Abstract
Highly skilled freelance workers are mainly depicted as a challenge to trade unionism because of their mobility, market power and specific interests in organisational support. The authors explore the manifestations of collectivism of highly skilled freelance workers on the basis of semi-structured interviews with 14 highly skilled freelancers and 35 representatives of intermediaries such as trade unions, professional associations, staffing agencies and cooperatives in medicine, IT and film in Germany. The results reveal new forms and dynamics of labour market collectivism arising from concurrent conflicts and negotiations of job access and working conditions.
Introduction
In recent decades, contingent forms of employment have gained significantly in importance in most Western European countries, indicating a trend away from contracts for work to contracts for employment (Streeck, 2005). Of particular interest is the increase of freelancers who work on a fixed-term or project basis in organisations (Connelly and Gallagher, 2004). According to the Labour Force Survey (Eurostat, 2020), the proportion of freelance workers in the EU’s total self-employed workforce has risen from about 47% (in 2007) to 72% (in 2019). In Germany, the share in 2020 was around 55% (Eurostat, 2021). While this increase may be partly due to firm-specific outsourcing strategies, individual-driven forms of solo self-employment occur in knowledge-intensive services, which are identified as the sectors with the highest prevalence of freelancers (Bonin et al., 2020).
In the face of declining unionisation rates and increasingly fragmented and flexible labour markets, consideration of contingent workers’ concerns and risk exposure can be decisive for unions (Ebbinghaus, 2006; Gumbrell-McCormick and Hyman, 2013; Hyman and Gumbrell-McCormick, 2017). However, freelance work challenges the German trade unions as it contributes to the increasingly blurred distinction between employees and employers (Murgia and Pulignano, 2018; Streeck, 2005). Freelance work also challenges systems of interest representation in which workplace-level co-determination matters crucially in organising workers into unions (e.g. in Germany, Frege and Kelly, 2003) or where the workplace has been the predominant focus of organising campaigns (Simms, 2012). Although freelance workers may operate within the organisational core, they are typically less attached to firm-internal communities (Wilkesmann et al., 2020). Additionally, they are not usually integrated into employment protection and social security systems (Murgia and Pulignano, 2018) and are particularly exposed to market risks (D’Amours and Legault, 2013; Saundry et al., 2012). Therefore, freelance workers’ interests may differ from those of trade unions’ core clientele. In fact, unionisation rates in this special segment of labour are low, and highly skilled freelance workers in particular seem to prefer individual negotiations (D’Amours and Legault, 2013; Heery et al., 2004; Pernicka, 2006; Simms and Dean, 2015). However, in light of the limited social protection, freelance workers may have to rely on various forms of collective support. Previous research with its focus on classic forms of interest representation may have underestimated the extent to which freelance workers rely on support to cope with (not necessarily be protected from) social and economic risks and develop new forms of collectivism. Therefore, a broader scope of labour market intermediaries (Autor, 2009; Koene and Pichault, 2021) and of forms of collectivism (Hyman and Gumbrell-McCormick, 2017) needs to be taken into account.
On a general level, current debates point to changes in the foundations of solidarity – from economic or class-based identities to identities unrelated to the sphere of work (Piore and Safford, 2006). Recent contributions also suggest the emergence of new, possibly less stable forms of solidarity, such as networks and temporary collaborations (Heckscher and McCarthy, 2014), solidaristic associations that acknowledge heterogeneity (Zoll, 2000), or new social movements (Hyman and Gumbrell-McCormick, 2017). More specific starting points for a study of labour relations of contingent and in particular freelance work may be inspired by the Labour Process Theory (LPT). While interaction at the workplace is frequently viewed as a prerequisite for collectivism (Refslund and Sippola, 2022), class-based identities interact with other orientations in myriad ways and contexts and collectivism takes various forms (McBride and Martinez Lucio, 2011; Marks et al., 2020; Stephenson and Stewart, 2001).
With their ambivalent relation to unions and their transient connection to workplace communities, highly skilled freelance workers represent a paradigmatic case in which new manifestations of collectivism should become visible. This article analyses the following question:
How and under which conditions does collectivism of freelance workers evolve?
Particular attention is paid to highly skilled freelance workers’ concerns arising from their work, their labour market position and employment relationship, and their embeddedness in broader communities and relations to labour market intermediaries, and the German industrial relations system.
The purpose of this article is to contribute to the understanding of interest representation in contingent employment relationships, especially in knowledge-intensive services. To this end, it connects insights from LPT perspectives on worker resistance and collectivism with research on the role of labour market intermediaries and social relations between these actors and freelance workers. The results of this study complement prior research on collectivism by shifting the focus from the workplace to dynamics in the labour market. This involves an analysis of how highly skilled freelance workers in medicine, IT and film articulate their central concerns, such as negotiating job access and working conditions, the ways in which labour market intermediaries relate to these concerns with different strategies, and how freelance workers associate in reaction to these. We contribute to research on labour market intermediaries by pointing to shifting functions and relational, partly conflictual, dynamics of labour market intermediaries and their relation to freelance workers. With regard to research on collectivism, the study shows how collectivism arises from conflicts around the employment relationship and mobility power, and relations to and between labour market intermediaries. Results reveal new manifestations of collectivism beyond the workplace – i.e. in recruitment and supply networks and negotiation constellations in labour markets – and reflect the implications of the LPT for labour markets with highly skilled knowledge workers.
The next section briefly summarises debates on collectivism in the LPT literature, which are then related to research on freelance workers’ individualism versus collectivism, and to recent research on labour market intermediaries. After describing the data and methods used in the analysis, presenting and discussing empirical findings and drawing a conclusion, the article finally suggests areas for further research.
Theoretical advances toward understanding collectivism and solidarity
Labour process theory is strongly interested in ‘conflictual relations between workers and managers in the production process’ (Smith, 2006: 389), with a specific interest in control strategies, worker resistance, skills and broader relations of production in capitalist economies, as well as in manifestations of collectivism (see, for an overview, Gandini, 2019; Smith and Thompson, 1998; Storey and Bacon, 1993). Collectivism is usually defined and measured by the recognition of trade unions and industrial democracy (Purcell, 1987). While organised labour should be a reference point in the study of collectivism, collectivism cannot be limited to trade unions as organised collective activity (McBride and Martinez Lucio, 2011; Martinez Lucio and Stewart, 1997; Stephenson and Stewart, 2001). Helpful in identifying further forms and causes of collectivism, the LPT recognises forms that manifest also in workers’ unexpected uses of (Dundon and Dobbins, 2015; Martinez Lucio and Stewart, 1997) or their resistance to employers’ strategies – such as organisational misbehaviour (Thompson and Ackroyd, 1995) – or in informal worker communities of coping (Korczynski, 2003).
Prior LPT contributions have prominently located identity work within interests and power struggles related to the employment relationship and the labour process, and within political economies (Edwards, 1984; Marks and Baldry, 2009; Marks and Thompson, 2010; Martinez Lucio and Stewart, 1997; Smith and Thompson, 1998; Storey et al., 2005; Thompson and Ackroyd, 1995). This points to the relevance of past power struggles and experiences in various contexts, both at work and beyond (McBride and Martinez Lucio, 2011; Martinez Lucio and Stewart, 1997). For instance, research has shown collectivism’s capacity to develop in everyday life and in regional, occupational, or professional communities (McBride, 2011; McBride and Martinez Lucio, 2011; Marks et al., 2020; Stephenson and Stewart, 2001), also in relation to community unionism as an avenue for trade union renewal (Stewart et al., 2009).
The relevance and existence of collectivism and solidarity are brought into question by the growth of the service sector, new forms of work arrangements (Atzeni, 2009; D’Art and Turner, 2002; Stewart et al., 2020) and individualist HRM policies (Gunnigle et al., 1998). In this sense, Taylor (2001: 13) points out the development of an ‘individualistic and self-centred culture of contentment that sees no virtue in forms of collective association and solidarity’. Although the majority of research rejects a decline in collectivism and solidarity (D’Art and Turner, 2002; Heckscher and McCarthy, 2014; López-Andreu, 2020; Purcell, 1987; Simms and Dean, 2015; Stewart et al., 2020), some authors have identified changes in their form, particularly with respect to solidarity. For instance, the rise of the gig economy (Stewart et al., 2020) and new ways of building and maintaining social networks through technological developments have fundamentally affected the relationship of self to society (Heckscher and McCarthy, 2014).
Solidarity is often cited as a key factor and even as the prerequisite for collective action and collectivism (e.g. Atzeni, 2009; D’Art and Turner, 2002; Gunnigle et al., 1998; Heckscher and McCarthy, 2014; López-Andreu, 2020). At its broadest, solidarity reflects a communal sense of obligation to support the interests and goals of a group for the sake of its advancement or, on a more general level, to support social change. In defining solidarity, Heckscher and McCarthy (2014) raise two important questions: What is the content of the obligation, and toward whom (e.g. what target group) is the obligation formed? ‘In other words, defining a group of people as sharing common interests which are opposed to another group (workers against managers) is central to the process of social identification and solidarity building’ (Simms and Dean, 2015: 176; see also Kelly, 1998).
The project-oriented nature of work of highly skilled contingent workers may aggravate solidarity and collective action (López-Andreu, 2020; Simms and Dean, 2015), which, according to mobilisation theory (Kelly, 1998), depends on the development of a collective sense of injustice and its attribution to the employer. Contracts that tie workers more loosely to one employer, such as those for freelance work or temporary agency work, call for an expanded view of collectivism (Smith, 2006). As freelance workers are only temporarily integrated into a specific firm and workplace, they are expected to exit a workplace rather than raise collective concerns. Mobility power, defined as ‘securing labour services for capital and securing labour processes for labour’ (Smith, 2010: 269), can be expected to be exceedingly important to freelance workers’ relationships with employers. Workers’ mobility choices are not necessarily purely individualistic and unrelated to collective outcomes of struggles over the organisation of the labour process or to ‘work-effort-bargaining at work’ (Smith, 2006: 391; Smith, 2010). For example, their social networks (Antcliff et al., 2005) and the threat of exit may strengthen the bargaining power of workers on the shop floor, which is related to the fact that the amount of effort cannot be fixed ex ante (hereafter effort power). As Smith (2006: 389) notes in raising awareness of the ‘double indeterminacy of labour power’, further research is required regarding the interrelationships of effort power and mobility power. Accordingly, further study is needed regarding the relation between negotiating job access and working conditions and the relationship between mobility and collectivism.
To summarise, conceptual considerations and research about collectivism with respect to the LPT allow for a fine-grained analysis of possibilities for collective action in the context of workplace changes, conflicts and power asymmetries. Although some study results show that union leadership is crucial to solidarity building and worker mobilisation (e.g. Atzeni, 2009; D’Art and Turner, 2002; López-Andreu, 2020; Simms and Dean, 2015), LPT-inspired analyses also suggest that understanding the implications of employment change for industrial relations requires extending attention beyond trade unions and how they relate to growing groups of contingent workers, such as freelance workers. These analyses also direct attention toward micro-level manifestations of collectivism and solidarity, which can develop in the workplace as well as in professional or regional communities, with only an indirect or no connection to a specific workplace. These manifestations are particularly important for workers who are less attached to a firm, such as freelance workers, for whom problems in finding work and negotiating working conditions are exacerbated and mobility power is essential.
Highly skilled freelance workers between individualism and collectivism
Although highly skilled freelance workers are often viewed as preferring individual instead of collective, union-led negotiations (D’Amours and Legault, 2013; Pernicka, 2006), individualistic and instrumental orientations may feed into collectivism (Furåker and Berglund, 2003; Hyman, 1992; Jansen, 2020). In this sense, previous research identified patterns of collectivism in such contexts.
The predominance of the highly skilled among freelance workers (Bonin et al., 2020) suggests that professional orientations may be important to facilitating collectivism (McBride, 2011; Marks and Baldry, 2009; Marks et al., 2020; Pernicka and Reichel, 2014). While members of established professions such as medicine may adopt values and orientations of their membership and not require identity work (Smith, 2010), others, such as IT professionals, may take on collectivist orientations from their professional communities as well as from their class position (Marks and Baldry, 2009). The class position of highly skilled freelance workers is less clear than that of employed knowledge workers, as their legal status is associated with more autonomy from specific employers, and as they face different risks depending on how they are integrated into welfare states (Murgia and Pulignano, 2018). For instance, freelance workers in Germany are only partially integrated into social security systems (Keller and Seifert, 2013; Schulze Buschoff and Schmidt, 2009). They have restricted access to unemployment insurance, and they are only rarely integrated into collective agreements and company-level co-determination. In addition, in most countries, freelance workers are more exposed than dependent employees to market risks such as changes in demand and underemployment (D’Amours and Legault, 2013). Since their income depends mainly on continued access to jobs (Blair, 2001) and the corresponding remuneration, freelance workers face two main challenges: that of securing jobs repeatedly and that of ensuring subsistence and employability by negotiating working conditions from job to job.
Accordingly, mobility power (Smith, 2006) matters crucially for workers with fixed-term employment or for freelance workers: access to jobs can be used to discipline workers if control over work effort and recruitment is exerted by the client or customer, and poor working conditions are accepted for access to projects that are helpful for building a career (Blair, 2001). A scarce skill supply may allow highly skilled freelance workers to influence working conditions in their own favour, but may also help them to exert voice (Hirschman, 1970) to defend professional standards against economic imperatives, with potentially collective outcomes for dependent employees within the respective organisations (Ruiner et al., 2020).
Importantly, however, the limited duration of work in a specific organisation usually requires negotiation of terms before the job starts, especially with regard to the duration, tasks and income. Therefore, negotiation of working conditions can become more closely tied to the access to jobs, and the labour market becomes an important locus of possible conflicts and struggles over work. For these reasons, work-effort bargaining may shift from the workplace to negotiations over job access. This directs attention to avenues of recruitment, the forms negotiations take, and the relational contexts and actors involved in recruitment and negotiation.
With regard to the relational context, undoubtedly, there is some evidence that freelance workers act collectively by participating in networks, to counteract the increasing individualisation of employment relationships and to cope with risks (Blair, 2001; D’Amours and Legault, 2013; Gottschall and Kroos, 2007; Merkel, 2019). In addition, regionally clustered friendship-like networks, which influence recruitment, could counteract individualistic orientations, isolation from peers and fragmentation (Blair, 2001; Merkel, 2019; Windeler and Wirth, 2004). In considering the relationships between employers and workers in the labour market, a useful approach may be to analytically separate networks into recruitment and supply networks, which employers use to search for workers and workers use to search for jobs (Tilly and Tilly, 1994). Their composition can vary significantly, as can their relationship to negotiation processes and collectivism. In sum, the multiple functions and heterogeneous composition of networks result in a complex web of relationships linking various stakeholders committed to the negotiation of working conditions.
This raises the question of what additional actors relating to the work of highly skilled freelance workers intervene in these labour markets. Particularly relevant is how trade unions approach contingent workers. In this context, Simms and Dean (2015) identified two barriers in the mobilisation process of highly skilled, unionised contingent workers. The first barrier lies in the identification and attribution of perceived injustice. Solidarity building is particularly impaired when people feel honoured to be employed in their occupational field. The second barrier is attributed to exclusion tactics initiated by some union sections. However, research has observed that trade unions may reorient their strategies to encompass the concerns and expectations of freelance workers (Gottschall and Kroos, 2007; Heery et al., 2004; Pernicka, 2006; Windeler and Wirth, 2004) or form an association with new social movements and communities (Hyman and Gumbrell-McCormick, 2017; Stewart et al., 2009) and the personal networks that are important in flexible labour markets (Apitzsch, 2013; Saundry et al., 2012). Therefore, the roles of different types of collective actors need to be considered to understand the various manifestations of collectivism.
In Germany, the growing importance of professional associations or craft unions marks a departure from the system of co-determination at the workplace level and sectoral unions that use a class-based approach of interest representation (Keller, 2018; Pfeifer et al., 2017). While these associations may be relevant for freelance workers, their role for this group in Germany requires further research (Gottschall and Kroos, 2007).
In addition, ‘labour market intermediaries’ receive increasing attention. While digital platforms for gig work have been identified as a ‘managerial figure . . . which remains hidden and inaccessible for workers’ (Gandini, 2019: 1051), the interreference with mobility may be at the centre of other labour market intermediaries’ activities, and be complemented with additional functions. Labour market intermediaries can not only shape cross-border labour mobility (Shire et al., 2018), but also perform different functions in sectoral labour markets, such as dealing with incomplete and asymmetric information, facilitating collective action and providing support in situations of legal uncertainty for freelance workers (Autor, 2009; Koene and Pichault, 2021; Lorquet et al., 2018). More specifically, staffing agencies can be important in mediating workers’ social relations and in dealing with careers or negotiation processes (Barley and Kunda, 2004; Bielby and Bielby, 1999; Bonet et al., 2013; Lizé et al., 2022; Windeler and Wirth, 2004). This research includes labour market intermediaries such as temporary help and staffing agencies, trade unions and cooperatives. It provides typologies in relation to their functions, the extent to which employers and employees engage with them voluntarily, their origin and their efforts to change the institutional context of contingent workers (Autor, 2009; Bonet et al., 2013: 343; Koene and Pichault, 2021; Lorquet et al., 2018). While this line of research adds valuable insights on types and functions of labour market intermediaries, it is only implicitly concerned with power relations, collectivism and conflicts in labour markets (e.g. by pointing to the voluntariness of involvement of workers and employers, Autor, 2009; or by analysing how staffing agencies help to circumvent labour protection, Shire et al., 2018). Therefore, this stream of research may be fruitfully complemented with LPT perspectives, and in particular the concept of mobility power (Smith, 2006).
To sum up, collectivism is expected to evolve through a variety of interactions, struggles, organisations and networks. Insights into the complexities and dynamics of social relationships and their connection to trade unions emerge from research on the media and television industries (Apitzsch, 2013; Blair, 2001; Saundry et al., 2012; Windeler and Wirth, 2004), and other studies investigating complex contractual arrangements and orientations of knowledge workers (Barley and Kunda, 2004; Marks and Baldry, 2009; Storey et al., 2005). The integration of research findings points to the relevance of the occupational and relational contexts for understanding concerns and control strategies related to both the production process and mobility. In addition, it can be expected that the workplace is not the most important location for developing collectivism, because mobility and effort power are closely linked in freelance labour markets, and a variety of actors, in particular different labour market intermediaries, and networks may be involved in struggles and strategies for controlling job access and negotiating working conditions. However, we have only selective knowledge of the manifestations and dynamics of collectivism of freelance workers in relation to these issues and in comparison of different industries and occupations. Recent research on labour market intermediaries is helpful for better understanding who possibly intervenes in freelance labour markets and supports freelance workers, and can be fruitfully combined with LPT’s focus on conflicts and power relations between workers and employers. In this sense, the increasing proportion of highly skilled freelance workers presents an opportunity to examine how the workers articulate their concerns in negotiating job access and working conditions to reveal conditions and manifestations of collectivism, with particular attention to concerns and dynamics of associations arising from freelancers’ work, labour market position and employment relationship as well as freelancers’ embeddedness in broader communities and relations to labour market intermediaries.
Methodology
Case selection
Various studies show that the exploration of collectivism in different realms requires attention to context (Hyman and Gumbrell-McCormick, 2017; McBride and Martinez Lucio, 2011; Marks and Baldry, 2009; Martinez Lucio and Stewart, 1997; Storey and Bacon, 1993). First, the historically developed industrial relations system is decisive. In Germany, sectoral unions with a class-based approach organised around the distinction between employers and employees have long been dominant. These unions are complemented with strong works councils that represent worker interests at the workplace level and informally aid the organisation of workers into unions that engage in collective bargaining (Gumbrell-McCormick and Hyman, 2013; Müller-Jentsch, 1985; Streeck, 2005). However, this system has undergone several changes, such as decentralisation of collective bargaining, declining unionisation rates and fewer firms with collective agreements and works councils (Ebbinghaus, 2006; Gumbrell-McCormick and Hyman, 2013; Streeck, 2005). Furthermore, the solidarity between workers with different skill levels and different labour market power is currently affected by the fragmentation of collective bargaining in Germany, with craft unions and professional associations in part additionally separated into those organising employed professionals and those organising freelance workers, which are becoming more important than industry unions (Keller, 2018; Pfeifer et al., 2017; Windeler and Wirth, 2004). Second, sectors and occupations with specific institutional regulation (Bechter et al., 2012; D’Amours and Legault, 2013) may affect worker concerns and strategies of interest representation. This study compares three industries in which highly skilled freelance work is important: medicine, IT and film.
In these three industries, heterogeneous organisations deploy freelance workers. While in medicine, mainly hospitals commission services of freelance or locum doctors, and in the film industry, production firms or broadcasters formally employ or commission media workers, freelancing in IT occurs in IT companies or any other organisation in need of specialised support. In addition, labour market intermediaries are relevant especially for the matching of organisations and freelance workers, and to what extent they intervene in these labour markets will be discussed in the findings section.
Compared to freelancing in IT (Süß and Becker, 2013) and the phenomenon of locum doctors (Wilkesmann et al., 2020), freelance work in the German film industry has a long tradition, especially for directors (Sydow and Staber, 2002), and uncertainties regarding the deployment of freelancers versus dependent employees have been largely settled. As a result, those with creative autonomy and with authority over the work of assistants and other departments, especially directors, typically work on a freelance basis and hold academic degrees, so that directors are of particular relevance as representatives of the film industry in our study. However, with regard to employability and employment stability – that is, the risk related to the need for continual access to jobs – the three fields differ markedly. For instance, in contrast to the IT and film labour markets, the doctors’ labour market is restricted by formal criteria relating to competition and access. In this sense, locum doctors’ formal professional qualification not only is central to their employability but also affects the control of work, which is predominantly based on professional standards.
In contrast, film and IT projects allow greater personal or direct control and control of outputs as well as self-control and identification with the product (see also Apitzsch, 2013; Blair, 2001; Smith and McKinlay, 2009). Struggles over the organisation of work also occur in medicine, such as those related to the role of economic and organisational versus professional criteria in medical decisions (Ruiner et al., 2020; Wilkesmann et al., 2020). In addition, the deployment of freelancers in hospitals and IT projects is less clear and more legally disputed because of freelancers’ strong integration into the organisational core. Thus, a judgement of 2019 declared doctors to be necessarily involved in organisational processes and hierarchies so that they cannot legally be considered freelance workers. Since that judgement points to specific regulatory requirements for hospitals (Ehlen et al., 2022), it is unclear whether this reasoning can be applied to freelancers in other areas. However, the former locum doctors are still allowed to legally commit themselves to an intermediary, such as a temporary employment agency or a cooperative, and are even likely to do this because of the better working conditions offered by these intermediaries. Thus, this development in jurisprudence does not contradict our consideration of collectivism among freelance workers in medicine but supports its relevance.
Freelance workers’ income and employment security may also vary among the three industries, leading to different needs for collective support. Freelance workers in the film industry maintain and enhance their employability through their track record and networks, but their market exposure and risks have grown in recent decades (Baumann, 2002; Blair, 2001; Bomnüter and Schulze, 2019; Storey et al., 2005; Sydow and Staber, 2002). An IT freelance worker’s employability depends heavily on the worker’s professional qualification and track record (Süß and Becker, 2013), and film freelance workers’ remuneration varies widely, depending on the worker’s employability (Bührmann et al., 2013).
Examination of the differences in the degrees of professionalisation and in the working conditions of these three industries promises to yield insights into individualistic or collective concerns and collective action of freelance workers.
Data and methods
Our analysis relies on two qualitative studies. In 2015, we conducted semi-structured interviews with 10 freelance workers and 19 representatives of intermediaries (of four trade unions, seven professional associations, seven staffing agencies and two cooperatives, with a head of one cooperative also representing a professional association) in medicine, IT and film in Germany. This study focused on the negotiation of working conditions of freelance workers. In 2019 and 2020, we conducted semi-structured interviews with four freelance workers in the IT industry and representatives of 16 intermediaries in medicine and IT (three trade unions, six staffing agencies, four professional associations, and three cooperatives with a head of one cooperative also representing a professional association). This second study focused on the effects of engaging freelance workers on the organisational and inter-organisational levels. A representative of one of the leading staffing agencies in IT and a head of a cooperative and professional association of locum tenens physicians, and a head of an IT cooperative have been interviewed in both of the studies. We gained access to all of the intermediaries through calls in relevant newsgroups and through telephone acquisitions. We focused on smaller and specialised staffing agencies in the respective fields as well as on generalists that are active in more than one field (e.g. IT and medicine). We interviewed representatives of professional associations and trade unions of relevance to the occupational groups under study. Freelance interviewees were approached through direct contact with gatekeepers, calls in relevant newsgroups and social networks, client companies, labour market intermediaries and through telephone solicitation. They were selected according to three criteria (Eisenhardt, 1989): respondents had to work mainly as freelance workers, had to hold an academic degree, and had to have at least three years of freelance work experience. These criteria increased the homogeneity of the sample, helping to concentrate the comparison on occupational and sectoral differences. In both studies, we used investigator triangulation during data collection and analysis to increase credibility and reflexivity: research teams (consisting of three researchers in Study 1 and five researchers in Study 2) continually discussed sampling and interview strategies and necessary adaptations.
We have combined these two studies because the specific focus on freelance workers and intermediaries in medicine, IT and film provides valuable insights for the underlying research question of exploring and understanding manifestations of collectivism. In addition, the second study allows validation of observations of the freelance labour markets in medicine and IT, especially with regard to ongoing disputes of the employment status and legal reforms of freelance work (Ehlen et al., 2022).
Interviews ranged between 41 and 130 minutes with an average of 71 minutes and were audio-recorded and transcribed. Interviews of freelance workers in Study 1 started with a narrative stimulus on how the respondents became self-employed, followed by questions on their cooperation with colleagues and the organisation of their work, on gaining access to jobs, and on perceptions of the employment situation and prospects, as well as on their expectations of and experience with intermediary actors in the labour market (see online Appendix for details). While in Study 2 the emphasis was on cooperation and conflicts with colleagues and supervisors in the hiring company, all of the aforementioned themes were also addressed by specific questions. Interviews with intermediaries in Study 1 centred on their evaluation of the labour market situation and working conditions of freelance workers, on their perception of needs and expectations for interest representation or services, and on their relation to other collective actors in the respective field. In Study 2, the topics mentioned above were expanded to include cooperation and conflicts with colleagues and supervisors in the hiring company. We discuss in the findings section where we observed changes in the fields under study over time.
The data analysis was based on qualitative content analysis (Mayring, 2000) using the software MAXQDA 2020. The analysis contained a deductive application and an inductive development of codes. Initially, codes were derived from the theory and applied to the interview material, such as the codes labour supply networks and recruitment networks, or professional identity. Codes were also developed inductively as key issues were discussed and subsequently condensed and systematised to codes. Inductive codes were, for example, specific concerns of freelance workers, such as criteria for choosing between job offers or evaluations of what is important at work and in negotiating working conditions, or bogus self-employment. For each code, we formulated definitions, found examples, and defined coding rules. We controlled for rater influence by involving five people in the data interpretation process. Since the (mis-)match of encoded codes was discussed in team meetings, the coding system was jointly developed, leading to its clarification. With regard to our questions of interest, we reached a sufficient degree of theoretical saturation and informational redundancy.
Empirical findings
The presentation of results is organised around the two main concerns of freelance workers – job access and working conditions – as the negotiation of both usually occurs jointly in these highly flexible labour markets. Within these sections, we show dynamics in the relation between different types and strategies of labour market intermediaries and the development of freelance workers’ collectivism.
Individualism and collectivism in recruitment and labour supply networks
In the case of medicine, locum doctors used a mix of labour supply networks comprising recommendations by colleagues, direct contact with former clients, and agencies. Both studies showed that agencies provide an important way to get a job, especially for locum doctors just entering the market, whereas senior locum doctors were less dependent on agencies to establish contact with hospitals, as a locum doctor pointedly described:
I prefer not to work with agencies since I learned that they add up to 20% to the price. I agree with agencies trying to earn something, that is ok, but these prices are too high. . . . In the end, this reduces my income. And I do not depend on them, so I try to get along without them. They were helpful for me, especially with regard to my first clients, to enter the market by working with agencies and gain more experience while having the agency as an additional security. But at the moment I get along completely without them. (Study 1, freelance worker medicine-03:75)
Interestingly, some locum doctors interviewed still used administrative and other services of agencies while seeking job offers and discussing requirements and qualifications with the hospitals themselves. This practice is partly a reaction to the risk of mismatches owing to overly broad descriptions of the services required, which can be hazardous for patient care as well as for locum doctors’ reputation.
The locum doctors interviewed noted that some agencies reacted to the growing competition for clients by establishing exclusive contracts with hospitals – a practice that crowded out other agencies and attempted to monopolise clients’ recruitment networks. However, hospitals still used a variety of recruitment networks and tended to rely on direct contact with locum doctors because they preferred to work with those they had worked with before. Also, locum doctors established a cooperative that matches requests from hospitals with their members, who mainly bring in contacts with clients, as a cooperative representative noted:
[T]his is how orders come about, so that the arrangements take place directly on the spot. For me, this is the very, very best. This shows that the doctor would like to work there, otherwise he wouldn’t have accepted the jobs, and the [hospital’s] department is totally satisfied with him, otherwise they wouldn’t offer him any more jobs. (Study 2, intermediary medicine-43:56)
Thus, open and direct contact between professionals and hospitals has a strong advantage over the matching through third-party agencies that could not always ensure the matching of qualification demand and supply. To ensure the qualification of the locum doctors, only peers were involved in matching, and the cooperative uses a certificate developed by the professional association of locum doctors.
The labour supply networks of IT freelance workers consisted of contacts established through internet platforms, direct contact with former clients, ‘cold calls’ and through colleagues who recommended them when they could not work for a client themselves or when there was need for additional personnel. However, as the quotation below illustrates, the interviewees estimated that this type of job search has become less relevant over the years, since large firms increasingly use the services of staffing agencies:
Formerly, clients commissioned services only on the basis of personal contact. This is not possible anymore. That means that getting in contact with someone, and development of job offers out of this contact is replaced by getting in contact with an intermediary. (Study 1, freelance worker IT-02:28)
Staffing agencies have apparently crowded out direct contacts between firms and freelance workers and established themselves as the primary recruitment network for firms, a development that even intensified since Study 1, partly owing to the legal uncertainty related to the deployment of freelancers and the increasing skill scarcity. However, in the experience of freelance workers, the quality of the matching varied, so that direct contact with clients was necessary to find out which services were requested. Thus, IT freelance workers often described staffing agencies as being unfamiliar with the technical issues relating to the engagement and perceived those agencies as gatekeepers that primarily monetise the contact to clients.
Well, they earn good money for making a little phone call twice. . . . They just have the contact to the customer. They often don’t have the slightest say in the matter. And, yes, that’s why they have a purely intermediary function. (Study 2, freelance worker IT-41:41)
As a consequence, some of the freelance workers interviewed relied even more on professional networks to explicitly bypass staffing agencies. Moreover, and similar to our observations in medicine, to achieve an even more effective alternative to profit-oriented agencies some IT freelance workers formed cooperatives, which mediated the contact between freelance workers and firms. In contrast to the cooperatives described in the literature (e.g. Cheney et al., 2014), these cooperatives did not function as an alternative for organising service production. Rather, they organised the matching and negotiation processes that formerly took place in personal networks and indirectly through professional associations, and then became increasingly mediated by staffing agencies. Freelance workers viewed the cooperatives as more transparent to both clients and freelance workers about the profits they gain and, while charging less for matching, as able to offer more qualified experts to firms and higher income for freelance workers. In addition, freelance workers viewed the cooperatives’ internal organisation as resting on more democratic principles than those of profit-oriented firms.
To secure access to jobs, film directors formerly relied on direct contact with producers and departments – relationships they had established during previous projects and maintained through activities such as calling and meeting them regularly and visiting events such as film festivals. Agencies in the film industry mediated labour supply and recruitment networks in light of their clients’ career goals and on the basis of intense contact with their clients, knowledge of their creative style, and clients’ personal characteristics, such as being ‘easy going’ or sticking to the given budget and schedules. They also used detailed and up-to-date knowledge of the market, for example about projects that were currently underway or being planned, and about other directors who potentially competed with their clients. However, the freelance workers interviewed noted that their own close relationships with colleagues and producers resulted in jobs more often than contacts provided by agencies, confirming earlier findings on the limited use of staffing agencies for matching in the film and television industry (Baumann, 2002).
Despite the differences in the formal regulation of labour market access, we observed interesting similarities in concerns related to recruitment. In all three fields, freelance workers are concerned not only with securing access to jobs, but also with the quality of matching processes, which also bears on their future employability. This concern, especially where staffing agencies attempted to monopolise recruitment networks, drove them to self-organise into cooperatives as labour market intermediaries, thus transforming individual concerns into collective interests.
Individualism and collectivism in negotiation processes
Freelance workers’ mobility led to the challenge to negotiate and administer working conditions and contracts from job to job. An initial look at central concerns of highly skilled freelance workers gives the impression of prevalent individualistic orientations. In both studies, of most importance for highly skilled freelance workers were autonomy and self-determination of work, and these aspects were reported to be the primary reasons they became self-employed. The following quotation illustrates this observation:
My job is just like − or in IT consulting in general is that − if you’re employed somewhere, then you can be sent all over Germany. I would like to be able to decide for myself when and where I am at home and so on. (Study 2, freelance worker IT-02:06)
The freelance workers interviewed perceived permanent employees’ working conditions as incompatible with their needs for autonomy and income and, as freelance workers, they experienced more leeway in negotiating their fees or work intensity and autonomy. However, especially for locum doctors and film directors, work intensity and autonomy within the labour process were related to professional norms, such as being able to make work-related decisions according to medical knowledge or, in the case of film directors, according to both aesthetic standards and expectations of originality. Consequently, occupational communities provided important reference points for freelance workers’ concerns and effort power irrespective of the degree of professionalisation in the sense of standardised training and control of labour market access, as well as control of professional practice (Stinchcombe, 1959). Such arrangements may provide sources for collectivism outside the firm. In IT projects, we observed the development of strong senses of shared interests in joint problem solving and cooperation, which side-lined differences in employment status and possible conflicts about autonomy at specific workplaces. Therefore, sectoral specificities such as professional orientations have a bearing on what is expected of matching and negotiation by labour market intermediaries.
In all three industries, the freelance workers interviewed considered the professional associations to offer essential support, such as opportunities for networking and for exchange about professional issues with colleagues. Of particular importance was information about the fees commonly paid. The possibility of orienting their expected hourly rate on what has been discussed among the members and colleagues was a key prerequisite for successful individual negotiations:
[I]n my experience, if you say a price too high, you calculate yourself out of the market, because these intermediaries usually call several freelance workers and there is always one who is cheapest. (Study 1, freelance worker IT-01:88)
In addition to considering their own income, the freelance workers interviewed referred to local or occupation-specific standards they try to maintain (Windeler and Wirth, 2004), so that individual negotiations indirectly influenced working conditions of others:
If a producer offers me less, say 10, or 5% less, then I would say ‘Sorry, then I won’t do it.’ Because you have to be careful to not beat down prices. (Study 1, freelance worker film-01:91)
In addition to providing consultations on prices, professional associations offered legal advice, especially on freelance workers’ social security status, information about working conditions, and model contracts or checks of individual contracts. In film, this service was considered especially helpful for individual negotiations, since the requests for changes in contracts can be presented as objective problems identified by third parties, providing a basis for negotiation of the most important changes. Information on fees and contracts and legal advice on questions regarding social insurance and bogus self-employment were important for freelance workers in all three industries. In addition, specific legal issues were of concern, such as liability insurance in IT and medicine and copyright law in film. While trade unions increasingly offer consultations on how to calculate fees and on legal and social insurance issues, the freelance workers interviewed considered the unions to more strongly favour dependent employment and to provide less expertise on the labour market and working conditions of freelance workers in specific occupations.
In contrast to the marginal role of trade unions, membership in professional associations indicates collectivist orientations and occupational identities. The associations were perceived as a manifestation of solidarity with like-minded colleagues, which freelance workers perceived as helping to strengthen the association and enabling them to engage in collective bargaining or to exert political influence:
I appreciate that [the professional association] takes care to engage broadcasters in revising collective agreements. . . . I could say I take an agency that helps me personally, which would be a more intense and individualised support. . . . But I consider this a question of solidarity, to be a member . . ., especially working in such a small professional community, to have a strong association. (Study 1, freelance worker film-02:79) It is an association of persons working for the same goal. (Study 1, freelance worker medicine-01:83)
While negotiations, apart from collectively defined minimum standards in some areas of the film industry, were primarily carried out individually, they drew on the support of professional associations or labour market intermediaries. However, the role of intermediaries in the negotiation of working conditions of freelance workers varied both between and within the industries observed.
In IT, interests regarding the distribution of the fees between freelance workers and staffing agencies frequently differed. For example, a cooperative negotiating high fees with firms would leave a higher share for the freelance worker, who could then adjust the contract with the client to his or her needs.
In medicine, staffing agencies were important not only for job access but also for administrative issues, for providing contract templates and, for some locum doctors, for negotiating pay, working hours, lodging, or liability insurance issues. Some of the locum doctors interviewed preferred to negotiate with clients themselves and used the information staffing agencies provided in job offers to get an idea of appropriate demands in their individual negotiations. Cooperatives were regarded as helpful in providing legal advice and consultations, an overview of appropriate fees, and template contracts, along with offering the possibility to exchange information about working experience or clients. Cooperatives in IT and medicine considered themselves as acting on behalf of freelance workers, whereas agencies viewed themselves as a neutral mediator but were sometimes suspected of acting on behalf of clients.
In the film industry, support in negotiating working conditions was one of the most important concerns freelance workers mentioned. Owing to their growing complexity over time, contracts involve complicated agreements with respect to not only working conditions and fees but also copyrights, liability and other issues. Negotiation of contracts requires legal expertise, detailed knowledge of further developments in the market, and knowledge of major sources of conflict in contract design and in everyday work in projects. This need for legal support goes beyond the questions on the legal status of freelance workers with regard to social insurance issues and points to how sectoral differences in the division of labour relate to legal uncertainty. Respondents noted that tension often exists between the broadcaster’s and producer’s budget and time constraints and creative personnel’s interest in maximum autonomy. For instance, film directors have a strong interest with regard to camera directors, actors, location and duration of the shooting, and other conditions that affect the quality of the film and the visibility of the creative input, and hence the future market position of the director. In these conflicts, agencies can clarify the needs and ideas of both sides and could intervene in conflicts. In contract negotiations and by intervening in conflicts at work, agencies in the film industry, much like professional associations, act explicitly on behalf of the freelance workers as their clients.
Summary and comparison: Collectivism in medicine, IT and film
Findings point to striking similarities, but also to differences in manifestations and dynamics of collectivism in medicine, IT and film. Most importantly, in all three industries we observed a need for collective support related to access to jobs and to negotiating working conditions. Especially in medicine and film, negotiation strategies and the search for autonomy have a bearing on employability and they relate to central professional norms, such as making medical decisions that are based on professional knowledge and standards or pursuing the ideal of artistic freedom and originality in film. Feelings of belonging to professional communities are also important. This is in line with recent accounts which emphasise that explanations of manifestations of collectivism cannot be restricted to interests stemming from an employment relationship or a specific market position (Marks and Baldry, 2009; Marks and Thompson, 2010). These concerns are strongly related to occupational communities, and thus to collectives that extend beyond the firm and the workplace. Freelance workers in IT shared the concern of matching processes allowing them to find projects which suit their qualification and allow them to stabilise their reputation, but apart from that they were more concerned with close cooperation in projects, and autonomy with regard to place and time of work.
Professional associations were essential as a venue to connect within the occupational community and for providing legal advice and support in negotiations that are tailored to the needs and occupational identities of the highly skilled freelance workers in each industry. In IT and medicine, they even separated into professional associations for freelance workers. Although trade unions reach out to freelance workers, the specific need for support we observed in different sectors points to limits of unions’ strategies. In our studies, freelance workers perceived trade unions as mainly providing general support with an orientation toward standard employment, in contrast to tailored support provided by other labour market intermediaries and personal networks. This tailored support requires in-depth knowledge of sectoral and occupational specificities which may go beyond trade unions’ capacities, e.g. to judge a director’s creative achievement in the context of the division of labour in film projects, or professional autonomy and risks of bogus self-employment in hospitals and IT projects.
Similarly, matching in all three fields, despite differences in the standardisation of professional skills, appears to require in-depth knowledge of occupational specialisation and organisational and personal characteristics. Consequently, the freelance workers interviewed preferred to get jobs through their own networks, specialised staffing agencies or cooperatives, and labour market intermediaries were more important for administrative support, for providing market information, or support in legal issues. The comparison also highlights the importance of constellations of negotiating access to work and working conditions, and of looking beyond conflicts at the workplace and with direct employers. While staffing agencies in the film industry negotiated on behalf of film directors vis-a-vis production firms and broadcasters, in IT and medicine, staffing agencies were partly perceived as populating clients’ recruitment networks and acting for their own profit or for client firms, thus spurring strategies of solidaristic self-organisation in networks and cooperatives and complementing traditional forms of collectivism such as professional associations or trade unions. Table 1 summarises constellations of negotiation and collectivism of highly skilled freelance workers in medicine, IT and film.
Comparison of negotiation constellations and collectivism of freelance workers.
Our studies support findings regarding the use of networks and interest organisations in relation to occupational identities and career goals, and regarding the need for collective support in setting standards in flexible labour markets. They more generally support observations of the variety of functions and support which labour market intermediaries can provide to contingent workers (Autor, 2009; Koene and Pichault, 2021; Lorquet et al., 2018; Saundry et al., 2012; Windeler and Wirth, 2004), and complement them by pointing to sectoral differences of requirements for matching and negotiation, of power relations, sources of conflict around effort and mobility power, and ideational bases of collectivism.
Discussion and conclusion
This research explores the evolution of collectivism in flexible labour markets in knowledge-intensive services in which workers are less attached to firms, and in which freelance workers are usually expected to hold individualistic orientations. Using the paradigmatic case of highly skilled freelance workers, we investigate manifestations of freelance workers’ collectivism related to concerns arising from workers’ experience at work, from their labour market position and employment relationship, and from their embeddedness in broader communities and relations to labour market intermediaries. By contextualising these within the national and sectoral industrial relations systems and the different types and strategies of labour market intermediaries, we attempt to specify conditions and dynamics which lead to new forms of collectivism.
Our research identifies a new form of collectivism that becomes visible when relating workers’ and employers’ strategies and struggles around mobility power to the specific concerns of freelance workers. Freelance workers contested existing labour market intermediaries and developed new forms of collectivism generating space for a better and more solidaristic representation of qualifications, needs and requirements, either through personal networks or through cooperatives. We define this form of labour market collectivism as patterns of association that relate to new developments and conflicts in the labour market, which we illustrate in Figure 1.
Labour market intermediaries and labour market collectivism in medicine, IT and film.
For our analyses, we complemented LPT perspectives by examining a broader range of collective actors and the conflicts arising at the intersection of recruitment and negotiation processes to expand knowledge of manifestations of collectivism which to date frequently focuses on the workplace as a locus of association and building solidarity. The high inter-firm mobility of freelance workers increases the importance of the intersections of labour supply and recruitment networks (Tilly and Tilly, 1994) which are populated by different intermediaries, and in which negotiations of job access and working conditions coincide.
The analysis points to dynamics that evolve from relations between concerns of job access and negotiating working conditions and intermediaries’ strategies of populating recruitment and labour supply networks. When staffing agencies attempt to monopolise recruitment networks and thereby depart from the model of voluntary participation which has characterised temporary help agencies (Autor, 2009), or when they are perceived to interfere with freelance workers’ job access and negotiation strategies, they elicit the establishment of cooperatives and a stronger reliance on personal contacts for matching labour supply and demand and defining working conditions.
In general, our study supports the findings of earlier labour process analyses regarding the dynamic interaction between employer strategies and workplace collectivism (Atzeni, 2009; Dundon and Dobbins, 2015; Martinez Lucio and Stewart, 1997; Storey and Bacon, 1993), and transposes them into labour markets. Especially when a growing workforce is increasingly detached from a single firm and is more mobile, the labour market and wider professional communities can become the locus of collectivism. Therefore, we support that communities outside of firms are relevant to collectivism (McBride and Martinez Lucio, 2011; Marks and Baldry, 2009) and that looking beyond the workplace and effort bargaining is critical to understanding the dynamics of collectivism of freelance workers vis-a-vis firm strategies. Mobility power (Smith, 2006) matters crucially, recruitment and negotiations of working conditions are strongly intertwined. Collectivism – either as networks or as cooperatives that assume the functions of labour market intermediaries – develops in reaction to employers’ or staffing agencies’ strategies. These findings rather contradict the argument that collective action follows the typical injustice–leadership–action process constituted by mobilisation theory (Kelly, 1998). In partial alignment with Atzeni (2009), we observe conflict dynamics in the relation between different types of labour market intermediaries resulting in solidarity formation in social networks and cooperatives.
In addition to expanding on LPT’s insights on labour markets and mobility power, this research contributes to knowledge on labour relations of contingent workers, labour market intermediaries and cooperatives. Research on contingent work, especially in creative industries, has provided valuable insights on the role of social networks in recruitment, and, in rarer cases, on how these relate to trade unionism (Blair, 2001; Saundry et al., 2012). Previous research also yielded insights into the functions of various intermediaries in flexible labour markets (Autor, 2009; Koene and Pichault, 2021; Lorquet et al., 2018), which, we argue, can be fruitfully combined with LPT perspectives on conflicts and power relations around the employment relationship.
Cooperatives have rarely been considered in studies of contingent employment, particularly in relation to alleviating job insecurity and providing venues engaging with relational activities to replace hierarchical interaction with distributed peer collaboration (Bajard, 2020; Charles et al., 2020; Resch and Steyaert, 2020). Related to this, we have shown that cooperatives can be considered a new form of collectivism that combines the functions and logics of membership-based and labour market intermediaries (Bonet et al., 2013). Therefore, we complement existing research on types and functions of labour market intermediaries with unanticipated insights into shifting functions of intermediaries and into dynamic – both latently conflictual and cooperative – relations between these intermediaries and with freelance workers around mobility power.
In addition to reflecting our findings in light of research and knowledge on collectivism and labour market intermediaries, we conclude with situating our observations in industrial relations research. While control of labour market access may connect to classical trade union strategies such as closed shops (Heery et al., 2004), this connection marks a significant departure from Germany’s system of interest representation. That system takes a class-based approach of organising and representing interests of a core workforce in dependent employment relationships via works councils and trade unions’ collective bargaining. As a result, it only indirectly influences labour demand through corporatist arrangements (Streeck, 2005). The findings of this study support the importance of collective actors’ approach to contingent workers (Ebbinghaus, 2006; Gumbrell-McCormick and Hyman, 2013; Simms and Dean, 2015), but also suggest taking a broader view on how different types of labour market intermediaries relate to central concerns of these groups of workers, how these are shaped by sectoral and occupational specificities, and what different constellations of negotiation and cooperation result from this.
The German industrial relations system with its class-based approach, its abstraction from specific occupational interests, and its reliance on workplace-related strategies of organising therefore seems to provide the institutional space for new forms of organisation which then contribute to the fragmentation of interest representation and bargaining. Yet, the new forms of labour market collectivism, in particular social networks and cooperatives, currently lack a broader engagement with the political and institutional context, thereby being close to the ‘embedded fixers’ and ‘pragmatic experimenters’ identified by Koene and Pichault (2021). In addition, they lack the approach of broader inclusiveness beyond the occupation, and the cooperation with trade unions identified in other new social movements in a ‘variable geometry of resistance’ by Hyman and Gumbrell-McCormick (2017). Thus, a reorientation of collective actors’ strategies (Heery et al., 2004) and research into labour relations in the context of employment change needs to take into account possible complementarities between different forms of support provided by different types of intermediaries, and of conflicts that unfold not only in the workplace but also in the labour market.
To better understand the role of the institutional setting for the development of new forms of collectivism, future research should address functions and forms of collectivism, and especially cooperatives, in other industries and other employment systems and validate findings of these explorations with quantitative studies. Also of interest would be the examination of how traditional and new forms of collectivism represent the interests of low skilled freelance workers in relation to their customary clientele, to compare occupational subfields and include online platforms as new intermediaries (Gandini, 2019). Such an analysis could help to explain the role of skill scarcity for the development of labour market collectivism. Finally, research regarding individualism and collectivism in a flexible labour market would profit from interdisciplinary approaches toward collectivism and worker identities, and from a more fine-grained analysis of the work process and the emerging solidarities between contingent workers and the more stable part of the workforce in everyday work situations.
Supplemental Material
sj-pdf-1-eid-10.1177_0143831X221120534 – Supplemental material for Labour market collectivism: New solidarities of highly skilled freelance workers in medicine, IT and the film industry
Supplemental material, sj-pdf-1-eid-10.1177_0143831X221120534 for Labour market collectivism: New solidarities of highly skilled freelance workers in medicine, IT and the film industry by Birgit Apitzsch, Maximiliane Wilkesmann, Caroline Ruiner, Mona Bassyiouny, Ronny Ehlen and Lena Schulz in Economic and Industrial Democracy
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
We gratefully acknowledge the funding of the research on which this article is based, granted by the Mercator Research Center Ruhr for the project ‘Collective individualisation – individual collectivisation? On the negotiation of working conditions in the field of high qualified solo self-employed workers’ (grant number Pr-2014-0041) and by the DFG/German Research Foundation for the project ‘Between cooperation and competition – external workers in knowledge-intensive fields’ (grant number 384705290).
Supplemental material
Supplemental material for this article is available online.
Author biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
