italy’ in Wolk and Szkalej (n 54) 214–17. 61 art 17(2) of the 1998 Brazilian Copyright Act affording ownership in ‘collective works’ to the ‘organizer’ apparently applies also to works generated in corporations, especially in the publishing industry, see Elisabeth Kasnar Fekete, ‘Brazil’ in Wolk and Szkalej (n 54) 638. 62 See Lionel Bently and others, Intellectual Property Law (4th edn, OUP 2018) 135–36. In the USA, the leading case is Community for Creative Non-Violence v Reid (1989) 490 US 730, where the Supreme Court relied on common law agency principles to decide whether the work at issue has been made by an employee or an independent contractor. The factors used in that test, by and large, resemble those applied by English Courts under the CDPA. Whether a work has been made within the scope of employment depends, in the USA and the UK, on the scope of employee’s duties (and hence essentially on the employment contract or custom), and whether the work has been made to serve the employer’s interests; see Avtec Systems, Inc. v Peiffer (1994) 21 F 3d 568, 571–72 (4th Cir). The case-law in Canada (see Siebrasse (n 55) 596–98) and Australia (see Monoti (n 55) 557–59) takes a similar approach. 63 See, for example, ss 79(3), 81(4), 82(1)(a)&(2) and 87 CDPA. In Australia, (retroactive and prospective) employee consent is commonly sought in employment contracts, see Monoti (n 55) 561–62; while waivers of moral rights are equally common in Canada; see Siebrasse (n 55) 600. 64 See, with regard to the performer’s property rights in the UK, ss 191B and 191C CDPA—whereas the non-property rights of performers (primarily for consenting to recording her life performance) are in principle not transferrable or assignable (s 192A CDPA). In Australia and Canada, on the other hand, rights in employee performances are initially vested in the employer; see s 100AD(2) of the Australian Copyright Act, as well as ss 13(3), 2 of the Canadian Copyright Act. 65 As mentioned in n 55, an exception is that under US copyright law, certain types of commissioning works also fall under the ‘works made for hire doctrine’ which therefore allocates ownership to the commissioner; see further Birnhack (n 54) 19–20. 66 See, for example, s 90 CDPA. 67 See, for example, R Griggs Group v Evans [2004] FSR (31) 673, [57]—confirmed on appeal (2005 FSR (31) 706 (CA)) and criticized by Bently and others (n 62) 137–39, arguing that this approach (of too readily assuming implied licenses or even assignments in order to ensure that ‘rights should follow money’) undermines the statutory scheme designed to enhance certainty and protect authors. A ‘business efficacy’ approach is also adopted in Spain where courts apply art 51 of Royal Legislative Decree 1/1996 broadly to include not only salaried authors in an employment relation, but also service contracts if a transfer of rights is necessary to allow the recipient / commissioner to conduct its business; see Luis-Alfonso Duran and Damaso-Jose Gallardo, ‘Spain’ in Wolk and Szkalej (n 54) 336. 68 See, for example, the US Supreme Court decision in Tasini where the Court held in a 7-2 majority decision that existing licences covering analogue uses for commissioned works did not extend to exploiting the works in digital formats, in particular inclusion in online databases and on CDs: The New York Times Company, Inc, et al, Petitioners, v Jonathan Tasini, et al. (2001) 533 US 483. 69 Post Tasini, The New York Times responded by drafting an ultimatum for affected freelancers: they could contact the publisher and agree to continued distribution of their works online, but only on conditions set by The New York Times, including that affected freelancers would not ask for any additional payment. Freelancers also had to release the Tasini decision's legal claim on the publisher. The decision also led to a significant shift in the licensing practice of publishers: freelance contracts after the Supreme Court judgment generally included terms that allowed the publishers to exploit the works in whatever ways the future may reveal—see the discussion on the aftermath of the Tasini judgment in the USA in Birnhack (n 54) 24–26 (stating that it ‘resulted in a sweeping change in contracts between commissioning parties and independent authors’); and generally: Maureen A O'Rourke, ‘Bargaining in the Shadow of Copyright Law after Tasini’ (2003) 53 Case W Res L Rev 605, 626–34; Robert A. Gorman and Jane C. Ginsburg, ‘Authors and Publishers: Adversaries or Collaborators in Copyright Law?’ in Benjamin Kaplan (ed), An Unhurried View Of Copyright Republished (And With Contributions From Friends) (GORGIN-9 2005; Iris C Geik and others, co-editors). Gorman and Ginsburg, ibid 9, for example, note that ‘In this tug-of-war between author and publisher, the former won the litigation battle but not necessarily the economic war’. While Terry concludes: ‘the aftermath illustrates that the victory has had severe negative effects on the rights and the livelihoods of freelance writers; see Amy Terry, ‘Tasini Aftermath: The Consequences of the Freelancers' Victory’ (2004) 14 DePaul J Art, Tech & Intell Prop L 231 <https://via.library.depaul.edu/jatip/vol14/iss2/3>. 70 See, for example, ss 7, 29 of the German Copyright Act (Urheberechtsgesetz), as well as art L.111-1 §3 of the French Intellectual Property Code (IPC). The underpinning concepts in Civil Law jurisdictions, in a nutshell, often converge around the idea that the work is an expression of the ‘personality’ of the author, and as a reflection of the author’s self, cannot be transferred to, or owned by, others. 71 See ss 31(1), (5) of the German Copyright Act, stating that in principle, rights are transferred to the extent necessary to give effect to the underlying contractual relationship and its purpose (‘specific purpose’ principle or Zweckübertragungsgrundsatz). s 43 confirms that the general rules concerning the transfer of exploitation rights applies also to authors in employment relations, unless their specific contract with the employer provides otherwise. Employment contracts then are generally understood to imply a transfer of all those exploitation rights in works created by employees in the course of their employment (ie when there is an ‘inner connection between the employee’s contractual work obligations and the creative process’) which are necessary to further the business objective of the employer. For works without that ‘inner connection’, some assume that employees have a duty to offer a licence to the employer if the work is capable to compete with the employer’s business; see Jan Bernd Nordemann and Christian Czychowski, ‘Germany’ in Wolk and Szkalej (n 54) 149–52, 154–55. 72 art L.131-3 IPC. Since 2006, there seems to be some allowance by French courts to limit the need for an explicit mention of all exploitation rights to the types of contracts listed in that provision—namely for performances, publishing and audio-visual productions, hence allowing implied licences for other type of agreements. For journalists of press publications (not TV or radio, etc), art L.7111-3(1) also implies the transfer of exploitation rights regarding the specific work created for a press publisher; see Michel Abello, Jerome Tassi and Estelle Vard, ‘France’ in Wolk and Szkalej (n 54) 118, 121. 73 An ‘employment work’ is one created by an employee for the purpose of fulfilling work-related tasks, while the works whose exploitation rights are directly allocated to the employer include technical drawings, maps and computer programs. In addition to the priority right, the employee is also must seek the employers consent if she wants to license ‘employment works’ to third parties within the first two years after their creation. If consent is given, the remuneration obtained from such licensing needs to be shared between the employee and employer; see Ella Cheong and Li Luo, ‘China’ in Wolk and Szkalej (n 54) 446–47. 74 See s 43 UrhG. See also the further discussion below. 75 For a concise summary, with a focus on how these provisions apply in the formal employment context, see Nordemann and Czychowski (n 71) 147–63 and Birnhack (n 54) 16–19. 76 See s 31(5) UrhG. 77 As discussed in n 71, in the specific employment context, the specific purpose principle will usually operate to assume a transfer of all exploitation rights related to the business activity of the employer. 78 s 39 UrhG—these assumptions however can be overcome by an express allowance to modify, etc. 79 ss 31(a) and 40 UrhG—importantly, the author cannot waive these rights to cancel and/or terminate. 80 ss 34, 35 UrhG. 81 See YouTube’s Terms of Service—as updated for users from the European Economic Area (EEC) and Switzerland—of 22 July 2019, available at <https://www.youtube.com/static?template=terms>. YouTube’s earlier terms of service (and those applicable elsewhere) include a more explicit, and potentially broader licence whereby users grant YouTube ‘a worldwide, non-exclusive, royalty-free, transferable licence (with right to sub-licence) to use, reproduce, distribute, prepare derivative works of, display, and perform that Content in connection with the provision of the Service and otherwise in connection with the provision of the Service and YouTube's business, including without limitation for promoting and redistributing part or all of the Service (and derivative works thereof) in any media formats and through any media channels’. 82 TikTok relies on slightly different Terms of Service for users from the USA, EU countries, Germany in particular, and other jurisdictions—where the main difference relates to the applicable law and jurisdiction of courts or arbitral tribunals—see <https://www.tiktok.com/legal/terms-of-use?lang=en>. In the following, the standard terms relevant for users in the EU are discussed. 83 See TikTok Terms of Service, s 7, stating that users ‘acknowledge and agree that we may generate revenues, increase goodwill or otherwise increase our value from your use of the Services, including, by way of example and not limitation, through the sale of advertising, sponsorships, promotions, usage data and Gifts (defined below), and except as specifically permitted by us in these Terms or in another agreement you enter into with us, you will have no right to share in any such revenue, goodwill or value whatsoever. You further acknowledge that, except as specifically permitted by us in these Terms or in another agreement you enter into with us, you (i) have no right to receive any income or other consideration from any User Content [elsewhere defined to include anything uploaded by the creator] or your use of any musical works, sound recordings or audiovisual clips made available to you on or through the Services, including in any User Content created by you, and (ii) are prohibited from exercising any rights to monetize or obtain consideration from any User Content within the Services or on any third party service (e.g. , you cannot claim User Content that has been uploaded to a social media platform such as YouTube for monetization)’. 84 ibid. The licence further clarifies that ‘the rights granted in the preceding paragraphs of this Section include, but are not limited to, the right to reproduce sound recordings (and make mechanical reproductions of the musical works embodied in such sound recordings), and publicly perform and communicate to the public sound recordings (and the musical works embodied therein), all on a royalty-free basis. This means that you are granting us the right to Use your User Content without the obligation to pay royalties to any third party’. 85 ibid. The attempt to prevent any reliance on moral rights is arguably because TikTok ‘or authorised third parties, reserve the right to cut, crop, edit or refuse to publish, your content at our or their sole discretion’. 86 See Instagram’s Terms of Use <https://help.instagram.com/581066165581870; and Facebook’s Terms of Service – online available at https://www.facebook.com/legal/terms>. 87 Facebook’s Terms of Service, s 3.1. 88 See <https://www.bbc.co.uk/news/technology-33362722>. Facebook however does generally not allow creators to use its services to upload video that create a ‘music listening experience’; see <https://www.facebook.com/legal/music_guidelines>. 89 See ss 32(3), 32a(3), and 32b UhrG. In the case of creative works licensed to platforms which allow their exploitation worldwide (or in all markets where the platform operates), accessing, viewing and sharing the work in Germany (that is by other users from Germany) would trigger the mandatory protections discussed here. Under the first alternative mentioned above, because of art 4(2) of the ‘Rome I’ Regulation (EC 593/2008) of 17 June 2008 on the law applicable to contractual obligations, German law would govern a licensing clause in the standard terms between users residing in Germany and a platform like YouTube only if one would consider the user as effecting the ‘characteristic performance’. While this generally is of course the platform, one might argue that, with regard to the licensing of work created by users, the user in fact effects the characteristic performance (depecage). Alternatively, in cases where a creator residing in Germany acts outside her trade or profession, the specific consumer protection rule under the art 6(1)b) of the Rome I Regulation would lead to the application of German (copyright) contract law since the platform is directing its activities to the country where the consumer has her habitual residence. Under any of the scenarios above, German courts would apply the mandatory creator protection rules of the German UrhG even if the parties had chosen another law. Whether a court in another jurisdiction would consider these rules to be international mandatory and give effect to them of course would depend on that court’s private international law, and arguably in most instances whether the case at hand shows a strong link to Germany. 90 For this reason, these protections are argued to be less relevant for employees, as their salary is meant to provide them with a basis for subsistence, and is generally seen to include remuneration for works created in the context of their employment. At the same time, employees can avail themselves of these protections if it can be shown that their salary does not adequately remunerate them for the exploitation rights afforded to the employer; see Nordemann & Czychowski (n 71) 158–59. 91 s 32(1) UrhG. 92 ss 36 and 36a UrhG sets out in significant detail who can participate in this collective bargaining, and how to arrive at an agreement (via an arbitration mechanism) over what is adequate in case of a dispute. 93 See ss 31(2) and 36b and 36c UrhG. 94 s 32c UrhG. 95 See s 32a UrhG (so-called ‘bestseller clause’). It is irrelevant, whether the amount of profit was foreseeable for the contracting partner, s 32a(1). Again, collective agreements which specifically provide for remuneration for such cases take precedence, s 32a(4) UrhG. 96 See ss 27, 20b(2), 26 UrhG. In addition, uses of copyrighted material that are permitted by law (in form of so called ‘exceptions and limitations’) and for that reason trigger compensation payments for authors can be seen in this context—see Nordemann and Czychowski (n 71) 162–63. 97 See TikTok Terms of Service for Germany, s 5c, <https://www.tiktok.com/legal/terms-of-use?lang=de>. 98 See TikTok’s Terms of Service, s 7—quoted in its most relevant aspect in n 82 above. 99 For an instructive discussion, see ‘How TikTok Gets Rich While Paying Artists Pennies’ <https://pitchfork.com/features/article/the-great-music-meme-scam-how-tiktok-gets-rich-while-paying-artists-pennies/>. As the principal appeal of short videos on TikTok (and lots of other online memes more general) is to use more edgy, ideally unknown music as background, TikTok’s appropriation, per its Terms of Service, of any creator content for this purpose hence is significant, and is likely to deprive creators whose music is used by others for making any gains from their creative works on this platform. 100 See the further discussion of the conflict of laws (private international law) dimension in n 89 above. 101 See the YouTube Revenue and Usage Statistics for 2019 <https://www.businessofapps.com/data/youtube-statistics/>. While the overall cut which YouTube offers seems to be around 50% for all, there are significant differences, depending on the type of content, what advertisers are willing to pay for their adds: while YouTube does not seem to provide detailed data, creators can earn as little as US$0.35 per 1000 views or as much as US$5, see <https://fortune.com/2018/02/27/youtube-success-poverty-wages/>. 102 Under s 32(2) UrhG, a key benchmark is what creators can legitimately expect as a fair share in light of relevant practices in trade and industry. The further criteria relate to the type of licence granted (which will in all cases be the same licence inYouTube’s terms of service, discussed above), as well as the agreed use to be made of the copyrighted work. 103 Ellis (n 30). 104 See, for example, Instagram’s Terms of Use, ‘How We Will Handle Disputes’, <https://help.instagram.com/581066165581870>. Facebook’ Terms of Service adopt the same approach in its section on ‘Disputes’, see <https://www.facebook.com/legal/terms>. 105 See ‘Updated Terms of Service FAQs’ <https://support.google.com/youtube/answer/9326895>. 106 See Tribunal de Grande Instance, Décision du 07 août 2018, 1/4 social N° RG 14/07300—discussed by Mathilde Pavis at <https://ipkitten.blogspot.com/2018/09/paris-tribunal-guts-twitters-t.html>. 107 See Paris Tribunal (Tribunal de Grande Instance), UFC-Que Choisir v Google Inc (12 February 2019) and Paris Tribunal (Tribunal de Grande Instance), UFC-Que Choisir v Facebook Inc. (9 April 2019)—again discussed by Mathilde Pavis on the IP Kat Blog <https://ipkitten.blogspot.com/2019/03/paris-tribunal-strikes-again-and-guts.html and https://ipkitten.blogspot.com/2019/05/things-come-in-threes-paris-tribunal.html>. Based on Pavis’ translation of the judgment against Google, the Court found these blanket and open licences to be contrary to the provisions of arts L.131-1, L.131-2 and L.131-3 of the French Intellectual Property Code, which require that a party benefitting from a right-assigning contract, defines the relevant content, the rights conferred, as well as the types of use allowed by the author of the content. 108 See Abello, Tassi and Ward (n 72) 118–19. 109 See Martin Kretschmer and Philip Hardwick, Authors’ Earnings from Copyright and Non-copyright Sources: A Survey of 25,000 British and German Writers (2007). 110 See Axel Metzger, ‘Zehn Jahre reformiertes Urhebervertragsrecht: Warum das Vertragsrecht ungeeignet ist, die soziale Frage der Urheber zu lösen' in Eva Inés Obergfell (ed), Zehn Jahre reformiertes Urhebervertragsrecht (De Gruyter 2013). Current figures from the social insurance for freelancers in the creative industries in Germany show that the income remains relatively low—see <https://www.kuenstlersozialkasse.de/service/ksk-in-zahlen.html>. 111 Metzger, ibid 112. In the case decided by the Supreme Court, a translator had requested to adjust the typical (and industry-common) lump-sum payment per page of translation. The Court held that only for sales beyond a volume of 5000 copies, an adjustment alongside a formula in a collective agreement between publishers and authors of works of fiction was warranted. Limiting the adjustment to sales beyond 5000 copies in a market where most books sell fewer copies eventually meant that only translators of very successful books are able to ask for an adjustment (which in the case decided amounted, for a book of which 10,000 copies were sold, to no more than 166 euros). 112 See BGH, GRUR 2012, 496 – Das Boot. 113 See Metzger (n 110). Various responses by author associations and unions to the recent consultation of the German Ministry of Justice on implementing the new EU rules on adequate remuneration support this—see for example the response by Ver.di, one of Germany biggest workers’ union involving creators <https://www.bmjv.de/SharedDocs/Gesetzgebungsverfahren/Stellungnahmen/2019/Downloads/091019_Stellungnahme_verdi_EU-Richtlinien_Urheberrecht.pdf?__blob=publicationFile&v=3>. 114 See ‘Why Success on YouTube Still Means a Live of Poverty’ (Bloomberg, 27 February 2018) <https://fortune.com/2018/02/27/youtube-success-poverty-wages/>. The article compares this to the US federal poverty line of around $12,000, and cites Alice Marwick, an assistant professor of communication at the University of North Carolina: ‘If you’re a series regular on a network TV show, you’re getting a good amount of money. Yet you can have half a million followers on YouTube and still be working at Starbucks.’ 115 Mathias Bärtl, ‘YouTube Channels, Uploads and Views: A Statistical Analysis of the Past 10 Years Convergence’ (2018) 24 The International Journal of Research into New Media Technologies 16. 116 ibid 30. 117 For a critical history of minimum wage legislation in the UK, see Zoe Adams, Labour and the Wage: A Critical Perspective (OUP 2020) and Adams (n 19). 118 In the EU, the CJEU judgment in Case C-476/17 Moses Pelham GmbH vs Ralf Hütter (CJEU, , 29 July 2019) paras 26–39 and 66–74, held that even the inclusion of very short samples (two seconds in the case at hand) fall under the exclusive right of reproduction of the producer of the sound recording from which the sample has been taken—unless the sample has been modified to an extent that it is unrecognizable. If the sample remains recognizable, its use may fall under the quotation defence if it is used in a way that it ‘enters into a dialogue’ with the work (here phonogram) quoted. One has to assume that the approach adopted for audio recordings will equally apply to video recordings (films), so that any use of short, recognizable snippets amounts to an infringement in the EU, unless it can be justified under an exception. In the USA, the Court of Appeals for the 2nd Circuit held a 37-second sample (used in a seven-minute song) to be fair use—primarily because of its transformative nature, the reasonable relation between amount and purpose of use and since the use did not serve as market substitute for the original work; see Estate of James Oscar Smith v. Cash Money Records, Inc., Summary Order, 3 February 2020. For a general discussion from a US perspective, see M Sag, ‘Internet Safe Harbors and the Transformation of Copyright Law’ (2017) 93 Notre Dame Law Review 518. 119 See art 14 of the Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market (hereafter E-Commerce Directive) and 17 USC. 512(c) (2018) (US) (introduced by the Digital Millennium Copyright Act 1998). 120 For an empirical account of the notice and take-down system in the USA, see J Urban, J Karaganis and B Schofield, ‘Notice and Takedown in Everyday Practice’ (22 March 2017) UC Berkeley Public Law Research Paper No 2755628 <https://ssrn.com/abstract=2755628>; see also K Erickson and M Kretschmer ‘“This Video is Unavailable” Analyzing Copyright Takedown of User-Generated Content on YouTube’ (2018) 9 JIPITEC 75. For a discussion on the EU system, see C Angelopolous, European Intermediary Liability in Copyright: A Tort-Based Analysis (Kluwer 2016) and M Husovec, Injunctions Against Intermediaries in the European Union: Accountable But Not Liable? (CUP 2017). 121 A general description of automated enforcement tools by one of the major providers (Audible Magic) is at <https://www.audiblemagic.com/compliance-service/#how-it-works>. Specific content protection tools—other than the one by YouTube discussed in more detail below—are for example discussed here (<https://variety.com/2018/digital/news/instagram-block-copyright-infringement-video-1202692290/> regarding Instagram) and here (<https://variety.com/2017/digital/news/facebook-rights-manager-copyright-videos-third-party-1202578122/> concerning Facebook). 122 To understand the basics of Content ID—which as per YouTube’s owner Google 2018 report now deals with 98% of all copyright issues on YouTube, see ‘How Google Fights Piracy’ (7 November 2018) 21, 25 <https://www.blog.google/outreach-initiatives/public-policy/protecting-what-we-love-about-internet-our-efforts-stop-online-piracy/> and have a look at the short (YouTube) video: https://support.google.com/youtube/answer/2797370?hl=en-GB. For a legal analysis, see the authors discussed below and also S Jacques and others, ‘Automated Anti-piracy Systems as Copyright Enforcement Mechanism: A Need to Consider Cultural Diversity’ (2018) 40 EIPR 218. 123 See, for example, <https://www.youtube.com/watch?v=l4HpWQmEXrM&feature=youtu.be>, a video of a surprise marriage proposal using an Indie-Pop song as background that, due to the video’s popularity, has been listened to by 14 million viewers. See also the ‘Harlem Shake’ example, discussed in Michael Soha and Zachary J McDowell, ‘Monetizing a Meme: YouTube, Content ID’ and the Harlem Shake, ‘Social Media + Society’ (2016) 1–12. 124 YouTube for example explains that since January 2014, Content ID claims have ‘outnumbered copyright takedowns by more than 50 to 1’ (see <https://support.google.com/youtube/answer/7002106?hl=en-GB>). 125 On copyright owner’s options to earn from ads through Facebook’s and Instagram’s automated enforcement system, see <https://www.facebook.com/help/publisher/2276498512619396 and generally https://www.facebook.com/help/publisher/1548693938521733>. On TikTok, see ‘TikTok App Navigating Copyright Laws’ (Klemchuck LLP, 20 December 2018) <https://www.klemchuk.com/ip-law-trends/tiktok-app-navigating-copyright-laws>; and Divij Joshi, ‘Is the Clock Ticking for TikTok’s Intermediary Liability Exemptions?’ (Spicy IP Blog, 2 September 2019) <https://spicyip.com/2019/09/is-the-clock-ticking-for-tiktoks-intermediary-liability-exemptions.html>. 126 See ‘How Content ID Works’ <https://support.google.com/youtube/answer/2797370>. The platform makes clear that it reserves to itself the final say on what enforcement tool it decides to provide to whom; see ‘Qualifying for Content ID’ <https://support.google.com/youtube/answer/1311402>. Individual creators hence are left with the option to sign up to rights management organizations that have access to Content ID, see <https://www.tubefilter.com/2019/05/14/digital-rights-startup-superbam-creators-millions-youtube-contentid-copyright-claims/>. 127 See the discussion here: <https://support.google.com/youtube/thread/1281991 and here: https://youtube-creators.googleblog.com/2019/08/updates-to-manual-claiming-policies.html> (indicating that despite some changes to improve the system for creators, any inclusion of someone else’s content will continue to afford the exclusive monetization options to the Content ID claimant). 128 This concerns subject matter eligibility, any threshold or conditions for protection under the applicable domestic copyright laws (which usually will be the law of the country for which protection is claimed—lex loci protectionis, see n 89), as well as ownership or other legal title in the material claimed. 129 While YouTube webpages state that, for example, content licensed under a Creative Commons (CC) license, ‘public domain footage’ or material ‘used under fair use principles’ is ‘ineligible for use in or as a reference’ (see ‘Content Eligible for Content ID’ <https://support.google.com/youtube/answer/2605065>), YouTube has no means of ensuring that Content ID only identifies matches that are indeed infringing, based on the applicable lex loci protectionis (see n 89). 130 While it warns not to abuse the system and threatens to act if it becomes aware of such abuse (including disabling specific reference files (…) and releasing all associated claims, disabling Content ID or even terminating YouTube partnership—see ‘Content Eligible for Content ID’ (n 129), YouTube eventually leaves it to the content owner to ensure that no wrongful claiming occurs. Occasionally, YouTube has stepped in when patterns of abuse have emerged—see <https://youtube-creators.googleblog.com/2019/08/updates-to-manual-claiming-policies.html> (where it took steps to prevent manual claiming of very short video clips contained in user uploads—with the effect of transferring all monetizing revenue to the claimant—by requiring timestamps for such claims and subsequently disenabling manual claiming for such cases altogether). 131 See generally Sag (n 118) 544–54. 132 When a user disputes or subsequently appeals a Content ID claim, she risks a copyright strike, see ‘Dispute a Content ID Claim’ <https://support.google.com/youtube/answer/2797454>. For the user, receiving just one ‘copyright strike’ serves as a warning and affects the user’s ability to monetize her own content. A user that has received three strikes will have her/his account terminated, all uploaded videos removed, and will not be able to create new channels on YouTube—see <https://support.google.com/youtube/answer/2814000>. 133 If a user disputes a claim, it is then entirely up to the claimant whether it ‘releases’ or ‘upholds’ the claim. In the latter case, the user normally will have a right to appeal—which is in principle no more than another plea to the content owner to reconsider its claim over the upload and in no instance involves an independent third party, see <https://support.google.com/youtube/answer/2797454>. While users can eventually file a counter-notification which then requires the claimant to file court proceedings, a user then risks being sued in the USA (as explained below). In addition, YouTube conducts an initial screening whether counter-notices ‘provide a sound rationale for reinstatement’ and rejects more than two-third of counter-notifications at this level; see ‘How Google Fights Piracy’ (n 122) 31. 134 A user filing a counter-notification (see <https://support.google.com/youtube/answer/2807684?hl=en-GB&ref_topic=9282678>) in order to counter the claimants decision to uphold the decision against the user’s appeal, submits herself to the US ‘notice and take-down’ mechanism under the Digital Millennium Copyright Act (DMCA). Under s 512(g)(3)(D) of the US Copyright Act, the contents of any counter-notification must include a statement whereby the user consents to the jurisdiction of US Federal District Court of her/his permanent residence; or—if the user’s address is outside the USA—to the jurisdiction of ‘any judicial district in which the service provider may be found’. In the case of YouTube, this arguably entails consent to be sued in California, where YouTube has its headquarters. Filing a counter-notification hence means for a user from outside the US that (s)he consents to the jurisdiction of a US court, and arguably risks being sued as well as receiving a default judgment. 135 Consider the telling examples provided by Stephen Witt, ‘When Your YouTube Video Becomes a Corporate Profit Center’ Los Angeles Times (27 June 2015) <https://www.latimes.com/opinion/op-ed/la-oe-0628-witt-youtube-copyright-20150628-story.html>, as well as the Harlem Shake meme by Soha and McDowell (n 123). 136 See the discussion in Soha and McDowell (n 123) 6. 137 This however appears to be the current practice on YouTube—ie a tiny amount of copyrighted content if claimed under Content ID generally prevents the user to monetize her/his videos—regardless how irrelevant that copyrighted material is for the success of the user’s video as a whole—see ‘What Kind of Content Can I Monetise?’ <https://support.google.com/youtube/answer/2490020?hl=en-GB> and the blog discussion at <https://support.google.com/youtube/thread/1281991> where YouTube admits that it is ‘clearly not ideal’ that a copyright owner is ‘getting all of the creator’s revenue for a few seconds of a video’. An exception exists for certain eligible cover videos, see <https://support.google.com/youtube/answer/3301938>. 138 The UK has recently declared it does not intend to implement this Directive—see <https://www.parliament.uk/business/publications/written-questions-answers-statements/written-question/Commons/2020-01-16/4371>. 139 Directive (EU) 2019/790 of the European Parliament and of the Council of 17 April 2019 on Copyright and Related Rights in the Digital Single Market and amending Directives 96/9/EC and 2001/29/EC (2019) OJ L130/92 (‘Digital Single Market’ or ‘DSM’ Directive). 140 See recital 62 DSM Directive. 141 See art 17:1, 2 DSM Directive. 142 See art 17:3 DSM Directive. 143 art 17:4 DSM Directive. The contested issue of what best efforts to prevent access to infringing works entails is further addressed in art 17:5 which sets out several factors, to be applied against the broader principle of proportionality. 144 See Section ‘Copyright Protection for Creators on Digital Platforms’. 145 More or less the same option of notifying a platform about infringing copies and then asking them to remove those or block access to them has already been on the table in EU Member States (and lots of other jurisdictions which adopt a similar notice and take-down mechanism) on the basis of art 8(3) of the Copyright in the Information Society Directive. In that regard, the ‘added value’ of art 17(4)(c) lies primarily in the ‘stay down’ obligation it imposes: in other words, platforms also must ensure that the same infringing work does not re-appear elsewhere on the platform. 146 See Section ‘Content Protection on Platforms and Its Impact on Creators’. 147 In full, art17:7 states: ‘The cooperation between online content-sharing service providers and rightholders shall not result in the prevention of the availability of works or other subject matter uploaded by users, which do not infringe copyright and related rights, including where such works or other subject matter are covered by an exception or limitation. Member States shall ensure that users in each Member State are able to rely on any of the following existing exceptions or limitations when uploading and making available content generated by users on online content-sharing services: (a) quotation, criticism, review; (b) use for the purpose of caricature, parody or pastiche.’ (Emphasis added). 148 As an approach based on the traditional conception of exceptions and limitations to copyright as defences to infringement claims by copyright owners would not be enough to afford creators with means to prevent demonetization, the implementation of art 17(7) would have to pursue new approaches that conceptualize those using copyrighted content in a privileged way as right holders or claimants with remedies against those preventing them to rely on such exceptions. 149 Sag (n 118) 555. 150 While a full discussion of art 17(9) is not possible here, this provision calls for a ‘effective and expeditious complaint and redress mechanism’ available to users in cases of disputes over ‘disabling access to, or removal of’ their content. It asks right holders to ‘duly justify the reasons’ for their blocking and/or removal requests, and demands that ‘decisions to disable access or to remove uploaded content shall be subject to human review’. 151 In its last sentence, art 17(9) however does impose an obligation on EU Members to ‘ensure that users have access to a court or another relevant judicial authority to assert the use of an exception or limitation to copyright and related rights.’ (Emphasis added). Assertion of an exception in this sense would arguably cover cases where the user complains about being subjected to monetization of her/his content that benefits from a copyright exception. One hopes that EU Members will implement this provision in such a way that not only disputes over blocking or removing content where users invoke copyright exceptions must end up in front of courts, but also those about monetization of content that falls under copyright exceptions. 152 See recital 72 of the DSM Directive. 153 Under recital 73, this includes the actual or potential economic value of the licensed or transferred rights, the author's or performer's contribution to the overall work or other subject matter, and all other circumstances of the case, including market practices or the actual exploitation of the work. 154 Most notably, perhaps, is the judgment of Justice Higgins in the Australian Harvester Case where he argued ‘The provision for ‘fair and reasonable’ remuneration is obviously designed for the benefit of the employees in the industry; and it must be meant to secure to them something which they cannot get by the ordinary system of individual bargaining with employers…. The standard of ‘fair and reasonable’ must therefore be something else, and I cannot think of any other standard appropriate than the normal needs of an average employee, regarded as a human being in a civilised community.’ (Ex parte H.V. McKay (1907) 2 CAR 1). A similar principle underpinned the Wages Councils model in operation in the UK following the second world war. For an analysis, see: Adams (n 19). 155 For an overview, see: Zivile Buinickaite, ‘Collecting Societies under European Competition Law Scope and the Contribution of the Commission and Court of Justice of the European Union in Applying Competition Rules in Collective Rights Management’ 41 and Simone Schroff and John Street, ‘The Politics of the Digital Single Market: Culture vs. Competition vs. Copyright’ (2018) 21 Information, Communication & Society 1305. 156 Initially, while the EU subjected collection societies to competition law, it also recognized that competition law could not be applied rigidly in this context. See, for example: Case C-395/87 Ministère Public v Tournier [1989] ECR-2521, para. 24; Lucazeau v SACEM ECR 2811; Case 7/82 GVL v Commission [1983] ECR 483, [1983] CMLR 645; Case 127/73 Belgische Radio en Televisie v SV SABAM ECR 51, 313. All of these cases involved the application of competition law and recognized that there are legitimate interests that can limit its application in practice. Also discussed in: Christoph B Graber, ‘Collective Rights Management, Competition Policy and Cultural Diversity: EU Law Making at a Crossroads’ (2012) I-Call Working Paper 2012, 6<http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2161763> accessed 15 April 2016). However, more recently, the EU has taken a much harsher approach, see: European Commission, Commission Decision of 16/07/2008 relating to a proceeding under Article 81 of the EC Treaty and Article 53 of the EEA Agreement (Case COMP/C2/38.698 – CISAC) 220–23 <http://ec.europa.eu/competition/antitrust/cases/dec_docs/38698/38698_4567_1.pdf> accessed 17 December 2015 (although overturned, this case is illustrative of a shift in attitude). 157 Under Directive 2014/26/EU of 26 February 2014 on collective management of copyright and related rights and multi-territorial licensing of rights in musical works for online use in the internal market, collection societies that manage creators' rights in musical works for the multi-territorial licensing of those rights for online uses are subject to several requirements adapted to the digital context, such as enhanced capability to process large amounts of data, accurate identification of the works used by platforms, fast invoicing, and timely payment to right-holders. Creators who have signed up with a collection society that does not offer to manage such rights for their musical works have the right to shift to other entities able to do so. 158 This is the approach thus far taken by the Polish and Dutch courts. See ‘The Contractual Protection of Authors in Copyright in the Digital Single Market Directive – Does the Reality Live up to the Expectations? - NIR’ <https://www.nir.nu/notis/NIR2019W16> accessed 13 March 2020. 159 Morten Hviid, Simone Schroff and John Street, ‘Regulating Collective Management Organisations by Competition: An Incomplete Answer to the Licensing Problem?’ (2017) 7 JIPITEC <http://www.jipitec.eu/issues/jipitec-7-3-2016/4507>. 160 It is notable in this respect that in a number of countries, the UK included, notwithstanding the absence of express exemptions, collective bargaining among self-employed creative workers has a long-history. In other countries, such as Japan, Slovenia and Canada, moreover, no conflict is deemed to exist between competition law and collective bargaining, such that were collective bargaining rights extended to the formally self-employed, this issue would not even arise. The situation is similar in Italy (the position under EU law not withstanding) because unions are not considered economic undertakings and so would not be subject to competition law in practice. Finally, it should also be noted that the International Labour Organisation (ILO) convention on collective bargaining is broad enough to include the self-employed. See: Shlomit Yanisky-Ravid, ‘Collective Bargaining and Competition Law: A Comparative Study on the Media, Arts, and Entertainment Sectors’ [2012] SSRN Electronic Journal <http://www.ssrn.com/abstract=2132660> accessed 13 March 2020. © The Author(s) (2020). Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Work and works on digital platforms in capitalism: conceptual and regulatory challenges for labour and copyright law
JO - International Journal of Law and Information Technology
DO - 10.1093/ijlit/eaaa017
DA - 2020-10-16
UR - https://www.deepdyve.com/lp/oxford-university-press/work-and-works-on-digital-platforms-in-capitalism-conceptual-and-FK5Vom6pmB
SP - 1
EP - 1
VL - Advance Article
IS - 
DP - DeepDyve
ER -