Migration and labour unrest during the pandemic: Studies from Germany and Austria

This paper presents the results of research, which highlights the situation during the pandemic in sectors characterised by low wages and a high turnover of workers. The empirical basis is formed by company case studies in the meat industry, postal services, and mask production in Germany and Austria. This paper discusses the significance of different locations (at and beyond the workplace) and forms ( ‘ exit ’ and ‘ voice ’ ) of labour unrest in sectors of the economy that are characterised by a predominance of the use of migrant labour. It questions how conflicts over migrant labour have been articulated and possibly changed in the pandemic, and what factors may have contributed not only to an upsurge but also to the containment, regulation, and repression, of labour unrest.


Introduction
During the COVID-19 crisis, a high degree of dependence on migrant labour became evident in many sectors of the European economies that are important for sustaining everyday life, from elderly care to parcel services and from agriculture and the food industry to retail (Askola et al 2021;Cook et al 2020;Leiblfinger et al 2021;Molinero-Gerbeau et al 2021;Birke 2022).While unemployment increased in some parts of the economy characterised by precarious employment and low wages, the tension between growing production pressure and the need to maintain elementary social functions on the one hand, and the restricted mobility of workers due to border closures and lockdowns on the other, led to considerable conflicts such as mass infections, as in the meat industry or online retail; the extension of working hours to the point of exhaustion, as in the care sector; and growing public criticism of the 'exploitation of workers' in these and many other cases.
We present in this text the results of research, which highlights the situation during the pandemic in sectors characterised by low wages, a high turnover of workers, and a predominance of the use of migrant labour.Our aim is to discuss forms of labour unrest, drawing on research on migrant labour in the German meat industry and mask production and postal services in Austria.In doing so, we will refer to the power resources approach as developed in international research on workplace conflicts (Brookes 2013;Schmalz & Dörre 2014;Wright 2000).Following a critical appraisal of this approach in the literature, we will also explore its limitations.Our main argument is that the issue of labour unrest (and its absence) in workplaces with exploitative working conditions, can only be understood with reference to the 'multiple precarity' of many migrant workers.
We understand the concept of 'multiple precarity' as a heuristic that makes it possible to comprehensively analyse the specific work and life situation of migrants, while at the same time referring to corresponding sociological concepts.Similar to, what for example, feminist sociology has theorised for an (expanded) understanding of work, the concept of 'multiple precarity' also suggests a redefinition of the field to be investigated: in contrast to a critique focussed only on work processes and labour relations, it requires the development of an understanding of how different precarisation processes interact.
By the concept of 'labour unrest,' which is also central to this paper, we mean, in a broad sense, everyday workplace conflicts that may or may not gain visibility in the general public and may or may not take different forms in the spectrum between 'exit' and 'voice' (and also 'loyality') (Hirschmann 1970).In this paper, we therefore focus on three questions: 1. How were conflicts around migrant work articulated and transformed in the pandemic?2. To what extent did the concrete forms of labour unrest express a 'multiple precarity' of migrant workers? 3. What factors may have contributed not only to an upsurge but also to the containment, regulation, and repression of labour unrest?
Empirically, this paper is based on company case studies in Germany and Austria.We conducted semi-structured qualitative interviews with employees in the meat industry in Germany and in postal services and mask production in Austria, and expert interviews with management and trade unionists and other experts from outside the studied companies.The interview sample consists of material from two projects, first a total of 27 interviews and participant observations of consultation situations and strikes in the German meat industry.These data are part of a much more comprehensive study that examined five industries between 2017 and 2021 with funding from the state of Lower Saxony (Birke 2022;Bluhm et al 2021).In Austria, research on migrant labour during the pandemic was conducted in the context of a project of the University of Vienna funded by the Chamber of Labour (Arbeiterkammer) in the field of temporary agency work in a parcel distribution centre and a mask production company (n = 18).Methodologically, the projects were oriented to the strategy of case studies (Yin 2017) and, more precisely, company case studies (Pongratz & Trinczek 2010)a strategy often applied in sociology of work.However, specific features of migrant work made it necessary to re-define this strategy in two important aspects.First, in both studies, the empirical material was mainly obtained on the basis of interviews conducted outside of the companies examined.The conversations had high biographical narrative components, and while our questionnaires addressed the work process, the condition of 'multiple precarity' also played a major role as a fundamental problem of the life situation of many migrant workers.Second, considering the fragmentation of companies and employment, the cases had to be defined in a broader sense to include temporary employment agencies and subcontractors.In both projects, content analysis methods were used to evaluate the interviews, which were supplemented by fine analyses for particularly dense passages (Schreier 2012;Witzel 1985, p. 143).
The starting point for our approach of comparing research carried out in two different countries was the observation of strong similarities that we found in our empirical data regarding working conditions during the pandemic.This was the case despite different labour and migration law regulations in Germany and Austria as well as different economic sectors where state regulations, the existence of trade unions, and contract bargaining play a different role.
Largely independent of such factors, which are otherwise considered crucial for industrial relations, the migrant workers interviewed seemed to face the same challenges: They were particularly affected by overtime, health hazards, and labour rights violations.Furthermore, at that time, the industries studied were characterised by economic upswings, while they were hotspots of mass infections and/or health hazards.And finally, all fields of work examined were characterised by labour shortages and a pronounced fragmentation of the labour force due to a high proportion of temporary agency work or subcontracting.This fragmentation overlapped with ethnicised/racialised divisions in the workplace, which led to a split workforce.
Based on these commonalities, our common interest was to understand how labour unrest was possible under such unfavourable conditions.Therefore, from the wealth of existing material in both studies, we focus here on data collected in the pandemic that specifically point to motives for the absence and/or articulation of labour unrest.In addition to the explicit passages in the corpus dealing with labour conflicts and resistance, we compared the material in more depth on the basis of two other criteria: first, the power resources available to migrant workers and second, the multiple precarity (due to housing situation, residence issues, or family situation).
We will proceed as follows.After a theoretical reflection on power resources and mobility, we will, in 'Power resources and mobility: theoretical reflections' section, discuss workplace conflicts and labour unrest during the pandemic, based on an analysis of the series of interviews mentioned above.We start with the German example ('Labour unrest in the pandemic: the meat industry in Germany'), followed by the Austrian case ('Working conditions and labour disputes in the Austrian postal service and mask production').Based on our empirical evidence, we will turn to the interconnected and interwoven character of power relations in the different spheres (labour market, labour process, residence rights, migration regime).Our thesis is that multiple precarity and the fragmentation of the working class are important elements to identify the dynamics of these struggles.Against this background, the text concludes with a reflection on the ambivalence of labour unrest and the mobility power of migrant workers and the subsequent question of (trade union) organising.

Power resources and mobility: theoretical reflections
For many years, research on power resources of workers vis-à-vis the employers has been conducted both in the global north and south and in different regulatory frameworks (Brinkmann et al 2008, pp. 30-32;von Holdt & Webster 2008).One of the most important starting points of such research was the old insight that workers' struggles cannot be reduced to trade union action and labour unrest is not limited to the point of production.In fact, this remains a central argument and probably one of the few common grounds of, among others, feminist works on social reproduction, approaches of operaismo and postoperaismo, and studies on informal labour or critical history.In this paper, we define 'labour unrest' as conflicts that do not necessarily involve trade unions and or institutional structures, although at the same time they have a decisive influence on the balance of power in everyday work and can be a starting point for publicly visible actions and strikes (Silver 2003;critical: Birke 2016).Furthermore, it is important to note that there is a conceptual conjunction between the notions of labour unrest and multiple precarity; both terms allow a reference to the power relations in the labour process, but at the same time understand it as also shaped by relations of reproduction.
Within the framework of a heuristic, in which the balance of power between labour and capital can be typologically captured, several authors developed a concept of 'power resources' (PRA) (Jürgens 1984;Silver 2003;Wright 2000, p. 962). Subsequently, Silver (2003) distinguishes structural power as primary power resource 1 into workplace bargaining power and marketplace bargaining power.While workplace bargaining power depends on the position of workers in the production process or at other points in the capital cycle (logistics) and is mobilised by the refusal to continue working (through, e.g., strikes, sit-ins, or sabotage), marketplace bargaining power is the product of a tight labour market.Marketplace bargaining power is particularly given in phases of low unemployment but is at the same time limited by fragmentation into core and marginalised workforces, as well as by racial and gender segmentation.In contrast to structural power, associational power requires the emergence and organisational process of collective actors to develop and implement strategies (Silver 2003, 13).Schmalz et al (2018) suggest adding two more power resources, institutional power and societal power, to Wright and Silver's original concept.While institutional power can be understood as the result of struggles and negotiation processes based on structural power and associational power, societal power (coalitional power and discursive power) encompasses 'the latitudes for action arising from viable cooperation contexts with other social groups and organisations' (Schmalz et al 2018, p. 121).
A critique of the PRA cannot be addressed in detail here.However, the constant expansion of the term 'resources' suggests a reification of power that seemingly exists beyond the social relationship between capital and labour, which is problematic (Birke 2016).Thus, when we speak of 'mobility power', this is not meant as a further differentiation of the typology.The point is rather to investigatein extension of the heuristic aspect of PRAwhether migration relations, and in particular multiple precarity, produce different forms of labour unrest.While the migration of capital has been strongly addressed in discussion of PRA, this has hardly been the case for the migration of labour so farin this respect, too, this paper aims to contribute to the further development of the debate.
Furthermore, in the debate on PRA, 'labour unrest' has been usually seen as a 'new' source of empowerment for trade unions, while in the understanding of Wright and Silver it is a source of empowerment for workers in the first instance.However, it is indeed necessary to distinguish whether one is talking about workers' power or trade union power (see also Nowak 2018, p. 353).Since trade unions have played a marginal role in the sectors studied, this is an important differentiation in relation to our research.
However, the debate mostly, and until recently, remained centred on the 'trade union question'.In this context, questions like which workers the trade unions represent, and what different strategies of everyday struggle might be used by different groups of workers, remained unanswered, at least in dominant IR studies.This is problematic insofar as labour conflicts cannot be adequately explained with an abstract reference to 'the workers'.While migrant work is a topic in Silver's analysis (2003, pp. 20-25), we do not find a detailed description of migrant struggles in her work but only a rather vague hint at the experience of 'boundary drawing' of one group against other groups of workers.Turning to our empirical findings, we will discuss below if and how processes of workforce fragmentation and specific forms of resistance against adverse working conditions are linked.In doing so, we draw on the concept of 'differential inclusion' of migrants (Mezzadra & Neilson 2013), which examines how different residence rights and social entitlements are related to unequal social positions in migrants' countries of destination.In this paper, we specifically explore the extent to which unequal positions in the labour market and in the division of labour at the shop floor refer to 'differential inclusion', with the consequence of different possibilities for the assertion of interests of workers.
In this context, Morrison et al (2014) have referred to the specificity of 'mobility power' as a central element of migrant struggles for better conditions, describing how workers use transnational differences and the scarcity of workers for advancement.This resonates with works of labour process theory, which reveal that the indeterminacy of labour and thus the uncertainty of management's control over workers imply that workers always retain a certain mobility power in, for example, changing employers (Smith 2006).Although 'mobility power' must thus be considered a fundamental resource of labour in general, studies have shown that migrants use it to a particular extent due to their transnational mobility (Alberti 2014;Morrison et al 2014).Hence, with the critical appraisal of the PRA in mind, we do not understand 'mobility power' as a kind of uniform substance that can be accumulated and/or exchanged but as both the basis and the result of social conflicts.A central aspect that structures this conflict is the restriction of mobility through the right of residence, official practices such as work bans, and the differentiation of social rights on the basis of citizenship (Kalbermatter 2020;Ruhs & Anderson 2010).The question of 'mobility power' in this respecthere unlike for people with German or Austrian passportsis at the same time linked to the restriction of fundamental residence and mobility rights and related social claims.
Speaking of 'mobility power' hints, with other words, to the fact that power relations are not only anchored in the workplace but also beyond.After all, if we think further on the notion of 'mobility power', analysis of labour unrest would not be sufficient and convincing, if it was not extended to a wider range of questions concerning social reproduction.To deepen this aspect, we introduce the concept of 'multiple precarity' into the analysis.While in the context of sociology of work, 'precarity' has primarily been understood as an expression of a change in employment and work processes (Castel 1995;Dörre 2019), we understand the concept in a broader sense based on a feminist perspective taking into account power relations in reproduction and household contexts (Winker 2010).With the focus on migrant workers, we also use the term 'multiple precarity' to capture the connection with the migration regime and the differential inclusion of migrants, in particular the aspect of insecure residence rights, but also the lack of social entitlements or precarious housing conditions (Neuhauser & Birke 2023).At the same time, this does not mean that the position and the scope of action in the work process will appear unimportant or subordinate.Rather, both aspects must be defined based on the idea of a 'mapping' of power relations in and beyond work processes (Herod 1997).
In the following, we will discuss the multiple forms of power and counter-power at play in those workplace conflicts we found in our investigations during the pandemic and beyond.We will relate those conflicts to the differential inclusion and social reproduction and thus, the multiple precarityof migrant workers.Based on an empirical evaluation of both aspects, we finally discuss if there is something that we might call a specific form of 'migrant labour unrest'.

Labour unrest in the pandemic: the meat industry in Germany
The meat industry in Germany has been expanding extremely strongly since the 2000s, particularly in comparison to other industries (Wagner & Refslund 2016).This applies both to the number of animals killed (especially in the area of pigs and poultry) and to economic key figures such as sales, profits, and export volume.The undisputed basis for this growth was the use of cheap, migrant, and precarious labour, which has not only contributed to a 'race to the bottom' in meat product prices, not least in European and Southeast Asian markets but has also led to the relocation of large companies from neighbouring counties to regions in Germany where wages are much lower.Furthermore, the boom has been associated with fierce competition, in which former municipal slaughterhouses and small-and medium-sized slaughterhouses and cutting plants have been substantially discarded in favour of the large, transnationally operating players.
The cheap labour, those firms use, was originally (from the 1990s) based on the system of posted work.Because of various scandalswage fraud, health and safety hazards, physical violence against workers, miserable housing conditions, etc.posted work was subsequently replaced by other forms of precarious employment after 2014 (Weinkopf 2018).While subcontracting was subsequently carried out under German law, the violence and control systems of the former recruitment chain remained in place (Birke 2022, p. 223) and were particularly exposed in the pandemic.
In the very first phase of the pandemic, as in many other countries around the world (Cook et al 2020), mass infections occurred in meat-processing plants in Germany.In May and June 2020, the virus targeted the workers of Europe's largest meat-processing plant, Tönnies, in Rheda-Wiedenbrück (North Rhine-Westphalia) (Bosch et al 2020).After more than 1,500 of the 6,000 employees of the Tönnies plant had been infected with COVID-19, the plant shut down for 4 weeks, and also a lockdown was proclaimed for the two surrounding counties of Gütersloh and Warendorf, where many of the workers lived (Birke 2022, p. 249).After schools and kindergartens had been closed, despite interventions by politicians and employers, who claimed that the workers themselves were to blame for the infections, the resentment of the local population was finally directed against the employer, who was accused of negligent handling of the safety and health conditions in the factory (Birke 2022, p. 252).This discourse and an export ban on Tönnies products by the Chinese state played a major role in ensuring that a few months later a ban on contracts for work in the slaughtering, cutting, and processing of meat was enforced at the national level.
The Occupational Health and Safety Control Act (Arbeitsschutzkontrollgesetz), in addition to prohibiting the employment of subcontracted workers from 1st January 2021 in factories with more than 50 employees, restricted temporary agency work from 1st April 2021.Referring to Schmalz and Dörre (2014), it can thus be argued that in relation to the regulation of the meat industry in the Corona crisis there has been an expansion of workers' 'institutional power resources.'Collective bargaining and works councils have been potentially strengthened by the compulsion of direct employment and the fact that agency workers can now only be legally employed if there is a contract with a trade union.For the first time since the 1990s and the introduction of neoliberal labour market politics, the federal government has completely prohibited a form of employment, which contributes to the excessive exploitation of workers.
However, whether the ban on precarious employment is also reflected in changed power relations at the workplace and/or in an expansion of trade union organisation remains a question to this day, which can only be answered on the basis of an account of the struggles over the implementation of the new directives.In the spring after the Occupational Safety and Health Control Act came into force, negotiations on the minimum wage took place for the first time in almost five years, with an increase in the same to around twelve euros (Birke 2022, pp. 317-318).The government declared the bargained minimum wage to be 'universally binding' for all employers of the sector, while, at the same time, the contract was rendered virtually useless due to an increase in the national minimum wage to almost exactly the same amount from October 2022.At the same time, union membership increased only moderately: Out of about 150,000 workers (conservatively estimated), the union was able to recruit 1,800, meaning that trade union density remained below 10%.
As many of the new union members are migrant workers formerly employed by subcontractors, they might be seen and actively be addressed as key activists in future organising drives.Overall, an expansion of institutional power does not automatically equate to an improvement in the position of workers.To find out why, we need to look more closely at the forms of (everyday) labour unrest in the meat industry.

Wildcat strikes and labour turnover in the German meat industry
To discuss the question of how power resources are used, we cannot consider them in isolation from individual and historical changes.In our interviews in the meat industry during the pandemic, we asked workers both about their biographical background and about their first day in the factory.One of the first workers we met was Mr. Mohn, 2 a former teacher who had migrated from Lithuania together with his wife.Mr. Mohn explained to us that he emigrated due to austerity measures in the Lithuanian school system.The reasons why people who have worked in an academic profession in their country of origin find themselves forced to work in the meat industry in Germany are manifold.Besides the difficulties in nostrification, another explanation is that even EU citizens can only receive limited social benefits in Germany (Riedner 2018, pp. 18-20).Because of the need to rent a flat in a tight and segregated housing market, getting involved in work in the meat industry (even with the intention of leaving it quickly) is an expression of this multiple precarity that also affects EU citizens.
When Mr. Mohn was recruited for a job in meat processing many years ago, his employment contract was still based on posted work.He describes the system of exploitation at the time when he started working in Germany.
Yes, and there we worked for a year.And [the recruiter/subcontractor] gave us [contracts] for three months, for three months.For example, we had to work 200 hours [per month], which is 1,000 euros, and 300 must go to [the recruiter] Lithuania. 3  Together with his wife, who also worked in meat processing, the couple's income was 2,000 euros, of which 600 euros had to be transferred as a 'fee' to the intermediary in Lithuania.In addition, there was about 400 euros for the apartment, which was also provided by a Lithuanian intermediaryleaving about 1,000 euros for the rest of the living expenses of two adults, hardly enough to live on in a region with comparatively high prices.
Mr. Mohn describes the first working day as follows.
[ After a few months, dissatisfaction with the ratio of working hours to wages led Mr. Mohn to look for another job.As the choice was very limited, he stayed in the meat industry.At least, by changing to another company in the same industry, he got rid of the subcontractor and no longer had to pay 'fees'.He found a temp agency specialised in Russian-speaking applicants.They placed him with a company that was about 70 km away from where he lived.There, Mr. Mohn worked on the meat sorting plant for more than twice the previous wage: 10 euros 50.However, the working hours remained extremely long.Moreover, it was shiftwork with an inconvenient time schedule.However, Mr. Mohn experienced a small advancement, he proved himself as a supervisor in the packaging of finished meat products.He also learned how to manage his working time better.For example, there was a requirement that 35 pallets of meat had to be finished per shift.He found that with practice, this target could be met even if more breaks were taken than the work schedule provided for, and that it is not advisable to finish more pallets because otherwise management will increase the target.Or he learned to call the mechanic instead of problems faults himself.In conversation, Mr. Mohn left no doubt that he was proud of this 'resourcefulness', a resource based on familiarity with the labour process and the division of labour and to that extent, therefore, an expression of primary power.However, as a reaction, management began to recruit new labourers with lesser experience and more precarious residence permits through another temporary employment agency.The following deterioration of working conditions finally forced Mr. Mohn to leave that factory after some more months (today he works in a warehouse).
Most of our interviewees describe their first work experiences in Germany as very negative, problematic, and detrimental to health.Over time, however, they find ways to make small improvements, which at first glance are purely individual stepsusing different temp agencies, deciding to work in a different company, being offered a different job, and so on.At second glance, however, these steps depend on collectively organised communication; someone has to know where to apply and how to decrease the workload and work speed.Mr. Mohn's 'resourcefulness' is therefore by no means his individual disposition only, and not only based on the named personal knowledge of the production process as such.It is also a kind of informally conveyed collectivity, distributed 'discreetly' in social networks, notably in Facebook groups, in the neighbourhood close to the meat factory, or through advisory centres and NGOs.The example also shows that there is a link between mobility and a less vulnerable position in the labour process.One expression of this is that even before the pandemic in the meat industry, the shortage of workers with certain skills drove up wages for corresponding jobs, which points to the aspect of marketplace bargaining power.All in all, Mr. Mohn's story illustrates the ambivalence of workers' structural power, which on the one hand brings bargaining power, but on the other hand can be broken by management strategies of fragmentation and recomposition of the labour force.
Moreover, exploiting mobility power by no means leads per se to greater equality for all.For example, workers who are young and childless are more likely to take advantage of travelling to Denmark or the Netherlands, where unions are stronger and wages are twice as high.Even when local companies improve working conditions, they often do so in an exclusive manner that only includes certain areas/professions of the company; wage increases enforced by strategies of exit (Hirschman 1970) are often accompanied by greater inequality between men's and women's wages (Birke 2022, p. 212).In a similar vein, wildcat strikes do not automatically lead to collective organisation in a broader sense, hence a rise of associational power.In another interview, Mr. Sulat, a Palestinian worker, describes a wildcat strike: 40 or 50 people went to the boss and said, 'Either you give us permanent contracts or we leave'.And then, when so many people leave at once, that is of course a problem for them.And then all of them together threw away their clothes, their work clothes, that is, they threw them down on the floor and then they left.And then they were sort of intercepted at the factory gates by the superiors.They went back, and they all have permanent contracts.And we in our team, we just couldn't manage it. 5 Throwing away one's work clothes is a gesture that implicitly marks the protest not only against the working conditions as such but also a factory-specific 'bar' marked by different colours of work dresses and hoodsdifferent colours for different subs with differing work tasks and conditions, and others for directly employed persons.However, as can be seen in the further course of the interview with Mr. Mohn, but also other interviews, even striking workers are often perceived as 'Poles' or 'Romanians', etc.The 'Romanian team', as reported by the interviewee, ultimately failed to carry out the same action as 'the Poles'.Thus, the picture of mobility power remains ambivalent: solidarity between workers, but also ethnicised attributions and 'exclusive' solidarity.The question is whether and how workers can manage, not only to throw the hoods and protective gear on the floor but also to wage a struggle against racism and create solidarity between all working people.
During the pandemic, wildcat strikes like the one described by Mr. Sulat became a permanent feature.Not least due to the discursive power gained through the public debate in the context of the infection scandals, the density and visibility of workers' actions has increased significantly and a connection with institutionalised trade union politics has become plausible for the first time since the beginning of the golden age of the German meat-industrial hyper-hub in the early 2000s. 6In November 2020, for example, a wildcat strike occurred at Danish Crown in Essen (Oldenburg).As a result, workers were presented with new temporary contracts that began with a probationary period, even though they had been employed for years.People downed tools.They called a legal adviser from an NGO who had previously helped them to bargain with the employer.Almost immediately, the new contracts were withdrawn, including a commitment to permanently employ all workers by 1st January 2021, which was later put into practice.The advisor named above told us of the underlying reasons: It's clear, people know they're needed, and if working conditions don't improve, they'll go to AMAZON, or there won't be any more workers coming from Romania, Bulgaria or Moldova to work in the meat industry either. 7  This example again shows that mobility power is enhanced by workers' awareness of their increased bargaining power in the marketplace in the context of labour shortages.Many workers told us that their aim was to leave the meat-processing industry for good, and many finally reached this goal (like Mr. Mohn, who is working at a logistic company today).Thus, 'exit' might be a strategy to leave the sector altogether and not a tactic to improve conditions locally.It must be emphasised, however, that on the verge of the pandemic, strategies of 'exit' and 'voice' were more openly combined by workers or at least by those interviewed in our sample.However, as the limited results of the bargaining round of early 2021 show, there is no automatic transformation of voice (struggling for better conditions at the workplace) and institutionalised forms of workplace conflicts.Due to labour market conditions, this was the case not despite but because of a 'silent' effectiveness of strikes that did not require the involvement of union representatives in the first place.However, the segmentation of labour means that some have power resources, both in the labour market and in the production process, and others do not.Spontaneous strikes tend to be, while not always, ethnicised and gendered, as outlined in the example told by Mr. Sulat.Moreover, the multiple precarity and permanent recomposition of the workforce provide reasons not to engage in collective action, as will be shown in more detail in the Austrian case.While the labour shortages escalated during the pandemic, such ambiguities were exposed in extreme ways.

Working conditions and labour disputes in the Austrian postal service and mask production
In Austria, too, the pandemic has temporarily drawn public attention to fields of work, in which an above-average number of workers without Austrian passports are employed and in which various forms of precarious employment exist.The topic was particularly debated when scandals in well-known Austrian companies became public.The infection clusters in parcel distribution centres in the province of Lower Austria and Vienna in May 2020 and the so-called mask scandal at Hygiene Austria a year later made it into the major headlines, at least for a short time.Even if the triggers of the public discourse were not the precarious working conditions themselves but foci of infection in the parcel distribution centres and the re-labelling of Chinese masks as 'made in Austria' in the production halls of Hygiene Austria, media attention to the 'slave-like conditions' became at least, like in the German case, a by-product.
What the two fields of work studied in our project have in common is that they experienced an upswing during the pandemic.The production and distribution of hygiene masks in Austria only came into being because of the pandemic.Hygiene Austria was entered on the commercial register on 24 March 2020, originally as a joint venture between the established Austrian firms Palmers and Lenzing AG, a textile producer and a raw material fibre manufacturer.The aim of the joint venture was to produce mouth-nose protection and then FFP2 masks 'made in Austria'.Since the company was founded, large sums of public money have flowed to Hygiene Austria. 8The company had close personal ties with the federal government and public clients and made a profit of 5.7 million euros in the crisis year of 2020. 9 Postal and parcel services, our second example, grew due to contact restrictions and lockdowns worldwide.In Austria, too, postal and parcel services experienced a veritable share price rally with all-time highs. 10Historically, the Austrian post, as a formerly fully public service, had undergone a process of privatisation and economisation at the end of the 1990s.Today, Post AG is a corporation listed on the Vienna Stock Exchange.However, the holding company of the Republic of Austria (ÖBAG) still owns 52.8% of the 67.6 million shares.As a former state monopoly, privatisation split the operation into clearly separated business unitspost offices, logistics, and delivery (Flecker et al 2014, 42).As a result, previously recognised work was devalued and allowed the poorly paid fringes of the workforce to grow more and more (Flecker et al 2014, 38).The outsourcing of tasks from the organisation and the related emergence of new providers (as subcontractors and temporary agencies) resulted overall in a heterogeneity of employment relationships in a sector, which used to be characterised by a high degree of homogeneity (Flecker 2016, p. 38).
In both cases, the predominant form of employment for migrant workers was temporary agency work, which in Austria, as in Germany, is increasingly dominated by migrants although far more regulated: in Austria, officially, workers must be employed under the same conditions as directly employed workers from the first day of their employment (Riesenfelder et al 2018).Nevertheless, strategies of outsourcing to subcontractors and the use of temporary employment agencies are established costcutting strategies to flexibly adjust the workforce to fluctuations in demand.As a result, working conditions of (migrant) temporary agency workers are often subject to double standards in Austria as well (Benvegnú et al 2018).Temporary workers, for example, are disproportionately confronted with insecure employment, stressful and unhealthy work situations, and a lack of representation of their interests in the company (Riesenfelder et al 2018).
The majority of the interviewed workers came from Syria and Iraq, from where they fled to Austria between 2014 and 2016.While Syrians were usually granted asylum relatively quickly, this was not the case with Iraqis, as with other groups of refugees.They often waited a long time for their right of residence and were then often only granted subsidiary protectiona status that disadvantages them in many respects, for example, in receiving only limited social benefits.Many of the refugees also spoke of a variety of experiences of (racist) discrimination, which related not only to the workplace but also to their search for housing or experiences with authorities.
The interviews from the parcel distribution centre reveal that different employment relationships are linked to hierarchies in the work process, including inequalities in working hours, wages, and health protection.Moreover, the fragmentation of employment intersects with racialised divisions, with refugee temporary workers from Arabic-speaking countries at the lowest end.Interviewees reported constant humiliation from supervisors or the inability to take sick leave without risking termination (Neuhauser et al 2021).The interviewed temporary workers at Hygiene Austria reveal similar working conditions.The fact that wage slips already showed a wage below the collective agreement makes the intention of wage dumping obvious, albeit this should be illegal under Austrian law.
In both cases, the extremely high work pressure prevented compliance with infection and health protection measures such as physical distancing.While at Hygiene Austria there were several accidents at work due to massive failures to put safety systems into operation, at the parcel distribution centre it was the lack of infection control that led to several interviewees contracting COVID-19 (Neuhauser et al 2021).According to media reports, in total 179 workers were infected with SARS-COV2 within a short period of time at the parcel distribution centres in Inzersdorf and Hagenbrunn, leading to further secondary infections in 50 families. 11With regard to the mask producer Hygiene Austria, the Chamber of Labour has conducted 123 wage theft proceedings for affected employees.However, if a company is insolventas is the case in many of the temporary work agencies concernedemployees can only claim their entitlements from the Insolvency Remuneration Fund, which is largely fed by public funds.In addition to the monthslong delay in payment to employees, this also means a shift of the burden from the responsible companies to the general public.

Exit and voice in the Austrian cases
The interviewees (as many workers) turned to the Chamber of Labour or the trade union to assert their rights only at a relatively late stage, usually after their employment relationship had ended.This raises the question of why the affected workers took so long to claim their rights, especially since they were well aware of the legal violations.Hussein, an interviewee who fled Iraq via Turkey to Austria together with his wife and three children in 2015 due to the constant threat of IS, explains why the Chamber of Labour is often the 'last resort' in his reflections: [ : : : ] During the employment [relationship], we couldn't go to the Chamber of Labour for these problems.If you go to the Chamber of Labour, you get fired, you lose your job.We all know that we are only working here temporarily.We could have gone to the Chamber of Labour, but everyone has a certain amount of time planned for himself in which he still wants to do this job.For example, two, three months and then I quit. 12  The interviewed employees did not go to the Chamber of Labour or to the trade union because they were afraid of being dismissed, but there was also a strategic aspect to the reason why they often did not seek institutional help to claim their rights.Violations of labour law were sometimes consciously accepted, since the employment relationship was seen only as a short-term one, precisely because of the exploitative conditions.In any case, it is remarkable that for the vast majority of interviewees, the lack of wage payments and no other labour law violations was decisive for claiming their rights.The fact that workers mostly had the impression that they could not take action against labour law violations during their employment is also related to the circumstance that they had no contact persons in the company, as they considered the works council to be responsible only for permanent employees.In the case of migrant workers, this is a particular problem in that they often lack knowledge of Austrian labour law, especially at the beginning of their stay.Since the interviewed workers mostly turned to institutional help only after they were dismissed, the question arises whether and, if so, how they resisted precarious working conditions during the employment relationship.There were at least some narratives, expressing resistance strategies at the rank-and-file level.The most frequently described strategy was that of workers threatening to quit or spontaneously walk off the job when labour rights were violated.For example, Laith, who also fled Iraq and did not have a permanent residence permit at the time of the interview, told us that he had struggled in this way when he was employed by Hygiene Austria as a mask picker and not as a packer as stated in his contract.The new task would have meant wearing gloves to which he was allergic.
Yes, I have an allergy to the gloves and my contract says that I am in packaging.And he wants me to take off masks.I told him, 'I can't do masks.I'm not allowed to.'He said, 'Then you can stay home.' [ : : : ] I said, 'Okay, then I'll go home.I don't want to work anymore.'I was at the door when the boss called me.He said, 'No, come back.[ : : : ].' 13   Similarly, Majid, who fled the war in Syria in 2015, recounted a situation at the parcel distribution centre, in which the shift supervisor wanted to deny him his break after more than 5 hours of hard work with very heavy packages.
I started working at about six thirty in the evening until the next break at midnight, and I took over [ : : : ] seven trucks myself, and I was sweating like water.My T-shirt was water.Because it's Amazon [a delivery for Amazon], everything has to be out by one o'clock max and I have to stay at work until five o'clock.So, then I sat quietlyand what does the shift supervisor say to me: 'Go to the other department!'I said 'I'm beat and have back pain from this work,' and he said, no that I should go there and he is the shift manager.I said, 'I'm not going there because you're the shift supervisor.'Then he told me I don't have a chance to work here anymore and I said, 'I don't want to work anymore, thank you!' I was going to leave, then the other group leader came and said, 'What's wrong, you're good, you are very hardworking, what's wrong?' [ : : : ] They always play. 14  The fact that, as Majid pointed out, they 'always play' implies that the interviewees unravel the constant threats of dismissal as a strategic move designed to discipline the workers.For the workers, playing the game could also mean to counter the employers' strategies by threatening to leave the workplace in order to expand their room for manoeuver, at least situationally.Morrison et al's (2014) insightful study on labour migration from Moldova to the Russian and Italian construction industries also reveals that 'management by fluctuation' is counteracted by the workers themselves threatening to leave the plants and to remigrate.Due to the very informal nature of the employment relationship and the lack of union support and organising, and thus of associational power, such negotiations are often conducted directly with the supervisors on site.For the workers interviewed, it was particularly when they experienced extreme humiliation at the hands of superiors that they began to resist.However, the interviews also contained stories of 'exit attempts' that no longer had a playful character.For example, workers in the parcel distribution centre experienced that defending oneself against exploitative working conditions, such as the prohibition of breaks, could actually lead to dismissal and that the constant threats were therefore not just empty words.This shows how preconditional 'exit' strategies are, given the risk of actually becoming unemployedwhich is often not an option, especially for employees with family obligations or in a precarious situation in terms of residence rights.
An example in this regard is when Iraqi Laith (mentioned above) wanted to apply for a Red-White-Red Card Plus, for which he needed 'proof of a secure existence' and thus a corresponding employment relationship.Laith therefore summarises his work at the mask manufacturer as follows: I worked there because I need it.I need three payslips for MA35 [immigration office in Vienna] [ : : : ].The boss knows my situation exactly, that's why I do all this. 15 Workers who wanted to apply for family reunification, which also requires a certain income, were also under particular pressure not to lose their full-time job or to upgrade from part-time to full-time employment.The experiences of the interviewees reveal the importance of considering the residence status of migrant workers as part of their multiple precarity, which extends far beyond the employment relationship.As stated by Laith, a precarious residence status increases one's exploitability and thus limits the possibility of labour unrest.
In total, in the Austrian case, resistance in the work process and negotiations with supervisors remained mostly at a situational and individual level.Attempts to achieve something in a joint action of solidarity rarely emerged.In addition to multiple precarity, this lack is also due to fragmentation policies that divide the workforce, such as the division into core and marginal workers.Moreover, shift supervisors discouraged practices of mutual support between workers at a very early stage, as Laith tells of the corporate strategies at Hygiene Austria, 'But if he sees us working together, he separates us.Yes, he doesn't want us to work together as team.' 16 Majid was one of the few interviewees who told of a spontaneous association of fellow workers against the arbitrary restrictions imposed by shift supervisors at the parcel distribution centre.When an Egyptian worker was forbidden to go to the toilet because he was only allowed to take a break after three and a half hours, the colleagues expressed their solidarity by all standing up at once to go to the toilet.Even if this collective action did not result in a direct improvement of working conditions, it shows that workers were also ready for spontaneous collective protest. 17Almir, who fled Somalia and had been applying for permanent residence for 9 years without being granted asylum, reported another example of a small collective actiona sit-infrom the time before the pandemic.
You know what happened to us in Ramadan?We were 100% all Muslims, so we were fasting.We started work fasting and asked for him [supervisor] to turn off the machines so we could eat [after sunset] and he didn't allow it, even though it affected all of us.So, we were forced to talk to the company [temporary agency], which also said we don't get a break for that.They said we should carry and eat at the same time.How is this supposed to work?We are not robots.Not even robots can carry the packages down with one hand and you expect me to do that?Some packages are 24 kilos, some more and some less depending on the container.The first day this happened, and we called the company and they didn't answer us, we gathered and sat down and ate.Then the temporary employment agency stepped in and said they would talk to the [management of the] post [distribution centre].In the end, they came to the decision that each employee should be allowed five minutes to eat and the machine should not stand still.We said okay, that's better than nothing. 18  What is interesting about this example is that this is the only wildcat strike we have been told about and that the strike was about the denial of physical needs arising from a religious tradition.These needs gave rise, at least in the short term, to an alliance based on the experiences of a religious-ethnic community, through which the temporary workers tried to assert their interests.With the exception of these few examples, we were not told about joint protests or collective organising attempts.The reasons for the lack of collective resistance to the working conditions at Hygiene Austria were stated by Anas, who fled Syria to Austria with his brother in 2015 when he was still a minor, as follows, 'There was not such a mood between all of us that would have allowed us to organise.And everyone has their own problems.That is, there is not just the one problem, but it is different'. 19 The fact that all workers have their own problems, in addition to precarious working conditions, indicates that the refugee workers have to struggle to varying degrees with difficult living conditions due to family responsibilities, poor housing conditions, or unresolved residence issues.In this sense, as elucidated above, we speak of the 'multiple precarity' of migrant workers, a concept that sensitises us to the fact that to understand the exploitation of migrant labour, an expanded notion of domination and control is central and needs to consider not only the labour market but also precarious residence rights, the lack of access to public goods and services, or the racial segmentation of the housing market.

Conclusion
The two cases presented reveal challenges that arise for collective organisation and resistance in the context of precarious conditions of migrant workers.Due to what we call 'multiple precarity', the experiences of migrant workers in Germany and Austria in this respect are remarkably comparable despite different regulations for temporary agency work (which is much more strongly regulated in Austria) or residence law.This observation in areas of precarious migrant employment is also a result of management strategies based on outsourcing and the permanent replacement and recruiting of workers, which constantly undermine legal standards.In sectors where large proportions of migrants are employed, such strategies are interwoven with what Roediger (2007) and Esch (2018) have called 'racial management'in which ethnically defined groups are placed in competition with one another.
At the same time, exploitative working conditions are, both in the German and the Austrian case study, an expression of the link between the multiple precarity already highlighted and what in the transnational debate is often called informalised labour relations (Munck 2013).This also explains the limited significance of the different national regulation of employment relationshipsdue to multiple precarity, formal labour regulations become secondary to informal power relations.Reference can thereby be made to informal working conditions in the global south, which point to similar hurdles for solidarity and collective action among workers (Mayer-Ahuja 2012).
Against this background, we asked how collective resistance nevertheless emerges and how the underlying power relations are structured.Although to different degrees, we recognise in both samples forms of protest, in which migrant workers try to achieve better conditions at the workplace based on the power resources they experience.In both countries (as elsewhere in Europe), the shortage of labour is currently very prominent, especially in fields of work like those examined here.In particular, the German case study reveals that the knowledge about scarcity of workers is shared and also used by the workers in workplace conflicts, an observation which is associated with 'marketplace bargaining power' by Wright, Silver, and others.During the pandemic, the infection scandals and government regulations in the meat industry acted as catalysts that increased the "power of scarcity" of labour, leading to both the emergence of strikes (including wildcat strikes) and increased exodus from the industry to other sectors.Thus, while the pandemic might not have substantially transformed conflicts and power relations, in some of the sectors studied, infection scandals and political regulations have led to a sudden public visibility of working conditions.This societal and to a high degree discursive power resource has to some extent favoured both exit and voice strategies, with a slightly improved situation for trade union organising drives.In other casessuch as those studied in Austriasuch a constellation could only be observed to a very limited extent.Forms of collectivity and resistance therefore remain on a rather discrete and less publicly visible level.
However, on the spectrum between 'exit' and 'voice,' in which sense can we speak of specific struggles of migrant workers?While some authors understand power primarily as a precursor to collective organising at the workplace and the formation of unions, mobility-power-approaches appreciate the importance of (migrant) mobility beyond these institutional policies and also outside the points of production.Thus, the mobility power approach is promising to capture the collective strategies of migrant employees by pointing to the importance of extended temporal-spatial perspectives on labour and migration (Herod 1997).However, in our view, this approach falls short if it views mobility one-sidedly as the source of migrants' collective organising.On the one hand, the described strategies of labour unrest are not migration-specific, but they are a very common source of social conflict, which is especially prevalent in fields of labour where a shortage of workers is loudly complained.Instead of 'migrantising' and specialising the struggles themselves, it would make sense to ask, what are the commonalities of labour struggles in fields that are differently characterised by multiple precarity.Our cases show that there is no reason for a 'romantic' interpretation of migrant struggles.The scope of 'exit' strategies must not be overemphasised, and the preconditions of 'exit' not be underestimated.For example, employees with uncertain residence status or care responsibilities are much more spatially bound in terms of mobility not only between countries and regions but also between different companies.Thus, although exploitation and racial management by employers are counteracted by the workers threatening to leave the plants and/or to remigrate, this strategy is also structurally limited.
Although we have argued that strategies driven out of the mobility of workers are not 'migrant-specific', we argue, on the other hand, that there are forms of organisation based on ethnicised alliances that should be taken seriously, analytically.Alongside the spatiotemporally broadened perspective on forms of resistance that takes into account the multiple precarity of workers, this 'ethnicised' character of protests must also be understood as a possibly important starting point for collective organisation.This also implies recognising the ambivalence of these forms: due to their restriction to specific nationally or religiously defined groupswhich has to be seen not least as a consequence of the divisions produced by racial managementthey produce exclusion of other workers on the one hand, while on the other, protests and gestures (like dropping the work clothes) always also express defence against racism and precarity and the demand for universal equality.
Finally, our research reveals that strategies of 'exit' and wildcat strikes, like those seen in all case studies, are often not transformed into coherent nor longer lasting forms of collective organisation.'Exit' and 'voice' are linked, but one does not necessarily lead to the other.This also means that an extension of these struggles, for example, within the framework of institutionalised co-determination (such as the election of works councils, the exercise of labour rights, or joining trade unions) is by no means automatically given but rather has to be actively pursued, not least by broadening the view of power resources and mobility in the face of increasing fragmentation and multiple precarity of workforces.
Funding statement.There are no funders to report for this submission.
Competing interests.There are no potential or perceived conflicts of interest.

Notes
1 It is referred to as a primary power resource because, unlike secondary power, it is available to workers and employees without collective representation of interests.2 This name and those of all other interviewees have been anonymised.3 SOFI_F 1_B 05,2,45. 4 SOFI_F 1_B 05,4,122. 5 SOFI_F3_E_01_T,4,[111][112][113][114][115][116][117][118][119][120][121][122][123][124][125][126][127][128][129][130] The statistical coverage of data on the meat industry in Germany is extremely poor: not even the number of workers is adequately recorded in the official statistics (one has to resort to data from the employers' liability insurance associations), let alone the number of strikes or the number of informal, wildcat strikes.When we speak of a 'clear condensation' here, this is based on an estimate determined on the basis of qualitative observations, which refers to