The outsourcing challenge

Organizing workers across fragmented production networks

—

Edited by

Jan Drahokoupil

The outsourcing challenge Organizing workers across fragmented production networks

The outsourcing challenge

Organizing workers across fragmented production networks

—

Edited by Jan Drahokoupil

European Trade Union Institute (ETUI)

Brussels, 2015

© Publisher: ETUI aisbl, Brussels All rights reserved

Print: ETUI Printshop, Brussels

D/2015/10.574/34

ISBN: 978-2-87452-366-3

The ETUI is financially supported by the European Union. The European Union is not responsible for any use made of the information contained in this publication.

Contents

Jan Drahokoupil

Introduction………………………………………………………………………….. 9

Part 1 Outsourcing, its extent and dynamics……………………………… 23

Stefan Kirchner

Chapter 1

Who performs outsourcing? A cross-national comparison

of companies in the EU-28……………………………………………………….. 25

Luca Giustiniano, Lucia Marchegiani, Enzo Peruffo and Luca Pirolo

Chapter 2

Business outcomes of outsourcing:

lessons from management research……………………………………………. 47

Carina Altreiter, Theresa Fibich and Jörg Flecker

Chapter 3

Capital and labour on the move: the dynamics

of double transnational mobility………………………………………………… 67

Part 2 The impact of production network fragmentation

on working conditions…………………………………………………. 89

Monique Ramioul and Geert Van Hootegem

Chapter 4

Relocation, the restructuring of the labour process and job quality…….. 91

Contents

Virginia Doellgast and Elisa Pannini

Chapter 5

The impact of outsourcing on job quality for call centre workers

in the telecommunications and call centre subcontractor industries………………………………………………………………………………………… 117

Anna Mori

Chapter 6

Outsourcing public services: local government in Italy, England

and Denmark………………………………………………………………………………………… 137

Nathan Lillie and Ines Wagner

Chapter 7

Subcontracting, insecurity and posted work: evidence from

construction, meat processing, and ship building………………………………………………………………………………………… 157

Part 3     Organizing in fragmented production networks: how to

establish voice mechanism for peripheral workers?……………………………………………………………………………. 175

Virginia Doellgast, Katja Sarmiento-Mirwaldt and Chiara Benassi

Chapter 8

Union campaigns to organize across production networks in

the European telecommunications industry: lessons from the UK,

Italy, Sweden and Poland………………………………………………………………………………………… 177

Bettina Haidinger

Chapter 9

Organizing peripheral workers in parcel delivery

and postal services………………………………………………………………………………………… 199

Sonila Danaj and Markku Sippola

Chapter 10

Organizing posted workers in the construction sector………………………………………………………………………………………… 217

Chiara Benassi

Chapter 11

From concession bargaining to broad workplace solidarity:

the IG Metall response to agency work………………………………………………………………………………………… 237

Contents

Nadja Doerflinger and Valeria Pulignano

Chapter 12

Outsourcing and collective bargaining in the recent crisis:

implications for employment in multinationals………………………………………………………………………………………… 255

Part 4     Addressing the outsourcing challenge:

statements from practitioners in trade unions……………………………………………………………………………. 277

Rudi Wagner, GPA-djp (Union of Private Sector Employees, Graphical Workers and Journalists), Austria

Goodbye TINA!………………………………………………………………………………………… 279

Clare Harrisson, UNISON, United Kingdom

Outsourcing in the public sector: how do we respond

and move forward?………………………………………………………………………………………… 281

Alan Tate, UNI Global Union, UNI ICTS, Switzerland

Overcoming challenges in organizing contact centre workers………………………………………………………………………………………… 285

Fidel Gavilan, ABVV Metaal/FGTB Métal, Belgium

Employee voice in the metal sector: trade unions play a key role………………………………………………………………………………………… 287

List of contributors………………………………………………………………………………………… 291

Introduction

Jan Drahokoupil

Outsourcing, or a shift of activities performed within a company to its suppliers, has become widespread. This strategic choice has become increasingly available not only to large companies and multinationals but also to smaller ones. Production networks in many sectors have thus become increasingly fragmented. Outsourcing has also been pursued by public sector organizations. Cutting labour costs by lowering pay, in- creasing work intensity and/or shifting flexibility costs to workers are just some of the motivations for outsourcing. But it can also be used to circumvent employee representation and collective bargaining systems within companies, and labour market regulations in general. Though such intentions may not drive the bulk of outsourcing decisions, any change in company boundaries is likely to impact employment, working conditions and industrial relations in the value chain. This edited volume focuses on the dynamics of outsourcing in Europe from the perspective of employees. In particular, it considers one insufficiently studied as- pect: the impact of outsourcing on working conditions and employment relations in companies. The book also collects lessons learned from the efforts of employees and trade unions to shape outsourcing decisions, processes and their impact on employment and working conditions.

Decisions to ‘buy rather than make’ may be motivated by a number of concerns. The advance of information and communication technologies (ICT) has enabled firms to coordinate and control interactions across markets and geographies to an extent that was previously impossible, too risky or prohibitively expensive. Management consultants have long advised firms to focus on ‘core competences’ and outsource other activi- ties (e.g. Domberger 1998). Globalization and the creation of the Single European Market may have indeed made it more profitable for firms to specialize in what they can do best (e.g. Meyer 2006). Financialization, including executive compensation based on stock options, has put pres- sure on management to maximize short-term returns through cost sav- ings (e.g. Milberg and Winkler 2013). Finally, as discussed in Chapter 5, liberalization and privatization have triggered cost competition and re- structuring in previously protected sectors such as telecoms.

The advantages of outsourcing can, from a company perspective, in- clude the pursuit of higher specialization, access to external expertise and other sources of strategic differentiation, quality improvements and operational cost savings. These gains, as discussed in Chapter 2, need to be weighed against the risks of losing strategic flexibility and control over knowledge critical to a company’s strategic advantage. Outsourcing is also likely to generate additional coordination and monitoring costs. Moreover, many costs only become apparent in the implementation phase. These include a sub-optimal use of human resources and the need for new processes to monitor and coordinate suppliers.

Outsourcing and a company’s boundaries

The debate on the nature and impact of outsourcing is complicated by the fact that it refers to a number of related, yet distinct, processes af- fecting a company’s boundaries. First, outsourcing may involve vertical disintegration of the value chain when a company focuses on a narrower segment of activities and transfers such activities as production or dis- tribution to other companies. This book, however, focuses mainly on the outsourcing of primary value creation activities within core business are- as (so-called strategic outsourcing). This also includes the unbundling of business support processes such as IT and customer services. It can also involve relying on workers hired through labour-market intermediaries (agency workers) or on the self-employed to perform core activities.

The second distinction is that between outsourcing and offshoring (see Table 1). Outsourcing refers to externalizing in-house activities to suppli- ers (the ‘make or buy’ decisions) who may be located in the same country (or even on the same production site) or abroad. Offshoring refers to the relocation of activities to other (lower-cost) countries, though the foreign suppliers may be part of the same company as the outsourcer.

Popular attention has for the most part been devoted to offshoring processes as they involve a direct transfer of jobs abroad. The impact of offshoring, or offshore outsourcing, on employment, inequality and productivity has been studied extensively, with most studies showing an inequality-enhancing effect, consistent with the theoretical expectation that trade will increase the relative wages of high-skilled workers. There is also overwhelming evidence from the United States (US) and Western Europe of the negative impact of offshoring on low-skilled workers in

Table 1 Outsourcing and offshoring

Corporate boundary decision

Insource

Location decision

Domestic                                    Abroad (offshoring)

Outsource

Outsourcing               Offshoring                Offshore outsourcing

Source: adapted from Sako (2005) and Olsen (2006)

terms of pay and employment. Some recent papers also show a nega- tive impact on high-skilled workers, especially in services,1 while the evidence regarding the overall demand for labour remains inconclusive. The productivity effects of offshore outsourcing reveal no clear pattern, with somewhat greater positive effects in services (Olsen 2006).

Both outsourcing and offshoring impact the effectiveness of existing worker representation structures and collective bargaining institutions and hence the ability of workers to exert control over their working con- ditions. While outsourcing involves the transfer of activities and workers outside the scope of company-level employment relations institutions, offshoring may go one step further, involving a shift to a different regula- tory regime. Moreover, transnational employment relations institutions, such as European works councils (EWCs) and transnational company agreements where they exist, become relevant in the case of offshor- ing to foreign subsidiaries. At the same time, the effects of outsourcing and offshoring in terms of their implications for industrial relations and the applicability of labour market regulations become blurred when the former is used to import workers from abroad (see Chapters 3, 7, 9, and 10). As discussed in Chapters 7 and 10, subcontracting to transnation- al labour market intermediaries has been used to contest host country employment regulations and to arbitrage between different regulatory

1. For a detailed overview, see Winkler (2013).

environments (on social dumping, see also Bernaciak, 2015). The chal- lenges involved in integrating migrant workers into national industrial relations structures have impacted their relevance (see Chapters 3 and 9).

The impact of outsourcing and offshoring on working conditions and industrial relations has featured less prominently in outsourcing/off- shoring research. In the US context, Weil (2014) recently demonstrated that businesses, and ‘lead firms’ in particular, have used subcontracting, franchising and supply chain management to cut labour costs by ridding themselves of the responsibilities inherent to managing their workforces (see also Kalleberg 2013; Warhurst et al. 2012, arrived at similar find- ings on the basis of evidence from across the world). In the European context, it has been argued that the negative impact of outsourcing and offshoring on working conditions may be exacerbated by a weakening of worker representation structures and by concession bargaining (e.g. Doellgast and Greer 2007; Flecker 2009). Indeed, as a number of chap- ters in this volume show, outsourcing decisions may be motivated by the attempt to circumvent company-level employment relations institutions and/or national labour market regulations.

The factors and processes that condition and mediate outsourcing and offshoring outcomes, including the role of institutions and actors, re- main understudied, making it difficult to draw lessons for policymakers and practitioners (see also discussions of literature in Chapters 1 and 2). Milberg and Winkler’s (2013: 190–204) analysis of OECD countries in 2000-2008 showed that public expenditure on labour market protec- tion increased the positive impact of offshoring on share of employee compensation in gross value added (labour share), and that employment protection legislation had no effect in this context (but it had a negative impact when the period from 1991 onwards was included). Their analy- sis also showed that offshoring is associated with less economic insecu- rity (and a higher labour share) in countries with more supportive labour market institutions (i.e. the ‘Rhineland model’, including Belgium, Ger- man, Austria, and, Sweden and the ‘flexicurity model’, including Den- mark, Finland, and the Netherlands). Union density had positive effects in all their models except for the Mediterranean countries. The posi- tive effect of unions appeared stronger, the more flexible labour market intuitions were.

Finally, outsourcing and offshoring may directly impact working condi- tions and job quality by affecting the way labour processes are organized.

Table 2 Outsourcing and integration/transformation of the labour process

Corporate boundary decision

Insource

Outsource

Organization of the work process

Do not transform                          Integrate/transform

Source: adapted from Gospel and Sako (2010)

In this context, a distinction between outsourcing and the integration of work processes is useful (see Table 2). It is particularly relevant for understanding the outsourcing of such business services as IT or the creation of shared service centres (business process outsourcing).2 As shown in Chapter 4, profound transformation of the labour process is required to allow certain tasks to be separated from others in the send- ing location. However, in their aim to achieve economies of scale and/or specialization benefits, companies may decide to centralize some activi- ties without actually outsourcing them. This can have a similar transfor- mational impact on work processes and job quality as outsourcing itself, whereby activities remain in-house although possibly shifted to another location. At the same time, companies may outsource, and shift, activi- ties such as HR-related processes without necessarily integrating them. Such separation is also likely to affect labour processes and job quality, and may represent a first step in creating an (external) integrated ser- vices centre (cf. Gospel and Sako 2010). However, some tasks are likely to remain fragmented even after an integrated service centre has been created (see Chapter 4).

This volume starts by providing an overview of the extent and dynam- ics of outsourcing. The second part focuses on the impact of production network fragmentation on working conditions in a range of sectors char- acterized by a high prevalence of outsourcing. The third part consid- ers experience with worker organizing in these fragmented production networks. Finally, the volume includes comments on the outsourcing

2. With the same logic and similar implications, one could also make a distinction between offshoring and the integration/transformation of the labour process.

challenges discussed in this book from trade unionists who took part in the project’s final workshop.

The dynamics and nature of outsourcing

Outlining the outsourcing landscape in Europe, Stefan Kirchner in Chap- ter 1 compares differences in the extent of outsourcing in EU Member States and between individual economic sectors. The analysis draws on the 2013 European Company Survey, a data set that allows a unique com- parison across all Member States. It shows that the extent of outsourcing is high in Nordic countries and the Baltic States and low in Central and Eastern Europe, whereby the latter seems to remain primarily a Euro- pean offshoring destination. The differences between countries remain even when controlled for company size, skill content and industry com- position. In fact, outsourcing levels across sectors in a given country are closely related. The manufacturing, ICT and construction sectors tend to outsource the most. The overall picture is thus of highly country-specific outcomes, apparently shaped by national institutions and regulations. These, for instance, structure incentives for companies in terms of access to alternative flexibility arrangements. The considerable differences in the extent of outsourcing between countries also provide opportunities for trade unions to learn from the experience of countries more exposed to the outsourcing challenge.

What are the implications of outsourcing decisions for company perfor- mance? Chapter 2 by Luca Giustiniano, Lucia Marchegiani, Enzo Peruffo and Luca Pirolo reviews management research addressing this question, paying particular attention to the impact of outsourcing on a company’s human resources, a key aspect conditioning the success or failure of outsourcing decisions. The authors point to a management paradox in which organizational change, including restructuring and outsourcing, is typically assessed by considering short-term financial and economic performance. However, outsourcing decisions are usually motivated by structural and strategic gains that can only be assessed in the long term. Moreover, any such gains can be cancelled out by strategic and organiza- tional problems, again only assessable over a longer timeframe. Finan- cial market reactions to outsourcing decisions tend to reflect the per- ceived transaction risks involved. However, transactions risks related to longer-term organizational effects are difficult to estimate ex ante. The costs related to organizational change and human resource utilization

are particularly difficult to anticipate and are, in the short run, de facto hidden. The organizational and human resource risks involved in out- sourcing can be addressed through participating employees and their representatives in restructuring decisions. The authors conclude that effective employee involvement can create the conditions necessary for achieving satisfactory outsourcing outcomes in terms of enhancing com- pany competitiveness.

The last chapter in the first part offering a birds-eye view of the dynam- ics and implications of outsourcing across sectors looks at the ‘double mobility’ phenomenon where mobile capital employs mobile workers. The chapter by Carina Altreiter, Theresa Fibich and Jörg Flecker ar- gues that ‘double mobility’ is a particular challenge for employment relations. In this context, the threat of relocation gives employers the upper hand. At the same time, labour mobility solves access to skills and hence undermines the ‘power of the place’, leaving workers divided and willing to accept inferior conditions and pay. Such a configuration facilitates ‘hyper-Taylorist’ work practices, extensive working hours, extended control over workers and extreme numerical flexibility. As also discussed in Chapter 7, ‘double mobility’ is enabled by outsourc- ing to labour-market intermediaries providing access to workers from outside the location. These also ‘immobilize’ foreign labour through spe- cific accommodation arrangements. The author concludes by discussing the difficulties in organizing mobile workers, pointing to their power in exercising the exit option and linking organizing opportunities to their settling in the locality. Chapters 9 and 10 focus specifically on organizing migrant workers.

The impact of outsourcing on workers

Opening the part on the impact of production network fragmentation on working conditions, Monique Ramioul and Geert Van Hootegem analyze the impact of relocation on labour processes and job quality in both sending and destination locations. Their sample presents a vari- ety of labour processes, ranging from highly-skilled knowledge work to semi-skilled administrative work. These were relocated and integrated to improve efficiency and productivity. The analyzed functions include order processing, customer service and IT. However, in none of the cases were entire functions relocated. Instead, workflows had to be redesigned to allow decoupling. Additional functions then had to be introduced to

deal with co-ordination problems. Few individuals in the sending loca- tion were unaffected. Relocations negatively impacted job quality and increased the risk of stress in both the sending and destination locations. The new labour processes entailed less satisfying job content with fewer opportunities for learning.

In the telecommunication sector, outsourcing followed the introduction of market competition and the privatization of the incumbent firms. In Chapter 5, Virginia Doellgast and Elisa Pannini analyse the impact of outsourcing call centres in major telecommunications firms in Denmark, Sweden and Germany. The outcome was declining pay and working con- ditions for lower-skilled workers and ones who were more easily out- sourceable. The chapter links the decline in job quality to two related trends: first, jobs were shifted to subcontractors with weaker collective bargaining institutions; second, benchmarking and concession bargain- ing exerted downward pressure on pay and working conditions in the incumbent firms.

Outsourcing has affected not only the private sector but also public ad- ministration. As Anna Mori argues in her study of outsourcing by local governments in Italy, the United Kingdom (UK) and Denmark, while decisions to outsource were driven by different factors, they shared the underlying rationale of trying to circumvent public sector collective agreements and job protections. As a consequence, those transferred experienced a deterioration of employment terms and conditions. En- compassing employment relations institutions (social dialogue and col- lective bargaining) and strong legislative protection for public employ- ment cushioned working conditions in Denmark and Italy, though did not prevent increased workload. The transposition of the EU Business Transfers Directive (Council Directive 2001/23/EC) was particularly im- portant in the UK where domestic institutions proved to be weak. The directive helped maintain working conditions during restructuring.

Hiring workers through subcontractors has become the way construc- tion, shipbuilding and the meat industry operate. These sectors are also characterized by a reliance on posted and migrant workers. Based on evidence from Germany, the Netherlands, Finland and the UK, Na- than Lillie and Ines Wagner argue in Chapter 7 that subcontracting to transnational service providers has been used to contest host country employment regulation and to arbitrage between regulatory environ- ments. Though the posting of workers is regulated by EU legislation

(Directive 96/71/EC), it has become controversial following a series of decisions by the European Court of Justice which, among others, effec- tively limit the right to strike at foreign service providers. The respective regulations, de jure, confine regulatory arbitrage to exploiting differenc- es in social insurance costs. However, as documented by Lillie and Wag- ner, enforcement of rules is hindered by the complicated mix of home and host country legal standards. Violation is thus hard to detect and the enforcement options of labour inspectorates are limited. Posting, as well as other forms of migrant worker employment, thus commonly involves circumvention of the Posted Workers Directive.

Lessons from organizing peripheral workers

What can be learned from the efforts to establish employee voice mecha- nisms to tackle the negative impacts of outsourcing on working condi- tions and job quality? Virginia Doellgast, Katja Sarmiento-Mirwaldt and Chiara Benassi provide lessons from the telecommunications industry in the UK, Italy, Sweden and Poland. Their chapter focuses on trade union campaigns to improve or maintain the pay and working condi- tions of workers – including those employed through temporary agency and freelance contracts – involved in the restructuring of call centres and technician and IT services in incumbent operators. Successful or- ganizing involved creative campaigns across the whole production net- work. The historic bargaining power of unions could be relied on in the telecom incumbents to negotiate limits on externalizations. Strategies included partnerships to improve productivity as well as protests and organizing to generate additional externalization costs through negative publicity and labour conflict. Successful campaigns extended collective representation and legal protection to externalized groups. This required mobilizations to organize the peripheral workers and campaigns to close loopholes in national legislation. In the end, some campaigns proved to be more effective than others. However, variation in outcomes could not be attributed to any choice of unions to represent peripheral workers (or not). Successes could be explained by the possibility of accessing en- compassing labour market institutions (i.e. regulations on equal pay and working conditions for peripheral workers) and the ability to develop member mobilization across the production network.

Parcel delivery is a particularly challenging environment for worker or- ganizing, with fragmented collective agreement coverage, the prevalence

of practices circumventing existing regulations on self-employment, and a reliance of workers who are difficult to reach through traditional means of trade-union organizing. Chapter 9 by Bettina Haidinger analy- ses these challenges in a number of European countries, revealing cases of resistance even in what might seem as most unlikely contexts. To suc- ceed, targeted organizing had to be developed to reach migrant work- ers, the long-term unemployed and cross-border commuters. The self- employed workers seemed indeed sceptical, at least initially, about trade union organizing, though they were interested in collective action. There was thus potential for trade unions to get involved in their (self-)organ- izing, but they had to adjust their standard agendas and articulate the actual demands of the peripheral workers. Public campaigns also proved to be an effective strategy. The successful solutions included the use of collective agreements to re-regulate employment relations by insourcing (albeit at the cost of concessions) and public policies such as minimum standards, rigorous checks on self-employment, and extended contrac- tor liability for offences against national insurance/labour law as found in the construction or cleaning sectors.

The theme of organizing migrant workers is developed further in the study by Sonila Danaj and Markku Sippola. Drawing on extensive evi- dence from Northern Europe and the US, the chapter takes stock of the lessons learned by trade unions successfully organizing migrant and posted workers in the construction sector. The repertoire of successful strategies targeting migrant workers found in the chapter is similar to that of organizing non-migrant workers, though the strategies need to appreciate their particular culture, interests and work situation. In turn, strategies towards posted workers are similar to those towards long- term migrants. However, to function effectively, this mobile and trans- national sector requires a transnational organization to be developed for protecting posted workers’ rights.

Beyond the dilemmas of concession bargaining

Outsourcing and offshoring have often allowed management to obtain concessions in collective bargaining by threatening job losses. In this context, trade unions chose to protect core workers by allowing the use of peripheral workers, typically employed at inferior conditions with lower employment protection. With the use of agency workers as a flex- ibility buffer now widespread in the automotive sector, the chapter by

Chiara Benassi evaluates experience with the use of peripheral workers in the German automotive sector, an institutional context that should be relatively favourable to protecting core workforce. Benassi, however, ar- gues that the concessions ultimately backfired on core workers because of the introduction of cheap and flexible labour in the companies. In the long run, the presence of workers employed on inferior terms threatens agreed standards and the future existence of the permanent workforce. The alternative strategy is to recruit agency workers and bargain on their behalf. The German experience shows that a public shaming strategy can be helpful in this context. To be effective, the inclusive strategy also re- quires stronger sectoral bargaining, international representation bodies such as the EWCs, and international company agreements. These trans- national institutions need to be used to set common standards and limit the room for concessions related to workplace competition.

The theme of concession bargaining is further developed in the final chapter by Nadja Doerflinger and Valeria Pulignano. It compares cri- sis-related collective bargaining and its effects on individual groups of workers – including core, fixed-term and agency workers – in MNC subsidiaries in the metal sector in Belgium and Germany. The outcomes were mediated by the skill levels of the workforce, with concessions more likely where low- and medium-skilled workers were involved. Agency workers were protected in Belgium by strong sectoral bargaining institu- tions (including a lack of opening clauses). Moreover, the study high- lights the importance of European Framework Agreements empowering works councils at company level to negotiate on more security for the workforce. The agreement in question also helped develop transnational bargaining coordination, ensuring communications between union and employee representatives.

Concluding remarks

The evidence presented in this volume emphasizes that effective worker involvement in outsourcing decisions, planning and implementation is a key precondition for satisfactory outcomes for workers. Moreover, their involvement is crucial for the success of the outsourcing processes from the perspective of the company. Effective worker voice would also help prevent abuse of the Single Market institutions and would also help tackle social inequalities related to social dumping. Worker voice thus in many respects represents a collective good. However, as this volume

shows, effective worker involvement institutions cannot be taken for granted in the context of changing company boundaries and increasing production network fragmentation. Trade unions thus have a key role to play in ensuring that the institutional base for effective worker rep- resentation remains relevant. While strategies focusing on protecting core workers have had only limited success, integrating peripheral work- ers into effective organizing has been a challenge. The lessons from the chapters in this volume highlight the importance of innovative strategies targeting peripheral workers.

At the same time, the successes discussed in this book typically com- bine effective organizing with a reliance on supportive institutions and regulations at national and European levels. For instance, strong secto- ral bargaining institutions proved effective in mitigating the effects of workplace competition. Ultimately, however, effective European worker participation institutions are needed to address such competition, as it commonly happens in a cross-border manner. The increase in outsourc- ing also makes the effective regulation of agency work, self-employment and the transfer of undertakings particularly relevant. Such regulations are an important resource for trade unions and employee representa- tives. Experience has also revealed that the effective enforcement of such legal standards often requires well-oiled worker voice institutions. Though agency work and the transfer of undertaking are regulated at EU level, the evidence presented in this volume shows that these regulations have failed to consistently guarantee worker rights. The degree to which the principles of the directives have been transposed in individual Mem- ber States varies significantly. At the same time, UK experience with the transposition of the Transfer of Undertakings Directive, discussed in Chapter 6, shows that the directives may have made a positive difference in some more liberal countries.

An important aspect of outsourcing discussed in this volume is that it facilitates the employment of migrant labour. Employment through labour market intermediaries has become a standard in many sectors, most notably construction. However, the formation of the pan-European labour market has not been matched by equal regulatory and enforce- ment mechanisms able to effectively protect workers’ rights. Exten- sion of chain liability provisions could address some of these regulatory failures. Moreover, effective worker voice in the sectors with the most mobile labour markets requires the establishment of a transnational trade union organization.

References

Bernaciak M. (ed.) (2015) Market expansion and social dumping in Europe, Lon- don, Routledge.

Doellgast V. and Greer I. (2007) Vertical disintegration and the disorganization of German industrial relations, British Journal of Industrial Relations, 45 (1), 55-76.

Domberger S. (1998) The contracting organization: a strategic guide to outsourc- ing, Oxford, Oxford University Press.

Flecker J. (2009) Outsourcing, spatial relocation and the fragmentation of em- ployment, Competition & Change, 13 (3), 251-266.

Gospel H. and Sako M. (2010) The unbundling of corporate functions: the evolu- tion of shared services and outsourcing in human resource management, Industrial and Corporate Change, 19 (5), 1367-1396.

Kalleberg A.L. (2013) Good jobs, bad jobs: the rise of polarized and precarious employment systems in the United States, 1970s to 2000s, New York, Russell Sage Foundation.

Meyer K.E. (2006) Globalfocusing: from domestic conglomerates to global spe- cialists, Journal of Management Studies, 43 (5), 1109-1144.

Milberg W.S. and Winkler D. (2013) Outsourcing economics: global value chains in capitalist development, Cambridge, Cambridge University Press.

Olsen K.B. (2006) Productivity impacts of offshoring and outsourcing: a review, STI Working Paper 2006/1, Paris, Organisation for Economic Co-operation and Development.

Sako M. (2005) Outsourcing and offshoring: key trends and issues. doi: 10.2139/ ssrn.1463480

Warhurst C. et al. (eds) (2012) Are bad jobs inevitable? Trends, determinants and responses to job quality in the twenty-first century, Houndmills, Palgrave Macmillan.

Weil D. (2014) The fissured workplace: why work became so bad for so many and what can be done to improve it, Cambridge, Harvard University Press.

Winkler D. (2013) Services offshoring and the relative demand for white-collar workers in German manufacturing, in Bardhan A., Jaffee D.M. and Kroll C.A. (eds) The Oxford handbook of offshoring and global employment, Oxford, Oxford University Press, 72-99.

Part 1

Outsourcing, its extent and dynamics

Chapter 1

Who performs outsourcing?

A cross-national comparison of companies in the EU-28

Stefan Kirchner

1.    Introduction

Outsourcing is an issue for many companies in the EU-28. It seems that many European companies have followed the call to ‘focus on their core business’ that echoed throughout the 1990s.1 This call urged companies to retain only their core activities and outsource the rest (Domberger 1998; Kakabadse and Kakabadse 2000). For many companies, outsourc- ing gained prominence in their search for greater competitiveness in an increasingly volatile market environment. Several scholars of business understand outsourcing as a make-or-buy decision (Williamson 1975; see also Gospel and Sako 2010), i.e. a company can either ‘make’ a prod- uct or service itself or can ‘buy’ the product or service from a subcontrac- tor. In this sense a company could rely on internal competences or resort to the market to produce a product or service. The growing importance of outsourcing indicates that companies increasingly favour the market over internal competences.

The increasing number of companies outsourcing services increases the relevance of outsourcing beyond the economic considerations of man- agers. A growing number of researchers point out that outsourcing is restructuring whole economies and altering traditional systems of in- dustry regulation and employee representation (Doellgast and Greer 2007; Ramioul and De Bruyn 2008; Flecker 2009; Batt and Nohara 2009; Doellgast et al. 2009). Others emphasize that outsourcing has a major impact on employment conditions and job quality (Warhurst

1. I would like to thank Jan Drahokoupil and Jörg Flecker for comments and critique. I also would like to thank Eurofound for early access to the ECS 2013 dataset and support concern- ing the technical aspects of the The author is also grateful for the support he received from ETUI, especially from Jan Drahokoupil. The author also would especially like to thank Mike Geppert.

et al. 2012). In a shift to so-called ‘low road’ approaches, companies of- ten resort to downsizing, using outsourcing as a way of cutting labour costs (Kalleberg 2011). Therefore these two aspects seem directly linked. The impact of outsourcing on job quality might be exacerbated by the power of employee representatives and unions being weakened and con- cessions being demanded in the process of outsourcing (Doellgast and Greer 2007; Flecker 2009). Thus, the growth of outsourcing provides a challenge to employee representatives and unions.

From a cross-national perspective on outsourcing there is a lack of in- ternationally comparable quantitative studies at company level, with the few existing ones focusing on specific types of outsourcing: e.g. outsourc- ing to call centres (Batt and Nohara 2009; Doellgast, Batt and Sørensen 2009), the outsourcing of HRM functions (Mol et al. 2014) as well as the international outsourcing of various functions to other countries (Ala- jääskö 2009). Moreover, these studies cover only some but not all EU-28 countries.

Against the background of this lack of cross-national research on out- sourcing this article seeks to map outsourcing in EU-28 companies and to provide a basic orientation, focusing on the key question of ‘Who per- forms outsourcing in the EU-28?’

The article focuses on four aspects of outsourcing: (a) differences across EU-28 countries, (b) differences across industries, (c) the relation be- tween these two sets of differences, and (d) differences across countries in partial and full outsourcing.

The article is structured as follows: First I introduce the data set and var- iables for the empirical analysis. I go on to present my empirical results according to the four aspects of outsourcing, before concluding with a discussion of the results.

2.    Basis of the empirical analysis

2.1                   The European company survey (ECS) 2013 data set

The basis for the following analysis is the 2013 European company sur- vey (ECS). The ECS was commissioned by the European Foundation for the Improvement of Living and Working Conditions (Eurofound 2013,

2015) and was conducted by Gallup Europe (Gallup 2013).2 Conducted in all EU-28 countries, the ECS involved a representative telephone sur- vey of human resource managers in establishments with 10 or more em- ployees in a country. The target for national sample size varies from 300 cases in Montenegro to 1,650 cases in the UK, reflecting country size.

To ensure a high survey quality, Gallup (2013) carried out various measures. These included extensive quality checks with qualitative and quantitative pre-tests, and careful translations to ensure standards in this cross-national survey. Additionally computed weights were used to correct the influence of any sample stratification. The average response rate for management interviews scored a comparably high value of 35 per cent, ranging between 71 per cent in Montenegro to 18 per cent in Austria, overall providing a sound basis for empirical analysis.

The analyses in this chapter are based on all private-sector cases from the EU-28 countries, i.e. 24,251 establishments.3 To foster the compre- hensibility of the analysis I will subsequently use the term ‘company’ for the establishments surveyed in the ECS.

2.2                     Measuring outsourcing

The ECS included a variable on the outsourcing of production and ser- vices, with interviewers asking the following question:

‘Is this establishment partly or entirely outsourcing each of the fol- lowing activities (this activity) to a third party that is not owned by your establishment or the company you belong to?

[‘Outsourcing’ refers to … the contracting of a business function or process to another, independent organisation.]’

One of the activities listed was ‘Production of goods or services’. The an- swers provide the information on outsourcing activities for the analyses in this chapter.

2. Additional information available at the Eurofound website: http://eurofound.europa.eu/ surveys/ecs /2013/european-company-survey-2013.
3. The subsequent analysis does not cover public sector companies because reliable weights are not available for these 2,829 establishments. An additional analysis of these cases is

How reliable are the outsourcing variables in the ECS? The ECS 2013 quality report provided by Gallup (2013) extensively documents qual- ity assurance practices concerning particular variables. According to this report there is no problem or limitation to the variables used. Also, only 3 per cent of all survey respondents reported that the outsourcing ques- tion was not applicable. About 1.5 per cent gave no answer or did not know.

3.    Analyzing outsourcing by companies across the EU-28

The following section discusses the results of the empirical investigation of outsourcing in the EU-28. In line with the first ECS 2013 reports by Eurofound (Eurofound 2013, 2015), my own analysis also shows that about 27 per cent of companies reported that the production of goods or services is partly or entirely outsourced. So about every fourth company in the EU-28 performs outsourcing.

In the next sections I will investigate outsourcing in four steps: first I will look at cross-national differences in outsourcing. Second, I will show the breakdown of outsourcing by industry. In a third step I will consider the interrelationship between outsourcing in different sectors across coun- tries. Finally I will explore the background to the cross-national differ- ences, looking at the cross-national difference between partial and full outsourcing.

3.1                   Differences across the EU-28

Existing empirical research provides many reasons to assume that levels of outsourcing differ substantially across countries. International com- parative studies support the importance of national conditions for out- sourcing.

For example, Flecker (2009: 261 pp.) found clear national variations in cross-national case studies, with companies in different countries re- sorting to different forms of flexibility, reflecting national institutional opportunities and constraints. Similarly, several studies that focus on single countries emphasize that outsourcing relates to national con- ditions (e.g. for France and Spain see Lallement 2011; for the UK see Kelly 2013). In this respect, Germany serves as a prime example. Here,

empirical findings emphasize the role of deregulation on a tradition- ally highly regulated economy (Doellgast and Greer 2007; Crouch et al. 2009; Holst 2014). Outsourcing appears to be an element of eco- nomic transformation and, in contrast to the traditional model of co- ordination, does not seem to be effectively constrained, suggesting that companies are using outsourcing to circumvent the constraints of the highly regulated German economy.

Cross-national research on outsourcing to call centres (Doellgast, Batt and Sørensen 2009; Batt and Nohara 2009) provides another strain of literature helping to understand outsourcing across countries. These extensive analyses of call centre outsourcing show substantial national diversity, reflecting differing national institutions: in our case the indus- trial relations system as well as the structure of national labour markets including wage dispersion. The studies in various countries show a gen- eral trend of outsourcing to call centres.

This call centre research also shows that such restructuring by out- sourcing undermines traditional mechanisms of employee representa- tion as well as union strategies. For some companies sub-contracting to call centres is a means of circumventing formal regulations. Further- more the call centre studies (Doellgast, Batt and Sørensen 2009) show that there is a variety of institutional avoidance. This means that compa- nies in different countries use country-specific regulation loopholes al- lowing them to evade national institutional constraints. Thus, call centre research provides a strong argument for country-specific outsourcing patterns.

An international comparative study on HRM outsourcing by Mol et al. (2014) provides another source of empirical evidence on European companies. Using the so-called Cranet dataset, Mol et al. investigated the cross-national differences in various forms of HRM outsourcing. The findings show that HRM outsourcing is present in many countries, as a result of which country patterns are consistent across different forms of HRM outsourcing. However, they do not find a dominant explanation for the cross-national differences. Again, this empirical study finds sub- stantial cross-national diversity, without however being able to explain it.

Eurostat conducted another empirical study that focused on the inter- national outsourcing behaviour of companies in 13 European countries

(Alajääskö 2009). The available survey reports documented the general breakdown, showing that on average about 15 per cent of the surveyed companies outsourced activities internationally. However, there are substantial differences between countries. Companies in Ireland and the UK are the most frequent users of international outsourcing with rates above 30 per cent. Two Nordic countries display high values above 20 per cent, whereas Germany exhibits only a below average value.

Overall, most of the existing empirical studies agree that national in- stitutional set-ups matter. However, the empirical findings provide a mixed picture. The results of my own empirical investigation tend to support this.

Figure 1 presents the percentages of companies that outsourced the pro- duction of goods and services by country. At first glance it is easy to spot the substantial differences across the EU-28 countries, with values rang- ing from 53 per cent of companies in Finland to just 15 per cent in Croa- tia, i.e. a huge spread of almost 40 percentage points.

Is there a logic behind the breakdown of outsourcing across countries? The three Nordic countries Finland, Sweden and Denmark are located on the top end of the figure. Interestingly, the three Baltic States Latvia, Estonia and Lithuania are also high in the figure, exhibiting very high values above 40 per cent.

Several Central and Southern European countries range in the middle of the figure. For example countries like Greece, France, Austria and Spain have percentages above the average value of 27 per cent but still below 30 per cent, while Portugal, Luxembourg, Belgium and the Netherlands exhibit values above 30 per cent. By contrast Italy and Germany are be- low average, with both showing values below 27 per cent.

Interestingly, Ireland and the United Kingdom (UK) come in well below average. At 16 and 23 per cent they range way below the top-end coun- tries. This is surprising as one would assume the liberal institutional frameworks in these countries do not constrain outsourcing (see Kelly 2013). Also, Central Eastern European countries exhibit similar below- average values, with Hungary, Poland, Czechia and Slovakia all situated at the lower end of the figure.

Figure 1 Outsourcing of production of goods or services in the EU-28

Finland Latvia Estonia Lithuania Portugal Sweden Luxembourg

Belgium Netherlands Denmark

Spain Cyprus Austria Malta Romania France Greece Ø EU-28

Slovenia

Italy Hungary

UK

Poland Germany Czech Republic

Slovakia Bulgaria Ireland Croatia

16%

15%

36%

36%

34%

34%

34%

30%

30%

30%

29%

28%

28%

27%

27%

25%

25%

24%

23%

22%

22%

22%

22%

21%

44%

43%

41%

53%

49%

0%         10%         20%        30%         40%         50%

Percentage of companies that outsourced the production of goods and services

Note: ECS 2013, own calculations, weighted by establishment;

N: 23,326; chi2 P = 0.0000; countries in ascending order of percentages; excluding public service cases.

Overall the analysis shows that country-specific institutional set-ups have a major influence on the prevalence of outsourcing, even when con- trolled for company size, industry and employee skill levels.4

3.2                     Differences across industries

Existing research indicates that companies differ in their outsourcing activities across industries. While in previous decades outsourcing dom- inated in manufacturing and IT companies (Kakabadse and Kakabadse 2000). Existing cross-industry case study research shows that outsourc- ing is gaining currency in many other economic sectors (Flecker 2009). Therefore outsourcing should no longer be limited to manufacturing and IT. Accordingly, outsourcing should be prevalent in all economic sectors.

Seminal studies on outsourcing have focused on specific industries, in- cluding manufacturing, the automotive sector or the telecommunica- tions industry (Kinkel and Lay 2003; Kinkel and Maloca 2009; Mol et al. 2005; Doellgast and Greer 2007; Doellgast, Batt and Sørensen 2009; Batt and Nohara 2009; Holst 2014). This research has provided rich and extensive findings especially on outsourcing to call centres. However, for many other industries such comprehensive research is currently lacking.

The findings on the sectoral breakdown of outsourcing presented in Fig- ure 2 reveal substantial differences across industries, with percentages ranging from 18 per cent to 38 per cent. The average for all industries is 27 per cent. Only three industries exceed the average by more than two percentage points: manufacturing; electricity, gas, steam and air condi- tioning supply; and construction. By contrast, the three industries with the lowest values, around 18 and 19 per cent, are: accommodation and food service activities; human health and social work activities; and arts, entertainment, and recreation.

Overall, these findings indicate that there are substantial sectoral dif- ferences. Companies in the industrial and construction sectors tend to

4. I computed additional regression models, including multilevel models. These models showed that the country differences remain constant even if company size, industry and employee skill levels are introduced into the This indicates that the cross-national differences in outsourcing are not a simple result of differences in these variables. By con- trast, the country differences seem to be due to differences in other dimensions.

outsource more than service sector companies. Nevertheless, at least one in five companies in a given industry performs outsourcing, underlining the prevalence of outsourcing in many different industries.

Figure 2 Industries and the outsourcing of the production of goods or services in the EU-28

38%

0%        10%       20%       30%       40%

Companies that outsourced

the production of goods and services

Note: ECS 2013, own calculations, weighted by establishment; N: 23,326; chi2 P = 0.0000; excluding public service cases.

3.3     Supplementary analysis: differences across industries including the public service sector

The previous analysis summarized in Figure 2 did not include cases from the public service sector. The analysis of public service sector cases with the ECS 2013 data set is difficult because it does not include weights for these cases. However, weights are important to correctly depict general distributions. As a consequence of this limitation due to weighting cases from public services are only included in this supplementary analysis of industry comparison. All other analyses in the article were carried out at company level without the public sector cases.

A workaround was used to include the 2,829 ECS sample cases from the public service industries (NACE codes O, P, Q). For these cases no reliable company-level sampling weights are available. However, an al- ternative weight based on the number of employees allows them to be considered, even though it can only be used as an analytical weight. Note that results with this weighting factor indicate the percentage of all em- ployees working in companies belonging to specific categories, i.e. only allowing approximations of the underlying company distributions and how many employees are affected by outsourcing.

The alternative depiction in Figure 3 includes values for the three public services industries. Due to representativeness issues, outsourcing activi- ties in these three industries can only be depicted as a percentage of em- ployees working therein. Please note that the values differ between the Figure 2 and Figure 3. In the latter one needs to consider that industries made up of many small companies employ on average fewer employees than industries with many large companies which employ on average more employees. However this difference is quite small, as company size has only a limited influence on outsourcing. Therefore figures weighted by establishments roughly resemble the figures weighted by employees. With this limitation in mind, the values of the three public service indus- tries are comparable with the other sectors.

The supplementary analysis in Figure 3 shows that the industry ‘public administration and defence’ belongs to the service sector industries with the highest rates of outsourcing. With a value of 31 per cent of all em- ployees in that sector it is situated closely behind the information and communication sector. The remaining two public service sector indus- tries exhibit values way below average. In fact ‘education’ has the lowest

value in Figure 3. Thus outsourcing in the public service sectors appears to be unevenly distributed.

Figure 3 Alternative depiction to include public service industries – employees working in establishments that outsourced the production of goods or services in the EU-28 (by industries)

0%      10%     20%     30%     40%

Percentage of all employees in companies that outsourced the production of goods and services

Note: ECS 2013, own calculations, weighted by number of employees in establishment; N: 25,834; chi2 P = 0.0000; * public service cases (NACE codes N, O, P) included in this figure, bars highlighted in black.

3.4      The relation between differences across countries and industries

The two previous sections show substantial differences across countries and industries. But how are they related? Only because values are high in a given country does not mean that values in a specific industry in this given country are also high.

Sectoral differences provide an interesting counterpoint to the national models. Bechter et al. (2012) emphasize that industrial relations vary substantially from one sector to the next, and called for a reconsidera- tion of national models as recent transformations question the domi- nance of general patterns. The authors argue that both national models and the sectoral level need to be taken into account. This is in line with other studies emphasizing the growing role of sectors on company-level patterns (Crouch, Schröder and Voelzkow 2009; Lane 2008; Haidinger et al. 2014). Thus, outsourcing differences between industries might be as important as differences between countries.

Figure 4 presents an analysis of the relation between outsourcing at country level and at industry level. To reveal the industry patterns I merged the companies in the industrial sectors and construction indus- tries into a single category – industrial and construction sector (similar approach by Eurofound 2015: 3). Similarly, I merged all companies from service sector industries into another category – service sector. For each country and sector category I computed the percentages of companies using outsourcing. The figure thus shows how closely outsourcing in the two sectors is related in a given country.

The 28 countries presented in Figure 4 neatly follow the dotted line, which indicates that across countries the percentage of outsourcing in one sectoral category relates closely to the other category. In short: if in a given country the level of outsourcing is high in the industrial and con- struction sector it will also be high in the service sector. Values for the industrial and construction sector are approximately 1.3 times higher than in the service sector. On average the sectors differ by about 9 per- cent points.

Overall, the results show that in a given country the levels of out- sourcing across sectors are closely related. While there are substan- tial transnational differences across industries there is a tendency

of alignment at country level, again indicating that national conditions have a great influence on the outsourcing activities of companies in that country.

Figure 4 Percentage of EU-28 companies using outsourcing by sectors

50%

45%

40%

35%

30%

25%

20%

15%

10%

5%

0%

0%          10%         20%          30%         40%          50%         60%               70%

Percentage of companies outsourcing from the industry & construction sector

Note: author’s own depiction; N: 28 countries; regression results R² value = 0.73; regression coefficient: 0.616, constant: 0.046; p > |t|: 0.000.

Country abbreviations: BE: Belgium; BG: Bulgaria; CZ: Czechia; DK: Denmark; DE: Germany; EE: Estonia; IE: Ireland; EL: Greece; ES: Spain; FR: France; HR: Croatia; IT: Italy; CY: Cyprus; LV: Latvia; LT: Lithuania; LU: Luxembourg; HU: Hungary; MT: Malta; NL: Netherlands; AT: Austria; PL: Poland; PT: Portugal; RO: Romania; SI: Slovenia; SK: Slovakia; FI: Finland; SE: Sweden; UK: United Kingdom.

3.5     Differences across countries in partial and full outsourcing

The findings on cross-national differences show a surprisingly low value for Irish and UK companies. To better understand this I performed a further analysis, using additional information in the ECS dataset to dis- tinguish between partial and full outsourcing. This additional informa- tion refers to the respondent’s statement of whether or not production is an activity at the establishment.5 By combining this information with outsourcing data I can distinguish between partial and full outsourcing (see Table 1; similar approach by Eurofound 2015).

Partial outsourcing occurs when production is outsourced in companies where production is an onsite activity. By contrast, full outsourcing oc- curs when the outsourced production is not an onsite activity. Accord- ing to this distinction about 20 per cent of EU-28 companies perform partial outsourcing whereas about 7 per cent perform full outsourcing (see Table 1). The distinction between partial and full outsourcing builds on an important attribute of companies. Not all companies are involved in the production of goods and services. Many companies perform non- production tasks or fulfil other functions such as administration or man- agement, product design or development, marketing and sales or retail activities. When companies report no production but the outsourcing of production they are seen to perform full outsourcing.

Table 1  Extent of outsourcing – distinguishing between partial and full outsourcing

Production is an on-site activity

Yes                              No

Outsourcing of production

Yes                              Partial outsourcing

(20%)

No                               Production without

outsourcing (40%)

Full outsourcing

(7%)

Neither production, nor outsourcing (34%)

Note: own depiction; percentages are per cent of companies in the EU-28, weighted, N: 23,240; the total is 100%, differences due to rounding.

5. Respondents were asked to indicate activities at the establishment: ‘Please indicate if any of the following activities are carried out at this establishment?’ One answer category read: ‘Production of goods or services’.

Figure 5 The extent of outsourcing: full vs. partial outsourcing (only companies that outsource)

0%                20%               40%               60%               80%                     100%

Full outsourcing           Partial outsourcing

Note: ECS 2013, own calculations, weighted by establishment; N: 7,354; chi2 P = 0.0000; only cases included that reported outsourcing of production of goods and services; countries in ascending order of percentages; excluding public service cases; […%] percentages of companies that perform outsourcing in a country from Figure 1 above reported in squared brackets for comparison.

How does this distinction between partial and full outsourcing play out at country level? Figure 5 displays the cross-national differences, look- ing only at companies that use outsourcing. The findings show a differ- ent picture to those presented above, with Ireland and the UK coming out on top. This indicates that companies in these countries tend to per- form outsourcing in a more radical fashion (please note that companies in countries may overall outsource less frequently. Yet when companies outsource they might resort more often to full outsourcing – as is the case in the UK and in Ireland).

Overall, the results show a considerable difference between partial and full outsourcing, alerting us to the fact that companies might differ by the extent of outsourcing. For employees it makes a huge difference whether production is only partially outsourced or whether the entire produc- tion is relocated outside the company. The results show that Ireland and the UK adopt a very different and more radical approach to outsourcing than all other EU-28 countries.

4. Discussion and conclusion

In this chapter I asked the question: ‘Who performs outsourcing in the EU-28?’, putting my focus on the outsourcing of the production of goods and services. Using company-level data from the European Company Survey (ECS) 2013 this chapter has provided a comprehensive overview of the extent to which outsourcing is used by companies in the EU-28.

About 27 per cent of all companies in the EU-28 use outsourcing, hand- ing over the production of goods and services to external providers. This result reveals that about every third company in the EU-28 performs outsourcing. This could be seen as a cue for a general trend. To map the extent of outsourcing I proceeded in four steps.

First, I investigated cross-national differences. My results show that national conditions greatly influence the extent of outsourcing. Finland tops the EU-28, with about 53 per cent companies practising outsourc- ing, whereas Croatia tails at about 15 per cent – a difference of almost 40 percentage points between the country with the fewest companies outsourcing and that with the most. Considering the positions of the re- maining countries the general picture appears fragmented at first glance. Large countries and small countries, new and old EU Member States

appear mixed up, without revealing any clear-cut picture. This seems to support the findings of Mol et al. (2014) that there are no general coun- try patterns explaining the cross-national differences in outsourcing. The general trend seems to be determined by country-specific patterns.

However, taking a second look at the country breakdown, an underly- ing pattern can be discerned. In general, Nordic and Baltic countries come out on top, while Eastern European countries as well as Ireland and UK are at the bottom. Southern and Central European countries tend to range in the middle, though the differences amongst Southern and Central European countries are quite large. Interestingly, a com- parable diversity of countries that are expected to be similar is also en- countered in other empirical fields (e.g. research on job quality types; cf. Holman 2013). Moreover, the substantial differences between these countries could also indicate the existence of country-specific regulatory loopholes. Several case studies reported that companies performing out- sourcing make use of such (Doellgast, Batt and Sørensen 2009). Com- panies differ in their outsourcing strategies in line with the alternatives available in their countries (Flecker 2009). Instead of outsourcing, com- panies may for example use fixed-term contracts as functional equiva- lents. So while a general grouping of countries emerges from the analysis there are country-specific influences that blur the overall picture of out- sourcing across the EU-28.

Second, I discussed the breakdown of outsourcing across industries. My analysis shows substantial differences across industries in their tenden- cies to outsource production. Companies in the manufacturing, infor- mation and communication and construction sectors make greater use of outsourcing than companies in other industries. Compared to these pioneers of outsourcing, other industries exhibit considerably lower val- ues, with differences of up to 20 per cent. This indicates that differences across countries and industries are important to understand outsourc- ing in companies. It could very well be the case that companies in spe- cific industries have high levels of outsourcing even though the country overall has lower levels.

Third, I undertook an additional step to understand the interrelation- ship between outsourcing in difference sectors across countries. Empiri- cally, the country and sector breakdowns could be aligned, but they do not have to be. For example, outsourcing levels might be similar in the manufacturing industry across different countries while other industries

exhibit disparate levels. My analysis shows a strong interrelation be- tween outsourcing across sectors within a given country, i.e. when the level of outsourcing in a given country is high in the industrial and con- struction sector it is also high in the service sector. This indicates that the tendency of companies to outsource is substantially shaped by the coun- try institutional set-up, regardless of the sector of economic activity.

In my fourth and last step I explored the background of these cross- national differences, investigating the respective levels of outsourcing by distinguishing between partial and full outsourcing. I undertook this step in particular due to the low outsourcing activity in the UK and Ire- land, as their low values conflicted with general expectations. This addi- tional analysis showed that full outsourcing is most frequent in UK and Ireland, indicating that companies in these two countries take a more radical approach to outsourcing.

Interestingly, a similar pattern emerged in a study of international out- sourcing (Alajääskö 2009). Here too, Ireland and the UK exhibit the highest international outsourcing rates. So it seems that the empirical results are more in line with general expectations when different types of outsourcing are considered. The substantial differences in partial and full outsourcing indicate that the influence of national institutional con- ditions might be decisive for the type of outsourcing.

Finally concluding, in this chapter I took a first step in mapping out- sourcing in the EU-28, providing an overview and charting general ten- dencies. However underneath a general tendency patterns seem to be very country-specific. Whether or not a company decides to outsource activities depends significantly on the country. While conditions in spe- cific industries matter to a certain extent, national conditions predomi- nate. The existing literature argues that outsourcing has a substantial effect on job quality and puts employee representatives under pressure. From this chapter two aspects emerge which practitioners may like to draw on when tackling outsourcing:

Firstly, the findings suggest that it might be worthwhile to search for solutions at a national level. To some extent the magnitude of outsourc- ing appears to be a national phenomenon. The strong relation between sectors in a country indicates that union representatives from different sectors should consider joint efforts in tackling outsourcing.

Secondly, considering countries where outsourcing levels are high might help to devise strategies on how to cope with the effects of outsourcing on employees. The differences across the EU-28 countries are substantial. From this it emerges that some countries have much greater experience in outsourcing than others. Accordingly, unions in different countries face the challenges of outsourcing with a different intensity. Union rep- resentatives could therefore learn about successful strategies by consid- ering solutions developed in countries with higher levels of outsourcing.

References

Alajääskö P. (2009) International Sourcing in Europe, Eurostat Statistics in Focus 4/2009, Luxembourg, Office for Official Publications of the European Communities.

Batt R. and Nohara H. (2009) How institutions and business strategies affect wages: a cross-national study of call centers, Industrial and Labor Relations Review, 62 (4), 533-552.

Bechter B., Brandl B. and Meardi G. (2012) Sectors or countries? Typologies and levels of analysis in comparative industrial relations, European Journal of Industrial Relations, 18 (3), 185-202.

Crouch C., Schröder M. and Voelzkow H. (2009) Regional and sectoral varieties of capitalism, Economy and Society, 38 (4), 654-678.

Doellgast V., Batt R. and Sørensen O.H. (2009) Introduction: Institutional change and labour market segmentation in european call centres, European Journal of Industrial Relations, 15 (4), 349-371.

Doellgast V. and Greer I. (2007) Vertical disintegration and the disorganization of German industrial relations, British Journal of Industrial Relations, 45 (1), 55-76.

Domberger S. (1998) The contracting organization: a strategic guide to outsourc- ing, Oxford, Oxford University Press.

Eurofound (2013) Third European Company Survey: first findings, Dublin, Euro- pean Foundation for the Improvement of Living and Working Conditions.

Eurofound (2015) Third European Company Survey – Overview report: workplace practices, patterns, performance and well-being, Luxembourg, Publications Office of the European Union.

Flecker J. (2009) Outsourcing, spatial relocation and the fragmentation of em- ployment, Competition and Change, 13 (3), 251-266.

Gallup (2013) 3rd European Company Survey: technical report, Dublin, European Foundation for the Improvement of Living and Working Conditions.

Gospel H. and Sako M. (2010) The unbundling of corporate functions: the evolution of shared services and outsourcing in human resource management, Industrial and Corporate Change, 19 (5), 1367-1396.

Haidinger B., Schönauer A., Flecker J. and Holtgrewe U. (2014) Value Chains and networks in services: crossing borders, crossing sectors, crossing regimes?, in Hauptmeier M. and Vidal M. (eds) Comparative political economy of work, Basingstoke, Palgrave Macmillan, 98-118.

Holman D. (2013) Job types and job quality in Europe, Human Relations, 66 (4), 475-502.

Holst H. (2014) ‘Commodifying institutions’: vertical disintegration and institu- tional change in German labour relations, Work, Employment and Society, 28 (1), 3-20.

Kakabadse N. and Kakabadse A. (2000) Critical review – Outsourcing: a paradigm shift, Journal of Management Development, 19 (8), 670-728.

Kalleberg A.L. (2011) Good jobs, bad jobs: the rise of polarized and precarious employment systems in the United States, 1970s to 2000s, New York, Russell Sage Foundation Publications.

Kelly J. (2013) The United Kingdom, in Frege C. and Kelly J. (eds) Comparative employment relations in the global economy, London, Routledge, 170-186.

Kinkel S. and Lay G. (2003) Fertigungstiefe: Ballast oder Kapital? Stand und Effekte von Out- und Insourcing im Verarbeitenden Gewerbe Deutschlands, Mitteilungen aus der Produktionsinnovationserhebung 30 (August 2003), Karlsruhe, Fraunhofer-Institut für Systemtechnik und Innovationsforschung.

Kinkel S. and Maloca S. (2009) Drivers and antecedents of manufacturing offshoring and backshoring: a German perspective, Journal of Purchasing and Supply Management, 15 (3), 154-165.

Lallement M. (2011) Europe and the economic crisis: forms of labour market adjustment and varieties of capitalism, Work, Employment and Society, 25 (4), 627-641.

Lane C. (2008) National capitalisms and global production networks: an analysis of their interaction in two global industries, Socio-Economic Review, 6 (2), 227-60.

Mol M.J., Brewster C., Wood G. and Brookes M. (2014) How much does country matter? A cross-national comparison of HRM outsourcing decisions, in Wood G., Brewster, C. and Brookes M. (eds) Human resource management and the institutional perspective, New York, Routledge, 200-220.

Mol M.J., van Tulder R.J.M. and Beije P.R. (2005) Antecedents and performance consequences of international outsourcing, International Business Review, 14 (5), 599-617.

Ramioul M. and De Bruyn T. (2008) Towards strategies for making offshore out- sourcing economically and socially sustainable, Work Organisation, Labour and Globalisation, 2 (1), 117-132.

Warhurst C., Carré F., Findlay P. and Tilly C. (eds) (2012) Are bad jobs inevitable? Trends, determinants and responses to job quality in the twenty-first century, Houndmills, Palgrave Macmillan.

Williamson O.E. (1975) Markets and hierarchies, analysis and antitrust implica- tions, a study in the economics of internal organization, New York, Free Press.

Appendix

Table 2 Answering patterns for outsourcing of production of goods or services

Note: ECS 2013, own calculations; full data set;

* public service industries not included in the general analysis

Chapter 2

Business outcomes of outsourcing: lessons from management research

Luca Giustiniano, Lucia Marchegiani, Enzo Peruffo and Luca Pirolo

1.    Introduction1

Outsourcing has been broadly recognized as an important strategic choice made by companies and other organizations to achieve a wide variety of goals. Many studies have focused on the economic and finan- cial impacts and on the relationship between outsourcers and outsour- cees. This chapter offers a comprehensive overview of actual outsourcing outcomes found in management research, including impacts on human capital.

Analysis of the evidence on outsourcing in the OECD STAN database (OECD 2011: 1970-2009) shows that both the number of transactions (deals) and their scope (activities involved) have increased constantly during the last 20 years. Over time, outsourcing popularity peaks have coincided with certain trends, such as business process reengineering, strategic focusing on core business, outsourcing/offshoring strategies, shared services and corporate downsizing (e.g. Brunetta and Peruffo 2014). Furthermore, as recent research shows, companies are expedi- tiously outsourcing non-core business processes and functions in order to maximize their profits. Business profits can be increased through re- ducing costs and/or via acquiring external sources of strategic differen- tiation (e.g. higher-quality raw materials or distinctive expertise/com- petences able to improve the overall quality of products and services, enabling companies to sell them at higher prices) (De Fontenay and Gans 2008; Gospel and Sako 2010; Angeli and Grimaldi 2010; Doellgast and Gospel 2012; Giustiniano et al. 2014). In such a scenario, multinational

1. The authors are very grateful to Howard Gospel, Jan Drahokoupil and the anonymous re- viewers for their careful and meticulous reading of the paper, as well as for their constructive

companies (MNCs) have a wider range of opportunities for outsourcing and offshoring activities due to their scope and international presence. On the other hand, MNCs are exposed to possible changes in employee relations models in the diverse labour markets in which they operate (Marginson and Meardi 2006; Sippola 2011).

At company level, the decision to outsource activities is linked to expect- ed structural and strategic changes which can be assessed through the adoption of a long-term perspective. While immediate results are related to purely economic assessments, long-term effects are more strategic and made up of opportunities (e.g. focus on core competencies and pur- suit of greater higher specialization) and drawbacks (e.g. less strategic flexibility). Nevertheless, in a company’s strategic plans, management focus is generally on the short-term results due to the (shorter, expected) timespan of their individual assignments and the contingent pressures of shareholders and financial markets.

The short-term nature of outsourcing assessments is also a traditional argument used by trade unions and emphasized by the widening geo- graphic scope of outsourcing and offshoring. While unions have been traditionally considered as opposing outsourcing and offshoring (e.g. Lommerud, Meland and Straume 2009), more recent studies have de- tected a significant shift from resistance to proactive strategies (Ramioul and De Bruyn 2006). Similarly, MNCs embracing outsourcing are better able to interpret the differences existing in national cultures, business practices, workplace representation systems and collective bargaining structures (Anner et al. 2006; Doellgast and Gospel 2012; Pulignano and Doerflinger 2013), contributing to a better local strategic responsiveness in their local subsidiaries (Almond et al. 2005; Arrowsmith and Margin- son 2006; Bartlett and Ghoshal 1992).

The aim of this chapter is to illustrate the main findings on the expect- ed outcomes as reported in management and financial literature. To achieve this, the chapter explains a number of management paradoxes related to outsourcing, as well as how companies relate to their stake- holders. Management paradoxes referred to include: a) the paradox of financial and economic vs. strategic and organizational outcomes; and b) the time paradox. The chapter draws on previous research conducted on outsourcing antecedents, processes and outcomes, both in general (Mar- chegiani et al., 2012) and applied to specific activities within the value chain (Gospel and Sako 2010; Doellgast and Gospel 2012; Giustiniano

et al. 2014; Brunetta, Giustiniano and Marchegiani 2014; Sorrentino et al. 2015). The focus is on analyzing the ‘lessons learned’ reported in management and financial literature to nurture implementation strate- gies, for use by workers representatives (e.g. Benassi 2011). Among these lessons, it seems useful to affirm that in order to maximize the positive outcomes of outsourcing strategies, organizations have to take into ac- count all the human implications of outsourcing, such as the effects on workers and their representatives. To illustrate this, the investigation on the outsourcing outcomes will be complemented by an analysis of cer- tain aspects related to organizational design (i.e. company boundaries, coordination mechanisms) and labour issues (i.e. human reactions: re- sistance to change, hidden costs and the loss of competences and com- petitive advantage).

2.    Outsourcing and management paradoxes

The literature analyzed in this chapter shows that the birthplaces of the so-called ‘trendy phenomena’ in which outsourcing decisions are tak- en are somehow related to certain management paradoxes companies might face (e.g. Andriopoulos and Lewis 2009). A management paradox is a situation involving the simultaneous presence of contradictory, even mutually exclusive elements (Cameron and Quinn 1988; Clegg, Cunha and Cunha 2002). Such elements may be related to the different na- ture of outcomes (which we call an outcomes paradox) or the short- vs. long-term orientation in choices (which we call the time paradox). With respect to the first, financial and economic outcomes can be offset by strategic and organizational effects. For example, cost savings may be cancelled out by strategic and organizational rigidity, such as a less ef- fective span of control over the activities and the formal and informal practice of power.

The ‘time paradox’ related to outsourcing refers to the fact that massive reorganizations of value chain activities call for organizational change which often overtakes the timespans considered when assessing out- comes. In short, while companies expect the organizational settings to adapt to the changes in the medium term, outcomes are generally as- sessed for the short term. While the short-term perspective focuses on comparing the cost of in-house activities vs. the price of purchasing ser- vices and goods from the market, the long-term perspective highlights the effects from a strategic and organizational viewpoint. An organization

needs to adopt the latter approach to genuinely estimate the risks linked to the outsourcing strategy, such as i) losing strategic flexibility; ii) sac- rificing knowledge; and iii) the ‘ex-post cost control’ needed to moni- tor the quality level of the outsourced activities. The situation is even more serious when top managers believe the organizational design will automatically adapt to the new post-outsourcing setting, without iner- tial constraints or negative reactions. As a result, where companies once sought order, clarity and consistency (as seen in old organization charts and procedures), outsourcing might create paradoxes by generating cha- otic contradictions and inconsistencies regarding organizational goals, structures, processes, cultures and even professional identities (Smith and Lewis 2011).

Nevertheless, the achievement of long-term goals depends on both the short-term coordination and control of activities (as they affect company results) and the long-term maintenance of relationships, both with the outsourcees and other stakeholders (e.g. Gittel 2004). As reported by other studies on the subject, a deep understanding of outsourcing re- quires us to move beyond the oversimplified ‘either/or’ discourse (typi- cal of the ‘make or buy’ perspective), complementing the analysis of the resulting trade-offs with the identification of ‘both/and’ opportunities (Lewis 2000), including the option to create partnerships with external organizations (Geyskens, Steenkamp and Kumar, 2006). Accordingly, this chapter aims to contribute to ‘re-booting’ the managerial mindset through a re-examination of outsourcing phenomena doing justice to their inner complexity. This is even more relevant in MNCs engaged in outsourcing and offshoring practices due to their presence in numer- ous countries with different institutional settings (e.g. Jarzabkowski and Sillince 2007; Marchegiani et al. 2010; Jay 2013; Perri and Peruffo 2014).

3.    Companies and their stakeholders

The idea that companies are subject to multiple pressures is not new. In fact any corporation is subject to different groups of ‘who or what really counts’ (Freeman 1984; Mitchell, Agle and Wood 1997) or ‘constituen- cies’ (Zammuto 1984). Such an assumption is opposed to the simplis- tic economic view by which only a company’s owners, or shareholders, count as the company, which is seen as a means to fulfil their aspira- tions (for example, create and increase value for them). By contrast, the

stakeholder approach argues that there are other parties involved, in- cluding employees, customers, suppliers (also outsourcees), business partners (for example, allied companies), financiers, communities, gov- ernmental bodies, political groups, trade associations and trade unions.

The inclusion of the unions in the framework in which companies might take the decision to outsource requires the illustration of certain concep- tual elements related to the way companies relate with their stakehold- ers, whereby we refer in particular to their identification and their role.

Donaldson and Preston (1995) introduce three distinct, though mutually supportive, approaches to identifying company stakeholders: descrip- tive, instrumental and normative. The descriptive approach explains company behaviour and characteristics, while the normative approach focuses on the function of the corporation, identifying the ‘moral or phil- osophical guidelines for the operation and management of the corpora- tion’ (Donaldson and Preston 1995, p. 71). Looked at from this angle, when it comes to outsourcing, the literature mostly describes companies as instrumentally oriented to financial and strategic goals (see later Par. 2.3, 2.3.1), with minimal consideration of other relevant stakeholders. A normative approach, instead, would suggest a more inclusive considera- tion of all stakeholders (including trade unions and workers’ representa- tives).

When it comes to the role that stakeholders can play, Mitchell, Agle and Wood (1997) propose a model combining the attributes of power (the extent to which one party has the means to impose its will in a relation- ship), legitimacy (socially accepted and expected structures or behav- iour) and urgency (time sensitivity or criticality of stakeholder claims) (p. 865-867). When a stakeholder bases his influence on the company solely on his power, he is classified as dormant. When facing a dormant stakeholder, managers are entitled to stay ‘cognizant’ (ready to react) but not fully responsive. On the other hand, a stakeholder is dominant when he has power and legitimacy, and definitive when he is also able to address urgent (temporal) claims. Viewing labour unions as claimants of compatible interests with stable links to companies could help them gain a louder voice in outsourcing decisions (e.g. Jones and Wicks 1999). In such a scenario, the role unions could play is particularly relevant in view of the evidence on outsourcing and offshoring affecting employ- ment dynamics and skills/competences over time (e.g. Timmermans and Østergaard 2014). Indeed, the outsourcing of activities does not

prevent companies from developing the related knowledge and compe- tences. For example, outsourcing IT still requires companies to maintain and update the related competences if they do not want to gradually lose control over their providers.

Within the described framework, unions can achieve a progressive cen- trality (relevance) in company outsourcing decisions by adopting proac- tive strategies. These would be based on the assumption that a deeper understanding of the ‘offshore outsourcing processes should have a long-term positive impact on the economic effectiveness of the company as well as the working conditions and quality of work of the employees in source and destination countries’ (Ramioul and De Bruyn 2006, p. 621).

4.    Understanding the outcomes of outsourcing

Outsourcing decisions involve complex iterative processes covering the whole company as well as external stakeholders. Such processes can however be broken down into three simple stages: 1) the decision to out- source: why and how firms decide to outsource certain activities; 2) im- plementation of the outsourcing decisions: how the process of outsourc- ing should be managed; and 3) assessment of outsourcing outcomes: what are the effects of the outsourcing decisions. Trying to assess the outcomes without considering the previous points would lead to only a partial understanding of the phenomenon. Hence, in dealing with out- sourcing outcomes we refer to a general framework – that proposed by Marchegiani et al. (2012) – extensively used in the literature: anteced- ents – processes – outcomes. The framework uses 10 boxes to classify the existing studies on outsourcing into three columns: i) antecedents,

1. ii) processes, and iii)

Within this framework, two groups of outsourcing studies can be iden- tified. The first contains studies that fit into a specific box; in general, these describe the relevant phenomena or variables. Within this group, outcomes can be viewed in different ways, ranging from an economic or financial view to a broader management vision. The outcomes can be classified into three categories: economic and financial, strategic and or- ganizational (see Par. 2.3.1, 2.3.2, 2.3.3). The description of such catego- ries also includes the relations between outcomes and such ‘antecedents’ as a company’s characteristics, its sources of competitive advantage and its corporate and business strategies.

Figure 1 Classificatory framework

Source: Marchegiani et al. 2012

4.1                   Economic and financial outcomes

Although outsourcing is generally pursued by companies seeking ef- ficiency gains, assessing the economic impact of outsourcing is not straightforward. From a transaction cost perspective, outsourcing deci- sions are taken when a comparison of make-costs and buy-costs is in fa- vour of the latter. Thus, in principle the economic outcome of outsourc- ing should refer to the difference between make and buy. In practice, organizational implications make such a comparison rather complex. Despite the emphasis on performance improvement in most of the liter- ature (e.g. Leiblein and Miller 2003), many studies evaluate this variable through perceptions of advantages, cost-cutting and efficiency, market share and overall exports (Bertrand and Mol 2013).

We present here the main indicators that can be used to measure eco- nomic and financial outcomes, together with their relationship to out- sourcing antecedents.

The relevant literature shows that economic and financial performance mainly refers to the reactions of financial markets to the announcement

of an outsourcing strategy and its effects on a company’s value. Oh, Gal- livan and Kim (2006) analyzed market perceptions in reaction to IT- related outsourcing announcements and found that investors tend to react favourably to outsourcing when the level of transactional risk is low, while reacting negatively when the arrangements imply high trans- actional risk. Transactional risks may be related to opportunistic behav- iours of providers, high switching cost for replacing them, the monitoring and control of the execution of the deals, increased resource depend- ency or suboptimal provider performance. Similarly, the study by Hayes, Hunton, and Reck (2000) found positive and significant market value gains for smaller companies compared to larger ones, and for service companies compared to non-service ones. In fact, such companies can overcome competitive gaps (for example, lack of expertise/competence) via the services acquired from the outsourcees.

4.2                     Strategic outcomes

The strategic outcomes of outsourcing have been analyzed from both an empirical and a theoretical standpoint. The conclusion drawn by Insinga and Werle (2000) seems still valid: outsourcing is motivated by growing pressure on management to remain competitive through ‘accomplish- ing more with less’. The attempt to achieve more with less is pursued by companies through organizational solutions such as restructuring, downsizing and the reengineering of activities (see for example Peruffo, 2013; Peruffo et al. 2014; Peruffo et al. 2014). Long-term innovation ad- vantages can be related to outsourcing when companies seek special- ized sources to complement and strengthen internal R&D (Bertrand and Mol 2013). Despite the number of studies on the effect of outsourcing on companies’ sustainable competitive advantage, actual findings on pro- ductivity and efficiency gains are still very fragmented and with a sig- nificance limited to specific industries/sectors (Giustiniano and Clari- oni, 2013). However, a number of interesting findings have been found at different strategic levels: a) corporate, b) business, and c) functional.

The effect of outsourcing on corporate strategies has been analyzed with respect to vertical integration. This corporate strategy involves a compa- ny integrating (value chain) activities which represent either inputs (for example, raw materials) or outputs (for example, distribution channels) for its core business. For example, a producer of musical instruments could acquire a producer of wood (input) or a chain of music stores

(output). Rothaermel, Hitt and Jobe (2006) investigated the relationship between strategic outsourcing and vertical integration and found that, by balancing vertical integration and strategic outsourcing, companies could enrich their product portfolio and boost product success, while in turn achieving a major competitive advantage. The latter is related to the degree and success of ‘selectiveness’, i.e. integrating those inputs/ outputs generating the best synergies with existing activities, and the ‘smart’ outsourcing of those services that can be delivered at better con- ditions by other companies (oursourcees). For example, in knowledge- intensive industries (for example, pharmaceutics, biochemistry and healthcare), selective outsourcing could occur in favour of specialized and focused suppliers and in business-related activities (Quinn, 2000). Following this path of study, the contribution of Parmigiani and Mitchell (2009) argues that the concurrent sourcing of complementary compo- nents becomes more common when companies have relevant knowledge of the components in conjunction with suppliers (inter-company exper- tise) and within the company (inner-company shared expertise).

Other studies have investigated the role of outsourcing strategies in pur- suing or enhancing competitive advantage at the business strategy level. For example, Gilley and Rasheed (2000) showed that company strategy mediates the relationship between outsourcing and performance. Spe- cifically, on the one hand, ‘cost leadership’ strategies foster a positive relation between peripheral outsourcing and financial performance and between core outsourcing and innovation performance. On the other hand, strategies of ‘strategic differentiation’ of products and services (for example, higher-quality products sold at higher prices than the competi- tors’ ones) have a negative relationship. The latter can be explained by the fact that while outsourcing can contribute to ‘cost leadership’ strate- gies via the acquisition of services at a lower cost, companies compet- ing on ‘quality/differentiation’ face more difficulties in nurturing their sources of competitive advantage from external partners. In fact, if the strategy is based on the final quality they should be able to acquire high- er-quality services compared to the internal provision thereof.

Similar to corporate and competitive strategies, researchers have also analyzed functional strategies in exploring the link between outsourc- ing decisions and expected outcomes. With respect to corporate strat- egy, Quélin and Duhamel (2003) studied the motives of corporate management in large European manufacturing companies engaged in outsourcing and the risks they perceive to be associated with strategic

outsourcing of ‘operations’. The study highlighted that the main issues companies might face are related to access to external expertise and the possible quality improvements, as well as the fact that operational cost savings must be in balance with the cost of monitoring suppliers. In a study of functional strategies with a focus on production and innovation activities, Murray, Kotabe and Wildt (1995) hypothesized a moderating effect of product-related factors (product innovations, process innova- tions and asset specificity) on the relationship between global sourcing strategy and the financial dimension of product market performance. The study summarized that achieving the financial targets set for a prod- uct subject to outsourcing depends on the levels of the specificity of the underlying assets, as well on the company’s product and process innova- tion.

In the research on company characteristics, various authors have fo- cused on the outsourced areas to identify the functions that might pro- duce more positive outsourcing outcomes. The majority of the literature in this area concentrates on IT, seen as the area where outsourcing strat- egy can be easily implemented with high potential benefits. However, work has also been done in other sectors. For example, Quinn (2000) assumed that innovation and R&D activities play an important role in outsourcing strategies, suggesting that a company can derive higher innovation returns through outsourcing the entire business process or process design activities that are not core competencies. Leiblein and Miller (2003) also investigated the link between innovation, viewed as a possible functional area for outsourcing, and the outcomes of an out- sourcing decision. They found that companies with more experience in a particular technological process are more likely to internalize manufac- turing activities than are companies lacking such experience. Similarly, companies with high levels of sourcing experience are more likely to out- source their production than those without such experience.

Finally, research on the characteristics able to explain or predict the suc- cess of outsourcing has not been limited to ‘internal’ features but has also included some ‘external’ factors, such as the quality of the relationship between a company and its partners. Linking this factor to outsourcing outcomes, Lee and Kim (1999) conducted an in-depth study to investi- gate whether partnership is an effective way of improving economies of scale and scope in traditional modes of organization. Their study em- phasizes that partnership is no guarantee for a desirable outcome. To achieve such, companies must pay particular attention to the quality of

partnerships resulting from outsourcing and ensure they are positively influenced by factors such as participation, communication, informa- tion-sharing and top management support.

4.3                         Organizational outcomes: the effect of outsourcing on human capital

In general terms, all outsourcing strategies affect the boundaries of a company, while the (global) distribution of value chain activities requires new coordination mechanisms (Srikanth and Puranam 2011). In partic- ular, Ceci and Prencipe (2013) posit that physical distance puts pressure on traditional coordination approaches, i.e. focusing on the ‘decompos- ability’ of activities into sub-activities and communication. Physical dis- tance, in fact, influences the knowledge boundaries created by the de- composition scheme. Conversely, the ‘decomposition’ of the value chain and the outsourcing of certain activities affect the division of labour within and outside company boundaries, even internationally. Such new configurations increase the inherent complexity of coordinating labour and knowledge, calling for new organizational practices. Furthermore, Gospel and Sako (2010) found that the ‘trajectory’ (outsourcing vs. in- sourcing – transformation vs. preservation of business processes) also affects the distribution of capabilities between users and suppliers.

The underestimation of the core importance of human capital in out- sourcing decisions can be also analyzed via other approaches. The indi- vidual and collective resistance to outsourcing could generate hidden (or less recognizable) costs, at least in the short run. Such costs are related to the sub-optimal use of human resources compensating for lacking coor- dination mechanisms, the sub-optimal allocation of human capabilities, poor preparatory training and inertia. The hidden nature of such costs is linked to evidence that actual costs (i.e. wages, bonuses, etc.) are gener- ated by resources unable to develop their full potential. More specifical- ly, such hidden costs are linked to some of the ‘dangerous myths about pay’ identified by Pfeffer (1998). In fact, managers tend to (wrongly) be- lieve that the cost of labour and labour rates are the same thing, and that therefore companies can reduce overall costs by cutting labour rates (i.e. by offshoring activities). The ‘managerial illusion’ related to this ‘myth’ is based on the under-consideration of real productivity (which on the cost side is the ultimate source of profits) in favour of a partial and myopic focus on pay (which is just the compensation/‘price’ paid for available

working hours). The effect of such myths might contribute to the blind- ness of MNCs (giants) when international strategies are driven by the tentative exploitation of low-cost localized advantages in labour mar- kets. Blind giants are generally defined as any social entity (for example, governmental agencies, regulatory bodies, leading organizations, lead- ing coalitions within companies, unions, other stakeholders) having the power to influence the future trajectories (David 1986) of ‘actors whose vision we would wish to improve before their power dissipates’ (Hanseth 2000, p. 68). All company stakeholders, including top management, can be trapped in this role when they try to promote or compare any interna- tional outsourcing initiative without critically assessing the effect of staff defending the ‘in-house’ activity on the whole business.

Despite the abundant literature on the strategic and economic impact of outsourcing, few works have yet focused on labour and workers’ per- spectives. Brooks (2006) focused on the potential effects of outsourcing on IT workers, finding that IT workers switch organizations, change jobs or exhibit different work-related behaviour (changes in motivation, in- volvement or commitment) according to the impact of outsourcing on individual perceptions of job alternatives and job-related satisfaction. By contrast, workers can decide to be loyal to their organization depending on how outsourcing impacts their perception of the profession, career- related opportunities and their ability to change careers. The conclusions drawn by Brooks (2006) can be generalized around the individual per- ception of outsourcing: where negative perceptions of outsourcing take precedence over an individual’s level of satisfaction and commitment, then the potential loss in performance, productivity and innovation can be detrimental to the company

Some other human reactions to outsourcing (of any kind of activity) are very similar to the ones observed by scholars analyzing IT infrastructure dynamics (Monteiro 2000; Hanseth 2000; Giustiniano and Bolici 2012). Following David (1986) in particular, it is possible to identify some typi- cal situations and actors as the angry orphans, namely groups of ‘users’ whose ‘routinized’ standards are changed. Any employees working in area which has any interdependence with an outsourced function might react with inertia or inefficiently to the change. Furthermore, additional and less easily recognized costs are hidden in the loss of competencies in activities believed to be non-core with regard to a company’s competi- tiveness (Giustiniano and Brunetta, 2015). Such hidden costs include the ex-post costs necessary to reacquire competencies and capabilities, or

the costs involved in controlling outsourced activities, with companies forced to hire specialist consultants to control and maintain relation- ships with providers.

The implementation of outsourcing strategies may also generate new organizational requirements, such as gateway roles, i.e. links between the internal and external sections of the same business process. Such roles may be taken on either by liaison/interface employees or by ex- employees of A who have been transferred to B in the course of outsourc- ing certain activities. Two scenarios might be relevant for human capital:

1) employees remaining with the outsourcing company may experience significant job enrichment through switching from operational duties to coordinating and controlling providers; 2) where employees are taken on by the outsourcee, they may suffer a temporary liminality possibly generating frustration and the loss of individual/organizational identity. The concept of liminality, taken from anthropology, refers to a state of being ‘betwixt and between’ different statuses (Garsten, 1999). Similarly to other contexts, when workers are transferred to another company in the wake of outsourced activities, they might temporarily suffer from cri- ses in their professional and organizational identities (e.g. De Bernardis and Giustiniano, 2015).

Furthermore, unions can play an important complementary if not ex- clusive role in supporting employees in the case of activities being re- sourced (Jussila, Gylling and Saarinen 2014). In fact resourcing, meant as the reacquisition of activities previously outsourced, requires the re- development of the related competences and skills by the company’s staff. This is even more relevant when resourcing also involves the re- hiring of previously transferred staff.

5.    Conclusion

The description of potential outsourcing outcomes emphasizes the role of organizations as social entities. Only proper consideration of the hu- man aspect (including workers and their representatives) of organiza- tions can help create the conditions necessary for achieving satisfactory outsourcing outcomes. With a view to explaining the possible outsourc- ing outcomes reported in management literature, we have tried here to unveil some of their inner paradoxes within a wider framework in which trade unions are considered as relevant stakeholders in a company’s

strategic and operational decisions. While most of previous literature focuses on the financial and strategic outcomes of outsourcing, we have tried to shift attention to the less consolidated evidence on its organiza- tional implication.

Insinga and Werle (2000) argue that the business environment pushes companies to perform certain functions in-house and the rest by aggres- sive outsourcing. Nevertheless, this strategic choice creates a number of dependencies which, in turn, can lead to unforeseen strategic vulnerabil- ity. Such vulnerability can be explained by referring to human capital.

On the one hand, outsourcing offers dramatic opportunities for a com- pany to enhance its competiveness, allowing it to reduce costs or to ac- quire external ‘quality’ from providers. On the other hand, the potential loss of distinctive competencies, embedded in a company’s human capi- tal, poses a major threat. Therefore, it is in the interest of the compa- nies to avoid such pitfalls. Conversely, unions can help them dodge such myopia.

References

Almond P. et al. (2005) Unraveling home and host country effects: an investiga- tion of the HR policies of an American multinational in four European coun- tries, Industrial Relations: A Journal of Economy and Society, 44 (2), 276-306.

Andriopoulos C. and Lewis M.W. (2009) Exploitation-exploration tensions and organizational ambidexterity: managing paradoxes of innovation, Organiza- tion Science, 20 (4), 696–717.

Angeli F. and Grimaldi R. (2010) Leveraging offshoring: the identification of new business opportunities in international settings, Industry and Innovation, 17 (4), 393-413.

Anner M., Greer I., Hauptmeier M., Lillie N. and Winchester N. (2006) The in- dustrial determinants of transnational solidarity: global interunion politics in three sectors, European Journal of Industrial Relations, 12 (1), 7-27.

Arrowsmith J. and Marginson P. (2006) The European cross-border dimension to collective bargaining in multinational companies, European Journal of Indus- trial Relations, 12 (3), 245-266.

Bartlett C.A. and Sumantra G. (1992) Transnational management: text, cases and readings in cross-border management, Homewood, Irwin.

Benassi C. (2011) The implementation of minimum wage: challenges and creative, Working Paper 12, Global Labour University.

Bertrand O. and Mol M.J. (2013) The antecedents and innovation effects of domestic and offshore R&D outsourcing: the contingent impact of cognitive distance and absorptive capacity, Strategic Management Journal, 34 (6), 751- 760.

Brooks N. (2006) Understanding IT outsourcing and its potential effects on IT workers and their environment, Journal of Computer Information Systems, 46 (4), 46-53.

Brunetta F. and Peruffo E. (2014) May parents inherit from heirs? Towards an un- derstanding of the parent-spun-off relationship, American Journal of Applied Science, 11 (6), 921-928.

Brunetta F., Giustiniano L., and Marchegiani L. (2014) Caring more by doing less? An enquiry about the impacts of outsourcing on patient care, American Journal of Applied Sciences, 11 (2), 273-279.

Cameron K.S. and Quinn R.E. (1988) Organizational paradox and transformation, in Quinn R.E. and Cameron K.S. (eds) Paradox and transformation: toward a theory of change in organization and management, Cambridge, Ballinger,

1-18.

Cameron K.S. and Quinn R.E. (2006) Diagnosing and changing culture: based on the competing values framework, San Francisco, Jossey-Bass.

Ceci F. and Prencipe A. (2013) Does distance hinder coordination? Identifying and bridging boundaries of offshored work, Journal of International Manage- ment, 19 (4), 324-332.

Clegg S.R., da Cunha J.V. and e Cunha M.P. (2002) Management paradoxes: a relational view, Human Relations, 55 (5), 483-503.

David P.A. (1986) Narrow windows, blind giants, and angry orphans: the dynam- ics of systems rivalries and dilemmas of technology policy, TIP Working Paper 10, Stanford, Stanford University, Center for Economic Policy Research.

De Fontenay C.C. and Gans J.S. (2008) A bargaining perspective on strategic outsourcing and supply competition, Strategic Management Journal, 29 (8), 819-839.

De Bernardis L. and Giustiniano L. (2015) Evolution of multiple organisational identities after an M&A: a case study from Europe, Journal of Organizational Change Management, 28 (3), 333-355. doi:10.1108/JOCM-05-2014-0096

Doellgast V. and Gospel H. (2012) Outsourcing and human resource manage- ment, in Bach S. and Martin E. (eds) Managing human resources: human resource management in transition, London, Wiley, 317-335.

Donaldson T. and Preston L.E. (1995) The stakeholder theory of the corporation: concepts, evidence, and implications, Academy of Management Review, 20 (1), 65-91.

Freeman R.E. (1984) Strategic management: a stakeholder approach, Boston, Pitman.

Garsten C. (1999) Betwixt and between: temporary employees as liminal subjects in flexible organization, Organization Studies, 20 (4), 601-617.

Geyskens I., Steenkamp J.B.E. and Kumar N. (2006) Make, buy, or ally: a transac- tion cost theory meta-analysis, Academy of Management Journal, 49 (3), 519-543.

Gilley K. and Rasheed A. (2000) Making more by doing less: an analysis of out- sourcing and its effects on firm performance, Journal of Management, 26 (4), 763-790.

Gittell J.H. (2004) Paradox of coordination and control, California Management Review, 42 (3), 101-117.

Giustiniano L. and Bolici F. (2012) Organizational trust in a networked world: analysis of the interplay between social factors and Information and Com- munication Technology, Journal of Information, Communication and Ethics in Society, 10 (3), 187-202.

Giustiniano L. and Clarioni G. (2013) The impact of outsourcing on business per- formance: an empirical analysis, Journal of Modern Accounting and Auditing, 9 (2), 153-168.

Giustiniano L., Marchegiani L., Peruffo E. and Pirolo L. (2014) Understanding out- sourcing of information systems, in Tsiakis T., Kargidis T. and Katsaros P. (eds) Approaches and processes for managing the economics of information systems, Hershey, IGI Global, 199-220.

Giustiniano L. and Brunetta F. (2015) Rethinking employability: new managerial competencies in a global labour market, Sociologia del Lavoro, 137, 17-35.

Gospel H. and Sako M. (2010) The unbundling of corporate functions: the evolu- tion of shared services and outsourcing in human resource management, Industrial and Corporate Change, 19 (5), 1367-1396.

Hayes D., Hunton J. and Reck J.L. (2000). Information systems outsourcing an- nouncements: investigating the impact on the market value of contract-grant- ing firms, Journal of Information Systems, 14 (2), 109-125.

Hanseth O. (2000) The economics of standards, in Ciborra C.U. (ed.) From control to drift: the dynamics of corporate information infrastructures, Oxford, Oxford University Press, 56-70.

Insinga R. and Werle M.J. (2000) Linking outsourcing to business strategy, Acad- emy of Management Executive, 14 (4), 58-70.

Jarzabkowski P. and Sillince J. (2007) A rhetoric-in-context approach to building commitment to multiple strategic goals, Organization Studies, 28 (11), 1639- 1665.

Jay J. (2013) Navigating paradox as a mechanism of change and innovation in hybrid organizations, Academy of Management Journal, 56 (1), 137-159.

Jones T.M. and Wicks A.C. (1999) Convergent stakeholder theory, Academy of Management Review, 24 (2), 206-221.

Jussila K., Gylling M. and Saarinen M. (2014) Case 2: the dynamics of make vs. buy decisions in a global economy – a firm study, in Slepniov D., Waehrens B.V. and Johansen J. (eds) Global operations networks: exploring new perspectives and agendas, Aalborg, Aalborg University Press, 307-334.

Lee J. and Kim Y.G. (1999) Effect of partnership quality on IS outsourcing success: conceptual framework and empirical validation, Journal of Management Infor- mation Systems, 15 (4), 29-61.

Leiblein M.J. and Miller D.J. (2003) An empirical examination of transaction- and firm-level influences on the vertical boundaries of the firm, Strategic Manage- ment Journal, 24 (9), 839-859.

Lewis M.W. (2000) Exploring paradox: toward a more comprehensive guide, Acad- emy of Management Review, 25 (4), 760-776.

Lommerud K.E., Meland F. and Straume O.R. (2009) Can deunionization lead to international outsourcing?, Journal of International Economics, 77 (1), 109-

119.

Marchegiani L., Giustiniano L., Peruffo E. and Pirolo L. (2012) Revitalising the outsourcing discourse within the boundaries of firms debate, Business Systems Review, 1 (1), 157-177.

Marchegiani L., Pirolo L., Peruffo E. and Giustiniano L. (2010) National corporate systems and outsourcing decisions: a cross-country analysis, Paper presented at the Strategic Management Society Conference, Rome, 11-15 September 2010.

Marginson P. and Meardi G. (2006) European Union enlargement and the foreign direct investment channel of industrial relations transfer, Industrial Relations Journal, 37 (2), 92-110.

Mitchell R.K., Agle B.R. and Wood D.J. (1997) Toward a theory of stakeholder identification and salience: defining the principle of who and what really counts, Academy of Management Review, 22 (4), 853-886.

Monteiro E. (2000) Monsters: from systems to actor-networks, in Braa K., Sorens- en C. and Dahlbom B. (eds) Planet Internet, Lund, Studentlitteratur, 239-249.

Murray J.Y., Kotabe M. and Wildt A.R. (1995) Strategic and financial perfor- mance implications of global sourcing strategy: a contingency analysis, Journal of International Business Studies, 26 (1), 181-202.

OECD STAN (2011) Organisation for Economic Co-operation and Development

– STAN STructural ANalysis Database. http://www.oecd.org/industry/ind/ stanstructuralanalysisdatabase.htm

Oh W., Gallivan M.J. and Kim J.W. (2006) The market’s perception of the transac- tional risks of information technology outsourcing announcements, Journal of Management Information Systems, 22 (4), 271-303.

Parmigiani A. and Mitchell W. (2009) Complementarity, capabilities, and the boundaries of the firm: the impact of within-firm and interfirm expertise on

concurrent sourcing of complementary components, Strategic Management Journal, 30 (10), 1065-1091.

Perri A. and Peruffo E. (2014) Knowledge spillovers from FDI: a critical review from the international business perspective, International Journal of Manage- ment Reviews. doi: 10.1111/ijmr.12054

Peruffo E. (2013) Verso una cultura del disinvestimento: strategia, governance e valore economico, Milano, Franco Angeli.

Peruffo E., Perri A. and Oriani R. (2014) Information asymmetries, family owner- ship and divestiture financial performance: evidence from Western European countries, Corporate Ownership & Control, 11 (4), 43-57.

Peruffo E., Pirolo L. and Nenni M.E. (2014) Spin-off and innovation in the phar- maceutical industry, International Journal of Engineering Business Manage- ment, 6 (17), 1-7.

Pfeffer J. (1998) Six dangerous myths about pay, Harvard Business Review, 76 (3), 109-119.

Pulignano V. and Doerflinger N. (2013) A head with two tales: trade unions’ influence on temporary agency work in Belgian and German workplaces, The International Journal of Human Resource Management, 24 (22), 4149-4165.

Quélin B. and Duhamel F. (2003) Bringing together strategic outsourcing and corporate strategy: outsourcing motives and risks, European Management Journal, 21 (5), 647-661.

Quinn J.B. (2000) Outsourcing innovation: the new engine of growth, Sloan Man- agement Review, Summer, 41 (4), 13-28.

Ramioul M. and De Bruyn T. (2006) The MOOS Project: a response from trade unions to offshore outsourcing, Transfer: European Review of Labour and Research, 12 (4), 621-625.

Rothaermel F.T., Hitt M.A. and Jobe L.A. (2006) Balancing vertical integration and strategic outsourcing: Effects on product portfolio, product success, and firm performance, Strategic Management Journal, 27 (11), 1033-1056.

Sippola M. (2011) Nordic subsidiaries in the Baltic States: is model transfer pos- sible?, Employee Relations, 33 (4), 356-374.

Sorrentino M., Giustiniano L., Depaoli P. and De Marco M. (2015) Shared services: exploring the new frontier, in Nóvoa H. and DrOgoicea M. (eds) Exploring services science, Springer International Publishing, 215-227.

Smith W.K. and Lewis M.W. (2011) Toward a theory of paradox: a dynamic equilibrium model of organizing, Academy of Management Review, 36 (2), 381-403.

Srikanth K. and Puranam P. (2011) Integrating distributed work: comparing task design, communication, and tacit coordination mechanisms, Strategic Manage- ment Journal, 32 (8), 849-875.

Timmermans B. and Østergaard C.R. (2014) Offshoring and changes in firms’ domestic employment: the case of Denmark, in Slepniov D., Waehrens B.V. and Johansen J. (eds) Global operations networks: exploring new perspectives and agendas, Aalborg, Aalborg University Press, 15-53.

Zammuto R.F. (1984) A comparison of multiple constituency models of organiza- tional effectiveness, Academy of Management Review, 9 (4), 606-616.

Chapter 3

Capital and labour on the move: the dynamics of double transnational mobility

Carina Altreiter, Theresa Fibich and Jörg Flecker

1.    Introduction1

The dynamics of transnational value chains and production networks have resulted in highly fragmented and geographically dispersed manu- facturing processes in a wide range of industries. In the literature on transnational companies and employment relations the focus is mainly on TNCs relocating or offshoring work with the aim of ‘regime shopping’ or ‘institutional arbitrage’. This implies that companies aim at tapping national, regional or local labour markets with lower wages, higher flex- ibility or less stringent health and safety standards. There are, however, cross-border restructuring processes that are not motivated by access to local labour markets but, on the contrary, rely on transnational mobile labour. We argue that both the movements of jobs and the movement of workers may undermine working and employment conditions through a dis-embedding of employment relations. Therefore, those circumstanc- es in which mobile capital employs mobile labour deserve our special attention.

In order to grasp the complexity of this dis-embedding of employment relations we look beyond the perspectives focussing either on the move- ment of jobs or on that of workers, and include both the issue of capital and labour mobility. This constellation, which we call ‘double mobility’, can be found within both the dynamics of complex transnational pro- duction networks and in logistic networks in the realm of distribution. The mobility of both capital and labour can be understood as a complex interplay of (spatial) fixity and mobility, and of societal and territorial embedding and dis-embedding.

1. We are grateful for the comments on earlier versions of the paper we got during the ETUI workshops. We are especially indebted to Jan Drahokoupil and Ines Wagner for their de- tailed suggestions.

We argue that constellations where mobile capital employs mobile work- ers may alter our assumptions about the consequences of foreign direct investment, outsourcing and offshoring. Moreover, such constellations need to be analysed in detail because new forms of working conditions, employment relations, divisions or solidarities between workers may emerge. It goes without saying that this is highly important for trade unions and their capacities to organize and represent workers.

To illustrate this ‘double mobility’ phenomenon, we draw on cases re- cently discussed in academic publications but also in the press. On the basis of literature reviews we present four cases: Foxconn, Amazon, the Chinese fashion industry cluster in Italy and the Danish slaughter indus- try. The paper starts with a short definition of terms (1) followed by the presentation of the four cases (2). It then discusses the consequences for management and control strategies, the role of labour market interme- diaries (LMIs) and the dis-embedding of employment relations (3). The paper concludes with a discussion of challenges for trade unions (4).

2.    ‘Double mobility’: key concepts and definitions

When discussing ‘double mobility’ the first concept that needs clarifi- cation is mobile capital. David Harvey (1982) pointed out the inherent tension between spatial ‘fixity’ and ‘mobility’ when it comes to analyz- ing capital in its various forms (as a commodity, in forms of money in- vested in production facilities). Capital, therefore, is always both fixed and mobile to a certain extent. Transnational mobility implies the ca- pacity and actual strategy to move assets across national borders, i.e. to undertake foreign direct investment or divestment, to relocate busi- ness functions or to offshore certain activities. Potential capital mobility is important in locational decision-making and in threats of relocation and divestments. Sassen argues that (1991, 2001, 2002) the mobility of capital cannot be reduced to a locational dimension. To capture recent global developments the ‘reorganization of sources of surplus value’, ‘the transnationalization of ownership’ but also the technologies that enable mobility and the sources of ‘maintaining control over a vastly decen- tralized global production system’ become crucial (Sassen 1991: 22f.). The increased mobility of capital has shaped employment relationships and international labour migration. Research on work, labour and in- dustrial relations has a long tradition of looking at the consequences of the transnationalization of capital (Marginson and Sisson 1994, 1996;

Ferner 1997; Edwards et al. 1999; Smith 2006; Arrowsmith and Mar- ginson 2006). Research questions range from changing power relations between capital and labour to the transnationalization of worker rep- resentation institutions and the development of work organization and human-resource policies in TNCs. Transnational companies have the potential to weaken local employment relations for three reasons: first, because of their enhanced power position based on their greater mo- bility and the opportunities for ‘coercive comparisons’ and ‘concession bargaining’; second, because they can introduce employment relations practices from abroad; and, third, because local management often has only limited powers of decision-making.

Mobile labour, the second key concept, is highly relevant in our context because migrant or otherwise mobile workers are often more vulnerable than others, since they are often up-rooted and not familiar with local laws, customs and practices, and because they are less likely to be organ- ized and in a position to resist deviations from local employment relation practices and thus the dis-embedding of employment relations. ‘Regime shopping’ is not only possible through offshoring jobs abroad but also by recruiting workers from other national labour markets (Lillie 2010; Wagner and Lillie 2014). Mobile labour may take various forms which are linked to different mobility or migration regimes and specific legal frameworks and regulations. Its meaning ranges from work migration to undocumented foreign workers but also covers cross-country temporary agency work or the posting of workers. As our cases are located in Euro- pean countries we will focus solely on relevant aspects for that region. At EU level there are three Directives that play a major role in transnational mobile labour. First, the Employer Sanctions Directive (2009/52/EG) which aims at combating illegal immigration into the EU by prohibit- ing the employment of third-country nationals without a residence and work permit. Second, the Posted Workers Directive (Directive 96/71/ EC) which covers workers being sent to another EU Member State in- cluding the posting of temporary agency workers.2 While the third one, the more recent Temporary Agency Work Directive (2008/104/EC), does not exclude cross-border temporary agency work, it does not ex- plicitly mention it either. However, the ‘different aim and legal basis’ of the last two Directives ‘are conflicting, creating discriminatory measures for cross-border workers’ (Schömann and Guedes 2012: 58).

2. Chapters 7 and 10 also address the posting of workers in the European

Hence, ‘double mobility’ denotes constellations in which transnational and potentially or de facto mobile capital does not make use of the lo- cal labour market but employs mobile workers. We can assume that the forces undermining employment relations will cumulate in such con- stellations. Transnational and mobile capital may transfer management strategies from other contexts or gain concessions from the workforce by threatening to relocate jobs. In such situations mobile labour, for several reasons, may be less likely to offer resistance and defend local regula- tions, customs and practices. We can assume that this will accelerate the dis-embedding of employment relations and hinder union attempts to ‘re-embed’ practices they encounter in transnational value chains.

3.    Capitalizing on ‘double mobility’ in manufacturing and logistics

Case 1: Foxconn Electronics in Europe

Foxconn Electronics, a subsidiary of the Taiwanese company Hon Hai Precision Industry, is one of the world’s biggest manufacturers of elec- tronic products and produces for such leading companies as Apple, Sony or Microsoft (Ruckus 2013). In recent years, Foxconn has been criticized for its bad working conditions, and in particular for several suicides of young workers in its Chinese factories in 2010 and 2011 (Andrijasevic and Sacchetto 2013). The company employs some 1.3 million people worldwide. Although most of them work in Foxconn factories in China or South-East Asia, Foxconn has also established production facilities in Europe, for example, in Czechia, Hungary and Slovakia, allowing it to label its products ‘Made in Europe’.

In its European plants, Foxconn relies only partly on local labour mar- kets. Instead, it is especially interested in recruiting young and mobile workers independent of their origins as they are considered to be used to adjust to different working regimes quickly and to be able to manage the fast-paced work. In the two Foxconn plants in Czechia (Pardubice and Kutná Hora) we are focusing on, more than 50% of staff are temporary workers, especially from countries such as Slovakia, Poland, Romania, Bulgaria, Mongolia, Ukraine or Vietnam (Andrijasevic and Sacchetto 2013), whereby the majority are recruited by temporary work agen- cies (TWAs). As the company is characterized by just-in-time produc- tion and high demand fluctuations, temporary workers ensure a flexible workforce adaptable to the specific order situation. In addition, forms of

management similar to those in China have been established. To achieve this, Foxconn hired predominantly Chinese and Taiwanese managers at first and sent Czech staff for training to Foxconn factories in China. To achieve production targets, employees face an intense pace of work and constant monitoring during their shifts (Andrijasevic and Sacchetto 2013).

The internal division of labour is differentiated by the types of contracts, by nationalities and gender. There is a clear separation between Czech workers and migrants and between permanent and temporary workers. Czech men often occupy better or leading positions while migrants and Czech women work on the production lines. As the working language is Czech, communication difficulties between workers are common. More- over, many migrant workers do not even know the content of their con- tracts (Andrijasevic and Sacchetto 2013).

Most of the temporary workers live in the residences provided by the TWAs, located near the production site. This saves time and effort for the workers as they do not have to look for accommodation on their own and do not have to worry about language skills which may be required when renting a local advertised apartment. Hence, there is no need to come into contact with the local community. Conversely, ‘ethnical’ divi- sions within the residences ensure a certain kind of social embedded- ness in a community of people with the same nationality. But, as a side- effect, these dormitories provide management with permanent access to the workforce. On this note, a temporary worker from Poland reported being picked out spontaneously for an additional shift in the evening by a coordinator. Furthermore, dormitories allow forms of staff supervi- sion going beyond work-related issues and entering the personal sphere, such as monitoring smoking or alcohol consumption (Andrijasevic and Sacchetto 2013).

Case 2: Amazon warehouses in Germany

Amazon.com is one of the biggest online retailers operating worldwide.3 To meet the demands of a growing online market, the company is con- stantly opening new sites but also relocating distribution and logistic centres. Locational decisions are guided by available infrastructure and

3. As there are currently no scientific studies available on that case we rely on newspaper arti- cles and reports from former workers.

the accessibility of markets. Fulfilment and distribution centres in Eu- rope are found in France, Germany, Italy, Spain and the UK, and most recently also in Czechia and Poland. In these two countries, the sites are located close to Germany, one of Amazon’s main markets in Eu- rope, yet are more profitable as wages are lower. Moreover, conflicts with unions are not expected, as workers are less unionized, with only 12% of workers in Poland and 17% in Czechia belonging to a union.4 In the following we will focus on a case in Germany which was broadly discussed in the media after the ARD broadcast a documentary on the working conditions of Amazon workers at its Bad Hersfeld distribution centre, where structures are similar to other sites in the UK or United States. Amazon warehouses have a high demand for workers. At the Bad Hersfeld centre, as elsewhere, only a small percentage of the workforce is recruited from the local labour market. Workers from all over Eu- rope (Spain, Hungary, Romania and Poland) are employed especially to cover distribution peaks. The company relies on TWAs, like Trenk- walder or Manpower, to provide enough suitable workers on time.5 In the case of Bad Hersfeld, foreign mobile workers are accommodated in nearby hotels, holiday villages or motels and bussed to their workplace and back again. As the ARD documentary shows, people have to live un- der degrading conditions, are completely isolated from the local commu- nity and under constant surveillance from a private security company in their dormitories.6 Workers report similar restrictive conditions at the workplace, with ‘pickers’ being tracked by GPS and constantly monitored and warned when they are too slow. Talking to co-workers, going to the toilet too often, not meeting expected performance targets or being late lead to workers being given points that can lead to contract termination.7

4. Frankfurter Rundschau, 09.2014 (http://www.fr-online.de/wirtschaft/baubeginn-fuer- versandlager-nahe-prag-amazon-liefert-kuenftig-aus-tschechien,1472780,28423394.html – 20.01.2015); union density and worker participation (Fulton 2013).
5. Amazon’s business model features only having a small permanent Therefore the majority of employees are temporary agency workers. Among those we can further distin- guish between foreign and domestic temporary workers.
6. Süddeutsche, 02.2013 (http://www.sueddeutsche.de/wirtschaft/arbeitsbedingungen-bei-

online-haendler-arbeitsagentur-belastet-amazon-leiharbeitsfirma-1.1605634 – 20.01.2015).

7. Frankfurter Allgemeine Zeitung, 21.02.2013 (http://www.faz.net/aktuell/wirtschaft/ unternehmen/hinter-den-kulissen-das-tagebuch-einer-amazon-packerin-12089481-p3. html, 01.2015); Focus, 25.02.2013 (http://www.focus.de/finanzen/news/unternehmen/ tid-29719/ueberstunden-und-fristvertraege-insider-packen-aus-so-springt-amazon-

mit-seinen-buero-angestellten-um_aid_926999.html – 20.01.2015).

Case 3: a Chinese fashion industry cluster in Italy

The third case is not a specific company but a business cluster in the fashion industry in Northern Italy. Back in the 1980s Chinese migrants established ethnic workshops employing Chinese workers to do (sub-) contracting work for Italian clothes manufacturers. Especially in Tus- cany but also in Milan and Veneto more and more Chinese workshops popped up as workers went into business themselves. Producing in Italy allows them to label products ‘Made in Europe’ (Ceccagno 2007; Wu and Sheehan 2011). However, the workshops do not rely on the local labour force but exclusively on workers from China. Ethnic networks function as labour market intermediaries (LMIs) for supplying Chinese workers who are then employed by the shop-owners in Italy. Legal regulations are often flouted and there is a high level of undeclared work. As Wu and Sheehan (2011) point out for Veneto, not all Chinese fashion enterprises are officially registered, companies shut down and re-open frequently, and many of the foreign workers do not hold resident permits (2011: 139). This seems to be – at least to some extent – tolerated by the lo- cal authorities. Gaining legal status for Chinese workers usually means they would have to pay (almost) all taxes and social security contribu- tions. Production schedules require almost non-stop availability, leav- ing hardly any room for activities and duties outside work, for example, childcare. As Ceccagno (2007) reports, workers often remain childless or send babies back to China to be taken care of by grandparents when they are unable to organize childcare in Italy.

To meet the short production cycles in the fashion industry and offer low-cost products, working conditions in these ethnic workshops are rough. ‘The Chinese workers work at night, or 15 or 16 hours in a row; in the high season even 30 hours. These stretches of time are only inter- rupted by short naps leaning on the sewing machine’ (Ceccagno 2007: 641). Due to the long working hours workers usually work and sleep in the same place, with regular meals provided by the workshop owners. The Chinese workers form a large ethnic group in these Italian cities and can be described as ‘closed communities’; a consequence of the high level of control and dependency. As a result, many of the workers who stay in Italy for a longer period of time – some even decide to stay in the hope of being able to start their own business someday – hardly blend in with the local community.

Case 4: the Danish meat industry on the move

Our last case is the Danish slaughter industry, with a special focus on Danish Crown Corporation, a key player in in the international pork in- dustry (Refslund 2012). About 10 years ago, due to increasing interna- tional competition, Danish Crown started relocating several production sites mainly to Germany and Poland, offshoring certain activities like deboning or cutting, as these countries have a reputation as low-wage countries in this industry.

The slaughter industry in Denmark is known for its established culture of unionization and militancy. The threat of wage cuts and job relocation sparked off a public debate and national protests, driven by the concern that concession bargaining could spread to other industries. Although the outsourcing threats weakened the unions’ negotiation position at that time (Gorm Hansen 2007), in the end, Danish workers and unions did not accept lower wages. Though this led to Danish companies relo- cating large parts of production to Germany and Poland, the remaining jobs in Denmark were paid without cuts in wages (Refslund 2012).

In its ‘new’ German slaughterhouses, Danish Crown prefers not to em- ploy the local workforce, with most employees being temporary workers from Eastern Europe subcontracted via TWAs. Refslund (2012) reports that less than 10% of the workforce at Danish Crown’s German produc- tion sites are employed directly by the company. As a consequence, Dan- ish Crown is able to pay lower wages and avoid paying social security contributions in Germany for most of their workers.

In the German slaughter industry mobile workers live in substandard shared apartments near the production sites and are under constant management surveillance (Wagner 2015). One newspaper report8 re- fers to two workers sharing a bed, using it alternately in line with their shifts. As labour organization traditions are weak in Eastern Europe, temporary workers in Danish Crown slaughterhouses in Germany are often reported to be sceptical about unions, a circumstance which makes it difficult for local unions to get in touch with them and to complain about bad working and living conditions. Nevertheless, Refslund (2012: 123) reports of Polish workers complaining about low wages, inciting

8. Die Zeit, 17.12.2014 http://www.zeit.de/2014/51/schlachthof-niedersachsen-

fleischwirtschaft-ausbeutung-arbeiter/seite-1, 17.12.2014 – 23.01.2015.

the slaughterhouse companies to substitute them for Romanian and Ukrainian workers. Many of the Poles have since moved on to Ireland or England.

4.    The consequences of ‘double mobility’

Though the cases presented above vary in terms of types of work, work- ers’ nationalities, outsourcing relationships, employment relations, etc., they show several similarities which we would like to explore in the fol- lowing. Before going into details about the consequences for trade unions we would like to first discuss some crucial aspects of ‘double mobility’ common to the cases presented above: management control strategies, the role of LMIs and the dis-embeddedness of workers.

4.1                   Management control strategies

In the illustrative examples of ‘double mobility’ presented above, man- agement control and deployment of labour are in many ways character- istic of the respective sectors and types of business activities. Electron- ics manufacturing, the meat industry, but also distribution companies are often strongholds of Taylorist work organization. But it also became clear that the transnationality and mobility of capital play a role. First, management strategies and the way managers are recruited and trained are influenced by corporate headquarters (Voss et al. 2010). Perfect il- lustrations of this ‘home-country effect’ are the Foxconn sites in Europe. Second, despite investments in factories and warehouses and thus a ‘spa- tial fix’ (Harvey 1982), capital tends to be mobile. Amazon could move its distribution centres from Germany to Poland, should the German trade union (ver.di) continue to be successful in organizing strike actions – as happened when Danish Crown relocated most of its production to Ger- many because workers refused to accept lower wages.

One very particular ‘spatial fix’ is the use of migrant mobile labour usu- ally without local family ties. ‘Having workers live next door to a fac- tory so that they might always be on call for work’ (Rainnie et al. 2010: 300) allows for high levels of temporal flexibility and extended working hours. Following Pun and Smith (2007) this specific setting can be de- scribed as a ‘dormitory labour regime’. The term was established in the context of today’s Chinese production regime, even though providing

accommodation for workers has a long history. The function of these dormitory labour regimes is to ‘capture single migrant workers for short periods of tenure in order to maximize the utilization of labour services during the working day’ (Smith 2003: 334). We similarly find this aspect of extended control in our cases when Foxconn workers reported be- ing picked up without notice for an additional shift (Andrijasevic and Sacchetto 2013), in German slaughterhouses (Wagner 2015) or in the Chinese workshops in Italy where due to long working hours workers usually sleep on the premises. Numerical flexibility is further enhanced by the services of LMIs who supply local and foreign labour to cover peak demand.

Control strategies and personnel policies are clearly interrelated: using a mobile and highly replaceable workforce does not allow high-road strat- egies for work organization and productivity. Instead, hyper-Taylorism – as in Amazon warehouses – is both possible and appropriate for cap- ital under such circumstances. It is possible because workers lack the power to oppose these control strategies, and it is appropriate because management does not have to draw on skills provided by a particular national vocational training system. In relation to management control and deployment of labour, ‘double mobility’ is thus characterized by ex- treme forms of asymmetry of power between capital and labour and by far-reaching dis-embeddedness with regard to the surrounding society and institutions. Naturally, both asymmetry of power and dis-embed- dedness are contested: trade unions are trying to alter the power rela- tions through organizing industrial action and, in this way, to re-embed the employment relations.

4.2                         Outsourcing recruitment and employment to Labour Market Intermediaries

In the cases presented above, Labour Market Intermediaries (LMIs) played an important role as the HR function responsible for recruit- ing and managing the employment contracts of the temporary labour were often outsourced to different kinds of agencies. This prominence of LMIs may be due to the selection of examples and thus a coincidence. We would argue, however, that there are reasons why LMIs are crucial facilitators of ‘double mobility’. What characterizes labour market inter- mediaries? LMIs broker the relationship between capital and labour, i.e. trying to bring employers and employees together. But LMIs also serve

other purposes, most importantly reducing transaction costs, shaping compensation levels, displacing risks and building networks (Benner 2003). LMIs are not a new phenomenon and we can distinguish be- tween public, membership-based and private ones (Benner 2002; Ben- ner 2003; Enright 2013). We assume that LMIs have a specific position in double mobility constellations, as they enable flexible and location- independent access to the labour force, thereby facilitating and possibly accelerating the double mobility of capital and labour. Generating their profits from selling their services to companies, private LMIs in particu- lar have become key players within transnational production networks as they determine how, where and which people are employed. Follow- ing Benner and Enright we distinguish three different types of private LMIs:

• Temporary work agencies (TWAs) and contract brokers:9 based on an ILO definition also used by the European Parliament we define TWAs as agencies who ‘provide temporary employees to user enterprises to cover employee absences, skill shortage and varying seasonal work- loads. Workers are employed and paid by the agency, but are con- tracted out to a client for either a prearranged fee or an agreed hourly wage’ (Eichhorst 2013).
• Executive search firms, head-hunters and employment placement agencies: these agencies only refer or place job applicants, acting as intermediaries but not as employers.
• Informal intermediaries: Enright (2013) equates these intermediar- ies with the so-called ‘gangmasters’ in the UK who broker unskilled workers especially in the agricultural sector. Typically, foreign work- ers are involved, some of them with illegal status and vulnerable to exploitation and forced Often there is no clear-cut distinction between this and the human trafficking of human beings for labour exploitation.

In our cases TWAs and informal intermediaries are the most important forms of LMIs. In all four cases, recruiting a suitable labour force is out- sourced to LMIs (either TWAs or informal intermediaries). But there are also other benefits for companies. As Amazon, Foxconn and Dan- ish Crown only have a small permanent work force and get the majority

9. It is difficult to find a general definition as it is highly dependent on national regulations and statutory

from TWAs, they can offload the risks and costs associated with regular employment contracts to the TWA, while remaining flexible themselves. In addition, LMIs and security firms play an important role in organizing the specific spatial fix of capital and labour in that they are responsible for the accommodation arrangements and, consequently, for ‘immobi- lizing’ foreign mobile labour in a strategy to achieve extreme levels of numerical flexibility.

4.3                     Disembedding employment relations

The similarities between the case examples above also relate to the rela- tive importance of the societal context of the business activities and, in particular, its impact on employment relations and terms and condi- tions of employment. The cases presented illustrate how transnational capital, outsourcing and the use of mobile labour have resulted in dis- embedded employment relations in the sense of minimising the influ- ence of national, regional or local regulations, customs and practices. As Wagner and Lillie (2014) state, blurred boundaries and deterritori- alization enable companies to exit from specific norms and industrial relation systems by using mobile non-domestic workers. But this phe- nomenon is double-sided: at the same time companies, in taking loca- tional decisions and threatening to offshore activities, basically decide where they employ their mobile workers. This double mobility leads to both societal and territorial dis-embeddedness (Hess 2004: 177f.). More precisely, when capital and labour are mobile, such institutions as cus- toms and practices or regulatory frameworks are less likely to shape the actions of individuals and groups, i.e. they are less attached to particular places. While the literature on embeddedness is more often concerned with the relation between companies and their societal environment, the concept may also be used for workers or, closer to the meaning Polanyi (1944) gave to the term, for the exchange relations between employers and workers.

Looking at the employer side we can draw on extensive research on transnational companies, outsourcing and offshoring and the transfer of forms of human resource management and work organization. De- pending on their organizational structures transnational companies are in a strong position to ‘regime shop’ in their locational decisions, to use ‘coercive comparisons’ between subsidiaries or, more generally, units within the production networks and to enforce uniform methods more

easily (Marginson and Sisson 1994, Marginson et al. 1995, Thompson et al. 1995, Edwards et al. 1999). The need for workers’ skills and commit- ment, however, may lead to a certain adaptation to local circumstances (Mueller 1996). As a consequence, there is also a certain ‘power of the place’ that mainly depends on the emergence of (collective) actors and their strategies (Flecker 2000).

The examples presented above have partially shown the home country effect frequently described in the literature. Turning to labour, mobile workers are per definitionem not rooted in the local community and have not been socialized into the traditions of the regional working class. They therefore lack the source of power local communities and workers’ organizations may provide (Rainnie et al. 2010: 302). While it would be mistaken to equate ‘places’ with communities, as today’s geographic places should rather be conceptualized as intersections of partly global circuits (Massey 1994), it is still safe to say that localities or geographical proximity support the transfer and thus convergence of workers’ demands and the emergence of mutual support or collective ac- tion. Recruiting mobile labour can thus be seen as a way to cut off such power resources. A case in point is the replacement of mining towns by ‘fly-in / fly-out’ (FIFO) labour regimes in the Australian mining industry (Mayes 2014). The effect of using mobile labour is enhanced by strate- gies to keep individuals or groups of workers divided. In the more ex- treme cases, mobile workers in particular places are not only separated by language barriers and cultural differences but also by the forms of accommodation and surveillance in, for example, the dormitory labour regimes described above. However, mobile labour is not only character- ized ex negativo by what they are lacking, and the concept of embedded- ness may also be used to denote what they carry with them. Referring to Polanyi, Hess emphasizes ‘the history of social networks and the cultural imprint or heritage of actors that influence their economic behaviour ‘at home’ as well as ‘abroad’’ (Hess 2004: 177). Disembedding thus should not mean that workers necessarily become powerless. Rather, there is still leeway for individual agency if only because workers may influence work processes through bringing with them particular ways of working, demands and expectations.

In their attempt to open up a geographical perspective for labour process analysis Rainnie et al. (2010: 303) argue that ‘it is evident that local la- bour control regimes and how work is organized are shaped significantly by the manner in which workers and capitalists are differently embedded

in the economic landscape. This is important, because when capital and workers are relatively fixed in particular places within the landscape, it is likely that certain traditions of work and ways of organizing the labour process will become congealed in those places, at least until massive eco- nomic restructuring and the creation of new spatial divisions of labour begin to erase them.’ It is obvious that ‘double mobility’ means low levels of embeddedness and thus weak traditions and ways of working.

5.    Challenges for trade unions

The mobility of capital and labour brings specific challenges for trade unions. They find themselves dealing with a fluctuating, fragmented, so- cially isolated and non-local labour force including the employment of temporary agency workers while also facing ‘coercive comparisons’ and pressures on their bargaining power due to actual or threatened reloca- tion or restructuring.

The EU regulatory framework is not only ambivalent in, on the one hand, opening up possibilities to recruit workers from other national labour markets while, on the other, providing some guidelines for regulating the employment of mobile labour. What is more, national differences in the transposition of directives in combination with national law create very diverse legal situations and set the scope for how companies can handle mobile labour. Research suggests that reality deviates from the statutory framework. Companies seem to be very sophisticated when it comes to finding and using legal loopholes to circumvent European regulations and national law (Staples et al. 2013; Wagner 2014). Even if employment contracts are correct ‘on paper’, in practice regulations are often circum- vented, for example by undercutting the hourly minimum wage through the manipulation of working hours or by withholding annual leave pay (Wagner 2014). In other cases, there are workers declared as ‘posted’ by their employers although this is not the case in a legal sense, and we also find very different types of employment relationships which are neither one nor the other: i.e. workers posted by subcontractors, workers posted by agencies, locally hired migrant workers, self-employed foreign work- ers or posted third-country nationals or others who have a temporary work permit (Berntsen and Lillie 2014). Often these practices go unno- ticed and unchallenged either because workers lack sufficient language skills and are unaware of the legal situation in the host country or be- cause they are afraid to lose their jobs if they complain.

As we know from various examples the threat of relocation or offshoring puts pressure on unions and can lead to forms of concession bargaining. Often unions yield to companies’ demands for flexibility and deregula- tion in the hope that they can preserve employment in the country. How- ever, there are different ways to handle this pressure. In Denmark for instance, the Danish slaughter industry is known for its high rate of col- lective agreement coverage and trade union membership. But, with un- ions and workers refusing to accept wage reductions, companies started to relocate most of their production to Germany, because of the specific German regulatory environment allowing the use of low-paid subcon- tracted staff. Recently, Amazon Europe has started to look for less un- ionized locations than Germany (but still close to its main markets), such as Poland or Czechia. This is the result of the pressure exerted by Ger- man unions in recent months.10

Additionally, the segmentation and fragmentation of a company’s work- force, including the employment of highly mobile foreign workers from different origins, is also challenging for unions. On the one hand unions are concerned that local employment, social security and wage standards are being undermined. On the other hand it is difficult to organize these mobile workers. As the cases show, the lack of language skills and social isolation make it difficult to get in touch.11 At Foxconn and Amazon, but also in the Chinese workshops, employees live in isolated locations near or on the production site. At the Amazon warehouse in Germany peo- ple are shuttled from work to their accommodation where they are kept under surveillance by a private security company. Chinese workers in Italy work and live in the workshops with windows blackened-out and without any contact to the local community (Ceccagno 2007; Wu and Sheehan 2011).

The cases also suggest that this workforce fragmentation leads to rivalry among workers (between different job positions, between different eth- nicities etc.) and makes solidarity difficult. Many workers hope to gain a better position or a permanent contract and therefore accept poor

10. Frankfurter Rundschau, 16.9.2014 (http://www.fr-online.de/wirtschaft/baubeginn-fuer- versandlager-nahe-prag-amazon-liefert-kuenftig-aus-tschechien,1472780,28423394.html – 01.2015).

11. Due to their lack of language skills workers often do not know the content of their con- There is a story of a Romanian worker in a German slaughterhouse who signed his dismissal without knowing it – a further sign that it is common to let workers sign docu- ments in a language they do not understand.

working conditions. A high turnover of workers at the sites and work- ers’ scepticism towards unions make it even more difficult for unions to organize employees (Schmidt 2006). Additionally, workers are usually highly dependent on the income which is often higher – although not much – than in their home countries. They are afraid of getting sacked and losing their residence permits if they complain or join a union. In the Danish-German slaughterhouses demands by Polish workers for higher wages led to Romanian and Ukrainian workers being employed instead (Refslund 2012).

In all our cases a flexible and cheap workforce is essential for the produc- tion process. Managers often praise migrant workers as ‘good workers’ with a strong ‘work ethic’ and compliance (MacKenzie and Forde 2009), willing to work flexibly, to accept low wages, manage fast-paced work and work extensive hours. For transnational companies they are thus a valuable alternative to the local labour force. Using temporary workers and thereby opening up the possibility of ‘a consistent renewal of the workforce is a strategy to maintain the favoured quality of work ethic’ (MacKenzie and Forde 2009: 150). In general, while capital mobility enhances capital’s power resources vis-à-vis labour as it, for example, enables concession bargaining, labour mobility tends to weaken collec- tive labour through dis-embedding labour and dividing workforces.

However, labour mobility does not necessarily mean that workers bring with them the desired work ethos. They may transfer traditions of both quiescence and militancy to other geographic regions (Rainnie et al. 2009). Employers might have a certain workforce in mind but there are always aspects of uncertainty about how people come to grips with work settings. Research shows that the ‘willingness to accept low pay and low-status work diminishes as individuals and migrants commu- nities become more settled in the receiving country (…)’ (MacKenzie and Forde 2009: 145). In our example of Chinese workers in Italy, new communication technologies like the social media and mobile phones support a mutual exchange of information, allowing mobile workers to disseminate ‘information about labour market and working conditions, and make judgments about whether a boss was good or bad’ (Wu and Liu 2014: 1403). Similar occurrences have been reported from Foxconn where we see a high labour turnover because mobile workers have no commitment to the company and are willing to make use of the options offered by the global labour market. Hence, labour mobility and the in- security it involves for employers may also be seen as a source of power

for workers which they can use to protect their rights and bargain for the improvement of working conditions (Smith 2006).

6.    Conclusions

In this chapter we first presented examples of cross-border production networks and outsourcing relationships in which transnational and thus mobile capital employs migrant mobile labour. Foxconn Electronics, one of the case examples, operates not only in China and South-East Asia but also has production sites in Europe. While Foxconn has taken over plants and employs a local workforce, more than half of its work- ers in Czechia are temporary workers originating from a wide range of countries. A rather different constellation is found in the Chinese ethnic workshops that act as (sub-) contractors for Italian clothing manufac- turers. There, Chinese workers are subject to extremely hard working conditions in particular caused by flexibility demands to achieve short production cycles. The third case described is the online retailer Ama- zon. To meet the demands of a growing online market the company is opening (but also relocating) warehouses and logistic centres all over the world. Newspaper and television reports document the appalling living and working conditions of a largely migrant mobile workforce. Success- ful industrial action organized by German unions has been answered by the company relocating warehouses from Germany to Poland. A fourth case in point is the Danish slaughter industry which has offshored sev- eral activities like deboning and cutting to Germany where subcontrac- tors employ low-paid Polish workers.

All these companies neither restrict themselves to the local labour mar- ket nor depend on it. Instead, they rely on a mobile workforce consist- ing of migrant workers from various European countries, China or other Asian countries. This enables them not only to lower wages, but also to introduce extreme forms of flexibility via a workforce accommodated in dormitories close to the sites and under close surveillance by the com- pany or a security firm. Transnational companies regularly outsource the recruitment, employment and often the accommodation of mobile workers to LMIs such as cross-border TWAs. Often more than half of a site’s workforce consists of temporary agency workers. In particular, just-in-time production and highly seasonal online retailing rely on ex- ternal numerical flexibility provided by TWAs. The boundaries between formal and informal, legal and illegal are sometimes blurred (see also

Wagner 2014). In the case of Foxconn, Amazon or the Danish slaugh- terhouses different TWAs recruit workers and also employ them. In the case of the Chinese workshops ethnic networks ensure a supply of new workers from China, though these are then employed by the shop-own- ers in Italy.

We argue that the ‘double mobility’ presented in this chapter is a particu- lar challenge for employment relations. Both the mobility of capital (in particular the threat of relocation and the ensuing concession bargain- ing) and the mobility of labour (in particular through its dis-embedded- ness and divisions along language and ethnic lines) point to high levels of dis-embedded employment relations. The challenge for trade unions is not only to take on footloose transnational companies but also to or- ganize mobile migrant labour. The situation is aggravated by outsourc- ing relationships, i.e. the fact that workers are not employed by the core companies but by subcontractors or TWAs.

References

Andrijasevic R. and Sacchetto D. (2013) Integration verschiedener Arbeitsregime? Foxconn in Tschechien, in Pun N. et al. (eds) iSlaves. Ausbeutung und Wider- stand in Chinas Foxconn-Fabriken, Wien, Mandelbaum Verlag, 219-246.

Arrowsmith J. and Marginson P. (2006) The European cross-border dimension to collective bargaining in multinational companies, European Journal of Indus- trial Relations, 12 (3), 245-266.

Batt R., Holman D. and Holtgrewe U. (2009) The globalization of service work: comparative institutional perspectives on call centers, Industrial and Labor Relations Review, 62 (4), 453-488.

Benner C. (2003) Labour flexibility and regional development: the role of labour market intermediaries, Regional Studies, 37 (6-7), 621-633.

Benner C., Leete L. and Pastor M. (2007) Staircases or treadmills? Labor market intermediaries and economic opportunity in a changing economy, New York, Russell Sage Foundation.

Berntsen L. and Lillie N. (2014) Hyper-mobile migrant workers and Dutch trade union representation strategies at the Eemshaven construction sites, Economic and Industrial Democracy, 1-17. doi: 10.1177/0143831X14537357

Ceccagno A. (2007) Compressing personal time: ethnicity and gender within a Chinese niche in Italy, Journal of Ethnic and Migration Studies, 33 (4), 635-654.

Coe N.M., Dicken P. and Hess M. (2008) Global production networks: realizing the potential, Journal of Economic Geography, 8 (3), 271-295.

Dicken P. (2005) Tangled webs: transnational production networks and regional integration, SPACES 2005-04, Marburg, Philipps-University of Marburg, Fac- ulty of Geography.

Edwards T., Rees Ch. and Coller X. (1999) Structure, politics, and the diffusion of employment practices in multinationals, European Journal of Industrial Rela- tions, 5 (3), 286-306.

Eichhorst W. (2013) The role and activities of employment agencies, IP/A/EMPL/ ST/2012-06, European Parliament. http://www.europarl.europa.eu/studies

Enright B. (2013) (Re)considering new agents: a review of labour market inter- mediaries within labour geography, Geography Compass, 7 (4), 287-299. doi:10.1111/gec3.12035

Ferner A. (1997) Country of origin effects and HRM in multinational companies, Human Resource Management Journal, 7 (1), 19-37.

Flecker J. (2000) Transnationale Unternehmen und die Macht des Ortes, in Dör- renbächer C. and Plehwe D. (eds) Grenzenlose Kontrolle? Organisatorischer Wandel und politische Macht multinationaler Unternehmen, Berlin, Sigma.

Flecker J. and Meil P. (2010) Organisational restructuring and emerging service value chains: implications for work and employment, Work Employment and Society, 24 (4), 680-698.

Fulton L. (2013) Worker representation in Europe. Labour Research Department and ETUI. Produced with the assistance of the SEEurope Network, http:// www.worker-participation.eu/National-Industrial-Relations.

Gereffi G., Humphrey J. and Sturgeon T. (2005) The governance of global value chains, Review of International Political Economy, 12 (1), 78-104.

Gorm Hansen G. (2007) Come Together? Organisational case study on production in the food industry – Denmark, Internal Working Paper, WORKS Project.

Harvey D. (1982) The limits to capital, Oxford, Blackwell.

Hess M. (2004) ‘Spatial’ relationships? Towards a reconceptualization of embed- dedness, Progress in Human Geography, 28 (2), 165-186.

Huws U., Flecker J. and Dahlmann S. (2004) Status report on outsourcing of ICT and related services in the EU, Dublin, European Foundation for the Improve- ment of Living and Working Conditions.

Lillie N. (2010) Bringing the offshore ashore: transnational production, industrial relations and the reconfiguration of sovereignty, International Studies Quar- terly, 54 (3), 683-704.

MacKenzie R. and Forde C. (2009) The rhetoric of the ‘good worker’ versus the realities of employers’ use and the experiences of migrant workers, Work, Employment and Society, 23 (1), 142-159.

Manky O. (2014) Negotiating the margins: the trajectories of subcontracting in the Chilean and Peruvian mining industry, Paper presented at the 32nd Inter- national Labour Process Conference, Kings College, London, 7-9 April 2014.

Marchington M. et al. (eds) (2005) Fragmenting work: blurring organizational boundaries and disordering hierarchies, Oxford, Oxford University Press.

Marginson P. and Sisson K. (1994) The structure of transnational capital in Eu- rope: the ermgering Euro-Company and its implications for industrial relations, in Hyman R. and Ferner A. (eds) New frontiers in european industrial relations, Oxford, Blackwell Business, 15-51.

Marginson P. and Sisson K. (1996) Multinational companies and the future of collective bargaining: a review of the research issues, European Journal of Industrial Relations, 2 (2), 173-197.

Marginson P. et al. (1995) Extending beyond borders: multinational companies and the international management of labour, The International Journal of Hu- man Resource Management, 6 (3), 702-719.

Massey D. (1994) Space, place and gender, Cambridge, Polity Press.

Mau S. (2007) Transnationale Vergesellschaftung: die Entgrenzung sozialer Lebenswelten, Frankfurt, Campus Verlag.

Mayes R. (2014) Fly-in, Fly-out work and the matter of choice: household deci- sions, labour agency and the tyranny of distance, Paper presented at the 32nd International Labour Process Conference, Kings College, London, 7-9 April 2014.

Mueller F. (1996) National stakeholders in the global contest for corporate in- vestment, European Journal of Industrial Relations, 2 (3), 345-368.

Polanyi K. (1944) The Great Transformation: the political and economic origin of our time, Boston, Beacon Press.

Pun N. and Smith C. (2007) Putting transnational labour process in its place: the dormitory labour regime in post-socialist China, Work, Employment and Society, 21 (1), 27-45.

Rainnie A., McGrath-Champ S. and Herod A. (2009) Geography and worker or- ganisation: do spatial approaches change our theories of union praxis?

http://www.ilera-directory.org/15thworldcongress/files/papers/Track_2/ Poster/CS2T_23_RAINNIE.pdf

Rainnie A., McGrath-Champ S. and Herod A. (2010) Making space for geogra- phy in labour process theory, in Thompson P. and Smith C. (eds) Working life: renewing labour process analysis, Basingstoke, Palagrave Macmillan, 297-315.

Refslund B. (2012) Offshoring Danish jobs to Germany: regional effects and chal- lenges to workers’ organisation in the slaughterhouse industry, Work Organisa- tion, Labour and Globalisation, 6 (2), 113-129.

Ruckus R. (2013) Das Focxonn-Imperium, in: Pun N. et al. (eds, in Pun N. et al. (eds) iSlaves. Ausbeutung und Widerstand in Chinas Foxconn-Fabriken, Wien, Mandelbaum Verlag, 219–246.

Sassen S. (1991) The global city: New York, London, Tokyo, Oxford, Princeton University Press.

Sassen S. (2001) The mobility of labor and capital: a study in international invest- ment and labor flow, Cambridge, Cambridge University Press.

Sassen S. (2002) Locating cities on global circuits, Environment and Urbanization, 14 (1), 13-30.

Schömann I. and Guedes C. (2012) Temporary agency work in the European Un- ion: implementation of Directive 2008/104/EC in the Member States, Report 125, Brussels, ETUI.

Smith C. (2003) Living at work: management control and the dormitory labour system in china, Asia Pacific Journal of Management, 20 (3), 333-358.

Smith C. (2006) Beyond convergence and divergence: explaining variations in organizational practices and forms, in Ackroyd S., Batt R., Thompson P. and Tolbert P.S. (eds) The Oxford handbook of work and organization, Oxford, Oford University Press.

Staples R., Trinczek R. and Whittall M. (2013) ‘Posted Workers’: zwischen Reguli- erung und Invisibilisierung, Arbeit, 22 (4), 271-286.

Taylor P. (2010) The globalization of service work: analysing the transnational call centre value chain, Thompson P. and Smith C. (eds) Working life: renewing labour process analysis, Basingstoke, Palagrave Macmillan, 244-268.

Thompson P. et al. (1995) It ain’t what you do, it’s the way that you do it: produc- tion organisation and skill utilisation in commercial vehicles, Work, Employ- ment and Society, 9 (4), 719-742.

Voss H., Buckley P.J. and Cross A.R. (2010) The impact of home country institu- tional effects on the internationalization strategy of Chinese firms, Multina- tional Business Review, 18 (3), 25-48.

Wu B. and Liu H. (2014) Bringing class back in: class consciousness and solidarity among Chinese migrant workers in Italy and the UK, Ethnic and Racial Stud- ies, 37 (8), 1391-1408.

Wu B. and Sheehan J. (2011) Globalization and vulnerability of Chinese migrant workers in Italy: empirical evidence on working conditions and their conse- quences, Journal of Contemporary China, 20 (68), 135-152.

Wagner I. (2014) Rule enactment in a Pan-European labour market: transnational posted work in the German construction sector, British Journal of Industrial Relations. doi: 10.1111/bjir.12053

Wagner I. (2015) EU posted work and transnational action in the German meat industry, Transfer: European Review of Labour and Research, 21 (2), 201- 2013.

Wagner I. and Lillie N. (2014) European integration and the disembedding of la- bour market regulation: transnational labour relations at the European Central Bank construction site, Journal of Common Market Studies, 52 (2), 403-419.

Part 2

The impact of production network fragmentation on working conditions

Chapter 4

Relocation, the restructuring of the labour process and job quality

Monique Ramioul and Geert Van Hootegem

1.    Introduction

Organizations restructure their activities and relocate work to other sub- sidiaries and companies, both in their own countries and across borders. The key question of this chapter is whether work is transformed when it is moved and with what effect. Do companies simply duplicate the pro- cesses at the destination or are the activities restructured along the way, leading to changes in the division of labour? And what is the impact of such changes on job quality?

A central aim of this chapter is to better understand the spatial divi- sion of labour and how this is restructured in relocation projects. In do- ing so, we aim to help fill a gap in current knowledge on the impact of globalization on work. Economic globalization, its key characteristic of growing global interconnectedness between companies and regions, and the intensified international division of labour have been investigated and debated in a variety of studies using a wide range of theories and perspectives. In economics and economic geography, the trade relations between economic players or regions, and the behaviour of companies as economic players, form the dominant perspectives. On the other hand, sociological theories investigating global value chains and the interna- tional division of labour tend to focus on employment relations and in- dustrial relations (Morgan, Kristensen and Whitley 2003; Bronfenbren- ner 2007; Edwards et al. 2007; Robinson 2010; Riisgaard and Hammer 2011) or on employment conditions (Marchington et al. 2005; Flecker 2010). The descriptions and explanations of the way in which work is organized tend to be underdeveloped or not extensively addressed. The inter-organizational and spatial separation of activities and tasks has been less systematically investigated or focuses on the organization of departments (Gospel and Sako 2010) or business functions (Huws et al. 2009). If we are to really understand the consequences that reloca- tions have on work, changes in the division of work need to be examined

closely. To the extent that the relocation of activities is accompanied by an increased division of labour and more fragmented labour processes and jobs, we can argue that relocations not only affect the voice of work- ers and their employment conditions, but also job content and intrinsic job quality.

To gain a better understanding of relocations from the perspective of changes in the division of labour, we need an in-depth understanding of what workers affected by relocations actually ‘do’: what tasks do they perform and with whom do they collaborate and how do these change after relocation? The focus is on the distribution of tasks between geo- graphically dispersed units and the relationships between these tasks. This means that our prime interest is in the labour process and how this is restructured when part of the process is moved to other domestic or international subsidiaries (in the latter case we speak of offshoring) or to domestic or international companies through a trade relationship (out- sourced).

These research questions were investigated in three in-depth case stud- ies, selected out of a broad sample of 58 cases undertaken within the WORKS project, a large European study on the restructuring of value chains.1 The cases represented a variety of labour processes ranging from highly-skilled knowledge work to semi-skilled manual and administra- tive work. The first case study focuses on the relocation of administra- tive order processing in the logistics of a food processing company. The second investigates the relocation of part of a customer service to a call centre. The third case study concerns the reorganization of the informa- tion technology (IT) departments of a number of regional health organi- zations.

For each case study, eight to ten semi-structured interviews with man- agement and employee representatives from the organization were con- ducted in order to reconstruct the relocation, how it had developed, what measures and strategies had been deployed, and what the final outcome was. The interviews were complemented by company documents, ena- bling a comprehensive picture of the relocation process to be drawn. In addition, seven to nine in-depth individual interviews with employees

1. The WORKS project, Work Organisation Restructuring in the Knowledge-based Society, 6FP funding from the EU, contract n°: CIT3-CT-2005-006193.

directly involved in the relocation were carried out, both at the source and the destination companies. The interviews were transcribed for analysis. The data from the cases were mainly interpreted by means of a within-case analysis, as the primary focus was on understanding the idiosyncratic rationale of each case study. Though we attempt to analyti- cally compare the three cases, the research approach taken implies that we make no claim for a generalizability of the conclusions.

2.    Theoretical outline

2.1                   Looking inside the black box of relocations

Our key aim is to understand whether and precisely how work changes when the spatial division of labour is altered. How is the production process of a good or a service, as a flow of material or information, or- ganized? How are the different functions and tasks designed and how do they relate to each other? Do these flows, tasks and relations change when they are entirely or partly reorganized at a distance? Answering these questions enables us to analyse in-depth the job content of restruc- tured labour processes and hence to draw conclusions on core dimen- sions of job quality.

One of the key reference concepts used today in the literature on off- shoring and outsourcing is the global value chain, emphasizing the eco- nomic value added by the different activities before a product or service is delivered to the market (Huws et al. 2009: 25).2 The different ‘nodes’, ‘boxes’, or ‘steps’ making up the global value chain are defined in the literature as activities or business functions. Porter (1985) focused on the technologically and economically distinct activities that a company performs to do business, each of which adds economic value to the end product (and thereby contributes to the company’s competitiveness). A number of studies mapping the employment impacts related to offshor- ing and outsourcing and the restructuring of global value chains build on Porter’s definition and approach, using the concept of business func- tion to classify the activities composing value chains for their empirical

2. In one of the books setting out the conclusions of the WORKS project, a historical account of the various underlying theories behind the concept of the global value chain, which date back to 18th century economists, is summarized by Huws et al. (2009: 11ff).

investigations (Geurts, Coppin and Ramioul 2007; Statistics Denmark et al. 2008; Sturgeon and Gereffi 2009; Huws et al. 2009; Vandekerck- hove and Ramioul 2015). Mostly, and as also used in the WORKS project from which our empirical data come, business functions are classified into primary activities –production, assembly, transport and distribu- tion, sales and customer services – and secon-dary or support activities, such as finance and accounting, HRM, training and product develop- ment. In WORKS, the offshoring and outsourcing of a number of busi- ness functions (R&D, production, logistics, customer services, IT) were investigated in a range of industries to better understand contemporary changes in work.

However, activities (as defined by Porter) or business functions as such do not necessarily offer clarity on the division of labour, nor do they en- able us to specify work organization, job content or the tasks making up those business functions (Ramioul 2012a, 2012b). For example, ‘cus- tomer service’ or ‘logistics’ as technologically and economically distinct activities do not provide any indication of what types of jobs are includ- ed, what tasks make up these jobs and how they relate to each other and to other activities. There is in reality a range of organizational diversity in ways of organizing business functions, even within a single sector.

A related problem is the underlying assumption that global value chain restructuring involves the complete externalization of entire business functions. It is conceivable that corporate restructuring projects may also involve hiving off only some of the tasks and that, consequently, the division of labour within the business function changes through the relocation process. It is also imaginable that externalization is simulta- neously preceded by or provokes changes in the division of labour. In such cases, it seems essential to look ‘inside the black box’ of business functions and analyse task composition at a lower level when seeking to identify changes in jobs and job quality in relation to relocation (ibid.).

2.2                     Understanding the division of labour

In order to analyse the task composition of business functions and changes in these we need an analytical framework suitable to observe and understand the division of labour. Such a framework is provided by the SocioTechnical Organizational Structures Approach (STOSA) which finds its roots in a specific strand of sociotechnical systems theories

developed amongst others by de Sitter (1981) and other scholars from the Netherlands and Belgium (Christis and Korver 1992; Van Hootegem 2000; Van Eynatten and van der Zwaan 1998; Achterbergh and Vriens 2009; Ramioul 2012a, 2012b). As defined by de Sitter: ‘An organisa- tional structure is the grouping and coupling of transformations into tasks and the resulting relations between these tasks relative to orders’ (cited in Achterbergh and Vriens 2009: 236). Variations in the division of labour are based on grouping principles of the different operations of a transformation process. These boil down to a choice between ei- ther a coupling of similar operations for a range of different orders or the coupling of all the different operations for groups of similar orders.3 For instance, ‘production’ can be organized, in broad terms, either by linking together all the technical operations (for example: sawing, mill- ing, drilling, painting, assembling,…) involved in producing one product or by decoupling these technical operations and bundling them for all products on the basis of their technical similarity. This results in what is commonly referred to as a product-oriented or order-based versus a process-oriented or operation-based production flow. Operation-based production flows form the basis for production structures designed ac- cording to Taylorist division of labour principles where tasks are highly fragmented into shortcycled sub-operations. Figure 1 shows the basic ways of differentiating the grouping of operations in transformation processes.

Figure 1 Coupling all different operations for each (group of) orders or coupling all similar operations for all orders

Operation 1                Op2                    Opn

Orderflow 1 Or1

Orn

3. Similar orders may refer to similar customers, similar products/services or similar projects (Christis 2009: 13).

By analysing the coupling or decoupling of the tasks of a transforma- tion process, it becomes possible to explain organizational differences in the production of similar products or services. Due to the differences in these groupings, tasks in similar transformation processes can range from extremely fragmented, simple operations to wide-ranging, inte- grated and complex jobs (Van Hootegem 2000: 67ff).

Increasingly diverse products and services, heterogeneous clients and markets, and (sometimes unpredictable) fluctuations in orders and workloads lead to increased complexity in transformation processes. This forces an organization to break down the production structure into different and separate workflows. Following the principles set out, it should be preferable to do this according to order types and to integrate all different operations for one order (or a set of similar order types) rather than taking an operation-based approach. The reason for this is that the tasks involved in producing a single order (product or service) are highly interdependent. If these different tasks are separated and spread across different workflows, this increases organizational com- plexity and interdependency between, rather than within, workflows.

A core argument for reducing the complexity of production structures is that when tasks for producing one product are coupled, any disturbances that inevitably occur during the transformation process will be attenu- ated because the number of relationships and interfaces in the network of tasks and the variability within these relationships are reduced (Ach- terbergh and Vriens 2009: 261ff).4 At the same time, such order-based coupling will strengthen the regulatory potential to solve these distur- bances. Especially in turbulent environments characterized by a high proneness to disturbances, an order-based structure is important. If, on the contrary, the different tasks involved in realizing a single product are assigned to separate flows, every product needs time to be passed on to the next flow which may generate disturbances. In addition, resolving such disturbances becomes more complex. Consequently, researchers in this area postulate avoiding the operation-based grouping of tasks and task fragmentation and instead striving for order-based flows generating so-called integrated tasks.

4. Christis (2009: 13) emphasises, however, that in the lean production model based on similar organizational principles and on just-in-time, the constant effort to achieve further reduction of in-process stock will increase the vulnerability of the production structure to

Based on these important consequences of the design of transformation processes, a next question is: what will be the effect of these changes on the degree of disturbances and on the regulation potential of workers? An advanced geographical division of labour may result in complex (in- ter-)organizational structures and increase the number of relationships and interfaces within the network of tasks. The distance between tasks is likely both to amplify disturbances between highly interdependent and geographically and/or contractually dispersed tasks and to decrease the potential to adequately resolve these.

2.3 The link with job quality

In order to link the way the transformation process is designed (i.e. the division of labour) with job quality we apply the job demands/job control model, developed by Karasek and Theorell (Karasek 1979; Karasek and Theorell 1990) and tested, fine-tuned and enlarged by a broad strand of research. Job demands refer to ‘the physical, social and organizational aspects of work that require psychological and/or physical attention and effort’ (Holman and McClelland 2011: 17). Job resources enable the per- son to manage these job demands, facilitate the achievement of goals, promote learning and fulfil basic human needs (ibid.). Essentially, job resources refer to decision latitude but the concept has gradually been enlarged to include other work aspects, including feedback, social sup- port, skills utilization and task significance (Demerouti et al. 2001).5

The link between the division of labour, as defined by the STOSA frame- work described above, and job quality is in fact easy to make. Processes designed as an operation-based grouping of operations and high levels of task fragmentation will confront workers with many disturbances and consequently place high job demands on them. At the same time workers lack the decision latitude, as a job resource, to resolve these disturbanc- es. Jobs with high job demands and low job resources are defined as high

5. In addition it is now seen as common sense in the quality of work literature also to take into account other dimensions, which can be broadly grouped under ‘quality of employ- ment’ (wages, working times, benefits, ) and ‘quality of empowerment’ (individual and collective voice and representation, access to training, etc.). Essentially, these additional

dimensions include the core aspects involved in the allocation of workers to tasks, as well as their participation in the organization (Bustillo et al., 2009; Holman & McClelland, 2011). Since our model places the emphasis on providing a coherent model of the organizational structure and its relation to job content, this broader set of dimensions is not included here.

strain jobs, characterized by concurrent high stress risks and low learn- ing opportunities. In other words, job quality is likely to worsen with high division of labour levels because adequate levels of regulation po- tential and decision latitude will be lacking at the place where problems occur (and are more likely to happen) (Achterbergh and Vriens 2009: 257ff). Based on this premise, job quality is fostered by a low degree of disturbances and high regulative potential (van Eijnatten and van der Zwaan 1998: 303ff).

In conclusion, the key questions to address are: ‘Will the relocation of activities be related to changes in the division of labour?; and secondly: ‘What are the effects of these changes on the degree of disturbances and on regulation potential as key determinants of job quality?’.

3.    Evidence from three disparate labour processes

3.1                   The FOODLOG Logistics Business Shared Services Centre

The first case study, nicknamed FOODLOG, concerns the relocation of the logistics administration of a subsidiary of a global company in the food industry to a dedicated centralized Business Shared Services Cen- tre (BSSC) (De Bruyn and Ramioul 2007). The rapid growth of FOOD- LOG and its evolution into a global company was accompanied by a huge number of subsequent comprehensive reorganization projects intended to optimize the dispersed activities and the links between them and har- monize processes and operations. The overall corporate strategy was to offshore and centralize operations into BSSCs. Due to the geographically dispersed nature of the various plants and activities and the global na- ture of the market, there was a vast and complex network of flows of raw materials, packaging, a variety of end products and flows of returned empty packaging between the different domestic and international sites, making logistics flows increasingly complex. Together with the logis- tics administration units of other European subsidiaries, the logistics department of the Benelux subsidiary – the subject of the study – was therefore relocated to a newly created BSSC. The original workflow is depicted in Figure 2. It shows that, before the relocation into the new BSSC, the work of the logistics administration departments was organ- ized in such a way that one employee was responsible for the entire order workflow for a number of customers.

Figure 2 The logistics workflow of the unit before relocation

Contrary to the original idea, not all logistic activities were offshored: the key middle task in order processing, order release, was decoupled from the transferred workflow and not centralized, remaining at the Benelux site. The restructured process involved the following steps: the central- ized BSSC entered the orders into the system, the Benelux site checked the availability of the different products at the different sites and released the order, which was then communicated back to the BSSC which in turn organized transport to the customer, calculated the required containers and processed the administrative formalities for the order. Order release was the most difficult task in the workflow. The main reason for this part of the process not being transferred was the fact that order release was based on IT inventory and production planning systems that were not integrated. Subdividing the workflow in this way meant, however, that the BSSC remained highly dependent on the Benelux site for processing its orders. For both parties, relocation complicated the whole process as an order had to move back and forth when being processed.

In addition to the incomplete transfer, work organization was funda- mentally redesigned and shifted from an order-based to an operation- based division of labour. In the original (Benelux) set-up each customer had had a single point of contact for the entire order process. In other words, all operations for processing one order were grouped. In the new BSSC set-up, by contrast, individual employees were each only responsi- ble for one subtask which they had to perform for a number of customer groups, as depicted in Figure 3. As a result, they had neither an overview of the whole logistics process for a specific order nor the potential to solve any problems occurring during order processing.

Figure 3 The logistics workflow of the restructured FOODLOG unit

This radically different work organization at the centralized BSSC led to considerable efficiency losses and unexpected costs: container loads were not correctly calculated, information on delivery schedules was missing, custom forms were missing, empty packaging went lost, com- munication with customers was inadequate, there were delays in trans- port, etc. These problems were more numerous and lasted longer than expected. The underperformance of the new unit was first and foremost caused by the lacking overview of the operation-based process and by the fact that not all order processing tasks were transferred. This in turn considerably reduced the decision latitude of BSSC employees, as ex- pressed by the local manager:

‘What you lose is customer focus and some process visibility because if order acceptance makes a mistake and it is not spotted, it has re- percussions on the next steps.’ (Supervisor Benelux export operations BSSC)

A second source of problems was the lack of experience among the new BSSC employees and the loss of contextualized, uncodified knowledge. The FOODLOG employees reported that, over the years, they had de- veloped their personal work practices to handle specific situations and problems in order processing. These included the setting up of specific, personalized systems and tools, such as small notebooks or electronic files, to keep track of order processing and to store information on cus- tomers, forms, customs officers, etc. These working practices were high- ly personalized and not shared.

The persistent problems in the restructured order processing depart- ment prompted management to make certain corrections. These were mainly geared towards intensified codification of workprocess knowl- edge and intensified monitoring: the introduction of key performance indicators and the systematic recording and analysis of errors. In addi- tion, they installed a ‘rescue team’ to fire-fight and temporarily help out the BSSC in order to minimize performance losses and partially compen- sate for the loss of uncodified knowledge.

3.2                     Separating questions and answers: CITYLIFE

The second relocation case considers the outsourcing of inbound tel- ephone customer services of an Austrian city’s public housing admin- istration (Schönauer 2007). The administration of the 220,000 city apartments was managed via a company, nicknamed CITYLIFE, which operated a range of decentralized walk-in customer service centres re- sponsible for all tasks relating to renting apartments: the administrative and legal procedures related to the tenancy (contracts, payments and de- posits), lawsuits in the case of conflicts (for example, over rent arrears), apartment sales, housing improvements, and maintenance and interven- tions in the case of technical defects. The main problem the management wanted to tackle was the poor quality of the telephone service: long wait- ing queues on the telephone and employees almost impossible to con- tact. The service centre employees were frustrated too because they were constantly interrupted by incoming telephone calls, which were most of the time very short requests for basic information (addresses, opening hours). These calls disturbed employees when handling more complex customer questions which required face-to-face, personalized meetings. The decision was taken to centralize and outsource the inbound tele- phone service of the service centres to a call centre, MULTICALL.

The management concluded that the best idea for reorganizing customer services was to subdivide them by decoupling what they called ‘obvious tasks’. The call centre management stated:

‘… and so the idea emerged of outsourcing on the basis of differences in knowledge-intensity: very specific questions can be shifted to the call centre because there are many areas in which services can be delivered 24 hours a day by non-qualified rather than skilled staff.’ (Manager MULTICALL)

All incoming phone calls from customers were handled at the call centre, which acted as a filter. If the customer query concerned an issue that required intervention by the service centre team, the call centre agent entered an electronic ‘call-back ticket’ or ‘technical intervention request’ and forwarded it to the employee who could then provide the answer when he or she saw fit. This ticket included the contact details of the customer and a basic description of the problem or the question. At the service centre, the employees prepared the telephone call based on the information on the ticket and made the calls at their own convenience and taking into account the other customers with whom they had (face- to-face) appointments. Figure 4 depicts the restructured workflow.

In order to separate the inbound customer questions and relocate them to the call centre, the entire workflow was analysed, documented and also digitized into a central knowledge database. This was then used to guide the MULTICALL agents through an algorithm of questions aimed at refining the customer request.

The restructuring of the telephone services meant that task components belonging to the same workflow, question and answer, were first decou- pled and then reconnected using an electronic ticket. A neat cut-off point between the two contrasting types of customer service was envisaged. For most simple questions, which concerned almost half of the 3,500 daily questions, the quality of service improved as a result. In the case of more complex customer requests, assessment was mixed. On the one hand, responses by the service centre agents could be better prepared and were no longer interrupted. On the other hand, decoupling the ques- tion from the answer created a complicated workflow which impacted service quality and increased the risk of errors. The link between the person receiving the question and the provider of the response turned out not to be entirely seamless in reality. In addition, there was no direct

Figure 4 The customer services workflow at CITYLIFE-MULTICALL

exchange or interaction between the respective workers. The call-back ticket turned out to be the Achilles heel. Mistakes in producing the ticket gave rise to additional iterative communications between customers, call centre and service centre employees. First, the time lag between ques- tion and response generated customer frustration since he or she had to wait for the call-back before the problem could be resolved. Next, the call centre agent had to rely on a personal interpretation of the customer’s message. This became apparent in that it was often difficult for the ser- vice centre employee to learn exactly what was needed from the elec- tronic ticket alone. A typical case was that customers exaggerated the urgency or seriousness of a defect, a situation difficult for the call centre

agent to assess. Problems were also related to the fact that the knowl- edge database contained a lot of errors and had to be constantly updated. Yet another important source of bugs in the workflow was the fact that the call centre agents were cut off from the decentralized service centres and the actions taken by the service centre employees. The latter were in turn not always informed either about actions taken by the agents or about what they had told the customer on the phone. In some cases, customers told the agents that they did not understand what the service centre agent had told them and asked whether the agent would be so kind to repeat this to them. However, the agent had no idea what the response was. The only solution in all those cases was to try to comfort the customer by preparing a new ticket with the same request. In other words: the outcome was a more complicated workflow with highly inter- dependent tasks and reduced capacities to adequately resolve problems.

3.3                     IT Health

The last case concerns the outsourcing of IT services by a number of publicly-owned hospitals and healthcare service organizations in a re- gion of Norway (Dahl-Jørgensen and Torvatn 2007). Originally nine independent hospitals and six health centres had their own IT units. The restructuring project included transferring these IT units to a new firm, ITHEALTH, which was still publicly owned but economically independent.

The fifteen IT units were centralized and their activities concentrated into a single unit process. In other words, rather than integrating the ITactivities into each decentralized health centre, which was originally the case, all IT activities were pulled out of these and coupled to serve all. The local health organizations transferred all their IT workers, equip- ment and contracts to ITHEALTH, which consequently owned, oper- ated, maintained and developed the whole IT-related infrastructure at the different health centres. This was expected to yield various improve- ments in costs and efficiency, such as centralized purchasing, coordinat- ed use of resources and standardization of work processes. The health centres negotiated specific contracts with ITHEALTH to receive the re- quired IT services. After a while, each of the health centres installed a specific IT procurement officer as their link to ITHEALTH, responsible for co-ordinating development and maintenance at their location. Later, some centres further divided the tasks of this liaison function between

Figure 5 Geographical centralization and concentration at ITHEALTH

a legal-contractual and a technical profile with the aim of co-ordinating the workflow between the health centre and ITHEALTH.

ITHEALTH was itself split up into four operational groups, each re- sponsible for a specific aspect of IT services at all health centres: man- agement, development, IT services and applications. The ‘applications’ team was responsible for all software development previously done lo- cally at the individual health centres. This team tailored the databases to meet the specific requirements of the different health centres and maintained these over their life-cycles (for example, electronic patient records, patient administration, laboratory computer systems, etc.). The ‘IT services’ unit included a centralized customer service centre sup- porting all health centres and responsible for resolving first-level prob- lems, while forwarding more complex ones to a central technical sup- port group. Before restructuring, contacts between the helpdesk and the staff at the health organizations were direct, while after the restructuring every query had to go through this central helpdesk. In addition, help- desk work organization was changed. Previously there was a job rotation

arrangement between first-level and second-level helpdesks, resulting in all staff having more versatile competences. After restructuring, IT con- sultants were assigned either to the first-level or the second-level help- desk with no possibility of switching.

The objective of full centralization of all IT activities was not achieved. The main reason why not all tasks were eventually shifted was the exist- ing IT infrastructures and related expertise required to maintain these, and the fact that some local health organizations still continued to devel- op their own IT applications. Moreover, the helpdesk could not be fully centralized and each location maintained its own helpdesk as there were still locally developed IT systems which required the specific skills of the former IT employees. The objective of a standardized, operation-based system and a fully centralized helpdesk was therefore not realized (at the time of the investigation).

4.    Analysis

4.1                   Relocations based on assessed knowledge requirements

A first observation is that in all cases knowledge requirements of the activities being restructuring were decisive for the design of the reloca- tions – an observation backed up by the literature (Schamp, Rentmeister and Lo 2004; Gereffi, Humphrey and Sturgeon 2005; Mirani 2007; Pe- tersen et al. 2008; Valenduc et al. 2008; Kletzer and Jensen 2009; Goos, Manning and Salomons 2009; Maenen 2010). The tasks that were least complex and required least interaction were decoupled and relocated. In all three cases, the common feature of the tasks retained at source was their complexity and/or required contextualization by workers, under- stood as the necessity to adapt the prescriptions by taking account of the specific and unique characteristics of orders or customers. At CITYLIFE, this divid¬ing line was relatively easy to identify according to manage- ment. Those parts of customer services where customer interaction had the greatest chance of being limited to a one-off encounter and referred to standardized answers were decoupled, while the more complex per- sonal contacts with customers remained in the service centers. But also at FOODLOG, order release as the most complex task of order process- ing was not relocated but decoupled from the transferred workflow.

4.2                     No lift and shift operations

Second, it seems that none of the cases were designed as a pure ‘lift and shift’ operation. All cases involved changes in the division of labour in re- lation to geographical relocation. These changes were deliberately planned as part of the relocation projects. In the three cases these included shifts from order-based to operation-based flows, implying a decoupling of the different operations of the order flows and a coupling of the similar op- erations for all order flows. In FOODLOG and ITHEALTH the expected benefits of centralization and the shift to operation-based workflows were rationalization, increased efficiency and productivity, and econo- mies of scale. Workflows based on functional groupings of operations are indeed expected to provide such economic benefits (Daft 2007: 91). In the case of CITYLIFE, the aim was to improve customer services pro- vision for a public service in a context of expanded service requirements.

But in addition to this deliberately planned division of labour, the practi- cal implementation of the relocation projects imposed additional inter- ventions in the workflow. These resulted from the fact that the relocation projects could not be implemented as originally envisaged. In the three cases, fully shifting all tasks of the activities to be relocated was inhib- ited by the specific characteristics of certain operations, information, technical infrastructures, data or competences. As a result, management had to re-organize the workflows taking account of the non-transferred operations.

ITHEALTH convincingly demonstrates how additional interventions were necessary to solve unforeseen problems. First, the centralization operation could not be fully accomplished due to the existence of site- specific technological infrastructures ànd the fact that the local sites continued to acquire and implement their own IT systems. This situa- tion generated the need to keep the previous employees on standby to provide support for the local infrastructures. The centralization of all IT activities further meant that new procedures were required to bridge the gap between the service providers and the various departments and to meet additional regulation needs of the spatially dispersed IT. This prob- lem was partially solved by creating a specific liaison function between the concentrated IT unit and the different health organizations: the IT procurers. This new function was further divided into contractual as- pects of collaboration with ITHEALTH and a technical liaison function to collaborate on IT projects on a day-to-day basis. In addition, it turned

out that the need for a helpdesk had grown as a result of the reloca- tion. As has been explained, tasks were also further subdivided. These examples demonstrate that a further fragmentation at the level of the individual tasks of the operation-based flows was applied in the course of the implementation.

4.3                     Growing workflow complexity

A third observation is that the relocations led to growing workflow com- plexity, associated with increased vulnerability in the case of errors and a reduced capacity to adequately resolve them. The reason for this is that in the relocated processes the decoupled tasks involved in producing a single order or customer request remained interdependent. Decoupled tasks need to be linked and coordinated in order to secure seamless pro- cesses. Geographical distance added complexity to these coordination requirements because remote communication and interventions are more complex and prone to misunderstandings than collocated inter- action (Cramton 2001). Looking closer at the errors and disturbances that occurred in each case after relocation, there are indications that a number of problems seemed not to be primarily caused by the distance between adjacent tasks, but rather by the way in which the relocation process was designed and implemented. The fact that not all tasks were transferred was one cause. The outcome was that work had to be shifted back and forth in an iterative process, often involving several loops be- fore a task could be completed. But the relocated jobs were also rede- signed in a way that the workers had to take care of only one step of the process rather than being responsible for an entire order or customer. This shift to more fragmented tasks and an operation-based division of labour resulted in the loss of a comprehensive overview of the workflow (which prevented correct diagnosis of any disturbances by the workers at the outset), broken links between decoupled but interdependent tasks and a reduction in regulative discretion. It was observed that these prob- lems triggered the need for additional organizational interventions in order to introduce new ways of coordination.

4.4                     Impact on job content and job quality

It is evident from the overview that, as a result of these combined interventions, the relocations left few individual jobs unaffected. In

particular, the relocations eventually resulted in smaller and fragment- ed tasks at the destination sites. This increased division of labour and the spatial dispersion of labour processes and tasks tended to affect job quality because workers were confronted with new problems and dis- turbances but lacked the resources to resolve them. Increased problem- solving requirements, as job demands, can only provide challenges and learning opportunities for workers if they are indeed accompanied by adequate problem-solving capacities related to decision latitude and support (as job resources). Since that was not the case, the result was a deterioration of job quality, with workers facing more stress risks and having less learning opportunities. This was the outcome at both the new destination units and the source companies.

4.5                     Managerial attempts to solve the problems

In all cases the companies attempted to reduce the (often unexpected and underestimated) performance losses associated with the relocation pro- ject, having to find solutions to the unforeseen problems that occurred during its implementation. In all cases, a range of interventions were intended to create additional regulation potential and establish func- tional coordination between decoupled activities organized remotely. These were often conceived from the start, but sometimes management also felt the need to intervene during the implementation of the restruc- turing process to solve the problems that had emerged. Overall, these interventions involved operational staff at both source and destination organizations.

Several of such coordination mechanisms identified in the literature (Mirani 2007; Flecker et al. 2008; Maenen 2010) are observed in the cases. These can be broadly grouped as follows: (1) implementation of information and communication technologies (ICT), (2) standardiza- tion and formalization of procedures and processes, (3) installation of boundary-spanning functions and (4) installation of dedicated commu- nication channels.

First, it appears that a lot was expected from ICT in terms of integrating informa¬tion flows and coordinating work, which is in line with obser- vations that restructuring is generally supported by dedicated ICT in- frastructures (Greenan et al. 2009). However, it seems that while ICT may perhaps show great promise in terms of increasing communication

capacities, coordination capacities are not necessarily improved.The role of ICT in bridging time and space was mentioned in several cases, but indeed was not necessarily described as unproblematic.

Second, original staffing capacity, for example a scaled-down ‘rescue team’, was kept in place at the source in order to solve the problems cre- ated by the relocation or pending full completion of the relocation pro- cess. This turned out to be necessary in particular because the required competences had not (yet) been acquired by the destination company, as in the cases of FOODLOG and ITHEALTH. Generally, this type of ad- ditional regulation mechanism was conceived as transitory.

Third, it seems that in several cases increased bureaucratization based on intensified standardization of processes, more procedures and moni- toring, such as the introduction of key performance indicators, the use of surveillance and control tools and so on, accompanied the relocation (Flecker et al. 2008). Such formalization and proceduralization were observed at both ends of the restructured workflows. In other words, it seems not to be the case that it was only the destination unit where a greater division of labour was implemented and which faced these or- ganizational challenges. The source units also took their share. At CIT- YLIFE, the team-based service centres were confronted with increased bureaucratization and monitoring of their work caused by the overall standardization of the workflow, ICT-based tools and data processing systems that underpinned the outsourcing and relocation of the inbound telephone service. Consequently they were subjected to the same detailed service levels, strict guidelines and surveillance as the call centre agents.

Fourth, functional coordination was achieved by creating additional lay- ers in the organization, such as specific coordination units or boundary- spanning functions. Putting specific liaison staff in place was a common solution specifically designed to establish operational links between dis- persed business functions and intended to safeguard opportunities for regulative intervention. In the three cases the boundary-spanning func- tion was assigned to the existing first-level hierarchy at the source.

5.    Conclusions

The first and most important conclusion is that the relocations inves- tigated were not simple lift and shift operations but seriously affected

division of labour. And while, in search of productivity gains, a greater division of labour was an intended goal of the restructuring to begin with, it was further boosted by the unexpected problems associated with the restructuring. In the relocations analysed, the work changed profoundly and generated a sharp contrast in outcomes at both ends of what was originally a single, integrated labour process. Especially at the destina- tion sites, tasks tended to be more divided and fragmented. The analysis further demonstrates that the separated activities often remained highly interdependent, leading to complex processes that were more sensitive to disturbances. This outcome can to a fair extent be explained by the combination of a shift to operation-based production structures, the in- complete transfer of tasks belonging to one workflow and the necessity to organize interdependent activities remotely. These increased the need for various forms of functional coordination. The increased division of labour and the spatial dispersion of labour processes and tasks tended to affect job quality because workers were confronted with new problems and disturbances, in other words more job demands, but lacked the re- sources to solve them.

From our research we have reached the conclusion that explicit aware- ness of the way labour processes and jobs are redesigned in restructur- ing projects may be just as important as fair compensation and decent alternative offers of employment. Sufficient time and resources should be allowed for true participation of workers, even in the design, prepara- tion and practical implementation of restructuring.

References

Achterbergh J. and Vriens D. (2009) Organizations: social systems conducting experiments, New York, Springer.

Barney J.B. (1999) How a firm’s capabilities affect boundary decisions, Sloan Management Review, 40 (3), 137-145.

Bronfenbrenner K. (2007) Global unions challenging transnational capital through cross-border campaigns, Ithaca, Cornell University Press.

Christis J. and Korver T. (1992) Sociotechniek, systeemtheory, werkgelegen- heidsverhouding, in Van Hees G.B. (ed.) Arbeids- en organisatieonderzoek in Nederland, Amsterdam, SISWO, 135-163.

Christis J. (2009) Wat is slim organiseren?, Rede ter gelegenheid van de installatie van dr. Jac Christis als lector Arbeidsorganisatie en productiviteit verbonden aan het Kennsicentrum Arbeid van de Hanzehogeschool Groningen.

Cramton C.D. (2001) The mutual knowledge problem and its consequences for dispersed collaboration, Organisation Science, 12 (3) 346-371.

Daft R.L. (2007) Organisation theory and design, Mason, Thomson South- Western.

Dahl-Jørgensen C. and Torvatn H. (2007) IT-Health in Norway: organisational case study on IT service providers in public administration – Norway, Internal Working Paper, WORKS project, Leuven, Higher Institute of Labour Studies.

De Bruyn T. and Ramioul M. (2007) Export the export. Restructuring of logistics in a global company in the beverage industry: organisational case study on logistics in the food industry – BENELUX, Internal Working Paper, WORKS project, Leuven, Higher Institute of Labour Studies.

de Sitter L.U. (1981) Op weg naar nieuwe fabrieken en kantoren: productie- organisatie en arbeidsorganisatie op de tweesprong [Toward new factories and offices: production organization and work organization on a cross-road], Deventer, Kluwer.

Demerouti E., Bakker A.B., Nachreiner F. and Schaufeli W.B. (2001) The job demands-resources model of burnout, Journal of Applied Psychology, 86 (3), 499-512.

Edwards P., Edwards T., Ferner A., Marginson P. and Tregaskis O. (2007) Em- ployment practices of MNCs in organisational context: a large-scale survey, Warwick.

Flecker J. (2010) Fragmenting labour: organisational restructuring, employment relations and the dynamics of national regulatory frameworks, Work Organisa- tion Labour and Globalisation, 4 (1), 8-23.

Flecker J., Holtgrewe U., Schönauer A., Dunkel W. and Meil P. (2008) Restructur- ing across value chains and changes in work and employment: case study evi- dence from the Clothing, Food, IT and Public Sector, Leuven, Higher Institute of Labour Studies.

Gereffi G., Humphrey J. and Sturgeon T.J. (2005) The governance of global value chains, Review of International Political Economy, 12 (1), 78-104.

Geurts K., Coppin L. and Ramioul M. (2007) The transformation of work? D9.2.1

– Tracing employment in business functions. A sectoral and occupational ap- proach, Leuven, Higher Institute of Labour Studies.

Goos M., Manning A. and Salomons A. (2009) Job polarization in Europe, Ameri- can Economic Review, 99 (2), 58-63.

Gospel H. and Sako M. (2010) The unbundling of corporate functions: the evolu- tion of shared services and outsourcing in human resource management, In- dustrial and Corporate Change, 19 (5), 1367-1396. doi: 10.1093/icc/dtq002.

Greenan N., Kaliguna E., Kocoglu Y., Makó C. and Csizmadia P. (2009) Employer’s use of technology and the impact on organisational structure, Leuven, Higher Institute of Labour Studies.

Holman D. and McClelland C. (2011) Job quality in growing and declining eco- nomic sectors of the EU, WALQING working paper 2011.3, Deliverable 4 of the WALQING project, SSH-CT-2009-244597.

Huws U., Dahlmann S., Flecker J., Holtgrewe U., Schönauer A., Ramioul M. and Geurts K. (2009) Value chain restructuring in Europe in a global economy, WORKS project, Leuven, Higher Institute of Labour Studies.

Huws U. and Ramioul M. (2006) Globalisation and the restructuring of value chains, in Huws U. (ed.) The transformation of work in a global knowledge economy: towards a conceptual framework, Leuven, Higher Institute of Labour Studies, 97-119.

Karasek R.A. Jr. (1979) Job demands, job decision latitude, and mental strain: im- plications for job redesign, Administrative Science Quarterly, 24 (2), 285-308.

Karasek R.A. Jr. and Theorell T. (1990) Healthy work: stress, productivity, and the reconstruction of working life, New York, Basic Books.

Kletzer L.G. and Jensen J.B. (2006) Tradable services: understanding the scope and impact of services offshoring, in Collins S.M. and Brainard L. (eds) Brook- ings trade forum 2005: offshoring white-collar work, Washington, DC, Brook- ings Institution Press.

Maenen S. (2010) Organizations in the offshore movement: a comparative study on cross-border software development and maintenance projects, Leuven, KU Leuven, Centre for Sociological Research.

Marchington M., Grimshaw D., Rubery J. and Willmott H. (2005) Fragmenting work: blurring organizational boundaries and disordening hierarchies, Oxford, Oxford University Press.

Mirani R. (2007) Procedural coordination and offshored software tasks: lessons from two case studies, Information and Management,44 (2), 216-230.

Morgan G., Kristensen P.H. and Whitley R. (2003) The multinational firm: organ- izing across institutional and national divides, Oxford, Oxford University Press.

OECD (2007) Staying competitive in the global economy: moving up the value chain, Paris, Organisation for Economic Co-operation and Development.

Petersen B., Pedersen T. and Sharma D. (2008) The role of knowledge in firm’s internationalisation processes: wherefrom and whereto?, LINK Working Paper, Copenhagen, Copenhagen Business School.

Porter M.E. (1985) Competitive advantage: creating and sustaining superior performance, New York, Free Press.

Powell W.W. (1990) Neither market nor hierarchy: network forms of organization, in Cummings L.L. and Staw B.M. (eds) Research in organizational behavior, Greenwich, JAI Press, 295-336.

Ramioul M. (2012a) The map is not the territory: the role of knowledge in spatial restructuring processes, Phd Thesis, Leuven, KU Leuven, Faculty of Social Sciences.

Ramioul M. (2012b) Die Umstrukturierung globaler Wertschöpfungsketten als Ursache für zunehmende Arbeitsteilung – wie man die ‘Black box’ Unterneh- mensfunktion öffnet, in Flecker J. (ed.) Arbeit in Ketten und Netzen. Die dynamische Vernetzung von Unternehmen und die Qualität der Arbeit, Berlin, Ed. Sigma, 231-258.

Ramioul M. and De Vroom B. (2009) Global value chain restructuring and the use of knowledge and skills, Leuven, Higher Institute of Labour Studies.

Riisgaard L. and Hammer N. (2011) Prospects for labour in global value chains: labour standards in the cut flower and banana industries, British Journal of Industrial Relations, 49 (1),168-190.

Robinson P.K. (2010) Do voluntary labour initiatives make a difference for the conditions of workers in global supply chains?, Journal of Industrial Relations, 52 (5), 561-573. doi: 10.1177/0022185610381564

Schamp E.W., Rentmeister B. and Lo V. (2004) Dimensions of proximity in knowledge-based networks: the cases of investment banking and automobile design, European Planning Studies, 12 (5), 607-624.

Schönauer A. (2007) Citylife – Property management for council housing. Organi- sational case study on customer service in the public sector – Austria, Internal Working Paper, WORKS project, Leuven, Higher Institute of Labour Studies.

Statistics Denmark, Statistics Finland, Statistics Netherlands, Statistics Norway and Statistics Sweden (2008) International sourcing moving business func- tions abroad, Copenhagen, Statistics Denmark.

Sturgeon T.J. and Gereffi G. (2009) Measuring success in the global economy: international trade, industrial upgrading and business function outsourcing in global value chains, Transnational Corporations, 18 (2), 1-36.

Valenduc G., Vendramin P., Krings B. and Nierling L. (2008) How restructuring is changing occupations? Case study evidence from knowledge-intensive, manu- facturing and service occupations, Leuven, Higher Institute of Labour Studies,

Vandekerckhove S. and Ramioul M. (2015) Using business functions to identify- ing new and growing jobs, in Holtgrewe U., Kirov V. and Ramioul M. (eds) Hard work in new jobs: the quality of work and life in European growth sectors, Basingstoke, Palgrave Macmillan.

van Eijnatten F.M. and van der Zwaan A.H. (1998) The Dutch IOR approach to organizational design: an alternative to business process re-engineering?, Hu- man Relations, 51 (3), 289-318.

Van Hootegem G. (2000) De draagelijke traagheid van het management, tendensen in het productie- en personeelsbeleid [The bearable lightness of management: tendencies in production and HR policies], Leuven, Acco.

van Welsum D. and Reif X. (2006) The share of employment potentially affected by offshoring:an empirical investigation, OECD Digital Economy Papers 107, Paris, Organisation for Economic Co-operation and Development.

van Welsum D. and Vickery G. (2005) Potential offshoring of ICT-intensive us- ing occupations, DSTI Infor¬mation Economy Working Paper, DSTI/ICCP/ IE(2004)19/FINAL, Paris, Organisation for Economic Co-operation and Development.

WTO (2005) Offshoring services: recent developments and prospects, in WTO (ed.) World trade report 2005: exploring the links between trade, standards and the WTO, Geneva, World Trade Organization.

Chapter 5

The impact of outsourcing on job quality for call centre workers in the telecommunications and call centre subcontractor industries

Virginia Doellgast and Elisa Pannini

1.    Introduction1

In recent decades, governments across the global North have liberal- ized their service markets and privatized major service providers with the aim of reducing prices, improving service quality, and enhancing competitiveness and innovation. Large firms have responded with re- structuring measures intended to drive down costs and increase returns to shareholders. One set of strategies involves shifting service jobs that were formerly performed in-house to subcontractors specializing in in- dustries such as IT, security services, catering and cleaning. These meas- ures can have significant effects on the pay and conditions of service workers within core firms and across their increasingly complex produc- tion networks, as jobs move to employers that both face stronger cost pressures and are covered by weaker institutional protections.

Debates concerning the labour market effects of service market liberali- zation often focus on outcomes such as job creation or economic growth (Mattoo, Rathindran, and Subramanian 2001). Policy pundits argue that legislation aimed at privatizing large firms and further opening markets to competition represents a ‘win-win’ for consumers and for workers, resulting in lower prices, increased consumer choice, and an expand- ing number of service jobs (Megginson and Netter 2001). However, re- search has shown that policy-driven changes in markets and ownership can also contribute to declining pay and working conditions for some groups of workers, particularly those who are lower skilled or whose jobs can be most easily outsourced (Hermann and Flecker 2012; Keune, Le- schke and Watt 2008). Within established firms and market segments,

1. This work was supported by the Economic and Social Research Council [grant number RES- 061-25-0444].

increasing cost pressures lead employers to seek reductions in labour costs. Meanwhile, industry areas and firms experiencing job growth tend to have weaker or no collective bargaining institutions, and to be char- acterized by both more market-driven wage setting and more flexible employment contracts than incumbent firms.

In this chapter, we examine the effects of production network fragmen- tation on job quality in the telecommunications and call centre subcon- tractor industries. These sectors are closely linked. Incumbent telecom- munications firms were pioneers in call centre technology and remain a major employer of call centre workers, both in-house and in subcontrac- tors. As telecommunications markets have liberalized, large firms have shifted a growing number of call centre jobs to subcontractors, while of- ten retaining a portion of these jobs in-house. They are thus a good place to look at the effects of subcontracting on job quality for both in-house and subcontracted workers, as well as the political dynamics of subcon- tracting decisions.

Findings are drawn from two research projects. The first is the ‘Global Call Center Project’, an international study of call centre management and employment practices based on establishment level surveys and case studies in 17 countries conducted in 2004-2005 (Batt, Holman and Holtgrewe 2009).2 Although this research is now a decade old, it con- tinues to be the most comprehensive survey of pay and working con- ditions in call centres worldwide, providing useful data for comparing differences across similar in-house and subcontracted workplaces. The second is a comparative study of employment restructuring in ten in- cumbent telecommunications firms, based on interviews, archival data, and site visits conducted between 2009 and 2014 (Doellgast, Sarmiento- Mirwaldt and Benassi 2013).3

2. The countries surveyed included Austria, Denmark, France, Germany, Israel, the Nether- lands, Spain, Sweden, Canada, Ireland, UK, US, Brazil, India, Poland, South Africa, and South The survey results include data from 2,477 call centres, with total employment of 474,942. More details on the survey can be found in the international report by Holman, Batt, and Holtgrewe (2007: 47-51).
3. This study included incumbent telecommunications firms from Austria, Czechia, Denmark, France, Germany, Italy, Poland, Sweden, UK, and USA. Findings are based on over 150 interviews, as well as site visits, comparative data drawn from collective agreements and surveys of union representatives, and archival material. Case study findings focused on restructuring in call centre and technician workplaces.

In the following sections, we first review outsourcing trends in the tel- ecommunications and call centre sectors, as well as associated changes in employment relations institutions. We then summarize the effects of these changes on pay and working conditions, focusing on two groups of employees: the in-house call centre workforce in large telecommunica- tions firms and the externalized workforce in call centre subcontractors. The central question we examine is: how has the growing availability and use of call centre subcontracting by telecommunications firms affected the quality of call centre jobs? We argue that job quality has declined due to two related trends. First, the shift of work to subcontractors has meant that a larger proportion of employees are working in an indus- try segment that faces stronger cost pressures and has weaker collec- tive bargaining institutions relative to in-house call centres. Second, the expansion of subcontracting also puts downward pressure on pay and conditions for remaining in-house call centre workers through a process of benchmarking and concessions. We conclude with a discussion of the role that employment relations and labour market institutions play in moderating these trends.

2.    Liberalization and production network fragmentation: telecommunications and call centre services

The telecommunications industry has transformed over the past two decades from one organized around national markets and dominated by state-owned telecommunications operators to a highly competitive information services industry. Technological changes led to falling costs associated with establishing competing networks and to the emergence of new market segments such as mobile, internet, and cable. National governments in Europe fully or partially privatized incumbent firms – in the mid-1980s in the UK and in the 1990s in the rest of Europe – and passed laws aimed at curtailing their market power. EU directives were central to the timing and form of these regulatory changes, as they re- quired member states to end monopolies and establish full competition by 1998.

The call centre industry has grown parallel to these changes, but also benefited more broadly from similar trends of liberalization and privati- zation affecting a wide range of service industries. While large firms often operate their own internal call centres, they also are able to take advan- tage of an increasingly sophisticated and diverse range of subcontracted

services offered by large multinational firms, such as Sykes, TeleTech, and Convergys. Convergys is an interesting case, as it was originally es- tablished through a merger of two call centre subsidiaries owned by the US-based incumbent telecommunications firms AT&T and Cincinatti Bell. Today Convergys is one of the largest firms in the call centre sub- contracting industry, owning 150 ‘service centers’ in 31 countries with 125,000 employees (Convergys 2014).

Telecommunications firms are among a wide range of clients in diverse industries that contract out call centre work. However, they represent a significant proportion of the subcontractor market. 51% of call centre sub- contractors included in the GCC survey reported that they served the tel- ecommunications industry (authors’ calculations). The telecommunica- tions and subcontractor industries also represent a large proportion of total call centre jobs. In 2011, an estimated 12% of agent positions in UK call cen- tres were in the ‘communications industry,’ while 10% were in ‘outsourcing and telemarketing’ – representing the largest sectors after finance and retail (Contact Centre Operations Labour Market Report 2012: 22).

The restructuring strategies adopted by telecommunications firms are thus closely linked to the emergence and expansion of the call centre industry. In the following sections, we review research findings concern- ing the effects of these restructuring strategies on job quality, drawing on survey data and case study examples.

3.    Job quality in call centre subcontractors

Job quality is most directly affected when large service firms move ex- isting or new call centre jobs to subcontractors that have lower pay and conditions compared to their in-house workforce. The decision to sub- contract call centre work is often based on a range of considerations. These newer firms can realize efficiencies through specialization and investment in new technologies. However, research findings suggest that a major source of cost savings is their ability to adopt lower wage structures and exercise more unilateral control over the flexibility and intensity of labour deployment compared to their clients.

Comparative findings show that pay and working conditions differ sys- tematically between in-house and subcontracted call centres. Across the countries included in the survey, typical wages for in-house agents were

on average 18% higher than those of agents in subcontractors (Holman et al. 2007). Based on an analysis of a subset of countries in the GCC survey, Batt et al. (2010) found that in-house call centres paid an 8.5% wage premium on average compared with subcontractors, controlling for the effects of human capital and other variables (see also, Batt et al. 2009; Van Jaarsveld and Yanadori 2011). In our ten-country compari- son of incumbent telecommunications firms, we also found systematic differences in pay rates for in-house and subcontracted call centre staff handling similar call types and customer market segments. The largest gap was in the US, UK, Germany and Denmark, where typical pay at sub- contractors was 20-50% lower than that enjoyed by the in-house work- force (Doellgast et al. 2013).

Other findings from the GCC survey show systematic differences in job quality across a range of measures. Subcontractors employed a larger proportion of part-time and temporary workers across diverse national and regulatory settings, and had higher dismissal and turnover rates – suggesting lower levels of job security (Batt et al. 2009; Van Jaars- veld Kwon and Frost 2009; Shire, Schonauer, Valverde and Mottweiler 2009). They invested almost 50% less in entry training, with significantly shorter training duration and qualification periods (Batt et al. 2009). In addition, jobs in subcontractors were characterized by lower complexity and discretion (Holman et al. 2007), and adoption of high-involvement management practices was less widespread, with significantly higher in- tensity of electronic performance monitoring (Batt, Doellgast and Kwon 2004; Doellgast, Holtgrewe and Deery 2009; Doellgast 2008). These practices are of particular concern for employee health, as more inten- sive performance monitoring in call centres has been found to contribute to higher levels of strain (Taylor et al. 2003), increasing rates of emo- tional exhaustion (Deery et al. 2002), depression and anxiety (Holman 2002), and musculoskeletal disorders (Sprigg et al. 2007).

What are the reasons for these differences in pay and working condi- tions between in-house and subcontracted call centres? Research find- ings from the two studies analysed in this chapter suggest three related explanations.

First, subcontractors face intense pressure from their clients to keep costs low and flexibility high. They must continually prove the value of their services relative to those of the in-house workforce – but also com- pared to other subcontractors offering similar services. Contracts are

often short, and can be withdrawn by client firms if targets are not met. This high level of competition means that margins are increasingly low, and cost-based considerations are central in decisions related to wage setting, work design, and scheduling practices.

Second, call centres are a ‘new’ sector with weaker (or no) collective bargaining institutions. Call centre jobs often cut across traditional in- dustries, with subcontractors serving clients as diverse as retail firms, financial institutions, collections agencies, and, of course, telecommuni- cations firms. In countries with low bargaining coverage overall, like the US and UK, call centre subcontractors often have very low or no union presence. Unions in continental European countries have succeeded in establishing collective bargaining institutions in this sector, but there are large differences across countries in the structure and coverage of sectoral agreements. For example, Austria, Denmark, France, Italy, the Netherlands, Spain and Sweden all have sectoral agreements either spe- cifically for call centre subcontractors or for more broadly defined service industries (Doellgast, Batt and Sorensen 2009; Doellgast et al. 2013). However, often these companies have some choice concerning which agreement to apply, resulting in ‘shopping around’ for the most favour- able terms and conditions. Where there is no automatic extension of col- lective agreements, such as in Denmark and Sweden, subcontractors are able to avoid applying sectoral collective agreements, giving them often unilateral control over pay and conditions. Germany represents a more extreme case where there are no sectoral agreements for the call centre or telecommunications industries, and only one major collective agree- ment at a call centre subcontractor.

Third, these problems are exacerbated by several characteristics of the international call centre industry that make it difficult for unions to or- ganize workplaces or to use collective action to improve existing terms and conditions. First, large multinational service providers increasingly dominate the industry, and these firms are able to offer clients remote services from a wide range of regions and countries. Corporate head- quarters are often located far from the locations where work is per- formed. The global strategies of these multinationals typically are based on creating a standardized ‘product’, in the form of efficient service and sales work. This is secured through intensive electronic performance monitoring and highly individualized performance management prac- tices, as well as avoiding unionization where possible. Second, call centre work is highly mobile. The ease with which work can be shifted ‘with the

flip of a switch’ provides these companies with a powerful threat they can use to discourage workers from seeking to unionize or worker represent- atives from attempting to push for improvements in pay and conditions. These problems are exacerbated in the case of offshore outsourcing, as cost savings associated with moving work from Europe to regions such as South Asia or North Africa can be substantial – from 30-50% of the cost of using domestic subcontractors. Third, high turnover rates further discourage union organizing, as it is difficult for unions to maintain sta- ble membership (Taylor and Bain 2001).

In sum, subcontractors face strong market- and client-based pressure that encourage them to adopt an employment model characterized by intensive performance monitoring, low pay and highly flexible employ- ment contracts. At the same time, they are better able to pursue this model than their in-house client firms because the characteristics of the industry make it difficult both to apply existing sectoral collective agree- ments and to organize and represent the subcontracted workforce.

4.    Challenges of organizing call centre subcontractors and improving job quality for the subcontracted workforce in ‘social European’ countries

Case studies from Denmark, Sweden and Germany illustrate the chal- lenges that labour unions face in two areas: first, in organizing call cen- tre subcontractors; and, second, where they have successfully organized works councils or collective agreements, in using institutionalized bar- gaining rights to secure substantive improvements in pay and condi- tions. These three countries are selected because they are all continental European countries with traditions of strong unions and encompassing collective bargaining. Thus, they represent cases in which we would ex- pect unions to have comparatively strong institutional resources to ex- tend gains won in major firms and industries to these externalized jobs.

In Denmark, the Union of Commercial and Clerical Employees (HK) is the major union responsible for the call centre industry. While call centres could follow white-collar sectoral agreements, most agreements with subcontractors were at company level, and a large number of sub- contractors did not adhere to collective agreements. HK was seeking to challenge this through organizing campaigns at several major call centre firms – with one focal campaign at ‘Go Excellent’, a major subcontractor

for the Swedish telecommunications firm Telia. A union organizer esti- mated that Go Excellent had starting pay levels of around 70-80 DKK/ hour, compared to 110-120 DKK/hr at Telia.

A distinctive challenge the union faced in encouraging the company to apply a collective agreement was presented by a historic agreement applying to employers who were members of associations negotiating white-collar agreements. This stated that labour unions could not strike or picket employers to encourage them to apply the collective agreement until they could document that at least 50% of the workforce were un- ion members. As Go Excellent was a member of an employers’ associa- tion, the union was bound by the 50% union membership rule. However, Go Excellent had established a new outbound department as a separate company, which was not tied to this association. This made it easier for HK to organize, and they got an agreement for a minimum salary after 6 months tenure of 129 DKK/hr – which was higher than other Go Excel- lent employees received after 5-6 years (Interview, HK Privat organizer, 26/4/12). At the same time, the union had not succeeded in organizing the larger company, and was not optimistic that it would pass the 50% membership threshold, as the workforce was young, turnover was high, and union members could be easily dismissed due to Denmark’s weak employment protections. Although it is illegal to fire a worker for union activities, union representatives noted that it was very difficult to prove this in the labour courts, particularly in sectors or jobs where turnover is high.

A second example is from Sweden. In contrast to the situation in Denmark, the white-collar union, Unionen, succeeded in organizing a sectoral agreement for call centre subcontractors, and many large subcontractors followed the agreement or negotiated company-level agreements. However, some call centres did not follow collective agree- ments. One of these was a small call centre company, Focus CRS, located in Northern Sweden – which had very low pay and no notice period for dismissals or layoffs. In 2013, after a colleague was refused sick pay, a group of employees organized with Unionen and sought to negotiate an agreement with management. The company refused, and workers went out on strike. The story of the strike gained wide media atten- tion and the intervention of the Swedish Minister of Industry and Trade. However, management responded by closing the centre and laying off its workers. The strike was viewed by the union as a success, in that it il- lustrated to other subcontractors the potential risk of operating without

a collective agreement in Sweden. At the same time, this case also dem- onstrates the limitations of collective action in an industry characterized by highly mobile work: management was able to respond to a successful organizing campaign by shutting down the call centre and moving jobs elsewhere.

A third example from Germany illustrates the large challenges unions and works councils face in improving pay and conditions at call centre subcontractors, even where they have successfully established collective bargaining (see Doellgast 2012: 95-108 for more detail). ‘Client Services Germany’4 is a major Germany-based subcontractor with several thou- sand employees. In the late 1990s, employees organized works council elections and negotiated a collective agreement with the support of the service union ver.di. Prior to the establishment of a works council, man- agement practices were driven by headquarters and clients, who were closely involved in monitoring and setting performance goals. The works councils used their co-determination rights to seek to negotiate a num- ber of improvements in working conditions, but always within the con- straints of maintaining low costs and flexibility. Working time accounts were introduced to provide increased income stability. However, man- agement maintained its discretion to increase or decrease staffing based on changes in call volumes. The works council negotiated strict rules prohibiting electronic monitoring of individuals. However, where these conflicted with client demands, the works council agreed to exceptions, negotiating several separate agreements to regulate how the company used data collected for each client within the framework of distinctive monitoring systems.

One of the most difficult challenges Client Services’ works councils faced was maintaining solidarity across locations in the face of competition for jobs. Management used differences in labour costs and practices to argue for desired changes to works agreements. In one example, a works council in one East German location negotiated a separate job security agreement that applied only to that location, and in exchange employees worked an additional 25 hours each year without pay, used in part for additional training. Works councillors noted that this competition made it difficult for them to seek improvements in conditions, particularly if they would create additional costs that could disadvantage their location.

4.   Case study name is a pseudonym.

This was combined with more general pressure to keep labour costs in line with those in the (mostly non-union) subcontractor industry.

These three examples illustrate the industry and political dynamics in the call centre subcontractor sector that contribute to lower pay and conditions, as well as the challenges faced by unions and workers seek- ing to use traditional collective bargaining institutions to improve job quality in this industry. These challenges were similar across Den- mark, Sweden and Germany, despite differences in industrial rela- tions institutions – with higher bargaining coverage and some secto- ral agreements for subcontractors in Denmark and Sweden. However, they also suggest that a common weakness in all three countries con- cerns the voluntary nature of adherence to collective agreements. De- spite relatively strong unions and collective regulation of work, the lack of state involvement in extending minimum terms and conditions is a considerable liability where there are strong cost pressures and thus significant incentives for employers to remain outside the formal bar- gaining system. The examples point to some potential sources of lever- age to organize works councils, use co-determination rights creatively, and use organizing and strike tactics. However, they also suggest that these face an uphill battle in the face of competition between locations for highly mobile work and strong pressures by clients to keep costs low and flexibility high.

5.    Benchmarking and concessions in in-house telecommunications call centres

The previous section discussed the shift of jobs to firms and workplaces characterized by lower pay and conditions relative to similar in-house jobs. A second concern is with how these trends affect job quality for the in-house call centre workforce within the firms that contract out work. As established incumbent firms expand their use of subcontracting, their in-house workers find themselves increasingly competing with a lower paid, more flexible externalized workforce. The emergence of a new market-based benchmark of pay and conditions associated with subcon- tracted services is often used as justification for changing or renegotiat- ing employment contracts for in-house workers. This can occur via con- cession bargaining or through more unilateral adoption of management practices mirroring those in subcontractors.

The political dynamics involved in ‘whipsawing’ or ‘coercive compari- sons’ between plants or workplaces competing for investment and jobs within multinational firms have been extensively documented (e.g. Edwards, Rees and Coller 1999; Greer and Hauptmeier 2014; Mueller and Purcell 1992). However, subcontracting has often been argued to be characterized by less direct or intense worker-to-worker competition. This is based on two assumptions, rooted in labour market segmentation theory or ‘insider-outsider’ arguments. First, jobs that are subcontracted are seen as qualitatively different from those retained in-house – includ- ing simpler tasks that are more easily codified and further from a firm’s core competency, or work that is subject to more fluctuation in demand. Second, ‘insider’ groups of permanent, core workers are viewed as hav- ing a shared interest with management in allowing this work to be out- sourced to firms with lower labour costs, to protect their own pay and conditions.

These assumptions have been challenged by recent research demon- strating the widespread use of benchmarking employment practices and labour costs between subcontractors and in-house workers. Sev- eral large international studies, including the Global Call Center project, have shown that this competition not only leads to declining working conditions ‘downstream’ in value chains, but can also undermine condi- tions for core workers as a broader range of jobs are potential targets of restructuring (Doellgast, Batt and Sorensen 2009; Flecker and Meil 2010; Flecker 2009; Hermann and Flecker 2012). Where unions are pre- sent, they often use diverse tactics to challenge outsourcing – which can include negotiating limits on the types and amount of work outsourced and extending collective agreements to subcontractors. However, these have varied success: under conditions of declining union density and bargaining coverage or increasing cost pressures, unions often find that the only effective strategy to halt or reverse outsourcing is to agree to re- duce pay and increase flexibility for certain employee groups (Doellgast 2008b; Pulignano and Keune 2014). Work is thus kept in-house under better conditions than in subcontractors, with stronger representation structures and protections. However, over time, this can lead to growing similarity in employment practices between the two industry segments.

The dynamics can be illustrated by two case studies, Deutsche Telekom in Germany and TDC in Denmark (see Doellgast and Berg 2015; and Doellgast, Sarmiento-Mirwaldt Benassi 2013 for more detail).

Deutsche Telekom (DT) reorganized its call centre jobs in several waves. Outsourcing happened gradually over the late 1990s to early 2000s, al- though this did not directly affect in-house staff due to strong job securi- ty agreements. In 2004, DT established a subsidiary called Vivento Cus- tomer Services (VCS) to handle call centre work both across the company as well as to compete in the subcontractor market. The union agreed that existing staff would have their pay cut by 8.75%, in exchange for an ex- tension of job security until 2008. Then between 2006 and 2008, DT transferred most of its VCS locations to two call centre subcontractors, Walter Services and Arvato. Those employees who moved to Arvato had the terms of their existing collective agreement secured through 2009, after which employees were asked to sign individual contracts with low- er pay and conditions – accompanied by the informal threat that jobs would be moved to other locations if employees refused to sign. Em- ployees who moved to Walter Services were automatically transferred to its (lower paid) collective agreement with ver.di, based on a provision in German law that provides an exception to transfer of undertakings rules when both firms are covered by agreements with the same union. In both cases, the transferred call centre workers experienced a pay cut of around a third of their former salary, increased monitoring, and de- clining control over their working time.

In the middle of this, in 2007, DT announced that it planned to shift 50,000 of its technical service, technical infrastructure and call centre jobs to three new subsidiaries. Ver.di responded with a six-week strike, with strong support from its membership. However, the union faced a number of challenges to building bargaining power in negotiations. One of the most worrying for ver.di was that DT had publicly threatened to sell the service subsidiaries to subcontractors if it was unable to get a favourable agreement. Under the agreement eventually reached, trans- ferred workers took pay cuts of more than 10 per cent, and new employ- ees were to earn 30 per cent below the former level. Variable pay also increased, and Saturdays were included in the regular work week in call centre operations. Management agreed to extend layoff protections until 2012, to refrain from selling the new service subsidiaries until 2010, and to offer 4,150 jobs at the subsidiaries to DT apprentices.

Despite lower labour costs, the proportion of call center work outsourced from the new subsidiary continued to grow. Internal call center employ- ment had declined from around 18,000 in 2007 to 12,000 in 2010. At the same time, 11,000 employees were working for subcontractors.

Management used the gap in pay and conditions between the two groups to benchmark costs and productivity. An agreement with ver.di and the subsidiary’s works council in 2011 introduced a series of changes aimed at providing more flexibility in-house, with the goal of reducing the use of subcontractors and stabilizing the internal workforce. One key com- ponent of this was to give management the flexibility to hire employ- ees for shifts for which subcontractors were more intensively used – i.e. late nights and weekends. The works council agreed that students and younger employees could be hired on part-time contracts to cover these shifts. All new call centre employees hired after 2011 were on these con- tracts working 20-30 hours/week, with shifts from 12 pm to 12 am. The proportion of part-time employees had been 20 per cent in the past; as a result of the change in hiring, this increased to 40 per cent by 2014. In addition, a pilot project was initiated, providing employees with three different options of alternative working time models – each with differ- ent degrees of choice and flexibility. The new ‘3 model’ system of work- ing time was rolled out on different business/lines every three months throughout 2014. In exchange, management undertook to stabilize in- ternal employment and reduce the use of subcontractors and temporary agency workers. Between 2011 and 2013, the number of subcontractor employees declined from 11,000 to 6,000.

TDC initially did not subcontract call centre jobs, instead relying on a combination of in-house centres, a dedicated call centre subsidiary and temporary agency workers. In 2011, TDC management announced that they were considering outsourcing call centre work, and asked the work- force to agree to changes in working time rules intended to reduce costs and increase flexibility to bring these more in line with those in the ex- ternal subcontractor market. In response, the union Dansk Metal nego- tiated an agreement in 2011-12 that introduced a new overtime policy, whereby employees would finish up with their final customer, with any overtime beyond their contractual working hours put into a ‘time bank’. Management could then send employees home early for an equivalent period of time when there was less demand. However, employees con- tinued to receive an overtime premium of 150% of typical pay for the first hour and 200% for every further hour.

This new working time model was successful in gaining a commitment from management to keep the jobs in-house for a period of two years. However, in 2014 management demanded additional concessions, again using the threat of outsourcing. This time the demands were more

extensive, including the elimination of paid breaks and lunches for new hires in their first two years; reduced pensions; cuts in overtime pay to a smaller 115% premium; and no pay for the first 5 minutes and last 10 minutes at work. A shop steward observed that management used comparisons with typical employment contracts and conditions at call centre subcontractors to argue for these changes (TDC shop steward, 30/10/2014).

When the union refused to accept the concessions, TDC announced that it would outsource 800 call centre employees to the US service provider Sitel, representing around half the call centre workforce across TDC. Workers transferred to Sitel have their conditions and workplace secured for one year, until the end of 2016. After that they will have to choose be- tween moving to, or remaining at, a call centre in Sønderborg, Denmark and moving to London – where Sitel operates a call centre staffed by Nordic language speakers. Those staying in Denmark would be moved to Sitel’s company-level collective agreement, which has lower pay and conditions compared to the TDC agreement, while the London location is not covered by collective bargaining. A shop steward pointed out that young people without any family commitments could take advantage of these new opportunities, but workers with families would find it difficult to relocate (Interview, 30/10/2014). Workers on temporary contracts at the time of the transfer – constituting a substantial portion of the work- force at some locations – will not have their contracts renewed.

These two case studies illustrate the negative effects subcontracting can have on pay and conditions for the in-house call centre workforce, through a process of benchmarking and concessions. Where concessions were not made, or deemed insufficient by management to close the gap in labour costs between the in-house and subcontracted workforce, call centre jobs and often the in-house workers in those jobs were moved to more poorly regulated subcontractors. This shows that worker develop- ments in new subcontracting industries are closely linked to changes in working conditions in traditionally ‘core’ workplaces – particularly for those jobs most at threat of further subcontracting.

6.    Conclusions

This chapter has discussed the effects that call centre subcontract- ing has had on job quality for call centre workers, with a focus on the

telecommunications industry. Understanding the political dynamics associated with restructuring decisions that shift jobs from established incumbent firms to subcontractors is crucial for evaluating the social consequences of changes in markets and ownership. Business interests often argue that competitive markets encourage more efficient allocation of resources via productivity improvements and product and technologi- cal innovations. However, research on the politics of and outcomes from outsourcing demonstrates that lower prices and expanding profits can also result from organizational strategies that allow firms to reduce wag- es and shift risk onto workers. In sectors and workplaces where the latter is a more significant driver of competitiveness than the former, service market liberalization is likely to lead to the expansion of insecure and low wage jobs. These effects should be taken into consideration when evaluating the economic and social impact of liberalization policies.

The research reviewed in this chapter demonstrates that the growth of a new, poorly regulated sector of call centre subcontractors has been as- sociated with declining job quality for employees performing call centre work. Subcontractors offer workers lower pay, less secure jobs, more use of variable ‘at risk’ pay systems, higher use of invasive monitoring prac- tices, and less worker control over schedules and breaks compared to similar ‘in-house’ call centres. As call centre work is outsourced, pay and standards across firms’ increasingly fragmented production chains suf- fer. In addition, the in-house workforce has to compete with outsourced and offshored employees via benchmarking of costs and performance. This can lead to a downgrading of job quality via concessions or the wholesale shift of workers to new employers.

While the findings discussed in this chapter are rooted in the specifici- ties of the call centre industry, with case studies drawn from European ‘coordinated’ economies, they apply more broadly to many subcon- tracted services in diverse institutional settings. In his book on the ‘fis- sured workplace’, Weil (2014) argues that a major source of the savings from subcontracting comes from its success in reducing labour costs via downward pressure on pay and working conditions. As wage setting is turned over to companies and individuals operating in increasingly com- petitive environments, the social aspects of wage setting in large firms are transformed into a pricing problem. Weil describes these develop- ments in the US, where institutional protections for many workers in large firms are minimal in international comparison.

These findings raise the question of how these pressures can be reduced, to expand the capacity of worker representatives to organize and rep- resent new service workplaces and to avoid concession bargaining in established industries undergoing liberalization. Several broad implica- tions for public policy and union strategy can be derived from the find- ings presented.

First, encompassing collective bargaining institutions are an important tool both for establishing greater parity of job quality across in-house and subcontracted workplaces and for maintaining favourable pay and conditions for in-house workers. Although outsourcing is not always transparently a cost cutting strategy, the outsourcing of call centre jobs from unionized workplaces does often have this motivation. Lower la- bour costs and more flexible employment practices are possible due to low union density and bargaining coverage in the subcontractor sector. Benchmarking of labour costs and employment practices then inhibits worker representatives in both segments from organizing for improve- ments – instead encouraging a strategic focus on retaining jobs at any cost. In continental European countries with traditions of sectoral bar- gaining, state policies and union organizing efforts should aim to extend these institutions to new sectors and to establish greater parity of terms and conditions for call centre workers in different sectors. The Swedish and Danish examples of union campaigns in call centre subcontractors described in this chapter provide evidence that unions are seeking to leverage institutional forms of bargaining power in new ways to organize within the sector. Further examples of these campaigns are discussed in Chapter 8 of this volume.

Second, as outsourcing exacerbates the power imbalance between la- bour and management, it is necessary to strengthen legislation aimed at remedying this imbalance. Transfer of undertakings legislation can increase the costs associated with transferring workers to subcontrac- tors and temporarily protect the employment conditions of transferred workers. At Deutsche Telekom and TDC, transfer of undertakings rules and agreements resulted in the short-term protection of pay and con- ditions as in-house workers were transferred to call centre subcontrac- tors. However, there were few lasting effects of these provisions: in both companies, pay and conditions were brought down to a lower level after the transition period ended. This suggests that transfer of undertaking legislation itself could be modified to support longer-term stability in collective agreements. For example, employees’ rights to maintain their

former terms and conditions could be extended on an indefinite basis; or legislation could require that subcontractors apply the collective agree- ment of their clients’ workforce.

Even strengthened legislation is likely to have limited effectiveness given the international scope of multinational service firms and international mobility of call centre work. The TDC case shows that even in Nordic countries, call centre jobs can be offshored to lower wage locations – in that case, to London. Thus, a third observation is that union campaign- ing and organizing efforts should seek not only to establish more en- compassing institutions within countries, but also to extend negotiated protections across core firms’ domestic and international production networks. On the one hand, bargaining power within large employers could be mobilized to negotiate provisions requiring subcontractors to maintain pay and conditions at a certain level. On the other hand, unions could use the transfer of existing members to subcontractors as a strategic opportunity to organize collective action aimed at securing improvements. At the international level, European Works Councils (EWCs) provide one lever for establishing cross-border communication between worker representatives and consultation with multinationals on their international restructuring practices. Most of the major European incumbent telecommunications firms have EWCs, as well as many major subcontractors – for example, Sitel. France Telecom is a distinctive case in which a world works council has been established, and a global agree- ment negotiated with the global union federation UNI-ICTS. Conditions in call centre subcontractors have been a limited focus to date – under- standably, given the already large challenges of improving conditions in subsidiaries in regions with lower pay and conditions such as Eastern Europe and Africa. However, these consultative bodies provide some scope for future framework agreements aimed at improving conditions in outsourced and offshored call centres.

Efforts to organize across the production networks of multinational ser- vice firms could complement current international union campaigns fo- cused on call centres. UNI-ICTS sponsors a ‘Global Call Centre Action Month’ every year and organizes international conferences that bring together call centre workers from diverse industries and countries to dis- cuss their shared challenges and lessons learned from successful cam- paigns. These serve to encourage exchange of best practices and build in- ternational solidarity among union activists campaigning for improved conditions in call centres around the world. These efforts face an uphill

battle, but are a necessary part of efforts to improve job quality in call centre subcontractors and avoid the erosion of terms and conditions for the remaining in-house workforce.

References

Batt R., Doellgast V. and Kwon H. (2004) U.S. call center industry report 2004 national benchmarking report strategy, HR practices & performance, CAHRS Working Paper #05-06, Ithaca, Cornell University, School of Industrial and Labor Relations.

Batt R., Holman D. and Holtgrewe U. (2009) The globalization of service work: comparative institutional perspectives on call centers: introduction to a special issue of Industrial & Labor Relations Review, ILRReview, 62 (4), 453-488.

Batt R., Nohara H. and Kwon H. (2010) Employer strategies and wages in new service activities: a comparison of co-ordinated and liberal market economies, British Journal of Industrial Relations, 48 (2), 400-435. doi:10.1111/j.1467- 8543.2010.00789.x

CfA (2012) Contact centre operations: labour market report 2012, London, CfA business skills @ work. http://www.oph.fi/download/145729_Contact_Cen- tres_LMI_2012_.pdf

Convergys (2014) Convergys company overview. http://www.convergys.com/company/company-overview.php

Deery S.J., Iverson R. and Walsh J. (2002) Work relationships in telephone call centres: understanding emotional exhaustion and employee withdrawal, Journal of Management Studies, 39 (4), 471-496. doi:10.1111/1467-

6486.00300

Doellgast V. (2008a) Collective bargaining and high-involvement management in comparative perspective: evidence from U.S. and German call centers, Indus- trial Relations, 47 (2), 284-319. doi:10.1111/j.1468-232X.2008.00521.x

Doellgast V. (2008b) National industrial relations and local bargaining power in the US and German telecommunications industries, European Journal of Industrial Relations, 14 (3), 265-287. doi:10.1177/0959680108094135

Doellgast V. (2012) Disintegrating democracy at work: labor unions and the future of good jobs in the service economy, Ithaca, ILR Press.

Doellgast V., Batt R. and Sorensen O.H. (2009) Introduction: Institutional change and labour market segmentation in European call centres, European Journal of Industrial Relations, 15 (4), 349-371. doi:10.1177/0959680109344366

Doellgast V. and Berg P. (2015) Negotiating flexibility: external contracting and internal working time control in German and Danish telecommunications firms, Draft paper.

Doellgast V., Holtgrewe U. and Deery S. (2009) The effects of national institutions and collective bargaining arrangements on job quality in front-line service workplaces, Industrial and Labor Relations Review, 62 (4), 489-509.

Doellgast V., Sarmiento-Mirwaldt K. and Benassi C. (2013) Alternative routes to good jobs in the service economy: employment restructuring and human resource management in incumbent telecommunications firms, Final Project Report ESRC Grant RES-061-25-0444. http://www.esrc.ac.uk/my-esrc/ grants/RES-061-25-0444/outputs/read/be077186-d4cf-43e9-88a5- e42ae32bb2e6

Edwards T., Rees C. and Coller X. (1999) Structure, politics and the diffusion of employment practices in multinationals, European Journal of Industrial Rela- tions, 5 (3), 286-306.

Flecker J. (2009) Outsourcing, spatial relocation and the fragmenta- tion of employment, Competition & Change, 13 (3), 251-266. doi:10.1179/102452909X451369

Flecker J. and Meil P. (2010) Organisational restructuring and emerging service value chains: iImplications for work and employment, Work, Employment and Society, 24 (4), 680-698.

Greer I. and Hauptmeier M. (2014) Management whipsawing: the staging of labor competition under globalization, Industrial and Labor Relations Review, 1–39 (Draft accepted).

Hermann C. and Flecker J. (eds) (2012) Privatization of public services: impacts for em- ployment, working conditions, and service quality in Europe, New York, Routledge.

Holman D. (2002) Employee wellbeing in call centres, Human Resource Manage- ment Journal, 12 (4), 35-50.

Holman D., Batt R. and Holtgrewe U. (2007) The global call centre report: inter- national perspectives on management and employment, Sheffield, Global Call Centre Research Network.

Keune M., Leschke J. and Watt A. (2008) Introduction: Liberalisation, privatisation and the labour market, in Keune M., Leschke J. and Watt A. (eds) Privatisation and liberalisation of public services in Europe: an analysis of economic and labour market impacts, Brussels, ETUI, 13-34.

Mattoo A., Rathindran R. and Subramanian A. (2001) Measuring services trade liberalization and its impact on economic growth: an illustration, Policy Re- search Working Paper 2655, Washington, DC, World Bank.

Megginson W. L. and Netter J. M. (2001) From state to market: a survey of empir- ical studies on privatization, Journal of Economic Literature, 39 (2), 321-389.

Mueller F. and Purcell J. (1992) The Europeanization of manufacturing and the decentralization of bargaining: multinational management strategies in the European automobile industry, The International Journal of Human Resource Management, 3 (1), 15-34.

Pulignano V. and Keune M. (2014). Understanding varieties of flexibility and security in multinationals: Product markets, institutions variation and local bargaining. European Journal of Industrial Relations, 1–17. doi:10.1177/0959680114527880

Shire K. A., Schönauer A., Valverde M. and Mottweiler H. (2009) Temporary work in coordinated market economies: evidence from front-line service workplaces, Industrial and Labor Relations Review, 62 (4), 602-617.

Sprigg C. A., Stride C. B., Wall T. D., Holman D. J. and Smith P. R. (2007) Work characteristics, musculoskeletal disorders, and the mediating role of psycho- logical strain: a study of call center employees, Journal of Applied Psychology, 92 (5), 1456-1466.

Taylor P. and Bain P. (2001) Trade unions, workers’ rights and the frontier of con- trol in UK call centres, Economic and Industrial Democracy, 22 (1), 39-66.

Taylor P., Baldry C., Bain P. and Ellis V. (2003) ‘A unique working environment’: health, sickness and absence management in UK call centres, Work, Employ- ment and Society, 17 (3), 435-458. doi:10.1177/09500170030173002

Van Jaarsveld D., Kwon, H. and Frost A.C. (2009) The effects of institutional and organizational characteristics on work force flexibility: evidence from call cent- ers in three liberal market economies, Industrial and Labor Relations Review, 62 (4), 573-601.

Van Jaarsveld D. D. and Yanadori Y. (2011) Compensation management in out- sourced service organizations and its implications for quit rates, absenteeism and workforce performance: evidence from canadian call centres, British Jour- nal of Industrial Relations, 49 (Supplement s1), s1–s26. doi:10.1111/j.1467- 8543.2010.00816.x

Weil D. (2014) The fissured workplace: why work became so bad for so many and what can be done to improve it, Cambridge, Harvard University Press.

Chapter 6

Outsourcing public services: local government in Italy, England and Denmark

Anna Mori

1.    Introduction

Over the past three decades, public administrations have been subject to far-reaching restructuring, not stopping at outsourcing a wide range of tasks and services. Under the pressure of increasing public debt, stricter spending constraints, shifting consumer preferences and the demand for higher value for money in an era of austerity, these organizations have differentiated the provision of services, opening up the traditional direct production and delivery systems to competition through the adoption of a wide set of market-type mechanisms, including public-private part- nerships, voucher systems, liberalization policies and contracting out (OECD 2011).

The introduction of market discipline in the provision of public services represents one piece within the broader jigsaw puzzle of public adminis- tration reform, an ‘unending wave of reforms’ (Pollitt 2001) that has ex- panded progressively since the 1980s across all European governments under the label of New Public Management (NPM) (Hood 1991), sug- gesting uniformity and communality. The aim of this doctrine has been to remove differences between the public and private sectors in a drive to increase efficiency and effectiveness: this has seen governments im- porting business-like tools and values into their public administrations, along with market-type mechanisms from the private sector (Pollitt and Bouckaert 2011).

Outsourcing is an instrument taken from the NPM-inspired toolbox, a market-type tool for saving public funds (Savas 2000) through a shift of service provision across public sector borders towards private organi- zations. The hope here is that opening up service provision to market discipline can lead to cost reductions, since private providers in a com- petitive regime are expected to realize economies of scale and to raise effort or productivity with a given input/workforce-combination. While

outsourcing has unquestionably promoted competition, gains are often largely based on the reduction of labour costs rather than on quality im- provement and innovation: empirical evidence increasingly shows that savings may simply correspond either to reduced employment levels or to a deterioration in employment terms and conditions (Flecker and Hermann 2011).

This process, leading to complex changes in the organization of work and employment conditions, has been subject to growing scrutiny in the last few years, given the magnitude of the emerging drawbacks (Her- mann and Flecker 2012; Petersen et al. 2012). It has been repeatedly assumed that the vertical disintegration of service provision, involving the dispersion of activities across organizational boundaries, has led to a deterioration in employment terms and conditions (Flecker and Meil 2010; Schulten and Böhlke 2012), since it triggers competition with a cheaper labour supply, shifting work from highly unionized and better sheltered public sector organizations to more vulnerable private com- panies subject to market fluctuations and where trade union power is low or non-existent. Moreover, the overlapping of blurred boundaries between the public and private sector is expected to translate into in- creased disparities and instability in terms of employment and working conditions (Flecker 2009).

This chapter discusses how outsourcing impacts public sector working conditions and employment structures from a comparative perspective. I will be exploring the issue through a case study analysis of public ad- ministrations in three countries – Italy, England and Denmark – with a specific focus on local government. Within each country one municipal- ity was selected. The local government sub-sector was selected as it has experienced extensive waves of outsourcing labour-intensive activities, driven by legislative interventions and budgetary constraints. The com- parative perspective also allows us to ascertain whether the implications for reconfiguring work across boundaries display a lowest common de- nominator for all countries, or conversely whether national distinctive- ness emerge. In conclusion, explanations of the outcomes are provided.

The overall picture across the three municipalities studied is domi- nated by a deterioration of employment terms and conditions for pub- lic employees who, depending on the cases, were either transferred to subcontractors or functionally seconded to the external providers. As a consequence organizational boundaries between the public and private

sectors and across organizations have become increasingly blurred and overlapping, leading to negative consequences for labour, both in terms of wages and allowances and in the organization of work (workloads, working hours and workplaces). However, beyond such general conver- gence, impacts take different forms based on the role played by national institutional arrangements, legislative set-ups and differing national in- dustrial relations regimes (Flecker 2010; Vraengbæk et al. 2013; Jaehr- ling 2014). Specifically, encompassing employment relations institu- tions and strong legislative protections for public employment, though unable to prevent public services being outsourced, have to a certain extent cushioned working conditions for public sector staff in Denmark and Italy; while, conversely, a market-driven employment relations re- gime, putting considerable emphasis on market adjustments and self- regulation and thus minimizing the role of employment regulation and organized labour institutions (Gallie 2007), has inevitably led to a broad deterioration of employment conditions in the UK.

The chapter is structured as follows. After providing a brief overview of the debate on restructuring through outsourcing and its implications for employment relations in the public sector, the empirical findings from the three-country analysis are discussed from a comparative perspec- tive. The chapter is then summarized in a brief concluding section.

2.    Restructuring through outsourcing and labour issues in the public sector

With reference to the public sector, outsourcing is defined as the trans- fer of public service provision to an external organization, while retain- ing public financing of the activity and keeping overall responsibility for performing the function in line with the accountability principle towards citizens (OECD 1993; Jensen and Stonecash 2005). Unlike private com- pany restructuring measures which increasingly involve a delocalization of tasks outside national borders, contracting out within a public admin- istration generally implies a restructuring of activities across organiza- tional boundaries, with the result that the borders between the public and private sectors are becoming increasingly blurred and overlapping.

Maximizing value for money in the provision of services constitutes the overriding rationale for outsourcing (Domberger and Jensen 1997; Sa- vas 2000): public administrations are likely to adopt such a policy on the

basis of economic evaluation, with a view to reducing their expenditure and boosting efficiency through opening up service provision to mar- ket competition. Despite this potential benefit, many drawbacks are re- ported. Savings are often lower than expected and may even decline in the long term (Domberger and Rimmer 1994; Roberts 2001); moreover scholars have started questioning not only the magnitude but even more importantly the actual source of savings (Jensen and Stonecash 2005) which may indeed be achieved at the expense of output quality (Quiggin 2002; Jensen and Stonecash 2005) or through reducing labour costs at the expense of employment conditions rather than improving input uti- lization (Domberger and Jensen 1997; Blöndal 2005).

Accordingly, restructuring policies represent a burdensome challenge for labour issues: while outsourcing is used on purpose to downsize the internal workforce (Alonso et al. 2013), it inherently leads to the creation of intertwined links between public and private organizations, triggering a shift in work terms and conditions.

A thorough and detailed account has been provided by some previous studies1 which repeatedly describe how outsourcing public services is al- ways to the detriment of labour (Vrangbæk et al. 2013), depriving work- ers of industrial relations institutions: the overall picture is dominated by a fragmentation of working conditions both in terms of wages (Flecker and Hermann 2009; Flecker and Meil 2010) and work-related benefits (pension schemes, maternity/paternity leave, sick-pay). Moreover it has led to numerically-relevant transfers of personnel to the private com- pany taking over as service provider (Flecker et al. 2008; Flecker and Meil 2010) which has often translated into a switch from a public sec- tor collective agreement to the private provider’s terms and conditions, usually less protective, more flexible and poorly paid. Redundancies are no exception. The demand for higher contractual flexibility and variable workloads has triggered a shift from permanent to temporary contracts as a way to recruit flexible staff (Flecker and Thörnqvist 2012). Moreo- ver, these studies have depicted a dominant trend towards the stand- ardization and intensification of work, making workers easily replace- able and undermining their bargaining power. Finally, understaffing has

1. For further details: WORKS (Work Organisation and Restructuring in the Knowledge Society) Project – Changes in Work (www.worksproject.be); PIQUE (Privatisation of public services and the impact on quality, employment and productivity) Project (http://www.pique.at).

necessarily led to appreciable work intensification, meaning longer work- ing hours, higher workloads and long shifts (Flecker and Meil 2010).

In the next sections, evidence from three local governments is described and discussed from a comparative perspective in an attempt to distin- guish similarities and divergences.

3.    Implications for labour and working conditions in local government

The way outsourcing has impacted public employees’ working conditions and employment structure was explored through a case-study analysis of municipalities in three countries – Italy, England and Denmark. The evidence gained in the study stems from an in-depth qualitative analysis based on 44 semi-structured interviews carried out between September 2013 and November 2014 and complemented by documental analysis of secondary sources such as legislation, annual reports, internal Power- Point presentations, collective agreements, practitioners’ reports, inter- nal employer association and union documents and newspaper articles.

3.1                   The Italian municipality: the response to austerity measures

Outsourcing in the Italian municipality was driven by twofold legislative pressure. On the one hand central government intervened proactively to boost contracting out as a policy tool to reduce public spending and increase management economies: during the 1990s legislation incen- tivized municipalities to either outsource service provision to external providers or to subcontract it to joint-stock or limited liability compa- nies fully or partly owned by the local authority (Law 142/1990; Leg- islative Decree 267/2000; Budget Law 448/2001). On the other hand increasingly strict austerity measures hit local government budgets in terms of both staffing levels and financial resources: starting with the 2008 annual budget law the government defined the overall financial targets for municipalities in an effort to contain public debt, setting le- gal constraints on their expenditure, including personnel costs (Decree Law 112/2008; Decree Law 78/2010; Legislative Decree 149/2011; Law 183/2011) (Bordogna and Pedersini 2013). Further legislative interven- tions directly targeted public sector headcount, allowing municipalities to only hire new employees in 2009 when expenditure on them did not

exceed 10% of the labour cost of those having retired in the previous year (‘headcount pruning’). This figure rose to 20% for 2010 and 2011, and 50% for 2012 (d.l. 112/2008) (Bordogna and Neri 2014).

This twofold pressure forced the municipality studied to outsource a wide range of services from the mid-1990s onwards, starting with blue- collar activities such as park and street maintenance, waste collection, cleaning and catering and slowly encroaching on core services in the 2000s such as social care, crèches and residential home services.

The overall picture that emerged is rather strange since, despite the law enabling the shift from public administration to private companies, the municipality held on to almost all staff involved, relocating the workers to different offices and only transferring a residual share to private pro- viders. When catering was outsourced in the early 2000s, including the meals-on-wheels and food service for crèches, the municipality at that time employed ancillary catering staff on temporary contracts and cooks on permanent contracts: as a consequence of the transition the former were all transferred to the subcontractor after a 3-month probationary period while the latter maintained their public status, being seconded to the private company. The number of catering staff with a public sector employment contract has since shrunk due to retirements, with replace- ments hired by the subcontractor.

The second transfer involved manual workers doing park maintenance: at the end of the 1990s the service was passed to a company in which the administration had a minority holding, and workers were given the voluntary option of transferring to it. The large majority, however, was relocated within the municipality, with only a few workers accepting the transfer. However, the collective agreement applied by private companies providing environmental services and public sanitation provided for high- er wages and allowances compared to the public one, though with less favourable working conditions such as longer shifts, different working time schedules including night and weekend work and prolonged on-call periods. A union official who monitored the transfer added that ‘working time was longer, they knew they had to work mornings and afternoons but they were content, concluding that ‘I work more but I earn more’ (CISL union official, 16 September 2014). Moreover ‘it was profession- ally gratifying because they were given greater autonomy, allowing them to feel responsible for their area of competence, thereby raising their level of professionalism’ (head of municipal HRM, 10 October 2014).

Such outcomes were achieved through the constant monitoring of un- ions during outsourcing measures, a strategy that allowed public person- nel to keep their jobs through being redeployed within the administra- tion, a move sanctioned by the municipal council: in addition legislative provisions enabled employment terms and conditions to be safeguarded, such as Article 2012 of the Italian Civil Code which guarantees the same contractual conditions in transfers between organizations. This provi- sion has since been incorporated into law 428/1990 which fulfilled ob- ligations arising from EU legislation including Directive 77/187/CEE relating to the safeguarding of employees’ rights in the event of transfers of undertakings. At the same time, severe public spending constraints for personnel and ‘headcount pruning’ cut staff from 1200 to around 700 between 2006 and 2014, hence heightening demands from the local gov- ernment to internally reallocate staff. Moreover the municipality always included social clauses in its calls for tenders: though not a compulsory requirement, bidders were encouraged to take on council workers, as this led to them being awarded extra points when the tenders were as- sessed (Head of procurement, 8 August 2014).

Outsourcing quickly led to the creation of a two-tier workforce, as in the case of the catering service. While the seconded cooks with their perma- nent public-sector employment contracts remained covered by the pub- lic sector collective agreement, the transferred staff, now covered by the private sector agreement, experienced a deterioration in their employ- ment conditions: the subcontractor not only drastically cut back work- ing hours and wages, but also dismissed a major slice of the workforce in summertime when schools were closed, only to rehire them in Sep- tember. Such job reconfiguration inevitably led to a higher workload for permanent public staff, on duty for the whole year. Other benefits and allowances such as maternity leave and sick pay remained basically the same since the collective agreement for workers in the tourist industry, the one applying to catering staff, explicitly protected employment terms and conditions on transfers between undertakings (Collective agreement for workers in the tourist industry 2010-13, Article 97).

However the case of the social workers probably represents the most em- blematic example of the two-tier workforce. The outsourcing of social services was driven by the need to replace retiring public sector staff, as succinctly put by the head of social services: ‘outsourcing here has to be seen as an alternative to the use of a temporary agency’ (15 October 2014). This inevitably led to a perverse situation where workers with the same

educational background performed exactly the same task and worked side by side with each other, yet were subject to great inequalities in con- tractual arrangements depending on the nature of their employer. The collective agreement for social cooperatives which applied to external workers provided for lower wages – up to €250-280 less a month – and maternity leave pay corresponding to 30% of wages, against the 100% for the first months stipulated in the public sector collective agreement. With regard to union rights, external workers were a lot worse off: ‘fewer hours for assembly, less leave for union work’ (CGIL union official, 21 July 2014). However this mixed management enabled the municipality to gain flexibility in terms of both labour costs and work organization (Head of social services, 15/10/2014).

3.2                     The English municipality: in the wake of Thatcherism

Such emotive bumper stickers as ‘government is too big’ or ‘we need to cut red tape’ (Asher 1987) well express the popularity of outsourc- ing in England as an alternative way of providing public services since the 1980s. The English municipality studied was subject to a series of decisive legislative moves which triggered the adoption of outsourcing through the regulatory regime of Compulsory Competitive Tendering (CCT): starting with the Local Government Planning and Land Act in 1980, which introduced CCT for a limited range of jobs, mainly blue- collar ones such as highway maintenance or building maintenance, CCT was extended in 1988 to cover almost all manual services, including re- fuse collection, street cleaning, school catering, ground and park main- tenance. As a third step, the tasks covered by the Local Government Act were extended in 1992 to include white-collar professional, financial and technical services. In addition to legislative pressure to contract work out, the sector was severely hit by budget cuts as well, with municipali- ties required to cut their spending by 27% in real terms between 2010-11 and 2014-15 (Bach and Stroleny 2014).

The case-study municipality first tested the market in 1979 for street cleaning, followed by refuse collection in 1982 and a wide range of blue-collar technical tasks during the rest of the decade: park and ground maintenance, vehicle maintenance, building cleaning, school catering and meals-on-wheels. From the 1990s onwards, the munici- pality started putting white-collar professional services out to tender. Beginning with IT and legal services, it moved on to urban planning

and engineering tasks, payroll management and home care for the el- derly.

When the municipality first embarked on competitive tendering for street cleaning, it asked all bidders to consider taking on council workers who were to be made redundant, while offering generous early retire- ment or severance terms to those workers not taken on by the contrac- tor, or redeployment within the municipality. This ‘soft’ approach turned out to be a once-only offer, with all subsequent outsourcing contracts featuring large sections of internal staff being made redundant. Moreo- ver, severance terms became increasingly less attractive, rapidly equal- ling the minimum national standard.

Employment terms and conditions represented a big issue. Key inter- viewees agree with splitting the historical trend into two parts, whereby the watershed is represented by the municipality’s implementation of the Transfer of Undertakings (Protection of Employment) Regulations (TUPE) in 1993.2 Before TUPE, workers transferred to external provid- ers did not enjoy any kind of protection of their employment terms and conditions. As a consequence, there was a general erosion of pay scales and benefits: ‘workers directly employed by the council were paid more for the same skills compared with the private companies’ (GMB union official, 22 October 2013).

The overall picture is dominated by longer and more flexible working hours, spread over 7 days instead of 5. Street cleaners lost their bonuses and allowances for work at night or weekends. Nurses providing home care were not paid for travel time between one patient and the next. Moreover, transferred staff were expected to be much more mobile: gar- deners who had previously maintained green spaces within the munici- pality were now required to cover 5-6 parks in different localities, travel- ling the whole day. Nurses, as well, had to cover several municipalities.

The application of TUPE from 1993 onwards changed the situation in the early stages, protecting employees’ wages and working conditions after the transfer. The most common complaints that workers reported after TUPE implementation concerned insecurity of work rather than

2. The TUPE regulation was enacted in 1981, implementing the European Community Acquired Rights Directive 77/187/EC. But it was only effectively implemented at a later stage in the public

wage insecurity: ‘to put it in a nutshell: jobs are more uncomfortable, there is greater inconvenience for the worker, less security in terms of what he or she would be expected to do’ (UNISON union official, 22 No- vember 2013). This led to higher staff turnover, above all among older workers, and a growth in cases of burnout. Private providers, as a way to circumvent TUPE constraints, often opted for hiring new low-paid workers rather than employing council staff under the same terms and conditions.

Pension schemes presented another thorny question since they were not covered by TUPE legislation. This resulted in most of the private pro- viders slashing pension contributions paid to workers. The issue was so severe and widespread that legislative intervention from central govern- ment became necessary. In June 1999 HM Treasury issued its Statement of Practice on Staff Transfers from Central Government ‘A Fair Deal for Staff Pensions’, providing guidance for staff transfers from the public sector to an external contractor: ‘the new employer offers transferring staff membership of a pension scheme which though not identical is broadly comparable to the public service pension scheme which they are leaving’ (A Fair Deal for Staff Pensions 1999). Despite this guidance, a number of problems emerged. The Fair Deal framework did not for in- stance apply to staff transfer from one private provider to another when contracts were re-tendered, as happened in several cases in the munici- pality in question. In 2013, HM Treasury revised its Fair Deal for Staff Pensions policy, establishing that staff belonging to a public adminis- tration pension scheme and compulsorily transferred out to a private company remain eligible for the public pension scheme. However, the municipality only admits workers transferred after 2008 into its pension scheme, denying access to most of staff previously transferred.

3.3                      The Danish municipality: a strategic response to administrative restructuring

Since the 1990s, the Danish public sector has been subject to a wave of reforms, introducing market-type mechanisms and outsourcing into the provision of services in a gradual and incremental manner (Petersen and Hjelmar 2013). Despite the more limited scope of the trend compared to the Anglo-Saxon tradition (Pollitt and Bouckaert 2011), the issue has triggered a heated and persisting debate at both political and academic levels (Greve 2001).

In 2007 the Danish government enacted a major structural reform, revo- lutionizing the structure of the decentralized levels of government: 273 municipalities were amalgamated into 99 and 14 counties were liqui- dated and replaced by 5 regions with a narrow range of responsibilities. The aim of the reform was to lever economies of scale and improve wel- fare services by reshuffling competences and functions between levels of government. Municipalities ended up as ‘net-gainers of areas of respon- sibility’ (Mailand 2014), taking on responsibility for welfare services for instance. Outsourcing was exploited as a strategic response to the in- creasing range of services the municipality had to provide with fewer resources to a larger population, and as a way of rationalizing the use of financial and human resources.

The municipality studied was the result of a merge of two municipalities following the 2007 administrative reform. One of the two had started restructuring service provision in 1996, outsourcing home care. In the subsequent period, cleaning, waste collection and roads maintenance were put out to tender. After 2007, contracting out was extended to the second municipality, including job centres in 2009 (re-insourced after 2 years) and park maintenance in 2010.

The impacts on working conditions show some common traits across all services outsourced. First the large majority of public staff involved was transferred to the private subcontractors, following the service. A second clear-cut consequence was the increase in workload. This led for instance to home care nurses having to visit more patients in the same amount of time; cleaners having to work faster in schools; street cleaners being re- quired by contract to cover wider surfaces in the same amount of hours. Work overload constituted one of the main challenges for union officials, who reported being much more concerned about workloads than wages. The latter indeed remained almost the same, protected by the Transfer of Undertaking (Protection of Employment) legislation – ensuring the maintenance of working conditions on staff transfers from public em- ployment to private companies – and collective agreements signed by most of the private companies (in the case study only one small com- pany in the nursing sector refused to negotiate). Moreover invitations to tender always contained social clauses requiring bidders to take on the public staff concerned under the same contractual conditions.

Though wages remained stable, other benefits were reduced, including maternity leave, sick- pay and pension contributions. Another repercus-

sion was the change in contractual working hours: nurses shifted from the traditional 37 to 25-30, with a subsequent drastic reduction in pay; cleaners conversely reached the cap of 40-47 hours, in many cases work- ing too much in a single week. Nevertheless, union officials pointed out that their members, in particular low-skilled workers, very often pre- ferred private sector arrangements, as these allowed them to boost their wages through overtime (cleaners) or through working on a piecework basis (road maintenance workers); moreover whereas they had felt ex- cluded in the public sector, they represented the core workforce of a subcontractor: ‘in the private company they constitute the company, in the public they are just a number’ (FOA union official, 15 April 2014). Moreover, in labour-intensive activities such as home care, the private companies (unlike the municipality) provided nurses with new technolo- gies helping them to work faster.

No redundancies were initially reported in the municipality, but in the case of the cleaning tender, once the collective agreement expired, the subcon- tractor fired half of the cleaners transferred (40 out of 80 employees). Furthermore, there were quite a few examples of workers leaving due to de- teriorating working conditions (such as workers in road maintenance), work overload and burnout (in park maintenance). Finally many key in- terviewees witnessed a widespread feeling of insecurity and instability.

Coping with these issues represented a challenging task for unions, deeply concerned with the sharp increase in workloads and the insecuri- ty of employment which often follow outsourcing processes. The overall picture, however, depicts a collaborative relationship between unions, workers’ workplace representatives, municipal managers and KL, the association representing all Danish municipalities. Key interviewees on both sides reported positive experiences, taking advantage of the long- standing social dialogue tradition which allowed conflict to be kept at the lowest level possible (strikes were in fact very rare).

4.    Lowest common denominator? A comparative perspective

The previous section discussed the consequences for public employ- ment following outsourcing processes within three municipalities in three countries. As ascertained, despite some convergent consequences for labour, working conditions developed in different ways based on

the broader institutional context in which restructuring was embedded: specifically, industrial relations regimes and institutions to which each country is bound, complemented by national regulation of public em- ployment, played a major role in shaping and moderating the negative effects of externalization on pay and working conditions. In addition, the EU Transfer of Undertakings Protection of Employment (TUPE) legisla- tion acted as a cushion, upholding public employees’ working conditions during restructuring.

TUPE regulation, indeed, played a very influential role above all in the UK where public employees not only do not enjoy strong protection from the national legislative framework, but even more importantly where a market employment relations regime prioritizes market adjustments and self-regulation, hence minimizing the role of employment regula- tion and organized labour institutions (Gallie 2007) and inevitably lead- ing to a widespread deterioration of employment conditions.

Conversely in Italy and Denmark the protections guaranteed by TUPE legislation are strengthened by additional national institutional arrange- ments (Doellgast and Gospel 2012). In Italy, reflecting the Mediterra- nean employment model, the state has traditionally played a relevant coordinating role (Gallie 2007), defining a set of strong legislative pro- tections both regarding public employment in general and specifically the transfer of workers as a result of outsourcing (Italian Civil Code art. 2012; Law 428/1990, Art. 40). Denmark’s consolidated tradition of so- cial dialogue and collective bargaining, associated with an inclusive em- ployment relations regime where regulations and policies are designed to extend employment and employment rights as largely as possible and to provide welfare and security to all workers while minimizing differ- ences within the labour market, has acted as a buffer for labour issues, alleviating perverse outcomes (Petersen et al. 2012; Jaehrling 2014). Though the encompassing industrial relations institutions in force in It- aly and Denmark were unable to prevent the outsourcing of a wide range of public services, they obviously played a crucial role in protecting em- ployment terms and conditions from erosion.

Beyond such differences in the extent to which externalization nega- tively impacted public employment, certain similarities were detected across the three municipalities, as summarized in Table 1. Outsourcing triggered the increased transfer of public employees across public-pri- vate organizational boundaries, functionally or even contractually, as re-

Table 1 Implications for labour following outsourcing in comparative perspective

ported resulting in a growing fragmentation of employment conditions and a deterioration of work, since employment has generally become more insecure and flexible in terms of both working hours and work- place flexibility. Greater demands are placed on work tasks, including increasing workload, a higher pace of work and the use of piecework. Additionally, employment terms and conditions established by private providers went downhill, not so much in terms of wages but with sharp cuts to pension schemes, maternity leave, sick pay and holidays.

In a nutshell, restructuring public administration through outsourcing has been exploited as a way to circumvent public sector collective agree- ments – offering greater benefits than private sector agreements – and to deregulate job protection, notably higher in the public sector.

While the main rationales underlying the decision to outsource differ across the three public administrations, the will to open up public ser- vices to market discipline based primarily on labour cost competition was rather straightforward, as the evidence suggests. Dependent on what national regulation allowed, public personnel was either directly transferred to external subcontractors as a way to reduce staff costs or seconded to private companies, thereby boosting flexibility in human resources management. Interestingly, a significant rise in workloads emerged as a consequence common to all the municipalities studied: unlike other institutional dimensions, this labour process-related aspect

remains unprotected by legislation, enabling private companies to lever- age as a way of gaining competitiveness and cutting labour costs.

As already anticipated, the consequences for public employment display diversified patterns reflecting national variations with regard primarily to the scope of impact and the extent to which staffing levels were in- volved.

In the Italian municipality, employment conditions only deteriorated for temporary workers transferred to external private companies. These only constituted a small share of total headcount in contrast to the share of permanent public employees redeployed within the administration under the same contractual conditions or, where feasible, seconded to the external provider. In the latter case, the employees enjoyed the same employment terms and conditions established in the public sector col- lective agreement, while being managed by the private employer. How- ever in many cases the shift in management led to increased workloads, a faster pace of work and increased job pressure. Nevertheless, redun- dancies were completely avoided, safeguarding employment for the bulk of workers involved.

In Denmark personnel transfer has largely involved public employees following outsourcing: this has meant a collective agreement switch from a public to a private sector one, which, though not so different in terms of wages, is more disadvantageous with regard to other benefits and allowances (maternity leave, holidays, pension scheme, sick pay). In addition it has led to a sharp increase in workload, a faster pace of work and growing flexibility in working hours and working places. Even if only a few cases of dismissal were reported following the transfer to the pri- vate provider, the deterioration in working conditions was noteworthy and involved a large group of public employees.

Evidence from the English municipality paints the most extreme pic- ture: outsourcing was adopted on purpose as a tool for downsizing pub- lic sector employment, in most cases with public service provision being transferred to private providers and with public workers soon dismissed, only to be replaced by cheaper workers directly employed by the com- pany. Those who kept their job in the private firm experienced a sharp deterioration of working conditions in terms of both wages and other allowances. Moreover, unlike Danish transferred workers who switched from public to private sector collective agreements, in the UK subcon-

tractors often refused to signed collective agreements, unilaterally es- tablishing detrimental employment terms and conditions for trans- ferred workers, at least before TUPE was implemented. Finally, pension schemes were subjected to major cuts.

In the light of this evidence, distinctive national trajectories are devel- oping, dependent on European and national legislative frameworks and employment relations institutional configurations (Flecker 2009; Vraengbæk et al. 2013).

5.    Conclusions

Restructuring public administration through outsourcing has become widespread in Europe as a way of cutting public expenditure and boost- ing value for money in the provision of public services. Empirical evi- dence from the three municipalities studied however sheds light on the actual source of savings, stemming from competition based on labour costs: outsourcing in fact has led to a fragmentation of working condi- tions within local government and a deterioration of employment terms and conditions for public personnel.

The analysis points to how encompassing employment relations insti- tutions, complemented by the strong protection of public employment guaranteed by national legislation, have played a noteworthy role in sheltering public employment, shaping and moderating the negative consequences of outsourcing (Jaehrling et al. 2014). In addition, the TUPE legislation is acting as a cushion, helping to maintain working conditions during restructuring, above all in those contexts where other forms of labour protection are lacking.

Understanding the significant function of such labour institutions is crucial for preventing marketization processes with a concealed politi- cal intention to promote social dumping, with detrimental repercussions on the weakest link of the value chain, namely on workers. As evidence suggests, outsourcing choices are often purposely adopted to circum- vent public sector collective agreements and regulation, allowing labour costs to be cut via downward pressure on pay and working conditions, often via a switch from public sector agreements to private sector ones. Meanwhile, the widespread importance accorded to the TUPE regula- tion needs to be exploited as a strategy within the union toolbox. While

not preventing restructuring, it has the potential to at least mitigate the erosion of working conditions, helping to bring up job value.

References

Alonso J.M., Clifton J. and Díaz-Fuentes D. (2015) Did new public management matter? An empirical analysis of the outsourcing and decentralization effects on public sector size, Public Management Review, 17 (5), 643-660.

Bach S. and Stroleny A. (2013) Public service employment restructuring in the crisis in the UK and Ireland: social partnership in retreat, European Journal of Industrial Relations, 19 (4), 341-357.

Bach S. and Stroleny A. (2014) Restructuring UK local government employment relations, pay determination and employee participation in tough times, Trans- fer: European Review of Labour and Research, 20 (3), 343-356.

Blöndal J.R. (2005) Market-type mechanisms and the provision of public services, OECD Journal on Budgeting, 5 (1), 79-106.

Bordogna L. and Neri S. (2014) Austerity policies, social dialogue and public services in Italian local government, Transfer: European Review of Labour and Research, 20 (3), 357-371.

Bordogna L. and Pedersini R. (2013) Economic crisis and the politics of public service employment relations in Italy and France, European Journal of Indus- trial Relations, 19 (4), 325-340.

Doellgast V. and Gospel H. (2012) Outsourcing and human resource manage- ment, in Bach S. and Edwards M. (eds) Managing human resources: human resource management in transition, 5th ed., Chichester, Wiley, 317-335.

Flecker J. (2009) Outsourcing, spatial relocation and the fragmentation of em- ployment, Competition & Change, 13 (3), 251-266.

Flecker J. (2010) Fragmenting labour: organisational restructuring, employment relations and the dynamics of national regulatory frameworks, Work organisa- tion, Labour & Globalization, 4 (1), 8-23.

Flecker J. and Hermann C. (2011) The liberalization of public services, company reactions and consequences for employment and working conditions, Econom- ic and Industrial Democracy, 32 (3), 523-544.

Flecker J. and Meil P (2010) Organisational restructuring and emerging service value chains, implications for work and employment, Work, Employment & Society, 24 (4), 680-698.

Flecker J. and Thörnqvist C. (2012) Outsourcing, tendering and changing working conditions in local public transport, in Hermann C. and Flecker J. (eds) Privati- zation of public services: impacts for employment, working conditions, and service quality in Europe, New York, Routledge, 74-88.

Flecker J., Holtgrewe U., Schönauer A., Dunkel W. and Meil P. (2008) Restructur- ing across value chains and changes in work and employment: case study evidence from the clothing, food, IT and public sector, WORKS Project Report, Leuven, HIVA. http://www.forba.at/data/downloads/file/362-Case_Study_ Report_D10.1.pdf

Gallie D. (ed.) (2007) Employment regimes and the quality of work, Oxford, Oxford University Press.

Greve C. (2001) New avenues for contracting out and implications for a theoreti- cal framework, Public Performance & Management Review, 24 (3), 270-284.

Hermann C. and Flecker J. (eds) (2012) Privatization of public services: impacts for employment, working conditions, and service quality in Europe, New York, Routledge.

Hood C. (1991) A public management for all seasons?, Public Administration, 69 (1), 3-19.

Jaehrling K. (2014) The state as a ‘socially responsible customer’? Public procure- ment between market-making and market-embedding, European Journal of Industrial Relations, 1-16. doi: 10.1177/0959680114535316

Jensen P.H. and Stonecash R.E. (2005) Incentives and the efficiency of public sector-outsourcing contracts, Journal of Economic Surveys, 19 (5), 767-787.

Mailand M. (2014) Austerity measures and municipalities, the case of Denmark, Transfer: European Review of Labour and Research, 20 (3), 417-430.

Mailand M. and Hansen N.W. (2013) Public service employment relations in an era of austerity: the case of Denmark, European Journal of Industrial Rela- tions, 19 (4), 375-389.

OECD (1993) Managing with market type mechanisms, Paris, Organisation for Economic Co-operation and Development.

OECD (2011) Government at a glance 2011, Paris, Organisation for Economic Co-operation and Development.

Petersen O.H., Hjelmar U., Vrangbaek K. and la Cour L. (2012) Effects of con- tracting out public sector tasks: a research-based review of Danish and international studies from 2000-2011, Copenhagen, AKF, Danish Institute of Governmental Research. http://www.epsu.org/IMG/pdf/5111_udlicitering- srapport_EN_FINAL_VERSION.pdf

Petersen O.H. and Hjelmar U. (2013) Marketization of welfare services in Scan- dinavia: a review of Swedish and Danish experiences, Scandinavian Journal of Public Administration, 17 (4), 3-20.

Pollitt C. (2001) Clarifying convergence: striking similarities and durable dif- ferences in public management reform, Public Management Review, 3 (4), 471-492.

Pollitt C. and Bouckaert G. (2011) Public management reform: a comparative analysis – New public management, governance, and the neo-Weberian State,

Oxford, Oxford University Press.

Quiggin J. (2002) Contracting out, promise and performance, Economic and Labour Relations Review, 13 (1), 88-104.

Roberts V. (2001) Managing strategic outsourcing in the healthcare industry, Journal of Healthcare Management, 46 (4), 239-249.

Savas E.S. (2000) Privatization and public-private partnerships, New York, Chatham House.

Schulten T. and Brandt T. (2012) Privatisation and the impact on labour relations, in Hermann C. and Flecker J (eds) Privatization of public services: impacts for employment, working conditions, and service quality in Europe, New York, Routledge, 136-152.

Vrangbæk K., Petersen O.H. and Hjelmar U. (2013) Is contracting out good or bad for employees? a review of international experience, Review of Public personnel Administration, 20 (10), 1-21.

Chapter 7

Subcontracting, insecurity and posted work: evidence from construction, meat processing and ship building1

Nathan Lillie and Ines Wagner

1.    Subcontracting and posted work

An explicit goal of European integration is to remove barriers to the free movement of labour. European Union citizens have broad rights to live and work in other EU member states, and, in a legal sense, to be treated as equals when abroad. Many Europeans, particularly those in the less prosperous new member states have taken advantage of labour mobil- ity. Alongside this migration of individuals from East to West, however, another migration is taking place: that of workers ‘posted’ as dependent employees of transnational subcontractors or subsidiaries. These work- ers, unlike those who migrate as individuals, are treated in part under the standards of the country they are sent from, rather than those of the country they are posted to. Instead of coming under the EU regulatory regime governing the free movement of labour, these workers are legally treated under the free movement of services. Providing agency labour or subcontracting on a construction or shipbuilding project, or in a meat packing plant, for example, constitutes a ‘service’ under EU law. Because of the centrality of the free movement of services under EU constitu- tional law, EU rules limit what governments and trade unions can do to regulate transnational service providers (Lillie 2010). It is now common to subcontract to transnational service providers in this way specifically because firms can contest many aspects of host country employment regulations, thus ‘arbitraging’ between regulatory environments.

Regulatory arbitrage is defined as strategizing over the regulatory treat- ment of a transaction in selecting a regulatory regime from among two (or more) alternatives (Fleischer 2010: 4). In practice, this can mean physically moving the transactions to a different territory by opening up

1. The insights in this chapter are based on the findings of the Transnational Work and the Evolution of Sovereignty (TWES 263782) project.

an office or factory there, or incorporating a new company in another territory with a view to transacting business (such as managing employ- ment contracts) under the law of that territory, even if the actual activity takes place elsewhere. It can also mean moving between different forms of regulation in the same geographic space, such as between labour laws governing temporary agency work or between metalworking and con- struction collective agreements. The issue of posted work is therefore closely linked to mechanisms of subcontracting and agency work, and to the strategies of employers for ‘regime shopping’. The precise legal and organizational formulations used and the reasons for them differ from one context to another, but behind these superficial differences and complexities are a set of very similar employer strategies and motiva- tions all based on regulatory arbitrage.

Regulatory gaps emerge in the transnational regulation of employment for posted workers within the EU. These gaps are both in the regulations themselves – i.e. there are regulatory loopholes created by European in- stitutions keen to foster competition in wages and social standards2 – as well as in the fact that the enforcement of any labour standards at all is inherently difficult in the context of complex EU regulatory interactions with national systems. Company practices interact with this complex and variegated regulatory environment, systematically seeking cheaper op- tions for employing workers. This means that practices like ‘worker post- ing’ exist both to fill a real need for companies sending workers abroad to perform a particular service, and as a form of regulatory avoidance. Posting is most prevalent in industries such as construction and ship- building where subcontracting practices are highly developed because there are already organizational processes in place to facilitate multi- employer worksites, but it also emerges in other types of workplace of- fering opportunities for regulatory arbitrage. Subcontracting combined with posting allows employers to access alternative regulatory regimes in a legal sense, as well as in an employment relations sense, in that company boundaries and practices, as well as the expectations of the subcontractors’ workers relate back to their home country. Thus, posting shapes and is shaped by the interaction of EU, home and host country regulatory environments, the industry-specific environment, and an in- dustry’s production arrangements.

2   This intentionality is made clear in the infamous Laval decision (Case C-341/05), discussed later herein.

The aim of this chapter is to explore the diverse ways in which employ- ers use transnational contractual arrangements to construct precarious employment in the EU. We explore the way in which certain kinds of contracting result from firms taking advantage of their ability to arbi- trage between European regulatory frameworks, exploiting regulatory gaps and avoiding regulatory enforcement. Following a discussion on the European labour mobility policy we examine the contractual ar- rangements and labour market regulations in three respective industries and countries: 1) the construction sector in Germany, the Netherlands, Finland and the UK; 2) the German meat industry, and 3) shipbuilding in Finland. We look at construction in the EU from a broad perspective, as it is well-researched in many EU countries and we can draw some general lessons and comparisons from this research. As the meat indus- try and shipbuilding have been less researched, we look closely at how cases have played out in the specific contexts of Germany and Finland. Within these case studies we discuss the use of contingent labour as part of a new strategic approach to flexibility and examine the peculiarities associated with the use of respective contractual relations in each case.

2.    European labour mobility policy

The EU politics of labour mobility establishes a rights regime for work- ers migrating as individuals and a separate regulatory strand for workers posted by their employers. This fact is an important driver for transna- tional subcontracting, since it creates differences between workers who are sent by firms to work abroad, and workers who move as individuals. Posted work falls under the free provision of services rather than the free mobility of labour, which is important because posting is subject to a different set of social and labour regulations than individual labour mobility (Dølvik and Visser 2010). Posted worker contracts must comply with the Posting of Workers Directive (PWD), which allows host states to apply their own standards in specific areas when these are higher than the sending state’s.

The Directive specifies the categories of employment conditions it regu- lates:

1. Maximum work periods and minimum rest periods;
2. Minimum paid annual holidays;
3. The minimum rates of pay, including overtime rates; this point

does not apply to supplementary occupational retirement pension schemes;

4. The conditions of hiring-out of workers, in particular the supply of workers by temporary employment undertakings;
5. Health, safety and hygiene at work;
6. Protective measures with regard to the terms and conditions of em- ployment of pregnant women or women who have recently given birth, of children and of young people; and
7. Equality of treatment between men and women and other provi- sions on non-discrimination (Directive 96/71/EC, Article 3.1).

While unions originally regarded the Posted Workers Directive as a minimal set of protections, designed to enable action by regulators to improve conditions, this collided with the ECJ’s interpretation of the Directive’s intentions, which follows more of a market-opening line of thought (Arnholz 2014). In its Laval decision (ECJ C-341/05), the ECJ decided that host country regulators – including private players such as trade unions – cannot oblige foreign service providers to respect la- bour rights that go beyond the minimum standards of the PWD be- cause these might constrain a firm’s competitive advantage, insofar as that advantage is based on national wage norms from a cheaper sending country. This formulation makes labour rights a competition parameter, undermining the ability of trade unions to bargain for pay and condi- tions above the minimum set by legislation. ‘Proportionate’ action is still allowed, as the ECJ’s Viking decision, issued together with Laval, estab- lishes that unions have the right to take action limiting firms’ freedom of movement as long as that action is ‘proportionate’ to the objectives to be achieved (ECJ C-438/05). The combination of Laval and Viking pro- duced a situation where unions are restricted from conducting free col- lective bargaining on behalf of posted workers, because strikes could be ruled as disproportionate, exposing unions to fines. On the other hand, these decisions did establish that unions have the right to conduct ‘pro- portionate’ action to enforce legal minimum wages, which, as the ECJ’s recent Sähköalojen ammattiliitto ry v Elektrobudowa Spolka Akcjjna decision confirms, includes enforcement of legally extended collective agreements, which may include inter alia wage rates which are higher than minimum depending on skill levels, as well as vacation and over- time payments (C-396/13). Although it had seemed clear from Laval that unions would be able to defend the pay rates in extended collec- tive agreements, the ECJ’s Advocate General expressed in an Opinion in 2014 that this should just apply in respect to the lowest wage brackets in

those agreements, so the result of the Elektrobudowa was a favourable surprise for the unions.

While posted workers are only entitled to equal treatment with regard to the PWD, workers who move as individuals have the same labour rights as workers from the host country state. The legal constraints aris- ing from the transnational provision of services only apply to posted workers from transnational subcontractors, or transferred within mul- tinational companies. Employers use transnational subcontractors and work agencies to be able to choose between the labour rights and social protection regimes of the host and sending countries, thereby enabling them to recruit migrant workers from locations where wage expectations are low and who are employed under the (partly) home-country (low) wage terms. However, although the Posted Workers Directive does not cover all conditions, it does cover the most important ones. The terms which can be applied to posted workers on the basis of extended col- lective agreements and/or minimum wages are often so close to what it would cost to employ native workers that there is not always any great la- bour-cost difference, although this varies according to local conditions. Nevertheless, differences in social insurance costs, which have to be paid in the home instead of the host country, can reduce overall labour costs. Moreover, a big problem is the way in which enforcement of local rules is hindered by the complicated mix of home and host country legal stand- ards, by the ability of transnational subcontractors to shift between ju- risdictions to avoid compliance, and by the unwillingness of most posted workers to confront their employers with their legal rights (Berntsen and Lillie 2015; Cremers 2013; Wagner and Lillie 2014).

The ECJ’s approach has generated a great deal of concern from a labour rights perspective because it enables companies to opt out of host coun- try labour conditions and collective agreements, if they are willing and able to use transnational subcontractors, and because it imposes limits on unions’ right to strike and on the enforcement options of labour in- spectorates in cases where intra-EU mobility is involved. In an attempt to create a firmer legal basis for enforcing existing labour laws, the Eu- ropean Commission initiated policy negotiations around an Enforce- ment Directive for the Posting of Workers Directive. The Enforcement Directive was a reaction to complaints from trade unions and from a Commission-financed report showing that the labour rights of posted workers were routinely violated (Houwerzijl and van Hoek 2011). It was not the first attempt to find a solution to the post-Laval regulatory

environment: an earlier regulation foundered when its content became so badly compromised that the unions decided to abandon the attempt (Arnholz 2014). The Enforcement Directive encountered similar difficul- ties, with unions struggling to have wording included to protect national labour inspection practices and chain liability provisions from possible future ECJ decisions. For example, in countries and industries where subcontracting chain liability exists, contractors can be held liable for non compliance with the labour law and social security contributions of their subcontractors when the latter fail to fulfil their obligations. Chain liability provisions vary from country to country, but in the German con- struction industry, it is possible to extend liability to all subcontractors in the construction chain. This ensures that workers can collect back-pay and/or social security contributions when firms go bankrupt and disap- pear. It also obliges main contractors to consider the reputations of their subcontractors and institute policies for checking up on them. While trade unions were pushing for a similar regulation in the Enforcement Directive, the text only specified, in Article 12 of the Enforcement Direc- tive in which chain liability is discussed, direct (first tier) subcontractor liability.

The content of the Enforcement Directive as finally adopted hardly goes beyond the codification of existing national regulatory measures, in some cases allowing scope for the ECJ to limit them in future. For example, the Directive contains frequent references to the need to limit enforcement actions to those which are ‘justified’ and ‘proportionate’. This reveals a deep concern for preventing national labour rights enforcement from interfering with the free movement rights of firms, leaving open the pos- sibility that member states could be held liable should they undertake ‘disproportionate’ enforcement action, whatever that may mean. The ac- tual outcome is likely only to become apparent when the ECJ interprets the wording of the directive as to whether specific union and government efforts to enforce labour standards violate the free movement rights of firms.

3.    Contractual arrangements and the way the labour market works

European Union rules on labour mobility and transnational service pro- vision interact with national labour market rules, while industry prac- tices of subcontracting and agency work shape the posted worker labour

market. Thus, posted worker employment systems and the European Union regulations governing them must be understood in the context of labour markets shaped by boundaries and hierarchies created by con- tracting chain power relations. The use of agency workers and subcon- tractors is not a given and constant phenomenon, but rather is driven by the opportunities and strategies of firms. Firms use contracting to access expertise which they do not have internally, to externalize risk to less powerful firms (Deakin and Walsh 1996), to fragment employment rela- tions, complicating and weakening worker representation (Wills 2009), and to place workers in peripheral operations under less protected regu- latory systems (Doellgast and Greer 2007).

Harvey (2003: 195-196) divides subcontracting into ‘cooperative’ and ‘competitive’. The cooperative model is compatible with an organization based on professional qualifications, good working conditions (pay, etc.) and the application of specialized skills. Competitive subcontracting, by contrast, is based on lowering costs through leveraging the differences in the market power of the agents participating in the process. A common practice in this model is the successive subcontracting of the same task, typical of the ‘hollowed-out firm’ (Harvey 2003: 197).

Posting as we encountered it in our interviews almost always goes hand in hand with the second form. In this application, subcontracting be- comes a cost-saving way of exploiting differentials between countries, sectors and workplaces and of increasing/decreasing production in line with demand (Flecker et al. 2008). Workers at subcontractors often ex- perience poorer working conditions, such as lower wages, higher work intensity, increased job insecurity and a higher reliance on non-standard employment contracts (Grimshaw and Rubery 2005; Wills 2009; Fleck- er and Meil 2010; Gautié and Schmitt 2010). They are often not repre- sented by works councils and are often pressured by management not to talk to unions or other forms of worker representation. This undermines union representation and drives the creation of two-tier labour markets (Wills 2009). Thus, this form of inter-capitalist relations has distribu- tive consequences, affecting the wage-effort bargain within a productive process, and the ability of workers to take individual or collective action to redefine this bargain (Grimshaw and Rubery 2005; Wills 2009).

The PWD legal framework encourages transnational subcontracting by ‘immunizing’ transnational subcontractors from local enforcement ef- forts. The fact that posting invokes foreign conditions and regulatory

frameworks is not always as important as the way the way employers use it as a cloak to prevent monitoring and enforcement. Through posting, employers make it plausible that they are complying with the rules, even if they are not, thus making it complex and labour intensive to check. In the legal definition an employer must be ‘established’(i.e. actually carry- ing out business) in the sending country in order to send workers from there, and the workers must be ‘habitually’ employed in the countries they are sent from (Houwerzijl and van Hoek 2011). We found that these conditions are often not observed. Employers sometimes hire migrants locally and label them as posted workers, and they may employ them un- der contracts from countries where neither employer nor worker has any real connection. Nevertheless, we are more concerned with the character of the employment relationship this implies, rather than whether a par- ticular worker fits the strict legal definition. The actual contractual rela- tionship of posted workers is often vaguely defined, and only becomes specific when host country regulators subject it to close examination – for example in cases where a lawsuit is pursued.

Posted workers exist in a diverse pan-European labour market of mobile workers employed under various kinds of similar arrangements. The com- mon thread is that employers look for contractual constructions which allow them to access desirable regulatory frameworks – while posting is a common framework there are other arrangements as well. In some countries (for example in the Netherlands, Germany or the United King- dom) the legal regulation of self-employment is such that employers are able to avoid collective employment regulation by classifying workers as self-employed. In these cases, it is common to encounter nominally self- employed migrant workers who are de facto dependent posted workers. Occasionally, we found workers posted in an organizational sense but not a legal one. As with posted workers, these workers were brought by an employer to work on specific projects and had their accommodation and travel arranged as if they were posted workers, but had local work contracts and social security. This practice has become increasingly common, for example in Finland, because unions have been fairly suc- cessful in using legal and industrial action in recent years to ensure that employers adhere to their legal obligations under the Posted Workers Directive (Lillie 2012). When legal standards are complied with, the cost difference between posted workers and native workers is not that large, so de jure posting is only used when there is a legitimate reason for it. Although the details of contractual arrangements are often important in terms of particular enforcement efforts, posting and organizationally

similar forms of migration define a single labour market, with the line between the various categories blurred through ignorance, legal indeter- minacy and management strategy (Berntsen and Lillie 2015).

The posted work phenomenon has emerged from the specific regulatory environment of the European Union, and from contracting practices in certain industries, most notably construction. Moreover, in an increas- ing number of industries, now including among others construction, meat processing, distribution and shipbuilding, hypermobile workers working and living in substantially deregulated social spaces now domi- nate the labour market. In this context, it has become a systematic and large-scale way for employers to (more or less legally) access cheap la- bour and circumvent national labour laws and collective agreements.

4.    Shop floor employment relations/sectors of posted work

4.1                      Case study: the construction sector in Germany, Finland, the Netherlands and the UK

Construction is the industry most affected by posted work, and has been the primary focus of regulatory efforts such as the Posted Work- ers’ Directive. The PWD is implemented differently in the respective EU member states: in some it covers the whole economy, while in others it only covers certain sectors. The high prevalence of posted work in con- struction means that construction subcontracting processes shape and are shaped by the posted work phenomenon. This, however, is not as much of a change as it may seem: high levels of subcontracting and site mobility have long been a normal situation in the industry, making for a volatile labour market even within national boundaries.

As main contractors, large construction companies do not usually post workers themselves but act as site managers on large construction pro- jects. However, they play a vital role in the development of subcontract- ing chains, engaging companies from lower wage countries which post workers to construction sites in order to fulfil a particular construction service. In the four countries examined – Finland, Germany, the Neth- erlands and the UK – large companies (in terms of turnover) function as main contractors or as building service providers while small and medi- um-size companies assume the role of the subcontractors and provide

the majority of workers (Berntsen and Lillie 2015; Lillie 2012; Bosch and Zühlke-Robinet 2003).

We found workers posted both as employees with long-term work re- lationships with particular subcontractors, remaining there after a par- ticular posting (an indicator of a posting being a ‘real’ posting under the meaning of the PWD), and workers hired just for a particular posting. Workers in the first group are to be found working for actual contracting companies specialized in particular types of construction work. Workers in the second group work for work agencies, or ‘labour only’ subcon- tractors which essentially act as work agencies. We found many posted workers in skilled jobs, sometimes with long-term stable jobs with the contractors they work for. Even so, these skilled workers often work for lower wages than local workers would expect for similar jobs. Often these wages are illegally low, although we also encountered workers ap- pearing to be paid in accordance with local collective agreements.

Illegal underpayment is sometimes accomplished through vague em- ployment arrangements, only clarified in the case of an enforcement ac- tion. For example, we found a case like this at the Rauturuukki metal construction materials factory in Raahe, near Oulu in northern Finland. Skilled Polish masons hired to repair a blast furnace were being paid 3-4 euros per hour under Polish contracts. The employer claimed the contractual payment only constituted partial payment and was supple- mented by payment in Finland. Neither the Finnish construction union nor the shop stewards at the factory found this plausible, and conduct- ed a two day walk-out. This resulted in an agreement by the employing company, Beroa, to pay the full Finnish wages set forth in the collec- tive agreement. In most cases, unions either never find out about the underpayment, or are unable to take action to fix it. For example, one Dutch unionist estimated that at the Eemshaven construction site in the northern Netherlands, some 80% of the workers were receiving below- standard pay (Berntsen and Lillie 2015).

Employers will label highly qualified workers ‘unskilled’, assigning them to the lowest pay category while still appearing to comply with the col- lective agreements. This means that to prove that a contractor is in viola- tion, the union or worker must prove not only that they worked a certain number of hours, but also that the work done matched that of a higher skill grade set forth in the legally extended collective bargaining agree- ment. The classification of posted workers into skill brackets higher than

the lowest one stipulated in the collective agreement was challenged by the ECJ in a recent court case. Despite an opinion given by the Advocate General that unions should only be allowed to apply the lowest pay rate in the applicable extended collective agreement to posted workers,3 the ECJ supported the application of skill brackets for the determination of posted worker minimum pay rates. From the union perspective, this is quite fortunate, as the application of extended collective agreement rates to posted workers has been a cornerstone of union strategy for example in the Netherlands (Berntsen and Lillie 2015) and Finland (Lillie 2012).

The most problematic groups are those working for transnational work agencies and labour-only subcontractors active within the lower levels of the contracting chains. There is a grey area in terms of the difference between a temporary agency and a subcontractor. In Germany, where temporary work agencies are, with a few exceptions, not allowed in the construction industry, employers turn instead to labour-only subcon- tractors supplying only labour, recruitment services and payroll man- agement, but not providing supervision or equipment. ‘Real’ subcontrac- tors engage labour from labour-only subcontractors and work agencies on flexibility and cost grounds; these firms supply extra labour when needed, and the labour-only subcontractor or agency workers work un- der the supervision of the ‘real’ subcontractor’s management much as if they were temp workers. Thus, the de facto and de jure employers are different. Usually, these workers are hired for the duration of the posting, managed in small groups, and have tenuous relations with both their de jure and their de facto employers. With these kinds of employ- ment arrangements, being paid below legal rates is the norm, there is very little job security, and social security contributions and insurance might not be paid. Furthermore, these workers will often have an ex- plicit or implicit agreement with their employer to work at illegally low rates and to not cooperate with enforcement efforts. Sometimes, even this agreement is violated by employers. In the infrequent cases where posted workers go on strike or seek assistance from the unions, it is of- ten because the employer has violated this informal agreement with the workers, cheating them of their agreed wages.

3. Opinion of Advocate General Wahl, Case C-396/13, Sähköalojen ammattiliitto ry v Elekrodubowa Spolka Akcjjna 18 Sept 2014.

4.2                         Case study: the meat sector in Germany

Germany is the number one pork producer in Europe (Fleischatlas 2014). Most slaughterhouses in Germany are small or medium-sized, though the smaller companies play only a minor role with regard to the overall total of slaughtered animals. In 2012, 55% of the commercial val- ue of the latter was in the hands of the four biggest slaughtering compa- nies – Danish Crown, Tönnies, Vion and Westfleisch. Since the Eastern enlargement of the EU in May 2004 the large German slaughterhouses have scaled down their core workforces to a minimum (NGG 2013), with posted and temporary workers from Eastern Europe now doing the majority of the meat slaughtering and processing in Germany. A recent works council survey indicates that in some meat processing companies posted workers make up 50-90 % of the factory workers (NGG 2012). Of the 30,000 workers in the slaughtering industry every third one is employed under a subcontracting contract (NGG 2013).

The EU Posting of Workers Directive has been transposed into the Ger- man Posting Law. However, the particularity of the regulatory frame- work for posting workers to Germany is that this law does not apply to the whole economy but only to the sectors listed in it. Until mid-2014 the meat sector was not included. This created a regulatory gap because posted workers in unlisted sectors can work according to the conditions and pay of their home country.

As a consequence meat producers in Germany increasingly subcontract- ed slaughtering and meat packaging operations to Eastern European subcontractors. Low labour costs have been a major driver allowing the German slaughter industry to grow quickly over the last 10 years. The cost advantage in Germany allows slaughterhouses to lower their pro- cessing costs, enabling them to compete for retail and food service cus- tomers by offering lower prices. Companies such as Danish Crown have relocated their production facilities to Germany specifically to employ lower-wage subcontractor labour.

In Germany, posted workers predominantly from Bulgaria and Romania received piece-rates between 1.30-1.60 EUR per slaughtered pig. Apart from the low pay, subcontracted employment relations also divided the workforce, with subcontractors subdividing tasks according to nation- ality. The separate teams were not allowed to talk to each other or ex- change information on wages and working conditions. In dire cases of

work-related accidents posted workers were sent home since their em- ployer had not arranged for accident insurance. Workers were faced with pay deductions for accommodation, travel costs and work gear that were previously not agreed, reducing their pay considerably.

Posted workers accept these working conditions because of their iso- lation from collective channels of representation and because of their temporary status which increases the likeliness of them ‘enduring’ these conditions temporarily. Nevertheless, there are efforts to improve the conditions of posted meat industry workers. After a change in govern- ment, the German national minimum wage of 8.50 € was agreed in De- cember 2013. It took effect in January 2015 and includes a transitional period before becoming mandatory for all employees. In January 2014 the social partners in the meat industry agreed on a sectorial minimum wage starting at 7.75 € with a regular rise up to 8.75 € until December 2016. The sectoral minimum wage was declared as generally binding through the German Posting Law. However, the establishment of the minimum wage is unlikely to reverse the trend to hire subcontractors – as trade unions, government representatives and employers’ organiza- tions agree – because it has become an institutionalized part of the meat industry. Moreover, employers still save on labour costs because social security contributions for posted workers are paid in the home coun- try at a rate usually substantially lower than in the country where they are posted to. However, the new legal situation will improve the terms and conditions for posted workers. While this is certainly a positive de- velopment, experiences from the construction sector show that posted workers hardly use legal channels to enforce their rights. Workers are afraid of being unable to find work if they do so. As collective redress does not exist in Germany, trade unions cannot represent workers in court and therefore workers always have to reveal their identity in ju- dicial proceedings. Experiences from the implementation of minimum wages show that there is often a gap between policy and implementation. In this sense, it remains to be seen to what extent the minimum wage will effectively alter the established subcontracting structures, as well as to what extent it will improve the working conditions of posted workers.

4.3                         Case study: shipbuilding in Finland

Shipbuilding is a highly subcontracted business, with large multina- tional companies owning the yards, but most of the actual employment

taking place at subcontractors. Finnish shipyards are no exception. As in the construction industry, this has made it easy to switch to a less expensive foreign workforce by engaging foreign contractors. Shipbuild- ing in Finland was until recently dominated by STX Finland, a subsidi- ary of the Korean STX, which owned a large shipyard in Turku (yards in Rauma and Helsinki having been recently closed). In late 2014, after difficulty in acquiring large orders since the delivery of three large cruise ships in 2008-2010, STX sold the Turku shipyard to German shipbuilder Meyer Werft.

STX followed a highly subcontracted production model in which its blue-collar workforce was brought in primarily from the Baltic States via subcontractors. Of the workers at STX Finland sites at the end of 2009, about half were Finnish and half non-Finnish, either posted by sub- contractors or as agency workers. Only a small number of non-Finnish workers worked for STX directly, with the majority employed by subcon- tractors or agency firms. During the post-2011 downturn, only a small number of foreign posted workers remained on the site, though numbers increased briefly when Viking Lines put in an order for a large ferry. The sale to Meyer Werft brought with it a substantial order book which is expected to revive the Turku yard. Employment at the yard is forecast to increase substantially in the coming period, perhaps returning to the levels and dynamics of the pre-2010 period.

In 2009 when STX had a full order book, there were around 3,800 workers on site, with a similar number working for on-site subcontrac- tors. Off-site numbers working for STX subcontractors were estimated at around 40,000 workers. Since the end of this boom period, employ- ment levels have been much lower, with numbers between 1,350 and 2,850 mentioned in the press, including both STX and its on-site sub- contractors. Most of the variation in employment levels is due to in- creases and decreases in subcontractor and temporary workers. Dur- ing the boom period, the Metalworkers Union, the labour inspectorate and the main contractor, which was concerned about main contractor liability and its reputation, all attempted to monitor labour standards in the supply chain, but the frequency at which groups of workers work- ing for different subcontractors came and went made it difficult for the unions to monitor conditions. There was some unease among the native workforce, who could expect unemployment when the contracts were finished, about the posted workers, who were seen as competing on an unfair cost basis. Unlike the Finnish construction workers union which

had an effective blacklist to sanction subcontractors with a history of non-compliance with collective agreements, the metalworkers had no overarching strategy to ensure employer compliance with its collective agreement. Instead, it left it to the initiative of shop stewards to devise their own strategies and to government labour inspectors to uncover il- legal conditions.

As a result, during the shipyard boom of the pre-2010 period, no one really knew whether posted shipyard workers were being paid accord- ing to Finnish minimum standards. Government inspectors routinely uncovered irregularities, and shop stewards claimed that underpay- ment was rampant, but there was an evident lack of capacity to enforce standards. Shop stewards and inspectors complained of the prevalence of shell firms, which disappeared if regulatory authorities took too close an interest in them; they often just changed their names and moved else- where. Ironically, it appears that the slowdown of business has made it easier to control conditions, because there simply have not been the same numbers of posted workers going in and out, with more workers working for ‘legitimate’ employers rather than fly-by-night labour-only contractors.

5.    Outcomes/conditions and problems for workers

In insular contexts of national industrial relations systems, employers would be unable to escape from national laws, regulations and informal norms, as these comprehensively regulate the entire space in which they operate. However, in the current environment, the ability to ‘cherry-pick’ between different regulatory frameworks allows management to shape the micro-politics of individual worksites, to isolate groups of workers from each other, and to weaken the ability of regulators to influence working standards. The result is a variety of contractual arrangements tailored to circumstances and the employer’s strategy. We can differ- entiate between posted workers on legitimate contracts, which in some IR systems might legitimately offer lower conditions than what native workers might expect, but are legal nonetheless. Then there are posted workers on contracts that are made to appear legal, but where posting is used only as a way to conceal illegal employment arrangements, and to make it difficult and expensive to pursue corrective legal action. Fur- thermore, there are many workers who are posted in an organizational sense, but who do not work under foreign contracts; these workers have

other, often dubious, employment arrangements, such as bogus self-em- ployment. They are part of the same labour market as posted workers. These regulatory possibilities interact with the subcontracting possibili- ties of industry-specific work processes to shape the employment rela- tions of posted workers.

Posted workers are mostly excluded from collective channels of work- er representation, and are mostly in contingent, insecure employment situations (Wagner and Lillie 2014; Berntsen and Lillie 2015; Lillie and Sippola 2011). They can potentially be somewhat ‘cheaper’ under the slightly lower conditions of employment sometimes possible under the PWD legal regime, but more frequently posted work is a form of ‘grey’ employment, under which it is very hard to detect violations of labour standards. Posted work has enabled and encouraged employers to cre- ate a segmented labour market in which the rights of posted workers are de jure as well as de facto different to, and more often than not lower than, those of native workers in the workplace. It is clear from many conversations with posted workers that many feel deceived and betrayed by recruiters, who often profit from travel and accommodation fees for the workers, do not pay some or all of the wages due, or fail to pay social security contributions, leaving workers without benefits. Many compare their situations to those of other, better treated, native workers. While often resentful, they also accept that this situation is part of the reality they face, representing the best option they have in an unfair world.

In construction, smaller domestic firms – subcontractors who compete directly with Eastern European subcontractors – have sometimes been supportive of stronger labour market regulation (Alfonso 2012). Many firms build their strategies around circumventing regulations, social se- curity payments and collective labour agreements by creative corporate structures. Trade unions and regulatory authorities have noticed that temporary employment agencies sometimes use countries such as Cy- prus or Portugal as places to incorporate their businesses; this use of an EU country as a ‘flag of convenience’ allows the firm to pay lower social security contributions. While it is commonly recognized that the main contractor is the main beneficiary and power holder, and that any strategy to control the labour market must ultimately hold them respon- sible, unions cannot help but be caught up in the shell game of tracking subcontractors, as the specific features of the cases they deal with inevi- tably revolve around which contractor did what for whom. In addition to the fluidity, contract documents in foreign languages and unclear and

unfamiliar legal constraints interfere with the ability of unions to moni- tor the construction labour market.

These country and industry case studies show the diverse ways in which employers use transnational contractual arrangements to construct pre- carious employment in blue-collar employment in the EU. The precari- ous employment situation results from the fact that employers are able to use the transnational deregulation of labour markets to their advantage while labour market regulation agencies are constrained from enforc- ing the rights of mobile workers (Wagner 2015). While a pan-European labour market exists, this market is not matched by equal enforcement mechanisms able to effectively protect workers’ rights. Firms use sub- contracting and temporary agency work in all three industry cases and across countries as a way to exploit regulatory gaps and avoid regulatory enforcement.

These regulatory gaps differ across countries and industries, requir- ing new efforts by unions and labour inspectors to find out where these workers are and how to represent them, but underlying it all is a Brus- sels policy which seems to be more intent on undermining national en- forcement systems than on seeking legitimate solutions to the enforce- ment problems which free labour mobility has brought. Unions can and should seek autonomous solutions for organizing and representing post- ed workers (see Danaj and Sippola, this volume). Nevertheless, this will remain very difficult unless a new line of policy is adopted in Brussels, giving due attention to the functioning of industrial relations systems and national traditions of wage determination.

References

Alfonso A. (2012) Employer strategies, cross-class coalitions and the free move- ment of labour in the enlarged European Union, Socio-Economic Review, 10 (4), 705-730.

Berntsen L. and Lillie N. (2015) Hyper-mobile migrant workers and Dutch trade union representation strategies at the Eemshaven construction sites, Economic and Industrial Democracy. doi: 10.1177/0143831X14537357

Bosch G. and Zühlke-Robinet R. (2003) Germany: the labor market in the Ger- man construction industry, in Bosch G. and Philips P. (eds) Building chaos an international comparison of deregulation in the construction industry, London, Routledge, 48-72.

Deakin N. and Walsh K. (1996) The enabling state: the role of markets and con- tracts, Public Administration, 74 (1), 33-47. doi: 10.1111/j.1467-9299.1996. tb00856.x

Dølvik J.E. and Visser J. (2010) Free movement, equal treatment and workers’ rights: can the European Union solve its trilemma of fundamental principles?, Industrial Relations Journal, 40 (6), 491-509.

Flecker J. and Meil P. (2010) Organisational restructuring and emerging service value chains: implications for work and employment, Work, Employment and Society, 24 (4), 680-698.

Flecker J., Holtgrewe U., Schönauer A., Dunkel W. and Meil P. (2008) Restructur- ing across value chains and changes in work and employment: case study evidence from the clothing, food, IT and public sector, WORKS Project Report, Leuven, HIVA. http://www.forba.at/data/downloads/file/362-Case_Study_ Report_D10.1.pdf.

Fleischer V. (2010) Regulatory arbitrage, Texas Law Review, 89, 227-289. Gautié J. and Schmitt J. (eds) (2010) Low-wage work in the wealthy world, New

York, Russell Sage Foundation.

Grimshaw D. and Rubery J. (2005) Inter-capital relations and the network or- ganisation: redefining the work and employment nexus, Cambridge Journal of Economics, 29 (6), 1027-1051.

Harvey M. (2003) Privatisation, fragmentation, and inflexible flexibilization in the UK construction industry, in Bosch G. and Philips P. (eds) Building chaos an international comparison of deregulation in the construction industry, London, Routledge, 188-209.

Houwerzijl M. and van Hoek A. (2011) Comparative study on the legal aspects of the posting of workers in the framework of the provision of services in the European Union – Report to the European Commission, Contract Number VT/2009/0541. http://ec.europa.eu/social/main.jsp?catId=471

Lillie N. (2010) Bringing the offshore ashore: transnational production, industrial relations and the reconfiguration of sovereignty, International Studies Quar- terly, 54 (3), 683-704.

Lillie N. (2012) Subcontracting, posted migrants and labour market segmentation in Finland, British Journal of Industrial Relations, 50 (1), 148-167.

Lillie N. and Sippola M. (2011) National unions and transnational workers: the case of Olkiluoto 3, Finland, Work, Employment and Society, 25 (2), 292-308.

Wagner I. and Lillie N. (2014) European integration and the disembedding of labour markets: transnational labour relations at the European Central Bank construction site, Journal of Common Market Studies, 52 (2), 403-419.

Wills J. (2009) Subcontracted employment and its challenge to labor, Labor Stud- ies Journal, 34 (4), 441-460.

Part 3

Organizing in fragmented production networks: how to establish voice mechanism for peripheral workers?

Chapter 8

Union campaigns to organize across production networks in the European telecommunications industry: lessons from the UK, Italy, Sweden and Poland1

Virginia Doellgast, Katja Sarmiento-Mirwaldt and Chiara Benassi

1.    Introduction

The fragmentation of production networks described in the earlier chap- ters in this book has created a number of challenges for labour unions. On the one hand, worker representatives are charged with maintaining pay and working conditions for their members in large firms or ‘core’ workplaces while preventing job losses via externalization (Hassel 2014). On the other hand, labour unions often seek to organize across production networks, with the aim of extending collective agreements to workers in peripheral jobs or more poorly organized workplaces (Mac- Kenzie 2010; Holtgrewe and Doellgast 2012). In this chapter, we argue that these two kinds of union objectives are increasingly interconnected. As externalization via outsourcing and temporary agency work expands and as bargaining coverage declines, unions find it is necessary to im- prove the standards of externalized jobs to both prevent the erosion of pay and conditions for their traditional members and expand union bar- gaining power (Doellgast et al. 2009; Benassi and Dorigatti 2014).

Our analysis asks under what conditions unions and other worker repre- sentatives are successful in improving or maintaining conditions across production networks, and examines the ongoing challenges they face in accomplishing these goals. Findings are based on four case studies of incumbent or formerly state-owned telecommunications firms in the UK, Italy, Sweden and Poland. These cases are drawn from a 10-coun- try study of restructuring in incumbent telecommunications firms in

1. This work was supported by the Economic and Social Research Council [grant number RES- 061-25-0444]. Some of the material from the case studies in this chapter was taken from Doellgast, Sarmiento-Mirwaldt, and Benassi (2015).

Europe and the US (Doellgast et al. 2013).2 Competition in national tel- ecommunications markets over the 1990s was encouraged by the devel- opment of mobile networks and the internet, as well as by changes in the regulation of traditional and emerging market segments. As incumbent telecommunications firms have faced growing competition in increas- ingly differentiated markets, they have sought to reduce labour costs and increase labour flexibility. One set of strategies for accomplishing this has been to externalize a range of service jobs – including call centre, technician and IT services – to subcontractors, service subsidiaries and temporary agencies.

In Chapter 5, we focused on one widely used externalization strategy: the outsourcing of call centre work. In this chapter, we examine two further strategies, with an expanded focus on call centre and technician work- places. The first is the use of different categories of contingent work, including temporary agency and freelance contracts. In some cases, in- cumbent firms outsource jobs to subcontractors, which are then able to further depress pay and conditions through using these kinds of contin- gent contracts. The second involves the transfer of a group of internal workers to subsidiaries and subcontractors via spin-offs and outsourc- ing. This often results in the shift of workers to new individual contracts or collective agreements with lower pay and conditions than they for- merly enjoyed within the incumbent firm – or in its core business units. Thus, both sets of strategies can be, but are not necessarily, connected to outsourcing to a third-party firm.

Similar to Chapter 5, our cases include detail on the effects these strat- egies have on job quality. However, our focus in this chapter is on evaluating the conditions for union success in responding to these dif- ferent forms of externalization. For this reason, we have chosen cases in which unions have succeeded in campaigns to improve or maintain pay and conditions in the face of these employer strategies. In the UK (BT) and Italy (Telecom Italia), we examine union campaigns to secure equal treatment for temporary agency and freelance workers. In Sweden (TeliaSonera) and Poland (Orange Polska) we examine union efforts to

2. This study included incumbent telecommunications firms from Austria, Czechia, Denmark, France, Germany, Italy, Poland, Sweden, UK, and USA. Findings are based on over 150 interviews, as well as site visits, comparative data drawn from collective agreements and surveys of union representatives, and archival material. Case study findings focused on restructuring in call centre and technician workplaces.

maintain pay and conditions for workers transferred through a spin-off to a service subsidiary or subcontractor.

We find that in all four cases, worker representatives have had some suc- cess in improving working conditions or resisting their downgrading, based on creative campaigns within core firms and across firms’ produc- tion networks. Their strategies have focused on two areas. First, worker representatives have used their historic bargaining power within large firms to negotiate limits on externalization. These strategies have relied both on partnerships aimed at improving the productivity of the in-house workforce as well as organizing strategies and protests targeted at gen- erating additional costs to employers associated with negative publicity and labour conflict. Second, worker representatives have sought to extend collective representation and legal protections to externalized groups of workers. This has occurred via campaigns to organize these workers and negotiate agreements with their employers, as well as through cam- paigns to close ‘loopholes’ in national legislation that allow employers to by-pass standard employment protections or equal pay rules.

Findings suggest that successful campaigns have drawn on heterogene- ous forms of bargaining power provided by national and sectoral labour market and industrial relations institutions. These institutions can be strengthened by European directives transposed into national law. In some of the cases presented here, directives concerning equal treatment for agency workers and transfer of undertaking rules have encouraged changes to national legislation. However, these laws are proving to be weak without strong union organization in incumbent firms and their subcontractors, subsidiaries and temporary agencies. Ideally, unions need to develop new strategies that mobilize both sets of internal, firm- based and external, institutional resources to succeed in regulating work across interconnected core and peripheral workplaces.

In the following sections, we present findings from the four case studies. We then discuss reasons for the differences in union strategy and out- comes across the cases, and conclude with a reflection on best practices and lessons learned.

2.    UK and Italy: improving the pay and conditions of agency workers and freelancers

2.1                      UK: BT

BT – formerly British Telecom – was privatized by the UK government in 1984, and the market was fully liberalized in 1990. BT negotiates agree- ments with two unions: the Communication Workers Union (CWU) which represents non-managerial grades, and Prospect which organizes first-, second- and third-line managers and professionals. While both unions’ members are affected by trends of externalization via agency work, we concentrate on the CWU as our focus is on the non-managerial workforce.

In the early 1990s, BT began to increase its use of temporary agency workers in many areas of work traditionally performed by permanent employees. At one point in the late 1990s and early 2000s, around 10,000 employees were on temporary contracts across the company, representing 7% of the workforce. These were concentrated in certain ar- eas, with a large number in call centres: one interviewee estimated that 50% of the call centre workforce in BT’s retail business unit were in tem- porary positions at this time. In the mid-1990s, BT set up telemarketing call centres composed entirely of agency workers.

The terms and conditions for agency workers were poor. Many agency workers were paid considerably lower rates than their permanent col- leagues, despite carrying out similar work. They received no company sick pay, no paid maternity leave, only statutory annual leave, and had no job security. Agency workers also had a much more flexible shift pat- tern, meaning that in some areas they worked more during unsocial hours and weekends. Conversions from agency to permanent positions were minimal, with some agency workers having worked for more than ten years at BT.

The CWU had long criticized the growing use of agency workers as well as the poor conditions they were working under. They conducted an in- ternal debate in 2000:

‘… it was a big internal debate about whether we recruited these peo- ple or whether we stood up and fought against the strategy of the com- pany, and I think that it came to a point where we had vociferously

tried to oppose the use of agency and it wasn’t getting us anywhere. It was becoming so that they were moving so quickly bringing agency people in, and some of these agency people wanted to join the union, that we had to, I suppose, adopt a two-pronged strategy which was to recruit and organize amongst the agency, but to also still deal with BT so as to get these people permanent contracts of employment. So we were not endorsing the use of agency. We were actually going to the company and saying that we wanted these people to be given proper contracts.’ (Interview, CWU official, 15/05/2012)

The union mounted a campaign ‘Justice for Agency Workers’ to organ- ize the agency workers and to work with the temporary agencies to im- prove conditions. This campaign was successful in increasing member- ship density among agency workers, which grew to around 50% – a high rate for this group of employees. However, this was lower than member- ship density among the permanent workforce, which was around 90%. Recruiting and retaining members proved difficult because of the high turnover among agency staff. The CWU developed good relations with the agencies and was successful in gaining recognition agreements in some cases. At the same time, these did not cover pay.

By the mid-2000s, BT had begun to reduce the number of agency work- ers across the company. One of the reasons for this was that manage- ment had begun to recognize that permanent workers provided higher quality customer service. Another was related to downsizing in other ar- eas of the company, and the need to find work for these redeployed em- ployees. For these reasons, and following union campaigns to decrease the share of agency workers, the CWU and BT reached an agreement in 2006 called the Retail Sourcing Strategy. This allowed for no more than 10% of the workforce to be composed of contractors, fixed-term workers and agency workers from 2007. In the same agreement, management committed to a cap of 2,200 jobs in offshored locations in India. As part of this drive, well-performing agency workers in some call centres were given BT contracts, although on a new grade that was ‘slightly below’ that of the existing workforce. For the individuals concerned, this change im- plied a salary increase of a few thousand Pounds and employment under a permanent contract. By 2013, around 35% of the call centre workforce were on these new grades.

Also in the mid-2000s, the CWU started to campaign for legislation guaranteeing equal pay and conditions for agency workers in the

House of Commons, as well as at EU level where the United Kingdom was blocking proposed legislation on this issue. In 2008, the UK gov- ernment reached a tripartite agreement with employer and worker rep- resentatives which stipulated that agency workers would be entitled to the same pay as colleagues in equivalent permanent positions after 12 weeks on the job. The government also agreed to the EU Tempo- rary and Agency Work Directive on equal treatment of agency workers, which was passed in 2008 and transposed into national law and imple- mented in 2011.

The new regulations ensured that agency workers on traditional agency contracts were entitled to terms equal to those of permanent workers concerning pay, annual leave entitlement, breaks and the use of facili- ties. At the time that the legislation was implemented, 4,000 BT agency employees qualified for equal treatment; while the CWU had over 3,000 members (across BT and other companies) who qualified. Union officials observed that most of these agency employees saw their pay increase, in some cases by as much as 20%.

However, the legislation included a clause known as the ‘Swedish Dero- gation’, which allowed a temporary work agency to be exempted from the requirement of equal treatment on pay provisions if it offers an agen- cy worker a permanent contract of employment and pays the worker be- tween assignments, or during the periods where they are not working and where there are no suitable assignments. Agencies immediately be- gan to exploit this ‘loophole’ in the legislation. In February 2011, prior to the implementation of the agency regulations in October, Manpower advised the CWU that they would be trialling ‘pay between assignments’ (PBA) contracts for agency employees at BT. By 2013, these had become the ‘default contract’ for this group of workers. Union officials observed that agency workers were required to sign these contracts without be- ing advised of their implications – particularly that they were ‘contrac- tually and legally signing away their rights to equal treatment on pay’ (Presentation, CWU official, 21/11/2013). As a result, pay gaps began to increase again, with agency workers in call centres paid between £2 and

£4 less per hour than similar permanent workers at the same locations. In addition, the proportion of agency workers in the call centres began to increase. For example, in 2013, BT employed 3,050 agency workers in its UK call centres, compared to around 8,000 permanent employees (Interview, CWU official, 17/10/2013).

The CWU initiated a new campaign after the passage of the UK legis- lation called ‘Closing the Loopholes: Justice for Agency Workers’. This campaign had several objectives, including securing amendments to the UK Agency Regulations closing this PBA and other loopholes; raising the profile of agency issues among CWU members and encouraging agency members to become active in the campaign; raising political awareness and support; engaging and working with the TUC and the global union UNI; and giving ‘consideration to challenging the UK Agency Regula- tions within the European Commission’ (Presentation, CWU official, 21/11/2013). In January 2013, the CWU organized an ‘Agency Action Day’, in which agency workers at a number of BT call centres demon- strated against low pay and inequality of treatment. In September 2013, the TUC made a formal complaint to the EC that the UK government had failed to properly implement the Directive, citing evidence – provided in part by the CWU – that tens of thousands of agency workers were being paid less than permanent staff doing the same job. This conflict was on- going at the time of writing.

2.2                     Italy: Telecom Italia

In Italy, the former state-owned operator Telecom Italia held a monop- oly in the fixed-line market until the late 1990s, when legislation liberal- izing telecommunications was passed in compliance with EU directives. The three main unions at both Telecom Italia and in the Italian telecom- munications sector are: 1) the SLC (Communication Workers’ Union), affiliated with the former communist confederation CGIL; 2) FISTEL (Union for Press, Entertainment and Telecommunication), affiliated with the Catholic confederation CISL; and 3) UILCOM (Italian Labour Union for Communications), affiliated with the socialist confederation UIL. These unions are joint signatories to both a company agreement with Telecom Italia, and a sectoral agreement with the employers’ as- sociation ASSTEL.

Telecom Italia started outsourcing call centre work during the 1990s. One of these subcontractors was Atesia, which became (in)famous be- cause most of its workers were on freelance, or so-called ‘co.co.pro’, contracts. ‘Co.co.pro’, or ‘freelance’ project-based work, was intro- duced by law (Pacchetto Treu) in 1997 and was regulated by the Biagi law up until 2015. The definition of a ‘project’ as well as of the degree of autonomy is very loose, thereby allowing companies to abuse these

contracts.3 When employees are employed under a co.co.pro contract, there are no obligations for the employers at the end of the ‘project’ and the so- cial contributions amount to the half of those for standard work contracts.

These contracts at Atesia were characterized by low levels of pay – main- ly based on performance – and social contributions that were lower than those set by the sectoral collective agreement. One union representative observes:

‘In call centres there is a legal fiction which allows the company to do it [use co.co.pros]. I (as the company) can say that you are formally – in reality, you are dependent and come every day for 6 hours in order to earn your monthly wage – but formally the call centre can say that you are a co.co.pro who organizes her time autonomously and make a certain amount of hours available for providing certain services. Thus, there is a completely different situation between law and real- ity, between what really happens and what’s written in the contract that should happen.’ (NIDIL/CGIL representative, 14/06/2011)

In the early 2000s, Atesia workers started mobilizing against this prac- tice. With the support of the unions, they organised a series of demon- strations and succeeded in getting media coverage. In 2005, in the wake of these protests, union representatives and Telecom Italia management negotiated an agreement with the aim of improving the precarious em- ployment conditions of Atesia workers. Telecom Italia purchased 20% of Atesia, and then integrated the workers into its subsidiary Telecon- tact under permanent contracts. The agreement also prescribed the transition of around 4,000 freelancer contracts for workers remaining at Atesia into standard contracts. Even so, Atesia continued to employ many workers on project-based contracts who did not benefit from the telecommunications sector agreement applied to the subcontractor’s permanent workers.

The unions mobilized around the issue of co.co.pros and call centres, bringing it to the attention of the public.4 By so doing, they were able

3. The recent labour market reform – called the ‘Jobs Act’ – re-defined more precisely the char- acteristics of the project-based work in order to reduce the number of co.co.pros by January We do not discuss these reforms in detail because the new regulation started applying only after the end of our fieldwork.
4. See, for example, the demonstrations on 20 October 2007 (RSU Almaviva 2007) and on 19

September (Vespo 2008).

to successfully lobby the government. The 2007 Financial Act improved the social security of co.co.pros, increasing employers’ pension contribu- tions and introducing the right to maternity and sick leave. In addition, two legal ordinances came into force in 2006 and in 2008, intended to limit the use of these contracts – the ‘Circolari Damiano’ or Damiano Or- dinances. The first ordinance prescribed that inbound call centre agents should be offered permanent contracts because the work did not reflect the characteristics of an ‘autonomous project’. When the first Damiano ordinance was passed, Atesia asked the confederal unions to bargain an agreement stabilizing the employment conditions of only part of the co.co.pros instead of all of them, because some call centre agents – such as those working at Atesia – were outbound. However, the confederal unions refused the compromise.

The confederal unions pushed then for the second ordinance, which con- cerned outbound call centre agents, as it was difficult to distinguish in daily practice between inbound and outbound agents. Sometimes out- bound agents were even assigned inbound activities so that the company could continue using the project-based contracts.5 The second ordi- nance does not require the stabilization of the employment conditions of outbound co.co.pros – as did the first ordinance for inbound call cen- tre agents – but it requires some evidence of ‘autonomous work’ when they are employed on project-based contracts. The Damiano ordinance gave call centre and telecommunications companies one year to bargain the transition from atypical to permanent contracts with unions, giving labour and management sufficient time to figure out the best arrange- ments with a view to limiting the cost burden on the company as much as possible. The outcome was that Atesia stabilised the employment con- ditions of all co.co.pros (around 6,000). Unfortunately, it started hir- ing co.co.pros again right afterwards, a practice contested by the unions through strikes.6 Although with ongoing contracts, Atesia was forced to transition the majority of its workforce onto permanent contracts. This created new challenges associated with responding to varying demand levels common to the subcontractor sector. However, the unions have sought to work with management to find alternative solutions to pro- vide flexibility: for example, in October 2012 unions and management signed an agreement on short-time work arrangements in order to avoid redundancies.

5. See, g. Cobas Almaviva (2008).
6. See, g. Bonaccorsi (2008) and Infoatesia (19.02.2008).

Parallel to these developments, the unions sought to reduce the use of agency work within Telecom Italia’s call centre subsidiary, Telecontact. In contrast to the UK, agency work is relatively well regulated in Italy, as legislation has been in place since the 1990s guaranteeing temporary workers the right to equal pay and equal treatment. The social partners also established a special fund for unemployment benefits and for train- ing: when agency workers have a permanent contract with their agency, they are entitled to 800€/month and to attend vocational training class- es while they wait for their next contract. However, equal pay refers to the wage levels set in the collective agreements; thus, agency workers do not benefit from company-level agreements, which include social ben- efits and variable pay.

The collective agreement negotiated with Telecontact at the time it was established in 2001 permitted 30 per cent of the workforce to be employed on temporary agency contracts. This was much higher than in other areas of the company. In 2008, the unions negotiated a new agreement to shift 300 of the company’s agency workers onto perma- nent contracts. By 2012, there were no agency workers in the call centre subsidiary.

2.3 Comparison

In both the Italian and UK cases, unions faced similar challenges, associ- ated with uneven or incomplete application of laws intended to ensure equal treatment of different categories of temporary work. The unions concerned used a combination of strategies focusing on two areas. First, they used lobbying, social partnership and protest to change or enforce national legislation extending protections to these contingent workers. Second, they sought to mobilize both their traditional members and newly organized groups of temporary workers within ‘core’ firms to ne- gotiate limits on temporary employment contracts and improvements in conditions for temporary workers.

These strategies had varied success – and depended on heterogeneous forms of bargaining power. In Italy, the unions relied on the mobiliza- tion potential of their numerous members within Telecom Italia and on their political leverage with the government. This was further bolstered by public support in the wake of the Atesia ‘scandal,’ which was widely covered in the media due to a large extent to union mobilizations. In the

UK, the CWU was successful in limiting the numbers of temporary work- ers and gaining improvements in pay and conditions via application of equal pay rules. However, campaigns to close the loophole created by the ‘Swedish derogation’ are ongoing, and their success will likely depend on either support from the European Commission or a change in the current government – which has proven unwilling to date to review the temporary agency regulations.

3.   Sweden and Poland: avoiding declining pay and conditions associated with spin-offs and subcontracting with employee transfer

3.1                   Sweden: TeliaSonera

The largest competitor in the Swedish telecommunications industry is TeliaSonera – formed in 2002 through the merger of the former Swed- ish state-owned monopolist Telia and the former Finnish state-owned Sonera. Sweden’s telecommunications market was one of the earliest to liberalize in Europe, with far-reaching reforms promoting competition coming into force in the mid-1990s. TeliaSonera’s workforce is repre- sented by two major unions, the blue-collar Swedish Union for Service and Communications Employees (SEKO) and the white-collar Unionen, as well as by several smaller unions in the academics and professionals confederation SACO. These unions were joint signatories to a corporate- level agreement with TeliaSonera and to a sectoral agreement with the employers’ association Almega IT Employers.

Similar to most incumbent telecommunications firms, Telia, and then TeliaSonera, established a range of subsidiary companies for different business lines, and later spun off a series of business and support ser- vices. In each case, there was no renegotiation of collective agreements or change in terms and conditions of employment: ‘they move all their working conditions, salaries, everything, right over to the new company, because they use the same collective agreement’ (Interview, SEKO of- ficial, 28 May 2012). The most significant ‘events’ involved the establish- ment and sale of subsidiary companies responsible for internal services and for installation and network maintenance activities between 2001 and 2007 – which affected close to 11,000 employees – most of whom were over time transferred to subcontractors.

In 2001, Telia formed two separate business units or ‘Groups’ made up of business areas that they planned to sell off as ‘non-core’ businesses. First, Telia formed the ‘Telefos Group’, made up of 9 formerly internal businesses in the ‘business services’ area with 5,600 employees – or 17% of Telia employees. The venture capital firm Industrikapital purchased 51% of the Telefos Group in 2001; and each individual company was sold between 2001 and 2007. Second, Telia formed the ‘Orbiant Group’, made up of six companies with 5,400 employees responsible for network maintenance and the installation and servicing of customer equipment. Flextronics purchased 91% of the shares in the Orbiant Group in 2001, and the remaining 9% in 2002. Then in 2005, Altor purchased a major- ity stake in Flextronics Network Services, with the two companies merg- ing under the name ‘Relacom’.

The creation and sale of subsidiary companies in the Telefos Group and Orbiant Group described above was the most significant case of out- sourcing at Telia/TeliaSonera. As a result of these sales, close to 11,000 employees were moved to subcontractors between 2001 and 2007. Af- ter that time, TeliaSonera purchased nearly all of its technician services from these subcontractors.

According to a union official, the local trade unions were closely involved at all stages of outsourcing, and were able to reach favourable agree- ments easing staff transfer and retraining:

‘All these persons, they were moving from Telia to the new com- panies. And they looked after it so everyone would have a job after they… when Telia sold it, all the people had these possibilities. But if they worked for TeliaSonera and were moved to this new company, and they don’t want to move to this new company, they had special solutions for them. For instance, early retirement, they had possibili- ties to be educated. And they also could receive money to go out of the Company and have two years…for instance two years of payment.’ (Interview, SEKO official, 28/5/2012)

Under Swedish transfer of undertaking rules, pay and basic employ- ment terms and conditions specified in collective agreements were pro- tected when employees were moved to the Telia-owned companies, and then when the companies were sold to third parties. Thus, the collec- tive agreements moved with employees – under Swedish law, employees continued to be covered by agreements for one year following a transfer

of ownership, after which they could be moved onto new agreements (but with the same basic pay scale, pension rights, etc.).

Some aspects of employment conditions and benefits changed, despite the formal portability of past terms and conditions. According to union representatives who had experienced the move from Telia to the subcon- tractors, pensions gradually became less generous, work content became narrower and opportunities for advancement shrunk. However, at the same time, there were no direct concessions associated with outsourc- ing: ‘We are not the same technicians that we were ten years ago. But in principle we have the same pay, it is not different in that way’ (Interview, SEKO official, 10/5/2012).

Following the outsourcing of these jobs, there was then further frag- mentation of ‘production networks’ as subcontractors replaced perma- nent contracts with temporary contracts. In 2005, the temporary staff- ing agency Manpower launched ‘Manpower Network Services,’ and the subcontractor Relacom downsized its staff – many of whom got jobs at Manpower and then were sent back to Relacom as agency employees on temporary contracts (Geary et al. 2010). This appears to be a model that all of the companies have followed. A union representative from one subcontractor estimated that 25-30% of their workforce was employed through temporary agencies (Interview, SEKO official, 10/5/2012). An- other union official interviewed worked for a temporary staffing com- pany through this arrangement, but was posted almost exclusively to her former employer: ‘[Subcontractor] let me go, and then I took a job with [staffing company] who rented me back to the company that sacked me’ (Interview, SEKO official, 9/8/2011).

However, her terms and conditions did not change significantly follow- ing transfer: her collective agreement with the agency required it to pay temporary workers the same salary as colleagues doing similar work with- in the subcontractor. The union was also able to negotiate a new collec- tive agreement at this subcontractor in 2010-11 in response to concerns with the large number of employees made redundant and then re-hired through temporary staffing agencies: ‘the new rule is that after a while when you’ve been laid off, if the company needs a worker, they need to ask you first.’ At other subcontractors, there was an agreement stating that the company could not hire temporary agency workers until 9 months had passed after layoffs; and within that time period, employees who were laid off had to be re-hired if there was a need for additional staff.

Temporary employees also had some degree of job and salary security, as they were on ‘permanent contracts’ with their temporary agency and re- ceived 90% of their salary during the time when no work was available in their area – again based on a collective agreement. This is equivalent to the ‘pay between assignments’ contracts discussed in the UK case study above – introduced as part of the ‘Swedish derogation’ in UK law. How- ever, these contracts were more effective in preserving equal conditions in the context of strong collective agreements. In effect, this meant that the agency shared some of the risk of business fluctuations or seasonal changes in labour demand with the employee.

TeliaSonera also outsourced call centre work, though to a smaller ex- tent. Several interviewees noted that call centres were viewed as a ‘core’ area of the business and therefore kept in-house. One union official esti- mated that around 2,000 call centre employees had been outsourced in 1998-1999 (Interview, SEKO official, 27/6/2012). However, over time the company reduced outsourcing, and by 2014 TeliaSonera had stopped subcontracting any of its call centre work in Sweden. In addition, in the early 2010s the company was using over a thousand temporary agency workers. While these workers were more flexible, labour costs were simi- lar to the internal workforce due to high collective bargaining coverage and equal pay rules. The union succeeded in convincing management to convert most of these employees to permanent contracts as well: ‘we said, employ them in-house instead. I think they, the management, de- cided to do that because they noticed that the quality that they get from the employees was bigger if they are employed by us.’ (Unionen repre- sentative, 27/10/2014)

3.2                         Poland

Telekomunikacja Polska was the incumbent telecommunications firm in Poland. In the early 2000s, France Telecom acquired a controlling stake in the company, which is now called ‘Orange Polska’. Around 20 unions are active at Orange Polska. Two representative industry-level unions are the largest: the National Section of Telecommunications Workers (SKPT) of the national Solidarity organisation, and the Federa- tion of Unions for Telecommunications Workers (FZZPT) of the national OPZZ. In 1998, the main unions negotiated a multi-establishment agree- ment at national level, which inter alia covers all direct Orange Polska employees.

Spin-offs with employee transfer to subcontractors were one of the most prominent forms of downsizing at Orange Polska. The EU Transfer of Undertakings Directive was transposed into Polish law, stipulating that pay and benefits under an employee’s former collective agreement must remain in place for at least 12 months. However, bargaining coverage and union density are lower than, for example, in Sweden. Thus, typi- cally after the 12-month period, transferred employees would be shifted to different – often less favourable – terms and conditions.

At Orange Polska, many spin-offs involved non-core activities. The first spin-off in 1997 involved building administration and cleaning services, with subsequent ones affecting drivers, couriers, security services, ac- counting and remote computer maintenance. These services are now performed by third-party subcontractors. In many cases, the affected former Orange Polska employees transferred to the outsourcing part- ners. Two spin-offs – of technician services in 2002 (a proper spin-off) and of call centres in 2010 (a subsidiary creation) – will be discussed in depth to illustrate the process.

Network services were spun off in 2002, with 6,000 service technicians transferred to external companies. As part of this spin-off, the union negotiated a programme ‘Work for the worker’: in return for standing orders from Orange Polska, these external companies agreed to employ former Orange Polska employees on indefinite contracts and to guaran- tee them work for 1.5-3 years, depending on the region of Poland. The new employment contracts were individually negotiated by workers who took a wage cut of 15% on average. However, Orange Polska paid fairly generous financial inducements (on average 20 thousand złoty). The de- cision to move was voluntary, but if technicians decided not to move, they would be made redundant. The majority of affected employees took the opportunity to leave because the terms offered went far beyond what is required by Polish law.

Some unionists expressed pride in this agreement: ‘The conditions for the workers I think we managed to negotiate well, as far as Polish con- ditions go.’ (Interview with two union officials, 07/04/2012). However, these measures were only temporary. According to one unionist, ‘After the employment guarantees expired, there were further wage cuts and redundancies in many companies that took over workers.’ (Email com- munication with a union official, 11/10/2012). According to another, the transitional period ‘was a camouflage, window-dressing. Because the

employer survives these 18 months and then he can do as he pleases, so it was only for a while’ (Interview with two union officials, 17/04/2012).

In 2010, the customer service call centres for both mobile and fixed line business units were spun off to the subsidiary Orange Customer Service (OCS). Unlike the situation with technician services, OCS remained fully owned by Orange Polska. Several waves of tough negotiations were as- sociated with this spin-off, and the union SKPT led a collective dispute. Following this dispute, the unions succeeded in extending the terms of the collective agreement for Orange Polska to the new subsidiary. Trans- ferred employees received a two-year employment guarantee and no re- duction in their salary. A similar pension scheme was also introduced, which constituted a gain for the former mobile business unit employees. One unionist described the negotiations:

‘When the new company [OCS] was founded it did not offer work- ers anything except for the things that are by law offered to employ- ees that move to another company, for example, it sets a period of employment. … This new company initially did not offer anything. Only these negotiations that the unions undertook have led to these successes that [union official X] mentioned.’ (Interview, union official and works councillor, 15/02/2012)

Orange Polska managers stressed that these provisions went far beyond what is required by law: ‘This is important for us, a rather large group of people. We would like to carry this out in social peace, we want to be fair, we offer this, this and this, and that we will keep’ (Interview with four TP managers, 10/07/2012). Union representatives interviewed at- tributed this success to high membership density and union strength in the call centre area. One unionist commented on the difference in union strength between technicians and call centre employees:

‘Technicians are a very fragmented group who are not all located in the same place. Especially since the network services were spun off in 2002, there have been fewer technicians who would put up a resist- ance to the company’s restructuring plans. … Conversely, there was definitely concern about the call centre employees, because there any walkout would be very noticeable for the company, because custom- ers could not be served completely. It is the customer’s only contact with the company. So there was a concern, so perhaps these protests, which took place then [when OCS was spun off in 2010] led to the

company … fulfilling all these demands. And it fulfilled them, all of them were fulfilled.’ (Interview, union official, 19/04/2012)

3.3                         Comparison

The Swedish and Polish cases are interesting to compare, because there was a similar development in both whereby technician work was out- sourced, with existing staff transferred to subcontractors, while call cen- tre work was kept in-house. First, we see different outcomes in the tech- nician area. At TeliaSonera, subcontracted technicians for the most part preserved their pay and conditions over time, even when they were shift- ed to temporary agencies. This can be largely attributed to encompassing collective bargaining institutions, as well as the ability of well-organized unions in temporary agencies to negotiate provisions improving job and earnings security for their members. At Orange Polska, technicians had some protection under transfer of undertakings rules, as well as a favourable job security agreement negotiated by the unions. However, they experienced wage cuts of 15% associated with being transferred to new employers, followed by more significant reductions in pay and con- ditions after an 18-month transition period.

Second, we see similar outcomes in the call centre area – but for differ- ent reasons. At Orange Polska, the union successfully opposed down- grading of pay and conditions associated with moving call centre work to a separate subsidiary, using mobilization of a well-organized mem- ber base in an area that was strategically important for the company. At TeliaSonera, call centre work was shifted in-house and temporary agency workers were moved onto permanent contracts in cooperation with the union, but also for broader strategic reasons concerning service quality.

A comparison of these cases shows that unions again faced similar chal- lenges – here associated with maintaining employees’ pay and conditions during and after their transfer to different employers or subsidiaries. Transfer of undertakings rules in national legislation provided a starting point for and some leverage in union negotiations over the terms of em- ployee transfer. However, similar to the UK and Italian cases, success- ful union strategies relied on member mobilization ‘across production networks’ to use these resources – here across increasingly disintegrated networks of subsidiaries, subcontractors, and temporary agencies. In

Sweden, more encompassing collective bargaining institutions covering these networked workplaces provided a substantially stronger resource for unions in maintaining pay and conditions after employee transfer, as well as arguing for the re-internalization of call centre jobs.

4.   Conclusions

Past research has asked why unions adopt more or less inclusive strate- gies to organize and represent externalized workplaces or worker groups, as well as under what conditions these strategies succeed or fail. Theo- rists examining trends of labour market dualism have argued that un- ions in many ‘coordinated market’ or social European countries focus on protecting the working conditions of the core workforce, relying on tra- ditional, institutionalized sources of bargaining power while neglecting new groups of service workers or those on contingent contracts (Palier and Thelen 2010) – possibly at the long-term expense of working condi- tions for their own members (Eichhorst 2014). Other researchers argue that unions in a range of institutional settings increasingly seek to organ- ize externalized groups of workers, as their core membership shrinks or comes under threat from the expansion of outsourcing to lower wage sectors and workplaces (Taylor and Bain 2003; Heery 2004; Gumbrell- McCormick 2011; Tapia and Turner 2013). However, these studies have shown that unions have uneven success in accomplishing these goals, despite numerous examples of innovative campaigning, organizing, and partnership approaches.

The case study findings presented in this chapter show that successful union strategies to regulate pay and conditions across production net- works depend on organizing and mobilizing members both within core firms and across externalized workplaces. Increasingly, maintaining job quality for core workers and the bargaining power of their unions depends on extending collective bargaining and legislated employment protections to peripheral worker groups. In all four cases examined here, unions sought to accomplish these dual goals using a range of strategies. We argue that variation in outcomes was due not to union enthusiasm or reluctance to pursue these strategies, but rather to differences in unions’ success in accessing and developing two sets of resources: encompass- ing labour market institutions and member mobilization across core and peripheral workplaces.

First, encompassing national and sectoral labour market institutions were crucial resources for unions in extending terms and conditions from ‘core’ to ‘peripheral’ worker groups and employers. In the cases examined here, these institutions included national legislation and col- lective agreements that required equal pay and conditions for temporary workers or those transferred to subsidiaries and subcontractors. Among the four cases discussed above, Sweden demonstrates the highest cov- erage of collective bargaining at sectoral level and across externalized workplaces. Collective agreements set equal pay for temporary work- ers and agency workers employed by subcontractors. This broadly pre- vented TeliaSonera’s strategies of outsourcing, temporary agency work and spin-offs from being associated with downgrading of pay and condi- tions. Unions at subcontractors and temporary agencies were then able to build on existing agreements to further strengthen pay and conditions through collective bargaining with these external firms.

By contrast, national regulation and sectoral bargaining in the other three country cases were weaker or characterized by loopholes allow- ing employers to circumvent encompassing rules. In the Italian and UK cases, despite implementation of the EU Directives on equal pay for agency and fixed-term workers, national labour market regulation allowed employers to hire certain categories of contingent workers at lower pay rates. In Italy, employers used freelance contracts to circum- vent the equal pay regulation. In the UK, the transposition of the EU directive on agency work into national law allowed companies to pay lower rates if they offered permanent contracts with pay between assign- ments. In both cases, unions were involved in campaigns that succeeded in strengthening collective bargaining and labour market institutions – although in the UK, employers were able to exploit further loopholes to again differentiate pay between permanent and temporary workers.

A second important set of resources unions drew on as they sought to im- prove pay and conditions across production networks was their capacity to mobilize members across core and peripheral workplaces. In the tel- ecommunications industry, unions are best able to organize strikes and campaigns within incumbent firms, due to their history of high union density and often militant unions. Orange Polska’s union used bargain- ing power from a well-organized membership base in the call centres to oppose plans to differentiate pay and conditions via the creation of subsidiaries. Unions also used residual power in the technician area to negotiate agreements providing better terms for workers transferring to

subcontractors than those required by national law. In both the UK and Italy, unions negotiated formal limits on the use of temporary agency work within the incumbent telecommunications firms. In these two cas- es, unions also organized temporary agency workers, building up mobi- lization potential beyond their core workforce.

Ideally, unions need both sets of resources to challenge management’s attempts to differentiate pay and conditions by moving work to more poorly regulated sectors and workplaces: encompassing labour market institutions and the capacity to mobilize members across increasingly heterogeneous production networks. Unions face many challenges as they seek to develop and access these resources. We have focused on cases in Europe, in which unions enjoy residual institutional bargain- ing power at national and sectoral level, as well as additional leverage to strengthen national laws through the process of transposing EU di- rectives. In countries or regions with lower union density and weaker labour market institutions, these challenges are even greater. We have also focused here on the externalization of work to temporary agencies or third-party subcontractors located within the same country as the core ‘client’ firm. The already substantial difficulties faced by unions in regulating work across production networks within these countries is significantly exacerbated when work is moved across national borders, via offshoring; or handled by subcontractors relying on workers posted from other countries. Certainly in the telecommunications industry, firms are able to offshore or ‘nearshore’ an expanding range of service jobs – including, most visibly, call centre work.

While unions have varying capacity to access the institutional resources discussed in this chapter, the case studies have broader lessons for un- ions as they seek to develop new sources of bargaining power in dis- tinctive institutional settings. One implication of our findings is that the national context remains an important level of action for unions as they seek to improve pay and conditions for externalized groups of workers. Union campaigns to mobilize across national borders in multinational firms and their networked subsidiaries, subcontractors, and temporary agencies can also benefit from incorporating different levels of action into their strategies. Bargaining power in MNCs, as in incumbent ser- vice firms, depends on success on two fronts: both seeking to close the loopholes in national institutions, as well as mobilizing workers across organizational boundaries to challenge the segmentation strategies of firms.

References

Benassi C. and Dorigatti L. (2014) Straight to the core – Explaining union responses to the casualization of work: the IG Metall campaign for agency workers, British Journal of Industrial Relations. doi: 10.1111/bjir.12079

Bonaccorsi M. (2008) Ad Atesia tornano i co.co.pro, Left, 4, 68-69. Cobas Almaviva (2008) Call center, via alla seconda circolare Damiano,

01.04.2008. http://www.cobasalmaviva.org/rassegna%20stampa/Call%20 center,%20via%20alla%20seconda%20circolare%20Damiano.html

Doellgast V., Sarmiento-Mirwaldt K. and Benassi C. (2015) Institutions, cost struc- tures, and the politics of externalization: explaining variation in contestation over boundary strategies for call center jobs, unpublished manuscript, London, London School of Economics and Political Science.

Doellgast V., Sarmiento-Mirwaldt K. and Benassi C. (2013) Alternative routes to good jobs in the service economy: employment restructuring and human resource management in incumbent telecommunications firms, Final Project Report ESRC Grant RES-061-25-0444. http://www.esrc.ac.uk/my-esrc/ grants/RES-061-25-0444/outputs/read/be077186-d4cf-43e9-88a5- e42ae32bb2e6

Doellgast V., Batt R. and Sorensen O.H. (2009) Introduction: Institutional change and labour market segmentation in European call centres, European Journal of Industrial Relations, 15 (4), 349-371. doi:10.1177/0959680109344366

Hassel A. (2014) The paradox of liberalization: understanding dualism and the recovery of the German political economy, British Journal of Industrial Rela- tions, 52 (1), 57-81.

Gumbrell-McCormick R. (2011) European trade unions and ‘atypical’ workers, Industrial Relations Journal, 42 (3), 293-310.

Heery E. (2004) The trade union response to agency labour in Britain, Industrial Relations Journal, 35 (5), 434-450.

Holtgrewe U. and Doellgast V. (2012) A service union’s innovation dilemma: limi- tations on creative action in German industrial relations, Work, Employment and Society, 26 (2), 314-330.

Infoatesia (2008) Sciopero articolato di un’ora – 20 febbraio 2008. http://in- foatesia.blogspot.de/2008/02/sciopero-articolato-di-unora-20.html

MacKenzie R. (2010) Why do contingent workers join a trade union? Evidence from the Irish telecommunications sector, European Journal of Industrial Rela- tions, 16 (2), 153-168.

RSU Almaviva (2007) Per la manifestazione del 20 ottobre, 01.10.2007. http:// www.rsualmaviva.it/comunicati/c_2007-ott-01_3.htm

Tapia M. and Turner L. (2013) Union campaigns as countermovements: mobiliz- ing immigrant workers in France and the United Kingdom, British Journal of Industrial Relations, 51 (3), 601-622.

Taylor P. and Bain P.M. (2003) Call centre organising in adversity: from Excel to Vertex, in Gall G. (ed.) Union organizing: campaigning for trade union recogni- tion, London, Routledge, 153-172.

Vespo G. (2008) Call center, oggisquilla la protesta, L’Unità, 19.09.2008. http:// cerca.unita.it/ARCHIVE/xml/275000/272719.xml?key=GIUSEPPE+VESPO- &first=671&orderby=0&f=fir&dbt=arc

Chapter 9

Organizing peripheral workers in parcel delivery and postal services

Bettina Haidinger

1.    Introduction

As early as 2000, Michael Taylor and Alan Hallsworth noted that ‘in re- cent decades the speed of growth of companies in the courier service industry has been phenomenal. Here is an industry at the very heart of economic change and time-space compression – a business pivotal to globalization, supply chain management and e-commerce.’ (2000: 245)

Some fifteen years later, the courier service industry is a booming and dynamic business, as illustrated by this example: In January 2015, Deutsche Post-DHL announced it would be creating up to 20,000 new jobs for couriers by 2025 due to sustained growth in the parcel deliv- ery business.1 On the downside of this optimistic news however, these jobs go with low-wage contracts in newly founded subsidiary companies subject to a different and worse collective agreement than the contracts currently offered for this kind of work – resulting in wage cuts of 20%.

This is a striking example of sector boundaries – in this case the bound- ary between logistics and transport and the postal services sector – be- coming increasingly blurred and markets for formerly separated prod- ucts and services becoming integrated, thereby giving employers the opportunity to reduce costs, i.e. labour costs, by ‘shopping’ for the most convenient employment regulations on offer.

Logistics and delivery chains have undergone significant and rapid technological and organizational changes in the last three decades (Coe 2014). In an overview of research on the logistics sector in the US, Bonac- ich and Wilson (2008: 3) refer to developments in the sector as a ‘logis- tics revolution’, with power within value chains shifting from producers

1. http://www.welt.de/wirtschaft/article136763560/Die-Plaene-der-Post-erinnern-an-das-

Modell-Amazon.html, accessed 2 February 2015.

to retailers. Fragmented and internationally dispersed production pro- cesses need smoothly operating logistics chains. This has led to tech- nological logistics innovations such as containerization and intermodal transport. Online Shopping has changed consumer habits and boosted parcel delivery. At the same time, postal markets have been liberalized. This has resulted in formerly state-owned incumbents strategically ex- panding both internationally and into new business fields of the logistics sector, while the deregulated and liberalized postal market has become open for competitors from the courier industry and third-party logistics providers (Hallsworth and Taylor 1999; Hermann 2014).

Such developments have had strong repercussions on the business and employment relations found in these increasingly integrated markets for delivery, transport and logistics. Postal services as well as parcel and express services essentially thrive on a wide range of activities and busi- ness ties among multinational logistics groups, temporary employment agencies, franchise companies, subcontractors and ‘self-employed’ cou- riers – an environment providing very diverse labour and contracting relations for workers, and hence complicating their appropriate repre- sentation over all workplaces.

This contribution will discuss the challenges faced by unions, workers’ representatives and workers’ strategic collectives when coping with these various positions of labour in highly integrated delivery processes featur- ing fragmented layers of subcontracting. After a brief description of busi- ness and employment trends in the European postal and parcel delivery market, the contribution describes initiatives taken by unions to reach out to and represent peripheral workers and to support their local work- place struggles, highlighting key factors for successful interventions.

Despite a seemingly raise-to-the bottom trend in pay and working condi- tions in parcel delivery and courier services examples presented in this contribution show that workers’ struggles become effective and may re- verse this trend when two strategies are combined. First, unions need to acknowledge and strategically take up the demands and spontane- ous resistance of peripheral workers ‘at the bottom’ of global delivery chains. There is an urgent need for unions to find unconventional and targeted ways to approach self-employed couriers or couriers in subcon- tracted entities and to offer them support useful to them in their im- mediate work situation. Struggles happening ‘unlikely places’ can con- stitute starting points for workers and unions becoming more powerful

– and for regaining influence in the sector(s)’ bargaining disputes. Sec- ond, fighting for institutional re-embedding of employment conditions in terms of social protection, pay or employment contracts and of in- dustrial relations can help re-establish a sectoral level playing field with respect to labour standards.

The contribution is based on findings from a sectoral study on parcel delivery as part of the logistics sector (‘Social Dialogue and Participa- tion Strategies in the Global Delivery Industry: Challenging Precarious Employment Relations’2). This study was carried out as part of a joint project between researchers and trade union representatives in Austria, Germany, Czechia and Hungary. Furthermore, findings from an assess- ment of the long-term consequences of the liberalization of postal ser- vices on employment and working conditions in Austria, Belgium, Ger- many, Greece, the Netherlands and Poland (Hermann 2014)3 were used. In addition, information came from the ITF/Uni Global Global Delivery Newsletter4 and press coverage on recent union initiatives.

2.    Major trends in employment and working conditions

The postal service and parcel delivery market has undergone significant changes in the last decades. With the abolishment of post monopolies in the European Union competition between the active companies, i.e. the former state-owned incumbents and the new competitors, has in- tensified. In addition, European postal markets as such have become integrated, as seen not only in the increasing importance of cross-bor- der mail and parcel delivery but also in the merging of different busi- ness fields: letter post, newspaper delivery, parcels, express services, as well as postal services, logistics and haulage more generally (Taylor and Hallsworth 2000: 243). As a result, former post monopolists com- pete with express service providers, forwarding companies, publishing houses and newspaper delivery networks, while in the second and third

2. SODIPER 2011-2012; funded by the European Commission, DG Employment and Social Af- fairs VP/2010/001/0226; Partners: AT, GER, CZ, H, ETF/UniEuropa; focus on employment conditions on the fringes or the end of commodity chains in the global delivery industry.
3. PIQUE ‘Update’ Project 2013; based on EU research project 2006-2009; funded by the EC; privatization and impact on employment and working conditions; postal services one of 4 sectors; update 2013 for Austria, Belgium, Germany, Greece, Netherlands and Poland.
4. http://www.itfglobal.org/en/cross-sectoral-work/supply-chain-logistics/in-focus/global- delivery-briefings.

segment of the delivery chain hundreds of small companies and inde- pendent deliverers compete for contracts.

While European postal markets and the courier industry are converging, employment relations have become increasingly fragmented – not only between countries but also within countries (Hermann 2014; Sowers et al. 2014). This development poses great challenges to workers’ repre- sentatives, unions and unionizing strategies.

As a rule, the former monopolists in postal services in their respective home markets still provide the best employment conditions in the sector. However, faced with increasing competition from low-cost competitors, they have started to cut back the wages of newly hired workers and to make greater use of ‘mini-jobs’ which are partially exempt from employ- er social security contributions. New competitors in the letter market (letters, newspapers and bulk mail) mainly operate with self-employed deliverers paid by the number of items they deliver and also excluded from mandatory health and social security coverage (Hermann 2014).

While in the letter market self-employment is mostly used by new com- petitors, in parcel and express services self-employment seems to have become the norm rather than the exception. The parcel delivery sector in Europe is characterized by a fragmented and multi-layered structure involving several players including hauliers, postal service providers and van operators of different company sizes. A variety of business types are to be found in the sector, such as transnational third-party logistics ser- vice providers, subcontracting firms or self-employed drivers, and also temporary employment agencies can be found as employers in global transport and local delivery chains. A small number of global competi- tors are responsible for coordinating the supply chain and providing the international transport backbone and logistics; and some of them still employ couriers to carry out the operational business as such – the col- lection and delivery of parcels from and to customers as well as sort- ing. The majority however award contracts to ‘service partners’, medi- um- and small-sized enterprises and self-employed couriers. A recent investigation by Stiftung Warentest (December 2014) in Germany of the five leading parcel delivery providers, i.e. DHL-Deutsche Post, Hermes, DPD, GLS and UPS, showed that four of them operate exclusively or mainly with subcontractors to deliver the parcels. Only Deutsche Post- DHL has set a maximum on the number of delivery districts to be out- sourced to subcontractors.

The rationale behind subcontracting is to devolve risk, costs (infra- structure, means of production) and flexibility demands down the value chain. Contractors enter short-term contracts with subcontractors (can- cellation periods of one month; payment per parcel irrespective of the ups and downs in business) and benefit from the imposed flexibility of their service partners. Subcontractors have hardly any organizationally substantial relationship to the third-party logistics service providers and are characterized by individualized and insecure work arrangements. Workers in each of the layers are integrated into the institutions of work regulation to differing extents (Haidinger et al. 2014).

As a result, unions and workers’ representatives active in the postal ser- vice and parcel delivery sector are confronted with a variety of employ- ment situations ranging from self-employed status in a subcontracting chain to fairly stable yet differing employment conditions when working directly for a former incumbent in its home market or for a new competi- tor providing third-party logistics. Workers’ individual problems differ with regard to their employment status, their union and social security affiliation or to the consequences for their social status and earnings possibilities when they are migrant workers. These varieties of work and employment and the diversity of workers found in the sector pose tricky challenges for unions and unionizing strategies, highlighting the fact that union adherence to the past principles of vertical representation of their more or less homogenous constituencies and membership policies will not suffice. If unions remain stuck in their own organizational strait- jacket, they will fail to adapt to the new challenges of this industry and to the specific needs of its workforce.

3.    Sector challenges for unions and industrial relations

This section discusses how the challenges of covering peripheral work- ers – i.e. the self-employed, migrant workers, workers at non-unionized subcontractors or people working under labour contracts without col- lective bargaining coverage – in unions’ sectoral industrial relations and representation policies unfold in practice. In what follows I will explain how unions and workers’ collectives react to these challenges and what new and promising paths for achieving an improvement in working con- ditions are available.

3.1                   Fragmented collective agreement coverage

Growing divisions among workers in the postal and parcel delivery sec- tor are underpinned by increasingly fragmented collective bargaining structures. Findings from the SODIPER research project suggest that either the diversity of collective agreements (as in Austria or Germany) or employer unwillingness to engage in collective bargaining (as in Ger- many, Czechia and Hungary) have made it impossible for labour organi- zations to maintain uniform wage levels and working conditions for the postal and parcel delivery sector (Haidinger et al. 2014).

In Czechia and Hungary only the former monopolist companies are covered by the strong industrial relations traditions of the postal sec- tor, while new competitors do not need to adhere to the same collective agreements but just conclude – if any – company agreements or indi- vidual contracts with their employees. In Germany and Austria, the quite comprehensive industrial relations systems are circumvented by the use of bogus self-employment. In addition, adherence to new and worse or no collective agreements (for example, van operators in Austria, logistics in Germany or transport in Czechia) is common when service providers set up subsidiaries or establish new job categories. In the Austrian postal and parcel delivery sector, as many as twelve different collective agree- ments – including agreements for the transport sector and for newspa- per delivery – exist.

In Germany, Deutsche Post DHL very recently announced its intention to set up subsidiary companies (called ‘Delivery’) where newly hired cou- riers will be employed and covered by a different and worse collective agreement (logistics sector) than that covering the couriers now directly employed at Deutsche Post DHL (postal sector).5 In the Netherlands, wage cuts at the former monopolists were linked to the creation of a new job category, auxiliary or assistant deliverers, earning about 40% less than the former postbodes (postmen). Since 2007, all newly hired deliv- erers at Dutch Post have been employed as postbestellers (post deliver- ers), to the effect that 80% of delivery staff received the lower wage rate in 2012 (Hermann 2014: 27).

5. http://www.welt.de/wirtschaft/article136763560/Die-Plaene-der-Post-erinnern-an-das-

Modell-Amazon.html, accessed 2 February 2015.

3.2                     Circumventing regulations: the logic of subcontracting

Traditional forms of representation embedded in the national industrial relations system are obviously ineffective in this sector. In Germany, Czechia and Hungary most couriers employed by the new competitors or their subcontractors are either not subject to collective agreements or subject to ones worse than those of the former incumbents. In Austria, collective agreements are often either not adhered to or are circumvent- ed by contracting out the delivery process to self-employed drivers. The lack of workers’ representation within the subcontracting companies has repercussions on de facto working conditions as, even when labour standards are established within collective agreements, compliance is neither guaranteed nor the rule. Regulations are simply ignored and not abided by. Therefore, when collective representation within a company is lacking, workers are individually responsible for the enforcement of minimum standards. However, in many cases workers do not know their rights or are intimidated by management when insisting on them.

Self-employed workers, who in some countries make up the major part of the workforce at the new competitors in postal services and parcel delivery, fall completely outside the bargaining system and are hardly represented by trade unions. Some unions are sceptical about the effectiveness of tar- geting the self-employed as a particular group due to organizational and financial constraints and because they do not regard the self-employed as their constituency. Others see a big and increasing potential in the self-employed. It is indispensable to access this group of workers as well – not only for ‘moral’ reasons but also to safeguard employment stand- ards and to include them as workers deserving equal rights.

3.3                     Vulnerability of peripheral workers

The peripheral workforce at the bottom of the parcel delivery chain and in the most competitive segments of the postal service sector such as newspaper and bulk mail delivery is bearing the main burden of adjust- ment, as reflected by the highly flexible employment relations. Often workers are recruited from vulnerable labour-market groups, such as migrants in Austria and the Netherlands or the long-term unemployed in Germany (Haidinger 2012; Holst and Singe 2011; Hermann 2014; in- formation provided by FNV Transport en Logistiek).

In Austria, due to the restricted labour market access of third-country nationals and their subsequent difficulty in integrating into the Austrian labour market a substantial number of drivers with migrant backgrounds either work in a self-employed capacity or are (precariously) employed by small service providers. Often their formal vocational qualifications are much higher than what is required for working as a courier. Reasons for this momentary or already long-lasting de-skilling are manifold: lan- guage barriers, non-recognition of diplomas, having paused too long in the original profession, bridging a career gap with this job as a driver before changing to another job or before moving on to another destina- tion country. The SODIPER research also showed that migrants without a work permit or having experienced discrimination in their former em- ployment relationships often resort to working as a deliverer on a self- employed basis.

In border regions such as the Eastern part of Austria with borders to Slo- vakia or Hungary, postal services such as the delivery of bulk mail are carried out by workers commuting on a daily basis from these coun- tries.

The – in some countries – significant presence of migrant workers or cross-border commuters poses additional challenges to unions, mean- ing that it is essential to address migrant workers in a comprehensible and supportive form and with tailored services. Comprehensibility not only must be taken into account with respect to language issues but also with respect to the particular understanding and interpretation of the national and the even more complicated cross-border system of workers’ rights and the regulation of social rights available to migrant workers.

4.    Organizing peripheral workers in the subcontracting chain

This section provides examples of how to deal with the above-mentioned challenges and particular problems unions and peripheral workers are facing in this sector. It is noteworthy that while the vulnerability of in- tegrated, time-sensitive logistics potentially provides workers with po- sitional power, the fragmentation of employment and casualization of labour hamper the formation of associational power (Wright 2000). At the same time the liberalization and privatization of the formerly highly regulated postal sector and its convergence with the logistics and

transport sector seem to weaken and undermine the regulatory power safeguarding decent working conditions. What then are the success fac- tors for organizing and improving employment standards? What are the unions’ strategies to reach out to and integrate peripheral workers in the postal and parcel delivery sector into their constituencies? What strate- gies directly target those working within these highly integrated delivery processes with at the same time fragmented layers of subcontracting? What political resources for re-regulating employment should be sum- moned to safeguard decent working conditions?

This section provides insights into three broad key factors and their real- world application in the postal and parcel delivery service that might help to answer these questions in a rather optimistic manner.

4.1                   Targeted organizing in subcontracting firms

Local unions often have difficulties in gaining access to workers in out- sourced links of the parcel delivery and postal services chain. Due to the combination of a flexible, integrated and centrally controlled delivery process and the markedly decentralized and often informal employment relations, traditional union methods of approaching workers via shop stewards are limited. Recruiting and organizing strategies focusing on more direct and targeted information and action for/with the couriers themselves turn out to be more successful.

One positive example in this respect was an organizing campaign initi- ated by the service and transport union vida (Austria) in 2010 to reach out to, inform and organize couriers working in subcontracting chains. One of the information instruments deployed to gain workers’ atten- tion was the public announcement and propagation of an international solidarity campaign with Turkish UPS employees who had succeeded in setting up a union at that time.6 Since a good proportion of the couriers themselves have a Turkish background, they were positively affected by

6. http://www.vida.at/servlet/ContentServer?pagename=S03/Page/Index&n=S03_16.1.2.a &cid=1282637197673, see also State-of-the-Art-Union Report Austria, pp.13; for the inter- national UPS campaign information can be found here: http://www.upsblog.org, accessed 27 April 2015. The international campaign was essentially driven and supported by Global Delivery Network, a joint initiative of Uni Global and the International Transport Workers Federation (ITF), http://www.itfglobal.org/en/cross-sectoral-work/supply-chain-logistics, accessed 27 April 2015.

this obviously international attempt to get them sensitized to workers’ concerns and the purpose of unionizing. The aim of the campaign was to establish initial contact with subcontracted employees through personal conversations and by two larger-scale actions in front of the UPS head- quarters and to learn something about the main problems they faced at work. They reported working hours of up to 12 hours a day and the non- payment of overtime. This direct confrontation with drivers showed that a good number of them were positive towards vida’s attempts to support them. However, some voiced their fear of being observed by their em- ployers in their contacts with unions and consequently of being intimi- dated or even dismissed.

Another method used in this organizing attempt was the direct and natu- rally time-consuming contacting of couriers ‘on the road’. Many drivers work from dusk till dawn and do not have the time and energy to engage in organizing activities. Therefore, union activists did not ‘wait’ in their offices for their potential constituency to show up or call but accessed them at times the drivers had time to talk, for instance early in the morn- ing, in the evening or virtually on the road, for example, at filling sta- tions. Organizers’ experiences show that personal and pro-actively sus- tained and continuing contacts to peripheral workers are crucial success factors in building trustful relationships with a new constituency and in being taken seriously as an organizer and supporter.

4.2                     Self-organization of peripheral workers and union support

Instead of being pro-actively approached by unions, more often than not self-employed couriers and peripheral workers are left to their own de- vices. When asked in the course of the SODIPER research project if they would appreciate a union’s or another organization’s active role in sup- porting them, most of them were interested but sceptical. This scepti- cism was, on the one hand, due to the unions’ previous absence from the field in all countries. On the other hand, many saw themselves trapped in the sector’s logic of subcontracted exploitation and were sceptical about effective strategies paving the way to decent working conditions.

Some were prepared to draw individual consequences from their unsat- isfactory working situations, reporting their intention to quit soon. Only few drivers reported on industrial action and wildcat strikes leveraging couriers’ particular strategic position in the distribution process to avert

deterioration in working or payment conditions or to enforce entitle- ments. Nevertheless, such actions do exist:

In one instance, some 15 employees working for a third-party logistics subcontractor in Austria joined forces in early 2010 to set a deadline for payment of their outstanding Christmas bonus. The conflict was settled in favour of the workers after they threatened not to deliver the parcels before the deadline expired. In Hungary, workers organized their own wildcat strike to prevent a restructuring of couriers’ wages. In another case workers managed to convince their employers to introduce rotating night shifts scheduled in advance instead of imposing them involuntarily on other workers (‘We joined forces to have our concerns listened to.’) (Haidinger 2012: 43).

The study also came across cases of self-employed couriers joining forces against a subcontractor who had not paid self-employed workers in Aus- tria. They decided to foreclose the subcontractor who was unwilling to pay and ‘eliminated’ him from the subcontracting chain. As a result, they struck a deal with the general contractor and took over the delivery dis- tricts previously assigned to the subcontractor.

These were examples of self-organization that succeeded without union support. However, self-organized initiatives can be opportunities for un- ions to leverage already existing unrest and organization, as shown by two examples from Germany and the Netherlands.

In Germany, DPD – a company bound to the collective agreement of the logistics sector – decided to outsource distribution and sorting at a particular centre to subcontractors, with the result that all employ- ment contracts were successively transformed into contracts with much lower remuneration. When a court decision objected to this transfer, DPD reacted by dismissing the workers with wages paid for the next four months. During and after this period the 52 dismissed workers picketed the plant’s entrance on a 24-hour basis. While some of the colleagues accepted compensation payments, others decided to accept for the time being the subcontractor’s working conditions. At the same time, many of the workers joined the union ver.di, which reached an organization level of 85%. An elected works council then began negotiations on a new collective agreement very similar to that of DPD. These were successful from the workers’ point of view, with wages for centre workers increased by 40%. Consequently, the subcontractor had to charge DPD more for

its services – with the result that outsourcing no longer paid off for DPD. Eventually, all subcontracted workers were re-insourced (Molitor 2014: 18).

Experiences from the Dutch union FNV Transport en Logistiek suggest that union presence and support for the struggles of self-employed cou- riers have to be prepared very carefully. ‘Traditional’ union demands such as better social insurance and employment protection may not be fully shared by self-employed colleagues. Many work part-time and have different social backgrounds ranging from students to housewives, el- derly people or workers with a migrant background, and therefore have different interests concerning labour and social protection claims.

First and foremost, higher piece rates topped the Dutch agenda, with the consensus of all workers. This goal was achieved by a collective strike against delivering parcels at the existing piece rates. The union tried to go beyond this demand by carefully collecting demands from rank-and- file workers, helping them to articulate, print and distribute them.

4.3                      Re-regulating and ‘re-embedding’ couriers’ employment relations

Bottom-up organizing and effectively targeting a new constituency of sub-contracted and self-employed workers is one promising strategy for unions to embrace and support the most vulnerable workers along the delivery chain. And there is no alternative: unions have to learn from and listen to the demands of workers in their immediate situations of need and to support them in their struggles against exploitative contrac- tors and employers. At the same time, these workplace-based struggles can act as a trigger for further-reaching claims influencing and improv- ing working conditions in the sector as a whole.

As mentioned above, the re-regulation and improvement of employment standards for peripheral workers in the postal and parcel delivery sector is complicated by the fact that, on account of liberalization and market convergence, there is no longer a clearly defined sector that could be regulated. I refer in this section to two attempts to close loopholes for employers to circumvent labour regulations and to re-regulate employ- ment standards in parcel delivery and postal services, even though they are sometimes little more than a ‘compromise’. Both try – one at sectoral

level drawing on the social dialogue power of industrial relations, the other one at the legislative level of labour law – to prevent and inhibit the use of self-employment and subcontracting in delivery chains.

4.4                      Insourcing of self-employment through ‘adapting’ collective agreements

One example of the re-regulation of the postal sector involving the in- sourcing of self-employed couriers via adapting collective agreements again comes from Austria. (Haidinger 2012: 17-18) The Austrian Post acts as both a contractor and employer of couriers. ‘Parcel deliverers’ employed by the Austrian Post distribute 30% of the total parcel volume, 40% is delivered via regular postmen, while the rest (30%) is delivered through ‘service partners’ and subcontracted couriers. Between 2004 and 2009, the Austrian Post increasingly outsourced parcel delivery to service partners on cost grounds – a ‘red rag’ from the Postal Union’s point of view. This began with the awarding of contracts in peak times and was later extended to whole segments. A new collective agreement in 2009 was a cornerstone in regaining terrain in the market segment of parcel delivery, securing or even creating jobs though at the expense of deteriorated employment standards, as conceded by an interviewed works council member:

‘Due to the new collective agreement it is easier to recruit more per- sonnel and at the same time remain competitive against hauliers (i.e. subcontractors). This means we are able to withstand them [hauli- ers]. However, we were not able to agree on a collective agreement as good as the old one. (…) The collective agreement constituted a trade- off, allowing the work [of parcel delivery] to stay within the Austrian Post. We really fought for this new labour regulation in order not to have the same bad working conditions as stipulated in the van drivers’ collective agreement. But all in all it is of course not as good as our old one.’ (Works Council Member, Austrian Post)

The new collective agreement was the result of hard negotiations with management. At the beginning of the discussions on a new collective agreement, the aim of the postal union was to conclude an agreement for the entire sector of parcel delivery. However, employer representatives,

i.e. the Federal Economic Chamber, were strictly opposed to this pro- posal. Obviously, the social partners were unable to reach agreement on

a response to the converging markets of postal services, haulage, parcel delivery and logistics that would have established a level playing field with respect to labour standards.

4.5                     Bogus self-employment and contractors’ liability

A major challenge for the re-regulation of the postal and parcel delivery sector is how to deal with self-employment. One positive example comes from the Netherlands. Up until 2008, new competitors in the Dutch let- ter market mainly used self-employed deliverers paid under piece rates with average hourly wages of 7 Euros or less, i.e. below the national minimum wage, and without pension and other social security rights or healthcare coverage. The FNV Bondgenoten trade union started a public campaign in 2007, publishing information on these working conditions and attracting considerable media attention. As a consequence, the so- cial partners – with major intervention from the government – reached a compromise (set forth in a new government decree and a new collective agreement) after several rounds of negotiations under which new com- petitors had until the end of 2013 to convert 80% of existing contracts with deliverers into regular employee contracts. As a result of these conversions, deliverers working for new competitors now receive wages in line with the national minimum rates, i.e. enjoying a major wage in- crease (Van Klaveren 2013).

In these two examples, unions were strong enough to put pressure on employers, policymakers and legislators, demanding and enforcing minimum standards for workers in letter delivery. A number of coun- tries such as Belgium and Austria have legislation against circumventing employment contracts via so-called ‘bogus’ self-employment contracts. In Belgium, for instance, very strict rules and checks are applied when registering as a self-employed worker with a view to uncovering bogus self-employment: people working solely for one contractor and therefore financially dependent on him cannot be labelled as ‘self-employed’ but are employees.

In Austria, similar regulations exist, though the checks are made after- wards. In the case of parcel delivery, almost anybody holding a driving licence can register as a courier with the Chamber of Commerce. There are no special checks required for registration. To trace bogus self-em- ployment, social security institutions need to have reasonable doubt

about the self-employed status of a worker, or the courier himself must claim his rights as an employee. Every single case then has to be proven separately – a procedure requiring time and effort. Recently, there was a non-committal discussion on whether contractor liability for offences against national insurance law or labour law as introduced for the con- struction and cleaning industries should be extended to other industries. Such regulations already exist in the Netherlands where contractors such as large mail order companies have to adhere to minimum employment and social standards when awarding a contract to a third-party logistics provider. This implies greater responsibility for the cascade’s contrac- tors to hire companies compliant with existing labour regulations.

5.    Conclusions

Against the background of the liberalization and internationalization of parcel delivery and postal services, employment standards in these sec- tors have been upturned. As studies (e.g. Hermann 2014) on the em- ployment consequences of the liberalization of the postal market have shown, not only have employment numbers declined throughout the sector. What is even more important is that atypical employment rela- tions, such as ‘mini-jobs’ or self-employment, are gradually replacing regular full-time jobs. These not only provide less employment security, but also pay significantly lower wages. Competition and the pursuit of higher profits have also encouraged the former monopolists to cut wages for newly hired workers or workers employed as assistant mail deliver- ers, or to outsource whole business segments such as parcel delivery or distribution to subcontractors. New competitors in postal and parcel de- livery services, often third-party logistics providers or former monopo- lists ‘from abroad’, generally offer poorer working conditions than the former domestic monopolists.

These findings suggest that increasing parts of the postal and parcel de- livery sector are becoming low-wage segments where pay is barely more than the national minimum wage or even below it in the case of self- employment. Are peripheral workers gradually becoming the norm in this sector? This concluding section provides a summary of two key fac- tors crucial for stopping this trend and re-balancing wages and working conditions: on the one hand, fighting for re-regulation and insourcing (or the prevention of outsourcing) can help re-establish a sectoral level playing field with respect to labour standards. On the other hand, unions

need to acknowledge and strategically take up the needs, demands and spontaneous resistance of peripheral workers in global delivery chains.

5.1                   Struggles for re-regulation and insourcing

The chapter has provided a number of examples at local and national level in various European countries that have been successful in estab- lishing new regulations for (parts of) the postal and parcel delivery sec- tor. In Germany, the outsourcing of business segments at DPD to low- wage subcontractors was reversed after a long-lasting struggle carried out by the affected workers with the support of ver.di to raise wages at the subcontractor. Once outsourcing was no longer profitable, the ser- vices were again provided in-house. In the Dutch case of turning self- employed postal deliverers into employed ones with a guaranteed mini- mum wage, the Dutch government played an important role in urging for this regulation; it was by no means just a social partner agreement but was heavily influenced by government pressure. Again the Nether- lands provide an example of strong government pressure when estab- lishing contractor liability for parcel delivery, obliging contractors such as big mail order companies to adhere to minimum employment and social standards when awarding a contract to a third-party logistics pro- vider. In the Austrian case, self-employed parcel delivery couriers under contract from the Austrian Post were insourced after several rounds of concession bargaining between the postal union and management.

A situation of current interest is the conflict between the German ser- vice sector union ver.di and Deutsche Post DHL. The latter is planning to outsource parcel delivery to its own subsidiaries, replacing part-time and short-term workers with full-time workers – though at a wage level 20% lower than before under another collective agreement that will be valid by then. Ver.di is protesting against this plan, accusing the compa- ny of circumventing the collective agreement (‘Tarifflucht’). It is threat- ening industrial action to thwart this scheme, should management stick to the plan.7

7. http://www.welt.de/wirtschaft/article136657666/Post-plant-Zweiklassen-Gesellschaft-

bei-Paketboten.html, accessed 2 February 2015. http://www.welt.de/wirtschaft/ article138307606/Post-rechnet-mit-Streiks-in-den-kommenden-Wochen.html, accessed 27 April 2015.

Re-regulation can happen and outsourcing can be prevented. Success factors seem to be a strong government commitment to decent labour standards in support of unions’ demands and /or a militant base of workers with the strength and endurance to follow their claims through.

5.2                      Self-organized picketing by peripheral workers and unionizing

What this chapter has also shown is that instances of worker resistance can be found in unlikely places: highly individualized self-employed workers will take industrial action without any union support in order to raise piece rates, claim outstanding pay or profit from eliminating superfluous entities within the cascade of subcontractors. Nevertheless, such collective action remains problematic in a setting of high competi- tion among workers and on account of the remoteness of those actually setting out the conditions under which people work in this sector: the transnational third-party logistics companies.

However, the examples mentioned indicate that under specific circum- stances associational power can be gained relatively quickly even if this can be expected to remain transitory and local. Established workers’ representation institutions could provide associational power, yet they show high levels of inertia in adapting to new economic structures – though with exceptions as has been shown in this chapter. This not only limits the support they can provide to parcel delivery workers; it actually limits the spontaneous resistance of workers against their intolerable employment and working conditions. This also means that institutional and organizational adaptation may greatly enhance workers’ associa- tional power and thus better position them to take advantage of the fact that the logistics processes are not only inherently vulnerable but also economically crucial for entire global production networks spanning a range of business processes and sectors.

One main lesson learned from these examples is the need to find appro- priate, unconventional and targeted ways to approach couriers and to offer them support useful to them in their immediate work situation and disputes. Struggles exist in unlikely places, and these have to be taken seriously as they can constitute starting points for further organizing and winning members – and for re-regulating the sector(s) for the better from a peripheral worker’s point of view.

References

Bonacich E. and Wilson J. (2008) Getting the goods: ports, labor, and the logistics revolution, Ithaca, Cornell University Press.

Coe N.M. (2014) Missing links: logistics, governance and upgrading in a shifting global economy, Review of International Political Economy, 21 (1), 224-256.

Cumbers A., Nativel C. and Routledge P. (2008) Labour agency and union posi- tionalities in global production networks, Journal of Economic Geography, 8 (3), 369-387.

Haidinger B. (2012) On the move in Global delivery chains: labor relations and working conditions in the parcel delivery industries of Austria, Germany, the Czech Republic and Hungary, SODIPER Synthesis Report, Vienna, FORBA.

Haidinger B., Schönauer A., Flecker J. and Holtgrewe U. (2014) Value Chains and networks in services: crossing borders, crossing sectors, crossing regimes?, in Hauptmeier M. and Vidal M. (eds) Comparative political economy of work, Basingstoke, Palgrave Macmillan, 98-118.

Hallsworth A. and Taylor M. (1999) The transport sector and protected postal services: regulating the activities of Purolator Courier Services in Canada, Transport Policy, 6 (3), 159-168.

Hermann C. (ed.) (2014) Die Liberalisierung der europäischen Postmärkte und die Folgen für Arbeits- und Beschäftigungsbedingungen, Vienna, FORBA. http:// media.arbeiterkammer.at/wien/PDF/studien/Liberalisierung_Postmaerkte_ Arbeitsbedingungen.pdf.

Holst H. and Singe I. (2011) SODIPER National research report Germany, Vienna, FORBA.

Molitor C. (2014) Wir haben Recht wiederhergestellt, Mitbestimmung, 10/2014, 18-19.

Sowers E., Ciccantell P. and Smith D. (2014) Comparing critical capitalist com- modity chains in the early twenty-first century: opportunities for and con- straints on labor and political movements, Journal of World-Systems Research, 20 (1), 112-139.

Stiftung Warentest (2014) Schnell, aber ruppig, Stiftung Warentest, 12/2014, 76-97.

Taylor M. and Hallsworth A. (2000) Power relations and market transformation in the transport sector: the example of the courier services industry, Journal of Transport Geography, 8 (4), 237-247.

Van Klaveren M. (2013) Liberalisation and consequences for employment in the Dutch Postal Sector, Internal Report.

Wright E.O. (2000) Working-class power, capitalist-class interests, and class com- promise, American Journal of Sociology, 105 (4), 957-1002.

Chapter 10

Organizing posted workers in the construction sector

Sonila Danaj and Markku Sippola

1.    Introduction

In this chapter we look at trade union strategies for organizing posted workers in high-income European Union countries from the perspec- tive of both unions and workers. We focus on construction as a sector characterized by highly fragmented labour relations, resulting from long and complex subcontracting chains and the use of agency workers, self- employment and transient employment. The nature of the industry also presumes high rates of non-indigenous workers and makes it one of the most likely ones for finding posted workers (for more on posted work see Lillie and Wagner’s chapter in this book).

Migrants have been classified under various criteria, among which the most relevant for this chapter are direction, distance and duration of stay in the country of destination (Heberle 1955). Based on these crite- ria we differentiate between long-term immigrants and posted workers. The former term refers to people who have come of their own accord from a different country, whether an EU member state or a third coun- try, and who live in the host country with the intention of settling there permanently or staying for a relatively long period. Posted workers, on the other hand, are EU workers sent by their company to work in another EU country for a definite period of time (up to two years) as providers of particular services, regulated by the Posted Workers Directive (Direc- tive 96/71/EC; Lillie and Wagner in this book). This differentiation is important for understanding the strategies employed by host country trade unions, and the way duration of stay (long-term vs. fixed- and rela- tively short-term) and the manner of migration (of their own accord vs. through posting) affect the shaping of these strategies and their outcome.

While there is a relatively robust body of research focusing on the ambiv- alent attitudes and dilemmas of national trade unions towards migrant workers (e.g. Penninx and Roosblad 2000; Wrench 2000, 2004; Dølvik

and Visser 2009; Hardy et al. 2012), there are fewer studies analyzing the factors behind successful union campaigning with regard to migrants in general (e.g. Eldring et al. 2012; Adler et al. 2014) and posted work- ers in particular (e.g. Berntsen and Lillie 2014; Lillie and Sippola 2011). This chapter seeks to contribute to this debate, asking the following questions: What are the strategies employed by trade unions to organize migrant workers in general and posted workers in particular within their structures? What lessons can be learned from past experience and what recommendations can we draw for the establishment of successful voice mechanisms for posted workers?

Section 2 provides an overview of the existing literature on initiatives taken by trade unions in countries such as Denmark, Finland, Norway, the United Kingdom and the United States to mobilize migrant workers in general and incorporate them in their structures. The literature sug- gests that a successful migrant organizing campaign involves a com- bination of leadership-based, bottom-up, ‘zonal’ and multi-stakeholder approaches as well as ‘innovation’ in organizing strategies (Eldring et al. 2012; Milkman and Wong 2000; Wills 2008). Successful organizing of migrant workers basically follows the same principle as that used in the successful organization of non-migrant workers, though it also re- quires local unions to understand the particular culture, interests and work situation of those approached (Sherman and Voss 2000).

In Section 3 we analyse union strategies in the specific case of posted work- ers. Our analysis is based on empirical qualitative data collected within the framework of two projects: an Academy of Finland funded project entitled ‘Transnational Unionism and Democracy in Global Governance’ (2006– 2008) and a European Research Council funded project entitled ‘Trans- national Work and the Evolution of Sovereignty’ (grant number TWES 263782). They involved the case studies of four large power plant con- struction sites in three countries: Finland (Olkiluoto 3), the Netherlands (Eemshaven) and the United Kingdom (Ferrybridge and Carrington) for the periods 2006–2008, 2010–2012 and 2014 respectively. The data con- sists of more than 200 qualitative interviews with posted workers, trade union officials, shop stewards and managers, accompanied by participant observations on each site recorded as field notes.

In Section 4 we examine the role played by trade unions on site to organ- ize and represent posted workers, the strategies they have employed to recruit non-indigenous workers, which initiatives were successful and

why, the attitudes of the workers themselves towards the local trade un- ions and the reasons behind their (non)involvement with them. We dis- cuss the success of certain strategies across countries, comparing these with overall strategies used to organize migrant workers in general and providing suggestions on how to establish successful voice mechanisms for posted workers in the construction industry.

We argue that the strategies employed by trade unions to recruit long- term permanent immigrants are relevant for the recruitment of posted workers. However, considering their relatively short-term stay, hyper- mobility and posting situation, there is a need for specific strategies and initiatives to be able to unionize them. In order to overcome the country- bound union jurisdiction we highlight the need for a transnational or- ganization for the protection of posted workers’ labour rights.

2.    Organizing migrants

Trade unions in host countries have shown contradictory attitudes to- wards migrants in general, regardless of whether they are long-term per- manent migrants or short-term and/or posted workers (see Penninx and Roosblad 2000; Wrench 2004; Krings 2009; Hardy et al. 2012). While migrants are sometimes considered as a threat to the interests of union members (Frege and Kelly 2004; Hyman 2001), early studies have also highlighted migrant involvement in local trade unions (e.g. Kosack and Castles 1973). As the migrant labour force continues to increase, it is becoming increasingly important for unions to engage with migrant workers and represent them inclusively (Adler et al. 2014; Martinez Lucio and Perret 2009; Holgate 2005; Lillie and Martinez Lucio 2004; Frege and Kelly 2004). In the face of declining member- ship among national unions and the increasing replacement of ‘core’ jobs by irregular, mobile, agency and other forms of precarious labour, unions have started addressing migrant workers and their issues in an attempt to regain strength (Adler et al. 2014; Schelkle 2011). This change of attitude has happened not only because unions have understood and internalized the concept of rights, but also because workforce diversifi- cation and the new dynamics of the global market have put pressure on workers’ collective representatives to change or perish (Schelkle 2011).

Although organizing migrants has not been an easy task as they face certain challenges local workers might not face, there are several relevant

examples of successful organizing already identified in the literature (e.g. Adler et al. 2014; Eldring et al. 2012; Fitzgerald and Hardy 2009; Wills 2008; Milkman and Wong 2000). Despite their obvious vulnerability to deportation and limits on their legal rights, migrant workers are by no means impossible to unionize (Milkman and Wong 2000). Indeed, the fact that they are relatively powerless might provide the potential for or- ganizing (Adler et al. 2014; Wills 2008). Furthermore, the intersections of class with gender, ethnicity and immigration have repeatedly created an explosive mix for organizing campaigns (Wills 2008). In this section we look at some of these successful campaigns involving long-term per- manent immigrants, using them as a benchmark for the initiatives taken for organizing posted workers discussed in the following section.

Free movement of labour within the EU, especially after the A8 and A2 enlargements, has led to an increase in the number of labour migrants moving either individually or through a posting contract from these countries to the original EU-15 states, despite restrictions adopted in some of the latter countries. We have selected a number of relatively successful examples from the existing literature on recent migrant or- ganizing campaigns in Denmark, Finland, Norway and the UK in the last fifteen years with a view to identifying the success factors for migrant organizing. We also consider case studies from the USA, thereby add- ing illustrative examples from other geographical contexts. From this literature and the case studies, we find that the outcomes of the cam- paigns varied in terms of unionization rates. The most successful case in this respect was Norway, where national figures for 2008 show that migrants constituted 20 per cent of total union membership in the con- struction industry (Eldring et al. 2012), while the proportion of migrant workers in Denmark and Finland merely constituted a few percentage points (ibid; Lillie and Sippola 2011). Similarly, the campaigning in LA/ Southern California resulted in thousands of migrant workers becoming union members (Milkman and Wong 2000).

Union experience with long-term migrants suggests that the key to any successful campaign lies in the effective combination of the two approaches – leadership-based and bottom-up – into a comprehensive

1. The Accession 8 (A8) countries are Czechia, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia and Slovenia which acceded to the EU in 2004, while Bulgaria and Romania (A2) became EU members in 2007.

strategy. For a campaign to be successful, both central and local levels of a union need to be proactively engaged in organizing migrants. In a now classic study on the union recruitment of migrant workers, Milk- man and Wong (2000) highlighted the importance of such a strategy in organizing Latino immigrants in the 1990s, particularly office janitors in Los Angeles and drywall hangers in the region’s residential construc- tion industry. Eldring et al. (2012) came to a similar conclusion when studying union strategies towards post-accession migrant workers in the construction sector in Denmark, Norway and the UK, that is, successful campaigns were based on intense interaction between the central and lo- cal levels in formulating and implementing strategies, as also seen in the LA/Southern California case (Milkman and Wong 2000). In Denmark, the central level formalized existing developments in organizing Cen- tral and Eastern European (CEE) workers at local level by, for example establishing a cross-union collective agreement for foreign enterprises, while in Norway, the local achievements in recruiting labour migrants served as a yardstick in a shift in federation-level policy towards invest- ing increased effort into recruiting migrants and tackling social dump- ing (Eldring et al. 2012). In the Finnish (Olkiluoto 3) case, there was a coordinated joint effort between the national and local levels, despite a number of contradictory interests existing between the regional and central level (Lillie and Sippola 2011).

A ‘zonal’ approach to organizing appears to be another form of a suc- cessful union campaign. Instead of progressing from one workplace to another, and from one employer to another, as it is usually done, mobili- zation is focused on a specific area (or zone), a method which was adopt- ed in the case of London cleaners (Wills 2008). Such a ‘zonal approach’ is particularly successful in the case of dispersed workplaces (one janitor per building, for example), or when employers are small-size companies. This territorial basis for organization was even used in the case of what Milkman and Wong (2000) described as the industry-wide organizing of Los Angeles and Century City janitors and the wider region’s residential construction industry.

However, sometimes the way migrant workers are approached and their numbers in union ranks are not enough. Studies show that a multi-stake- holder approach is one of the cornerstones of campaign success. The campaigns reported by Milkman and Wong (2000) involved building al- liances or coalitions with the wider community and other stakeholders such as civil society or community organizations, and the use of highly

creative tactics to win public support and put pressure on employers. Similarly, the success of the London cleaners’ campaign involved ‘or- ganising the industry, the clients, London’s politicians, the media and a diverse alliance of community groups as well as the cleaners themselves’ (Wills 2008). One important success factor was the holistic, class-based approach adopted by the East London Communities Organisation (TEL- CO)2 that went beyond the realm of work in defining the interests of the workers. In its capacity as a community organization, TELCO sought to find common ground around issues such as job quality, housing, welfare, immigrant rights and street safety (Wills 2008). In other words, class interests were read through the lens of community, immigration, race and religion.

The concept of ‘innovation’ in organizing strategies has also been re- peatedly identified as a success factor by scholars of union (recruitment) campaigns (see, e.g. Adler et al. 2014; Eldring et al. 2012; Milkman and Wong 2000; Wills 2008). Innovation is necessary to manage the complex new employment relations unions are facing today. For Sherman and Voss (2000) innovative union tactics arise from a combination of three factors: a crisis within the local union, support from the international union and the presence of innovative staff from outside the labour movement. The first factor, a crisis within the local union, often stems from a shift in em- ployment patterns, with non-unionized immigrants increasingly replac- ing native-born, unionized workers. To deal with the problems this section of the working class faces, there is a need to shift away from traditional forms of organizing and to be open to new initiatives.

The innovative strategies in Finland involved a particular union club established for Estonian and Russian-speaking members in the early 2000s under the auspices of the Finnish Construction Workers’ Union. Similarly, in their study of Denmark, Norway and the UK, Eldring et al. (2012) list different innovative strategies used in each of their country cases. For example, in Copenhagen the association of local unions hired a Polish-born officer to assist Danish officers on matters related to mi- grant workers. At the same time, they established a Polish club in which meetings and social activities took place. As a next step, five more Polish-

2. In the case of cleaners in London, much of the effort for securing living wages, improved holiday and sick pay entitlements for foreign-born cleaners in Canary Wharf was kick-start- ed from outside the trade union movement by the East London Communities Organisation (TELCO).

speaking ‘consultants’ were hired to assist local unions. In Norway, in- novative methods included the temporary suspension of the rule that union assistance was provided only to those who were already members of the union until a foothold in migrant communities was gained. In the UK, methods involved site-based lay representatives approaching newly arrived workers, displaying the national Working Rule Agreement3 in larger site canteens, requesting the help of a TUC-seconded Solidarność national organizer for a short period of time, and conducting visits to workers’ homes in the Polish community (Fitzgerald and Hardy 2009; Eldring et al. 2012). In the case of the LA campaign, the unions also en- gaged in high-profile campaigning in the background (for example, J for J, that is, Justice for Janitors) involving college-educated organizers and researchers (Milkman and Wong 2000).

Alongside the features of a successful campaign listed above, one cannot underestimate the importance of unions’ media strategy. Public expo- sure through media channels has been an integral and difference-mak- ing element of many campaigns targeting LA janitors, drywall hangers in Southern California (Milkman and Wong 2000), London cleaners (Wills 2008), and Olkiluoto 3 construction workers (Lillie and Sippola 2011).

In other words, union strategies have aimed to increase union leverage by trying to gain an organizational critical mass (leadership-based com- bined with bottom-up), support from the public at large (multi-stake- holder or holistic approach and media), and access to management (zon- al approach and media). Innovative approaches have been commonly used in all cases, as individual cases have demanded tailored strategies specific to the needs of each case.

3.    Posted workers and union strategies

In all the case studies discussed in the chapter, local trade unions have worked to mobilize the non-local workforce. Certain elements of the aforementioned general recruitment strategies targeting migrant work- ers can be employed when organizing posted workers. However, there

3. The Construction Industry Joint Council Working Rule Agreement: a norm-establishing agreement even for enterprises that have not signed this agreement, it resembles the type of agreements common in Scandinavian countries.

are some specific aspects that must be taken into account when ad- dressing posted workers. In this section we present the approaches and initiatives used by local unions to mobilize posted workers in four con- struction sites in three geographical contexts: Finland (Olkiluoto 3), the Netherlands (Eemshaven) and the UK (Ferrybridge and Carrington).

All four cases involve the construction of power stations located in re- mote areas. The main contractors outsourced a considerable amount of the construction work, and thus worked with a large number of subcon- tractors, who in turn relied considerably on agency and posted workers. Although collective agreements are not recognized at the same levels in each country (at industry level in Finland and the Netherlands, but at workplace level in the UK), the cases that we will be discussing here all operate under collective agreements, and therefore recognize unions on site. It is worth noting that in the Finnish case, the French-German main contractor did not seek to follow the consensual Finnish model of in- dustrial relations – albeit recognizing the industrial agreements – as it did not start out from a position of cooperation with the Finnish unions, instead seeking to keep them off site and undermine their activities, while the unions gradually gained ground at the site through grassroots activity.

The initiatives taken by the unions to promote the unionization of posted workers can be grouped into the following four categories:

1. Making unions available for the workforce, that is, accessibility
2. Approaching the workforce directly, that is, pro-activity
3. Gaining the trust of the workforce, that is, trust-building, and
4. Cooperation with other stakeholders outside the workplace (includ- ing media coverage), that is, community outreach.

Union accessibility and availability have been considered as important organizing factors. In all four cases, unions were recognized on site. However, having a unionized workplace is not enough, as workers will not automatically become members. Posted workers have little or no knowledge of local unions and the way they operate and they usually have limited opportunities to leave the workplace and ‘look for them’. To facilitate contact and eventually recruitment, trade unions have tried to be literally there for the workers. At Olkiluoto shop stewards organ- ized site visits accompanied by a Polish-speaking union official; in Eem- shaven they established office hours on site, and in the UK, when there

is union recognition on site, there is a full-time office and two designated senior shop stewards available to workers at all times. Furthermore, union meetings were held in the on-site cafeteria during lunchtime, meaning that other workers could observe a union meeting without the pressure of membership and participation. In this way they were given the possibility of familiarizing themselves with the local unions’ actions and attitudes towards the issues of workers on site. At the same time, shop stewards were able to make contact and communicate with them directly.

In other words, availability was combined with a pro-active direct com- munication approach. Shop stewards approached the posted workers directly, trying to inform them about the unions, the benefits of mem- bership and how they could help them solve their problems. In Finland, union representatives went to the workers’ rest places on site and hung up posters with information on the basic conditions the workers were supposed to enjoy. In the Dutch case, unions distributed leaflets in the various languages of the workers and organized visits to their places of accommodation in an attempt to speak with them directly. In the Brit- ish case, the senior shop stewards managed to get themselves included in the induction process for new workers. Apart from health and safety induction, the collective agreement, local trade unions and their on-site services were introduced to new workers upon arrival. Union member- ship leaflets in different languages were also distributed on that occa- sion. For workers not speaking English, interpretation was provided. To encourage membership, the unions presented themselves as open-door service providers for handling individual grievances as well. As unions were only able to represent their members, posted workers first needed to become union members to address any specific grievance they might have, such as wrongful wage deductions.

Approaching workers is the first step towards gaining their trust. The shop stewards interviewed underlined the importance of such trust. Be- ing accessible, available and providing assistance in tackling individual grievances helps trade unions build up trust among the non-native work- force. However, the workers’ hyper-mobility works against the unions, as trust-building requires time and most of these workers only spend a relatively short period at each site. Limitations to communication due to the language barrier constitute a further obstacle to building trust. Em- ployers have been eager to provide interpreters, and in some cases, such as in the UK, the unions have reluctantly accepted though remaining

sceptical of this type of mediation, whereas in Finland the unions re- jected the offer, arguing that it would have jeopardized confidentiality and direct communication. To overcome the language barrier and gain the trust of the posted workers, the Dutch and UK unions channelled their efforts towards workers speaking the host language and enjoying the respect of their ethnic group.

In the Finnish case, there was a Metalworkers’ Union official who spoke Polish as a contact person. The Polish workers on site had greater con- fidence in a person who shared the same language and cultural back- ground. Moreover, Polish workers had already been supported by the Finnish Electrical Workers’ Union at a site close to Olkiluoto in a case which turned out a victory for the workers. This success was attributed to the union’s involvement and eventually helped boost the union’s cred- ibility in the eyes of the Polish workers in Finland. Similar examples of a trust-building strategy based on members of the ethnic group who spoke the host country’s language were found in all cases and it has already been successfully used by unions in their work with long-term migrants.

The final strategy used by unions to enhance their chances of success is that of cooperation. Cooperation is understood both as collaborating with other unions present on site (in sites where union membership is based on trades) as well as with other stakeholders outside the work- place. Collaboration with media and community organizations has given unions the opportunity to raise awareness at a larger scale and put pres- sure on management to comply with workers’ demands. This strategy reflects what Milkman and Wong (2000) call a ‘multi-stakeholder ap- proach’ and has similarities with what Wills (2008) call a ‘holistic, class- based approach’.

Collaboration among unions, community organizations and the media has strengthened them in their actions against the main contractor and their subcontractors by simple strength of numbers. The Finnish example is very telling: The Finnish Construction Workers’ Union tried to go it alone in defending Polish workers on site in 2007–2008, as did the Met- alworkers’ Union at a later stage, but their efforts were feeble. Eventually an ‘alliance’ between the Finnish Construction Workers’ Union, the Met- alworkers’ Union, the Electrical Workers Union and Trade Union Pro (the union for clerical workers) was set up, which seems to have been more successful in recruiting and representing non-Finnish workers com- pared to what each of them could do individually. Similarly, in the British

case GMB and UNITE work jointly on construction sites in representing engineering construction workers though there is little interaction with the unions representing civil construction workers, such as UCATT.

Furthermore, both the Olkiluoto and the Eemshaven sites have received a good deal of media coverage. In Finland as in the Netherlands, the media reported on the poor working and living conditions of non-native workers. Such media exposure mainly affects the main contractor and the client more than the subcontractors who usually hire the posted work- ers. However, to avoid bad publicity the main contractor will eventually pressure subcontractors to comply with the collective agreements in place, threatening them with contract termination. Fear of bad public- ity seems to have been effective not only in the Finnish and Dutch cases where the media were actually involved in exposing poor industrial rela- tions, but also in those where they were not. In the British case, unions pressure the clients and the main contractors to agree to union recogni- tion on site, and these efforts are particularly effective in cases where companies have actually suffered negative publicity on the poor treat- ment of posted workers in previous projects. Therefore, the threat and use of the media can be a major union instrument of leverage against employers.

4.    Challenges in organizing posted workers

The organizing of posted workers has been a far from easy task for the trade unions involved in our cases. At Olkiluoto 3 in 2009, when the construction works were at their peak, posted workers coming mainly from Poland comprised a third of a workforce of 4300, of whom just 100-200 were unionized; by the end of 2014 there were a few hundred Poles in the Finnish electricians’ union. In the Eemshaven sites in the Netherlands the unions managed to unionize 170 workers working for a Polish subcontractor. At Ferrybridge the number of unionized non-UK workers was slightly above 100 (that is, 15–20 per cent of the workforce), while no non-UK workers at Carrington joined the local unions, despite accounting for 20 per cent of the workforce in September 2014.

Numbers are small and vary from one period to the next because of the transient nature of the construction industry, that is, workers are hired for different spells ranging from a few weeks to several months. Work- ers hired via employment agencies might even just stay a few days. This

hyper-mobility of posted workers is the result not only of the transient nature of the construction industry but also of their posting situation (Berntsen and Lillie 2014). Time constraints inhibit the ability of unions to properly stand up for posted workers’ rights and leave workers with little possibility to familiarize themselves and engage with local unions. This in turn leaves room for social dumping and worker exploitation (Cremers et al. 2007). It is thus paramount for unions to circumvent such obstacles and to be able to offer representation and protection for such hyper-mobile posted workers.

In the four cases presented in the chapter, trade unions have tried to tackle these time constraints by making themselves available and acces- sible to workers at their place of work, that is, on the construction sites. Whether through site visits (in Olkiluoto), office hours (in Eemshaven) or a full-time open-door office (in Carrington and Ferrybridge), they have made themselves accessible to posted workers who might other- wise find it difficult to familiarize themselves with local trade unions and to individually contact them, especially considering that they work long hours and remain in the country only temporarily.

As workers show signs of scepticism and indifference towards union membership due to their presumably short-stay employment on site, unions have pro-actively approached them through opening direct com- munication channels, through which they introduce themselves and try to gain the trust of the workers. In the British cases, the unions were introduced to workers from the beginning through the induction process during which they met senior shop stewards, were informed about un- ions and the collective agreement and were offered support for address- ing individual grievances whenever they needed to. In Finland, posters with information on worker rights were hung up in the communal areas. In the Netherlands, workplace engagement was combined in some cases with visits to workers’ accommodation away from the prying eyes and intimidating looks of management, thus providing unions and workers with an opportunity to get to know each other. To expedite trust-build- ing, unions in all four cases started by approaching and mobilizing those members of the non-indigenous workforce who spoke the host country language and enjoyed the respect of their fellow countrymen.

Time constraints also force unions to attune their recruitment strate- gies more towards posted workers, at times deviating from those used for long-term migrants. In the Norwegian case (Eldring et al. 2012), we

saw local union offices temporarily suspending the rule obliging workers to become members prior to being offered assistance by the unions. With the removal of the pressure to join and the financial burden that might come with the payment of membership fees (especially for people employed in low-paid jobs), migrant workers would be more likely to approach unions. In the long run, leniency regarding the membership conditionality gives unions the possibility to gain migrant workers’ trust and eventually persuade them to become members. But the same strat- egy would be counter-productive in the case of posted workers, because once they receive union services without a prior commitment they may lose the incentive to join. In fact, presenting unions as service providers has helped local unions recruit posted workers.

The hyper-mobility of these workers is combined with a rather limited jurisdiction of the trade unions on the construction sites. For example, in the UK case, the elected shop stewards were only able to represent workers hired by the same company. The shop stewards, therefore, had no jurisdiction over other companies. There were two senior shop stew- ards on site (one representing each of the collective agreement signa- tory unions), both of whom had the right to represent all workers within the engineering construction covered under the National Agreement for the Engineering Construction Industry (NAECI), but had no jurisdiction over the civil construction workers, who were covered by another collec- tive agreement and a different trade union altogether.

In the case of Olkiluoto 3 in Finland, we found shortcomings in the coordination between the central and local levels. There were mutual misunderstandings as both the local union and the posted workers were pushing in separate ways. In August 2007, Polish members of the Finn- ish Construction Workers’ Union (Rakennusliitto) employed by an Irish- based employment agency appealed to the union with regard to collec- tive agreement and labour code violations. However, the shop stewards and Rakennusliitto regional office did not take any action, leading to the workers appealing to the Warsaw office of the European Migrant Work- ers Union (EMWU). The EMWU contacted Rakennusliitto and offered to assist in organizing the Olkiluoto construction site by, for example, supplying Polish-speaking organizers, but Rakennusliitto rejected this proposal. Despite this shortcoming, impromptu industrial action was taken by the Poles and Rakennusliitto issued a strike threat. In the ensu- ing situation, both Rakennusliitto and the Polish workers were involved in negotiations with the employer on unexplained deductions for social

security contributions to a third country. However, the results of the negotiations proved unsatisfactory for the Poles, leading to most of them resigning from the union.

Misunderstandings have also been influenced by a general scepticism to- wards trade unions based on previous bad experiences or perceptions of trade unions in the country of origin. Spanish workers at both UK sites, for example, reportedly showed dissatisfaction with unions back home, labelling them as corrupt, while Polish workers refused to comment and kept on saying that they ‘did not need them’ or ‘did not want anything to do with them’ when asked about unions. Moreover, the language barrier remains an impediment in all countries despite efforts to facilitate com- munication through leaflets translated into the languages spoken by the workers on site, as well as the use of the services of an interpreter, most of the time one of the workers who spoke either the language of the host country, that is, Dutch or Finnish, or English.

Management has played their part in minimizing posted workers’ in- volvement with the unions. Worker hyper-mobility is not so much based on the will of workers to move from one workplace to another as on the transient nature of the construction industry and the fact that many posted workers are transferred regularly by their employers. In the Dutch case, when efforts were made to organize posted workers, the unions returned to the site only to find that some of the people they had been talking with had left, either returning to their country of origin or moving on to the next job. Similarly, in the British cases civil construc- tion workers might only be working on site for a few days at a time, and, especially when hired via employment agencies, one could never know if they would show up the next day. In one particular episode, posted workers reported unequal pay to the senior stewards. The complaint was brought to the attention of the main contractor, yet the following day the workers were gone, with a resignation letter emailed to the unions. The stewards suspected the workers had either been threatened or paid off, given the fact that it was the workers themselves who brought the issue to the unions’ attention. In the case of the posted workers at the Car- rington site, workers said they could not stay for more than six month for ‘tax purposes’. EU legislation indeed states that if a person works for more than six months in a country other than the one where they pay their social security contributions, they are obliged to pay these contri- butions for that period in that country. However, management decisions to move people out every six months and bring in new ones to complete

the work creates the impression among workers that they are staying for too short a time to become affiliated with the local unions.

Another management strategy involves the separation of non-indigenous workers from the local workforce on site and in their accommodation. In the Finnish case, the posted workers lived in a ‘barrack village’ near the site with little interaction with the local community. In one of the British cases, each company was given its own space for tea and lunch breaks, with entrance reserved solely for employees of that contractor, thus allowing workers to interact only with their fellows at subcontrac- tor level and having little contact with anyone else. Employers may thus seek to utilize labour control strategies (Peck 1992) combining a variety of labour processes, labour supply and patterns of control within a single site. As a result, the possibility for direct communication with unions and other workers is reduced, negatively affecting workers’ efforts to fa- miliarize themselves with organized labour on site and for unions to gain their trust.

Conclusion

The first important lesson drawn from the academic literature on mi- grant workers and unions is that such workers (whether long-term or temporary posted) can no longer be ignored by trade unions. The overall decline in union membership and increase in precarious employment, with work often outsourced to companies relying on migrant labour, make the latter important for union revitalization. One can already learn from the existing literature on successful migrant organizing campaigns that to protect workers’ rights unions need to maintain or increase their critical organizational mass and gain the support of other stakehold- ers and broader society. In order to achieve their goals, holistic, zonal and multi-stakeholder approaches in association with strong leadership commitment and bottom-up initiatives are a good combination for or- ganizing worksites with a high number of migrants. Bundling strategies and incorporating innovative and context-specific initiatives further in- crease the chances of success.

Secondly, as shown in the four case studies – Carrington, Eemshaven, Ferrybridge and Olkiluoto – discussed in this chapter, the approaches used to organize migrant workers are suitable to a certain degree when it comes to posted workers. Their hyper-mobility however calls for

strategies and initiatives taking into account their relatively short-term service-based employment situation. In our four cases, unions have pro- actively addressed the issue of time and mobility by becoming more ac- cessible for the workforce on site, by initiating direct communication within and outside the workplace, by gaining the trust of the workers through direct communication and the mediation of one or more of their peers, and by cooperating with other organizations, the broader com- munity and the media.

Their success has, nevertheless, only been modest. The number of post- ed workers joining local trade unions has been relatively small, and in one case (in Carrington) none of them joined despite union efforts. In the other cases, success was short-lived, as union members would move away a few months after becoming members, often to another country, thereby interrupting their relationship with the local union. Once work- ers are out of the country, local unions can no longer help them, as their jurisdiction is bound to national boundaries (see for e.g. Berntsen and Lillie 2014; Lillie and Sippola 2011; Lillie and Greer 2007). Although mi- grants can take traditions of militancy developed in one place to new places (Rainnie et al. 2009), their continuous cross-border hyper-mo- bility presents a challenge for spatially-bound organizational practices.

In response to the new labour relations configurations transcending na- tional borders, a transnational union representing workers across na- tional borders within the European Union has been proposed (see also Rainnie et al 2009; Lillie and Greer 2007; Turnbull 2006). The idea has already been tested with different results. For example, the European Migrant Workers Union (EMWU) was founded in 2004 under the aus- pices of IG Bau in Germany. Unfortunately, the project, which aimed at representing workers from CEE states, did not prove successful (Greer et al. 2011). Despite its transnational mission, the main reasons the project failed were the participating unions’ insistence on defending existing ju- risdictions, the slowness of inter-union cooperation, and its own inabil- ity to develop into an independent organization. Another multilateral project, the European Construction Mobility Information Network (EC- MIN), established in 2009 and coordinated by the European Federation of Building and Woodworkers (EFBWW), has proven to be more sustain- able. Its mandate, however, is limited to the provision of information on working conditions. Nevertheless, the initiative has been applauded as a way to build national union support for a transnational approach (ibid). More than ten years have passed since the establishment of the EMWU,

and that time might have helped change the attitudes of national unions and make them more open to transnational cooperation.

Unions’ experience with migrant workers shows that in order to successfully organize them it is necessary to bundle different approaches and find innovative ways of recruiting them. Although none of the cases studied perfectly met the tremendous challenge of organizing posted workers in particular, and transient labour in general, the comparative analysis of the various innovative approaches of the national unions to recruiting the workers can provide us with ‘tools’ for unions to be used in other national contexts as well as at the transnational level. The ‘tools’ we suggest to be benchmarked across European borders are crystallised in four keywords: accessibility, proactivity, trust-building and cooperation.

References

Adler L.H., Tapia M. and Turner L. (eds) (2014) Mobilizing against inequality: un- ions, immigrant workers, and the crisis of capitalism, Ithaca, Cornell University Press.

Berntsen L. and Lillie N. (2014) Hyper-mobile migrant workers and Dutch trade union representation strategies at the Eemshaven construction sites, Economic and Industrial Democracy. doi: 10.1177/0143831X14537357

Castles S. and Kosack G. (1973) Immigrant workers and class structure in Western Europe, London, Oxford University Press.

Cremers J., Dølvik J.E. and Bosch G. (2007) Posting of workers in the Single Market: attempts to prevent social dumping and regime competition in the EU, Industrial Relations Journal, 38 (6), 524–541.

Dølvik J,E. and Visser J. (2009) Free movement, equal treatment and workers’ rights: can the European Union solve its trilemma of fundamental principles, Industrial Relations Journal, 40 (6), 491–509.

Eldring L., Fitzgerald I. and Arnholtz J. (2012) Post-accession migration in construction and trade union responses in Denmark, Norway and the UK, European Journal of Industrial Relations, 18(1), 21–36.

Fitzgerald I. (2010) Improving best practices on the working and living conditions of posted workers: country report UK, National expert report for the European Federation of Building and Woodworkers.

Fitzgerald I. and Hardy J. (2009) ‘Thinking outside the box’? Trade union organis- ing strategies and Polish migrant workers in the United Kingdom, British Journal of Industrial Relations, 48 (1), 131–150.

Frege C. and Kelly J. (eds) (2004) Varieties of unionism: strategies for union revi- talization in a globalizing economy, Oxford, Oxford University Press.

Greer I., Ciupijus Z. and Lillie N. (2011) Organising mobile workers in the enlarged EU: a case study of the European Migrant Workers Union, CERIC Working Paper 13, Leeds, Centre for Employment Relations Innovation and Change.

Hardy J., Eldring L. and Schulten T. (2012) Trade union responses to migrant workers from the ‘new Europe’: a three sector comparison in the UK, Norway and Germany, European Journal of Industrial Relations, 18 (4), 347–363.

Heberle R. (1955) Types of Migration, Southwestern Social Science Quarterly, 36 (1), 65–70.

Holgate J. (2005) Organising migrant workers: a case study of working conditions and unionization at a sandwich factory in London,Work, Employment and Society, 19 ( 3), 463–480.

Hyman E. (2001) Understanding European trade unionism: between market, class and society, London, Sage Publications.

Krings T. (2009) A race to the bottom? Trade unions, eu enlargement and the free movement of labour, European Journal of Industrial Relations, 15 (1), 49–69.

Lillie N. and Greer I. (2007) Industrial relations, migration and neoliberal politics: the case of the European construction sector, Politics & Society, 35 (4), 551–581.

Lillie N. and Martínez Lucio M. (2004) International trade union revitalization: the role of national union approaches, in Frege C. and Kelly J. (eds) Varieties of unionism: strategies for union revitalization in a globalizing economy, Oxford, Oxford University Press, 159–180.

Lillie N. and Sippola M. (2011) National unions and transnational workers: the case of Olkiluoto 3, Finland, Work, Employment and Society, 25 (2), 292–308.

Martinez Lucio M. and Perret R. (2009) The diversity and politics of trade unions’ responses to minority ethnic and migrant workers: the context of the UK, Economic and Industrial Democracy, 30 (3), 324–347.

Milkman R. and Wong K. (2000) Organising immigrant workers: case studies from Southern California, in Turner L., Katz H.C. and Hurd R.W. (eds) Rekindling

the movement: labor’s quest for relevance in the 21st century, Ithaca, Cornell University Press, 99–128.

Peck J. (1992) Labor and agglomeration: control and flexibility in local labor markets, Economic Geography, 68 (4), 325–347.

Penninx R. and Roosblad J. (eds) (2000) Trade unions, immigration and immi- grants in Europe 1960–1993: a comparative study of the actions of trade unions in seven West European countries, New York, Berghahn Books.

Rainnie A., McGrath-Champ S. and Herod A. (2009) Geography and worker or- ganisation: do spatial approaches change our theories of union praxis?, Paper presented at the 15th ILERA Congress, Sydney, 24–28 August, 2009.

Schelkle W. (2011) Reconfiguring welfare states in the post-industrial age: what role for trade unions?, European Journal of Industrial Relations, 17 (4), 301–314.

Sherman R. and Voss K. (2000) ‘Organize or die’: labor’s new tactics and immi- grant workers, in Milkman R. (ed.) Organizing immigrants: challenge for unions in contemporary California, Ithaca, Cornell University Press, 81–108.

Turnbull P. (2006) The war on Europe’s waterfront: repertoires of power in the port transport industry, British Journal of Industrial Relations, 44 2), 305– 326.

Wills J. (2008) Making class politics possible: organising contract cleaners in Lon- don, International Journal of Urban and Regional Research, 32 (2), 305–323.

Wrench J. (2000) British unions and racism: organisational dilemmas in an unsym- pathetic climate, in Penninx R. and Roosblad J. (eds) Trade unions, immigration and immigrants in Europe 1960–1993: a comparative study of the actions of trade unions in seven West European countries, New York, Berghahn Books.

Wrench J. (2004) Trade union responses to immigrants and ethnic inequality in Denmark and the UK the context of consensus and conflict, European Journal of Industrial Relations, 10 (1), 7–30.

Chapter 11

From concession bargaining to broad workplace solidarity: the IG Metall response to agency work

Chiara Benassi

1.    International competition and company-level concessions

Market integration at a global level has opened new opportunities for companies to expand into new markets and to shift their production abroad, leading to growing international competition (Buckley and Ghauri 2004; Rees and Edwards 2010). The resulting pressure on costs has affected the dynamics of collective bargaining, especially in export- oriented manufacturing companies in Europe and other highly devel- oped countries where labour bargaining power has increasingly come under pressure (Tilly 1995; Raess 2006; Greer and Hauptmeier 2008).

Leeway for bargaining high salaries and good working conditions has inevitably declined as a result of employers’ focus on cutting labour costs to improve company competitiveness in international markets (Sisson et al. 2003; Anner et al. 2006). Moreover the public discourse is put- ting further pressure on labour’s attempts to resist the deterioration of workers’ standards (Piotti 2009). Unions have been publicly accused of endangering the jobs that companies threatened to outsource if condi- tions for improving a plant’s flexibility and cost competitiveness were not met. For instance, the German metal union IG Metall was portrayed as a ‘job killer’ when it initially rejected a management proposal to cre- ate 5,000 new jobs at lower wages and working conditions than those enjoyed by the existing workforce (Greer and Hauptmeier forthcoming). Similarly, the Italian metal union FIOM was accused of being too ‘ideo- logical’ when it refused to sign agreements reducing breaks and increas- ing overtime without compensation for the whole workforce at the FIAT plants in Pomigliano and Mirafiori in 2010 (La Repubblica 2010).

As a result of this pressure to improve competitiveness, bargaining con- cessions have been spreading across Europe. Under the threat of closing

down plants or outsourcing production units, labour representatives have agreed to lower standards for wages and working conditions in or- der to save production sites (Hancké 2000; Pulignano 2006; Greer and Hauptmeier 2008; Greer and Hauptmeier forthcoming). Typically, the core workforce has taken pay cuts or accepted working time reductions or overtime without compensation as in the cases of Opel in Germany (Schulten 1998), PSA Peugeot-Citroën in France (Turlan 2012) and SEAT in Spain (Albarracín 2005). However, concession bargaining has not only affected the working conditions of the core workforce, as labour representatives have also agreed to concessions at the expense of new hires or the peripheral workforce. Such concessions included the dual- tier wage systems for new hires at VW in Wolfsburg in the early 2000s (Dribbusch 2004) and at the US auto suppliers Delphi and Visteon (Slaughter 2003); the outsourcing of (generally lower-end) production units to subcontractors without collective agreements as in the German automotive, chemical and telecom industries (Doellgast and Greer 2007; Helfen 2011) and the employment of a cheap and flexible workforce in the German and Korean automotive industry (Jürgens and Krzywdzin- ski 2006; Yun 2011; Benassi 2013).

This chapter will focus on concessions on the use of peripheral workers, examining in detail labour strategies towards agency workers at compa- ny and sectoral level in the German metal sector. It has been argued that these concessions constitute an instrument protecting the standards of the core workforce against increasing competitive pressure. Through agreements at the expense of the peripheral workforce unions have, on the one hand, managed to maintain their bargaining power over the standards of their core members; on the other hand, they have also con- tributed to the dualization of labour markets (Lehndorff 2012; Hassel 2014; Eichhorst 2015). However, the findings presented in this chapter suggest that these concessions backfire on the core workforce in the long run because the increasing presence of a cheap and flexible workforce at company level threatens collectively agreed standards and the future existence of a permanent workforce. German unions have thus recently started resisting employers’ segmentation strategies, often justified by reference to international competition pressure, and have instead adopt- ed inclusive strategies towards the peripheral workforce. In particular, this chapter looks at the campaign IG Metall launched in 2008 to recruit agency workers and bargain on their behalf.

This chapter is structured as follows. The next section discusses the relevance of studying the German case from a research and policy per- spective. The third section illustrates the spread of concession bargain- ing on agency work in Germany and its consequences for labour bar- gaining power. The fourth and fifth sections illustrate the response of IG Metall and the works councils to the growth of contingent work. The empirical evidence is mainly based on the analysis of union reports and documents and on fieldwork conducted in the German metal sector be- tween 2011 and 2013, which included interviews with human resource managers and labour representatives. The last section summarizes the findings and draws conclusions for labour strategies.

2.    The relevance of the German case

The German metal sector represents a particularly relevant case for studying the spread of concession bargaining at the expense of the pe- ripheral workforce. First, works councils and unions in the sector benefit from an institutionalized bargaining position and from the most devel- oped codetermination rights at workplace level in Europe. Neverthe- less, German labour representatives were not fully able to counteract the spread of concessions despite such formally strong bargaining in- stitutions (Bispinck and Dribbusch 2011). This phenomenon reveals the great pressure on industrial relations in Europe stemming from neolib- eral policies and international competition.

Second, the IG Metall campaign for agency workers shows that tradi- tional collective bargaining institutions need to be re-interpreted and revitalized in the face of new challenges and under the pressure of em- ployers’ casualization strategies (Greer 2008; Turner 2009; Benassi and Dorigatti 2014). The case of the German metal sector suggests that concession bargaining at the expense of the peripheral workforce erodes overall labour bargaining power in the long run despite the continuing existence of formal institutions.

Third, the growth of contingent work in the German metal sector and the resulting union reaction have shown that the competitiveness of the internationally acclaimed German export model has up to now depend- ed on labour concessions increasing inequality within the workforce (Lehndorff 2012; Baccaro and Benassi 2014). These findings raise some scepticism about the recent proposals to export the ‘new’ German model

of industrial relations, especially decentralized bargaining, to Southern European countries in order to improve their economic performance (Anderton et al. 2012; Dustmann et al. 2014).

3.    The German case: eroding labour power through company-level pacts for competitiveness

3.1                      Seeking competitiveness in a fragmented industrial relations landscape

The German economy has always relied on its export-oriented manufac- turing sector. Research in the eighties and early nineties showed that la- bour costs were partly protected from international market competition thanks to German companies being specialized in upper-end market segments (Sorge and Streeck 1987; Streeck 1991). Since the mid-nine- ties, however, employers have increasingly focused on reducing labour costs in the face of upcoming competitors – such as US and Japanese manufacturers – starting to produce for high quality markets at lower prices than German companies thanks to the adoption of new technolo- gies and lean management strategies (Herrigel 1997; Jürgens 2004; Herrigel 2014).

In response to the cost pressure, German employers heavily restructured manufacturing production, especially with regard to integrating the value chain. Production was organized into modules which could be performed outside the core company by subcontractors (Jürgens 2004; Doellgast and Greer 2007; Greer 2008); or even outsourced to cheaper production sites, particularly in Eastern Europe (Kinkel and Lay 2003; Jürgens and Krzywdzinski 2006). Indeed, the Fraunhofer Institute reports that over 40 per cent of companies in core manufacturing sectors outsourced part of their production abroad between 1999 and 2001, mainly with the aim of reducing production costs (Kinkel and Lay 2003: 4).

These trends led to a public debate on the future of Germany as a pro- duction location (Standortdebatte), contributing to a consensus on the necessity to cut labour costs in order to re-gain competitiveness and helping legitimize the introduction of opening clauses at workplace level (Upchurch 2000: 76; Silvia 2010: 223). In this context works councils were made co-responsible for maintaining the investments in their plant and pushed to sign workplace agreements amending sectoral standards.

These so-called Pacts for Employment and Competitiveness (PECs) quickly spread across sectors and soon became an institutionalized in- strument for co-management (Hassel and Rehder 2001; Rehder 2003). These workplace agreements mainly included concessions on wages, working time and other working conditions, and the use of temporary work and subcontractors (Jürgens and Krzywdzinski 2006; Haipeter 2011).

3.2                      The consequences of concessions at the expense of the peripheral workforce1

Since the mid-nineties use of contingent work and subcontractors has spread under the tacit consensus of works councils attempting to use temporary workers as a cheap flexibility buffer, while protecting the standards of the core workforce from competitive pressure. A survey conducted among works councillors for an IG Metall internal research project in Berlin-Brandenburg-Saxony reported that the vast majority of the interviewed works councillors saw agency workers as an instru- ment for protecting the standards of the core workforce (IG Metall study reported in Aust et al. 2007: 263). In a similar survey conducted among works councillors in the metal sector, 43% of the them agreed on us- ing agency work as a flexibility buffer and only one out of three pur- sued equal pay and equal treatment as bargaining aims (Wassermann and Rudolph 2007: 15–24). As a result, provisions for agency workers in company-level agreements tended to shift costs from core to agency workers, boosting workforce segmentation in terms of working time (for example, overtime and unsocial working hours) and pay (for example, no access to company-level bonuses and benefits) (Zumbeck 2009: 15–40).

The attitude of works councils and unions towards agency workers start- ed changing a few years after the implementation of the Hartz reforms which deregulated the use of agency work. These lifted most limitations on the use of agency work: companies could hire workers on agency con- tracts without specifying the reason for the fixed term and without offer- ing any guarantee of a permanent job afterwards. Dismissal protection was lowered as agencies could employ workers on contracts lasting only

1. This section relies on the article published in the British Journal of Industrial Relations by Benassi and Dorigatti (2014).

until the end of their assignment at the hiring company. The equal pay principle, though existing in law, could be amended by collective agree- ment (Bundesagentur für Arbeit 2011). Due to unfair competition from the Christian union (whose collective agreements were declared unlaw- ful by the labour court in 2013), the DGB special bargaining body for agency workers signed a collective agreement setting wages at a much lower level than in the metal sector.

In 2011 the German government changed existing legislation to comply with the Directive on Agency Work, approved by the European Parlia- ment in 2008. The resulting legislation introduced a minimum wage for agency workers in order to limit potential wage dumping due to the in- flow of migrant workers from Eastern Europe after the barriers to the German labour market fell in 2011. However, the amended law still al- lows exceptions to the equal pay principle through collective agreements. Furthermore, it does not regulate the transition from an agency contract to a permanent position, stating simply that agency workers should be employed only temporarily (vorübergehend). Unfortunately, this vague term can be loosely interpreted by employers (Frankfurter Rundschau 2013; IG Metall 2011).

Against this deregulation background, employers’ use of agency work increased dramatically – Baccaro and Benassi (2014) used IAB data to calculate that in 1996 agency workers constituted, on average, 2% of the workforce in 45% of companies in core manufacturing sectors. In 2008, however, over 90% of these companies employed an average of 8% of their workforce on agency contracts (p. 361), with peaks of even 20–30% in big manufacturing companies (Bellmann and Kuehl 2007; Holst et al. 2009; Benassi 2013).

Because of this change in employers’ casualization strategies, IG Met- all started perceiving agency work as an attempt of ‘conscious creation of cheap workforce’ (IG Metall 2007: 23 cit. in Benassi and Dorigatti 2014) with the potential to threaten the standards of the core workforce in the long run. The interviews conducted with union representatives and works councillors and the analysis of IG Metall reports (IG Met- all 2007, 2009) suggest that the performance of core workers is often benchmarked against agency workers who tend to work harder and at a faster pace because they want to be hired permanently. At the same time, however, they are cheaper and therefore their presence allows employers to question the existing standards of the core workforce. Furthermore,

the increasing presence of agency workers limits labour mobilization strength: first because core workers are afraid of being substituted; and secondly because it is difficult to organize agency workers to go on strike. Finally, the intensive use of agency work reduces the overall impact of strikes as it can prevent production from stopping. This happened for instance in the Lower Saxony machine tool manufacturing plant of Atlas where the company used agency workers to maintain production dur- ing strikes. IG Metall managed to stop this practice in 2010 (IG Metall 2010).

4.    Extending solidarity to regain bargaining power

Until the mid-2000s IG Metall unsuccessfully tried to organize workers within their staff agencies (Weinkopf and Vanselow 2008: 26), leaving the regulation of agency work to the bargaining abilities of company- level works councils and avoiding any direct intervention. This strategy, however, proved to be detrimental because works councils agreed to the use of a cheap peripheral workforce under pressure to cut labour costs (Benassi and Dorigatti 2014: 12).

The ‘Same Work, Same Wage’ campaign started in 2008 as a response to the dramatic growth of agency work following the Hartz reforms. Its goal was to unionize agency workers and to set homogenous and fair standards for them at sectoral level. To this aim, it included initiatives both at company and sectoral level, supported by a campaign targeting national media.

The media campaign was oriented both towards the members of the la- bour organization and the wider public. First, it made union representa- tives and works councils aware that regulating agency work was a prior- ity for the union and had to be included in the bargaining goals of both works councils and union officials. Furthermore, an interviewed IG Met- all official from Berlin-Brandenburg suggested that the public shaming of employers’ casualization strategies helped works councils to discuss more openly their ambiguous attitude towards the peripheral workforce, as they could denounce the heavy pressure on cutting labour costs they were exposed to.

Second, the campaign wanted to raise awareness in the public at large towards the working conditions of agency workers in order to increase

unions’ leverage in the coming bargaining rounds and to lobby politicians to improve legal regulation. It presented agency work as a new employer instrument for exploiting workers, with potential negative consequences for German society in general in the long run. The campaign included a truck that toured different German cities, postcards and e-cards which could be sent to members of local and national parliaments to lobby for better regulation of agency work, and several billboards highlighting the gap in wages and working conditions between agency workers and regu- lar employees. Furthermore, IG Metall produced leaflets and brochures in which works councillors and the workers themselves reported on their direct experiences with agency work.

At company level, IG Metall tried to change how works councils under- stood and performed their workplace representation mandate. It pro- vided specific training workshops for works councils to raise their aware- ness of the issue of agency work and provide them with information on the legal framework of agency work and their codetermination rights in this regard. Through the seminars and media campaign works councils were also encouraged to actively engage in the issue of agency work at company level. Furthermore, IG Metall started intervening in support of works councils in difficult bargaining rounds on agency work. These interventions were usually linked to the sectoral campaign in order to increase pressure on management. For example, the works council of the BMW plant in Munich requested the support of IG Metall during the company-level negotiations on equal pay because management was re- fusing to reach an agreement. The local IG Metall threatened to position a truck with a sign reading ‘BMW-Temple of Slave Labour’ in front of BMW Welt, the company’s museum. Afraid of the potential damaging ef- fect this initiative could have on BMW’s reputation, management agreed to pay agency workers in accordance with the metal agreement (Benassi 2013). Similarly, IG Metall support helped the Siemens general works council to successfully re-establish equal pay for agency workers in 2012 after the company had announced its intention to withdraw its com- mitment, for example, by organizing protests outside Siemens plants in Frankfurt (IG Metall 2012a).

At sectoral level, IG Metall focused on recruiting agency workers. As part of the campaign it stressed the advantages of joining the union, includ- ing financial support in case of accidents and legal counselling, avail- able also through a hotline set up specifically for agency workers. Above all, IG Metall included the issue of agency work in the metal bargaining

rounds. First, IG Metall advocated equal pay for agency workers so that companies in the metal sector would pay agency workers in accordance with the metal agreement rather than the agreement reached between the DGB bargaining body and the associations of staff agencies. Second, IG Metall pushed for limiting the contract duration of agency work- ers, who were to be guaranteed permanent hiring after a certain period of time. Third, IG Metall advocated the expansion of codetermination rights to cover the use of agency work.

5.    The outcomes of the campaign

The outcomes of the campaign were positive. At company level works councils actively worked to achieve agreements gaining better working conditions for agency workers – by 2011, over 1,200 companies were covered (IG Metall 2012c). Moreover, the works councils interviewed during the fieldwork had all managed to achieve agreements improving the wage levels and working conditions of agency workers, even though there were differences in the extent and coverage across plants. The Ford plant in Cologne was the plant with the most wide-reaching agreement as it included quotas, equal pay and specific rules regarding transition from a temporary to a permanent contract. As argued elsewhere (see Be- nassi 2013), the Ford works council was able to achieve this result pri- marily because bargaining took place in early 2003, i.e. before the Hartz reforms deregulating the use of agency work came into force. The other plants, by contrast, only managed to regulate agency work several years after the Hartz reforms, also thanks to the IG Metall campaign.

The threat of outsourcing to Eastern Europe played a major role at the BMW plant in Leipzig, which had the highest rate of agency workers (around 30%). In this plant, the works councillor interviewed consid- ered the concessions on the use of contingent workers and subcontrac- tors inevitable as they constituted a pre-condition for setting up the plant in Germany rather than in cheaper production sites in Eastern Europe. Moreover, the BMW labour representative in Munich suggested that works council concessions on labour standards might affect relocation decisions; at the same time, however, he believed that these decisions were also ‘about politics’ rather than exclusively efficiency calculations. The works councillors at VW in Wolfsburg and at Ford in Cologne were even more confident that the use of agency work and the following re- regulation would not impact employers’ relocation plans. While the VW

works councillor believed that the Wolfsburg plant was not in any im- mediate danger of relocation at the time of the interview (maybe because of the political and strategic importance of the Wolfsburg headquarters), the Ford works councillor was sceptical about the council’s ability to in- fluence managerial decisions in this respect:

‘Management has always said that the plant is in danger, that business units need to be outsourced…Whether they do it or don’t, whether they outsource, is not something we need to think too much about. If a company has taken the strategic decision to set up plants in Eastern Europe or Asia, they’ll do it. […] This is why I’ve never been particu- larly impressed by this management rhetoric.’ (Ford works council- lor, 19.04.2012 cit. in Benassi 2013: 24)

In the plants in Cologne, Munich and Wolfsburg, the works councillors mainly highlighted the importance of concessions on internal flexibility (for example, the use of working time accounts in times of low demand) because the cooperation and technical skills of works councils consti- tuted a source of comparative advantage for the company in compari- son to other production sites abroad. At the same time, however, they suggested that employers were taking advantage of the lack of regula- tion of contingent work and of works council weakness in counteracting managerial decisions because they were more ‘blackmailable’ than the union which was able to represent workers’ interests as an independent political actor. Therefore, they welcomed the initiative of IG Metall to re-establish sectoral regulation of agency work and standards of wages and working conditions.

At sectoral level, IG Metall achieved major recruiting gains, with over 50,000 agency workers joining the union by 2012 (IG Metall 2012c). It also achieved important contractual results: In the steel sector an in- dustry-wide agreement on equal pay for agency workers was achieved in September 2010. In May 2012 the new collective agreement for the metal and electronics industry was signed. This strengthened co-deter- mination rights in the hiring companies, as works councils were now able to bargain with management over the justification of hiring agency workers and their period of assignment. Moreover, it set rules ensuring that agency workers were hired permanently after a certain period of time: it is encouraged after 18 months of continuous assignment and be- comes compulsory after 24 months. In the same year IG Metall negotiat- ed an agreement with the employer associations of staff agencies setting

wage bonuses for agency workers, with the aim of reducing the wage gap between agency workers and permanent workers. 15% additional wag- es were to be paid after six weeks of continuous assignment, gradually increasing to 50% after nine months (IG Metall 2012b). Furthermore, IG Metall negotiated the new collective agreement in the agency sector together with the other sectoral DGB unions, which set a sectoral mini- mum wage of 8.50€/h and a 9.6% increase in wage levels in Western Germany and 12.8% in Eastern Germany (IG Metall 2013).

6.    Conclusion: the need to focus on fairness and solidarity

In Germany concession bargaining spread as a consequence of increas- ing international competition and the weakening of sectoral bargaining structures. In this hostile context, works councils initially tried to pro- tect, at least to a certain extent, the standards of the core workforce by al- lowing management to employ increasing shares of contingent workers. These workers are usually employed at lower wage rates and constitute a convenient flexibility buffer for the company because they can be easily dismissed. However, this bargaining strategy did not prove to be effec- tive and sustainable in the long run in protecting the jobs of core work- ers from outsourcing or cost-cutting, especially in a context of labour market deregulation. Indeed, both works councils and unions reported that the increasing presence of a cheap workforce was slowly eroding collectively agreed standards and labour’s bargaining power, contribut- ing to a downward spiral which the concessions had originally aimed to avoid. For this reason, IG Metall started including agency workers in its bargaining scope and offering support to works councils in an attempt to avert employer pressure for concessions.

Even though these findings rely solely on the German metal sector, they provide important lessons for labour in other countries. Increasing pres- sure on workplace employment relations is a phenomenon affecting all European countries (Keune 2011; Marginson 2015). Similarly, labour market reforms have all focused on improving labour market flexibility in the last twenty years, meaning that contingent workers are increas- ingly employed across skill levels and occupations (Rhodes 1997; Gebel and Giesecke 2011; Marx 2011). To overcome these challenges, unions need to open their representation domain to new workforce groups, as a united labour front is of fundamental importance for reducing workforce

inequality. Besides the case of IG Metall, this is confirmed by recent empirical research comparing workers’ outcomes in the telecom sector across different European countries, which found that labour can play a fundamental role in reducing workforce inequality under cost-cutting pressure when it manages to adopt encompassing bargaining strategies (Doellgast et al. 2015; Benassi et al. 2015). Furthermore, bargaining strategies focused solely on maintaining standards for the core work- force were found to be unsuccessful in the long run, leading to an overall decline in wages and working conditions (Doellgast et al. 2015).

The findings therefore suggest that concessions at the expense of the peripheral workforce represent only a short-term response to the chal- lenges posed by the increasing pressure on workers’ representation and collective bargaining. Though it has been argued that concessions on working time and wages might contribute, in the short run, to prevent- ing plant relocations or closures – even though the evidence is mixed on the effectiveness of such concessions (Massa-Wirth and Seifert 2005; Chaison 2012) –, in the long run, however, they lead to deteriorating working conditions for the whole workforce (Pulignano 2014; Doellgast et al. 2015; Jürgens and Krzywdzinski 2008), and therefore do not rep- resent a sustainable strategy.

The key strategy to prevent concession bargaining seems to be to reverse the process of bargaining decentralisation, which allows employers to negotiate wages and working conditions both for permanent and tem- porary workers at workplace level, where labour is weaker. First, the labour movement should try to shift labour-management conflict from the workplace level to the sectoral and national level, as witnessed by IG Metall which engaged in sectoral bargaining on behalf of agency workers in order to re-establish a common bargaining floor for works councils. The campaign also made strategic use of the media, exposing employers’ casualization strategies to public discussion. By publicly shaming these practices, IG Metall put pressure on employers: the campaign helped re- frame the public discourse on flexibility, previously solely focused on the necessity of improving the competitiveness of Standort Deutschland.2 The campaign refocused public attention on the issue of fairness at work and of employers’ responsibility for negative societal outcomes such as

2. Germany as a production

increasing inequality. Both the union’s bargaining efforts and its active intervention in the public discourse were of fundamental importance in taking the pressure off works councils to maintain plant competitive- ness, allowing them to resist employers’ segmentation strategies.

Second, international, and particularly European, workers’ representa- tion bodies should play a more active role in setting common interna- tional standards, in particular regarding the regulation of contingent work, as workplace concessions (on external flexibility) are closely linked to increasing international competition and transnational com- panies’ strategies (Pulignano 2014, 2015). Of the automotive plants involved in the study, only Volkswagen’s International Works Council and European Works Council managed to negotiate with corporate man- agement the conclusion of the ‘Charta of Agency Work’. Its provisions on wages, training and the transition to permanent contracts apply to agency workers in all VW Group plants worldwide (IndustriAll 2012). The agreement will hopefully pave the way for further cross-national regulation at company level or even at sectoral level, helping to prevent the spread of concessions at the expense of the peripheral workforce by removing labour costs from competition.

Reference

Albarracín D. (2005) Workforce reduction plan agreed at Seat. http://eurofound.europa.eu/observatories/eurwork/articles/workforce- reduction-plan-agreed-at-seat

Anderton R. et al. (2012) Euro area labour markets and the crisis, Occasional Paper 138, Frankfurt am Main, European Central Bank.

Anner M. et al. (2006) The industrial determinants of transnational solidarity: global interunion politics in three sectors, European Journal of Industrial Rela- tions, 12 (1), 7–27.

Aust A. and Holst H. (2007) Moderner Sklavenhandel? Gewerkschaftliche Strat- egien im Umgang mit Leiharbeit, in Pernicka S. and Aust A. (eds) Die Unor- ganisierten Gewinnen, Berlin, Sigma, 231–312.

Baccaro L. and Benassi C. (2014) Softening institutions, hardening growth model: the transformation of the German political economy, Stato e Mercato, 34 (3), 369–396.

Bellmann L. and Kuehl A. (2007) Weitere Expansion der Leiharbeit? Eine Be- standsaufnahme auf Basis des IAB-Betriebspanels, Düsseldorf, Hans Böckler Stiftung.

Benassi C. (2013) Political economy of labour market segmentation: agency work in the automotive industry, Working Paper 2013.06, Brussels, European Trade Union Institute.

Benassi C., Doellgast V. and Sarmiento-Mirwaldt K. (2015) Institutions and in- equality in liberalizing markets: explaining different trajectories of institutional change in Social Europe Unpublished manuscript, London, London School of Economics and Political Science.

Benassi C. and Dorigatti L. (2014) Straight to the core – Explaining union responses to the casualization of work: the IG Metall campaign for agency workers, British Journal of Industrial Relations. doi: 10.1111/bjir.12079

Bispinck R. and Dribbusch H. (2011) Collective bargaining, decentralisation and crisis management in the German metalworking industries since 1990, WSI Discussion Paper 177, Düsseldorf, Institute of Economic and Social Research.

Buckley P.J. and Ghauri P.N. (2004) Globalisation, economic geography and the strategy of multinational enterprises, Journal of International Business Studies, 35 (2), 81–98.

Bundesagentur für Arbeit (2011) Zeitarbeit in Deutschland – Aktuelle Entwick- lungen. Stand Juli 2011, Nürnberg, Bundesagentur für Arbeit.

Chaison G. (2012) The new collective bargaining, New York, Springer. Doellgast V. and Greer I. (2007) Vertical disintegration and the disorganization of

German industrial relations, British Journal of Industrial Relations, 45 (1), 55–76.

Doellgast V., Sarmiento-Mirwaldt K. and Benassi C. (2015) Institutions, cost structures, and the politics of externalization: explaining variation in boundary strategies for call center jobs, Unpublished manuscript, London, London School of Economics and Political Science.

Dribbusch H. (2004) Agreements on cost-cutting and job security signed at Volkswagen. http://www.eurofound.europa.eu/eiro/2004/11/feature/ de0411203f.htm

Dustmann C. et al. (2014) From sick man of Europe to economic superstar: Ger- many’s resurgent economy, Journal of Economic Perspectives, 28 (1), 167–188.

Eichhorst W. (2015) The unexpected appearance of a new German Model, British Journal of Industrial Relations, 53 (1), 49–69.

Frankfurter Rundschau (2013) Richter begrenzen Zeitarbeit nicht, 10.12.2013. http://www.fr-online.de/wirtschaft/bag-urteil-richter-begrenzen-leiharbeit- nicht,1472780,25582804.html

Gebel M. and Giesecke J. (2011) Labor market flexibility and inequality: the changing skill-based temporary employment and unemployment risks in Eu- rope, Social Forces, 90 (1), 17–39.

Greer I. (2008) Organised industrial relations in the information economy: the German automotive sector as a test case, New Technology, Work and Employ- ment, 23 (3), 181–196.

Greer I. (2008) Social movement unionism and social partnership in Germany: the case of Hamburg’s hospital, Industrial Relations: A Journal of Economy and Society, 47 (4), 602–624.

Greer I. and Hauptmeier M. (2008) Political entrepreneurs and co-managers: labour transnationalism at four multinational auto companies, British Journal of Industrial Relations, 46 (1), 76–97.

Greer I. and Hauptmeier M. (forthcoming) Management whipsawing: the staging of labor competition under globalization, Draft accepted at Industrial and Labor Relations Review.

Haipeter T. (2011) ‘Unbound’ employers’ associations and derogations: erosion and renewal of collective bargaining in the German metalworking industry, Industrial Relations Journal, 42 (2),174–194.

Hancké B. (2000) European works councils and industrial restructuring in the Eu- ropean motor industry, European Journal of Industrial Relations, 6 (1), 35–59.

Hassel A. (2014) The paradox of liberalization: understanding dualism and the recovery of the German political economy, British Journal of Industrial Rela- tions, 52 (1), 57–81.

Hassel A. and Rehder B. (2001) Institutional change in the German wage bargaining system: the role of big companies, MPIfG Working Paper 01/9, Cologne, Max Planck Institute for the Study of Societies.

Helfen M. (2011) Tarifpolitisch Parallelwelten, Mitbestimmung, 7+8, 20–33. Herrigel G. (1997) The limits of German manufacturing flexibility, in Turner L.

(ed.) Negotiating the new Germany: can social partnership survive, Ithaca, Cornell University Press, 177–205.

Herrigel G. (2014) Globalization and the German industrial production model, Journal for Labour Market Research. doi: 10.1007/s12651-014-0170-5

Holst H., Nachtwey O. and Dörre K. (2009) Funktionswandel von Leiharbeit: Neue Nutzungsstrategien und ihre arbeits- und mitbestimmungspolitischen Folgen. Eine Studie im Auftrag der Otto Brenner Stiftung, OBS-Arbeitsheft 61, Frank- furt am Main, Otto Brenner Stiftung.

IG Metall (2007) Sozialreport Zeitarbeit, IG Metall Bezirksleitung Berlin-Branden- burg-Sachsen.

IG Metall (2009) Schwarzweißbuch Leiharbeit, Frankfurt am Main, IG Metall Vorstand.

IG Metall (2010) Einsatz von Leiharbeitskräften als Streikbrecher verhindert, 26.10.2010. https://www.gleichearbeit-gleichesgeld.de/initiative/aktuelle- meldungen/2010/arbeitskampf-bei-atlas-einsatz-von-leiharbeitskraeften-als- streikbrecher-verhindert/

IG Metall (2011) So kann der Missbrauch von Leiharbeit nicht gestoppt werden, 25.03.2011. http://www.igmetall.de/bundestag-abstimmung-ueber-arbeit- nehmerueberlassungsgesetz-7269.htm

IG Metall (2012a) Erst verlieren, dann verloren?, 30.10.2012. http://www.dialog. igmetall.de/artikel/datum/2012/10/30/titel/erst-verliehen-dann-verloren/

IG Metall (2012b) Mehr Rechte für Leihbeschäftigte und Betriebsräte, 21.05.2012. http://www.igmetall.de/tarifabschluss-sichert-faire-leiharbe- it-10140.htm

IG Metall (2012c) Über die Kampagne ‘Gleiche Arbeit – Gleiches Geld‘. https:// www.gleichearbeit-gleichesgeld.de/initiative/ueber-die-kampagne/

IG Metall (2013) Tarifnachrichten: Tarifergebnis für die Leiharbeit erzielt. http://igmetall.de/internet/IGM_FB_Leiharbeit_low_a2f3f11c584d34f2e- 280bee5879978a52cda05b9.pdf

IndustriAll (2012) Charta der Zeitarbeit im Volkswagen Konzern. http://www. industriall-union.org/sites/default/files/uploads/documents/GFAs/Volk- swagen/precarious_agreement_Nov_2012/charta_der_zeitarbeit_deutsch_ ohne_unterschriften.pdf

Jürgens U. (2004) An elusive model: diversified quality production and the trans- formation of the German automobile industry, Competition and Change, 8 (4), 411–423.

Jürgens U. and Krzywdzinski M. (2006) Globalisierungsdruck und Beschäftigungs- sicherung: Standortsicherungsvereinbarungen in der deutschen Automobilin- dustrie zwischen 1993 und 2006, WZB Discussion Paper SP III 2006-303, Berlin, Wissenschaftszentrum Berlin für Sozialforschung.

Jurgens U. and Krzywdzinski M. (2008) Relocation and East–West competition: the case of the European automotive industry, International Journal of Auto- motive Technology and Management, 8 (2), 145–169.

Keune M. (2011) Decentralizing wage setting in times of crisis? The regulation and use of wage-related derogation clauses in seven European countries, European Labour Law Journal, 2 (1), 86–94.

Kinkel S. and Lay G. (2003) Fertigungstiefe: Ballast oder Kapital? Stand und Effekte von Out- und Insourcing im Verarbeitenden Gewerbe Deutschlands, Mitteilungen aus der Produktionsinnovationserhebung 30 (August 2003), Karlsruhe, Fraunhofer-Institut für Systemtechnik und Innovationsforschung.

La Repubblica (2010) ‘Conosco le fabbriche meglio di loro’: Fassino risponde alle critiche Fiom, 30.12.2010. http://www.repubblica.it/econo- mia/2010/12/30/news/fassino_fabbriche-10698779/

Lehndorff S. (2012) German capitalism and the European crisis: part of the solu- tion or part of the problem, in Lehndorff S. (ed.) A triumph of failed ideas: European models of capitalism in the crisis, Brussels, European trade Union Institute, 79–102.

Marginson P. (2015) Coordinated bargaining in Europe: from incremental cor- rosion to frontal assault?, European Journal of Industrial Relations, 21 (2), 97–114.

Marx P. (2011) The unequal incidence of non-standard employment across oc- cupational groups: an empirical analysis of post-industrial labour markets in Germany and Europe, IZA Discussion Paper 5521, Bonn, Institute for the Study of Labor.

Massa-Wirth H. and Seifert H. (2005) German pacts for employment and competitiveness Concessionary bargaining as a reaction to globalisation and European integration?, Transfer: European Review of Labour and Research, 11 (1), 26–44.

Piotti G. (2009) Cost reduction through relocation, or the construction of myths in discourse, Competition and Change, 13 (3), 305–326.

Pulignano V. (2006) Still ‘regime competition’? Trade unions and multinational restructuring in Europe, Relations industrielles/Industrial Relations, 61 (4), 615–638.

Pulignano V. (2014) Negotiating flexibility and security in multinationals in Europe: the case for extending European Framework Agreements, ETUI Policy Brief 1/2014, Brussels, European Trade Union Institute.

Pulignano V. (2015) Transnational flexibility and ‘shareholder capitalism’: a case for extending EWC competences to cover flexibility, ETUI Policy Brief 2/2015, Brussels, European Trade Union Institute.

Raess D. (2006) Globalization and why the ‘time is ripe’ for the transformation of German industrial relations, Review of International Political Economy, 13 (3), 449–479.

Rees C. and Edwards T. (2010) Globalization and multinational companies, in Ed- wards T. and Rees C. (eds) International human resource management: globali- zation, national systems and multinational companies, Harlow, Pearson, 11–29.

Rehder B. (2003) Betriebliche Bündnisse für Arbeit in Deutschland: Mitbestim- mung und Flächentarif im Wandel, Frankfurt am Main, Campus Verlag.

Rhodes M. (1997) Globalisation. Labour markets and welfare states: a future of ‘competitive corporatism’?, RCAS Working Paper 36/1997, San Domenico di Fiesole, European University Institute.

Schulten T. (1998) New site pact signed for west German Opel plants. http://eu- rofound.europa.eu/observatories/eurwork/articles/other-working-conditions- social-policies/new-site-pact-signed-for-west-german-opel-plants [Accessed 26.01.2015]

Silvia S.J. (2010) Mitgliederentwicklung und Organisationsstaerke der Arbeit- geberverbaende, Wirtschaftsverbaende und Industrie-und Handelskammern, in Schroeder W. and Weßels B. (eds) Handbuch Arbeitgeber- und Wirtschaftsver- bände in Deutschland, Wiesbaden, VS Verlag, 169–182.

Sisson K., Arrowsmith J. and Marginson P. (2003) All benchmarkers now? Bench- marking and the ‘Europeanisation’ of industrial relations, Industrial Relations Journal, 34 (1), 15–31.

Slaughter J. (2003) Auto union embraces two-tier wages, Labor Notes, 30 Sep- tember 2009. http://labornotes.org/2003/09/auto-union-embraces-two- tier-wages

Sorge A. and Streeck W. (1987) Industrial relations and technical change: the case for an extended perspective, Berlin, Wissenschaftszentrum Berlin für Sozialforschung.

Streeck W. (1991) On institutional conditions of diversified quality production, in Matzner E. and Streeck W. (eds) Beyond Keynesianism: the socio-economic of production and full employment, Aldershot, Elgar, 21–61.

Tilly C. (1995) Globalization threatens labor’s rights, International Labor and Working-Class History, 47 (Spring), 1–23.

Turlan F. (2012) New competitiveness agreement safeguards jobs at car plant. http://eurofound.europa.eu/observatories/eurwork/articles/industrial-rela- tions-other-working-conditions/new-competitiveness-agreement-safeguards- jobs-at-car-plant

Turner L. (2009) Institutions and activism: crisis and opportunity for a German labor movement in decline, Industrial and Labor Relations Review, 62 (3), 294–312.

Upchurch M. (2000) The crisis of labour relations in Germany, Capital & Class, 24 (1), 65–93.

Wassermann W. and Rudolph W. (2007) Leiharbeit als Gegenstand betrieblicher Mitbestimmung. Anforderungen und Arbeitsressourcen von Betriebsräten in Betrieben mit hohem Leiharbeitnehmeranteil. Ergebnisse einer Befragung von Betriebsräten in Betrieben des Organisationsbereichs der IG Metall im Früh- jahr 2007, Arbeitspapier 148, Düsseldorf, Hans-Böckler-Stiftung.

Weinkopf C. and Vanselow A. (2008) (Fehl-)Entwicklungen in der Zeitarbeit?, Bonn, Friedrich- Ebert-Stiftung.

Yun A. (2011) Building collective identity: trade union representation of precari- ous workers in the South Korean auto companies, TRAVAIL, capital et société, 44 (1), 155–178.

Zumbeck C. (2009) Zeitarbeit und befristete Beschäftigung: Analyse und Hand- lungsempfehlungen, Frankfurt am Main, Bund-Verlag.

Chapter 12

Outsourcing and collective bargaining in the recent crisis: implications for employment in multinationals

Nadja Doerflinger and Valeria Pulignano1

1.    Introduction

The chapter investigates the effects of crisis-related collective bargaining on different contractual groups in four subsidiaries of two multination- als (MNCs) in Germany and Belgium. Several studies have highlighted how collective bargaining played a major role in the recent economic and financial crisis in safeguarding jobs (Glassner and Keune 2010). However, little attention has been paid to its effects on different groups of workers, although different approaches towards different contractual groups of workers may be expected within the context of increasing la- bour market fragmentation (Holst 2014).

We use the term ‘crisis-related collective bargaining’ to emphasize that local negotiations on flexibility and security for different groups of work- ers were held in economically difficult situations, with both bargaining parties under pressure. ‘Concessions’ on pay and working conditions re- flected the desperate attempt of local unions to save jobs in response to MNC ‘social dumping’ and benchmarking strategies (Pulignano 2015). However, as the European Trade Union Confederation (ETUC 2010:3) stated, ‘concession bargaining risks to cause a renewed increase in ine- qualities’. Hence, it is crucial to understand the social processes entailing these workplace inequalities. This involves studying the effects (rather than just the causes) of crisis-related collective bargaining on differ- ent contractual groups of workers (temporary and permanent workers) and comparing them within and across countries. The study uses both company-level factors (for example, skills and such company-contextual features as technology) and country-/European-level structural factors

1. This is a jointly written chapter and both authors equally contributed to it; names are in alphabetical

(for example, national collective bargaining and framework agreements at EU level) as variables explaining the effects of crisis-related bargain- ing for different groups of workers.

The chapter uses in-depth comparative case study analyses of four sub- sidiaries of two multinational companies (MNCs) operating in the metal sector in Belgium and Germany,2 looking at the period from when the crisis hit the plants until their economic recovery. We explicitly examine crisis-related collective bargaining and its outcomes for different con- tractual groups of workers. Comparing companies in two countries is informative, since we can investigate whether the crisis produced simi- lar or different effect within and/or across countries. Furthermore, it enables us to assess whether and to what extent a country’s institutional system mediates crisis effects. Comparing Germany and Belgium is in- teresting, since both countries feature different union traditions and sys- tems of workplace representation despite their coordinated nature.

Drawing on a rigorous comparison of four MNC subsidiaries in Belgium and Germany, the study offers two contributions. Firstly, it contributes to the debate on understanding how workplace inequalities are pro- duced. In particular, it argues that inequalities are the product of social imbalances and examines the conditions (i.e. a more or less encompass- ing collective bargaining system, the presence or absence of European Framework Agreements, the high or low skill level of the workforce) un- der which those imbalances occur. In so doing, the chapter adds to the literature debates which see workplace inequality either as imposed uni- laterally by management or as the effect of unions focusing on their core membership. Secondly, and more specifically, by looking at the afore- mentioned conditions under which crisis-related collective bargaining leads to more positive or less positive outcomes for different groups of workers, the chapter contributes to providing relevant policy-related in- sights specifically for trade unions.

The chapter is structured as follows. We start by examining the role of collective bargaining during the crisis, before moving on to compare the

2. The data was collected between autumn 2011 and autumn The analysis is based on 25 semi-structured interviews with strategic and operational HR managers and works council- lors and trade unionists (at workplace- and sector-level). Moreover, documentary material such as company-level agreements, annual reports or relevant newspaper articles comple- mented the primary data collected.

relevant crisis-related measures adopted in Germany and Belgium at different levels. Having explained research design and methodology, we present the empirical findings and engage in a comparative analysis.

2.    Collective bargaining in the recent crisis

In the context of the US recession in the 1980s concession bargaining described a company’s strategy to cope with structural changes in the economy – the result of intensified international competition, a grow- ing pace of technical progress and cost pressures (McKersie and Cap- pelli 1982). In exchange for such concessions, US unions could push for more security in several cases, for example through no-layoff policies or income protection schemes (Cappelli 1985). In this respect, concession bargaining was considered as a way to exchange labour cost moderation for improvements in job security in economically difficult times.

Likewise, the recent crisis has not left the collective bargaining land- scape untouched. In the initial crisis years, workers and employers in the private sector often tried to jointly cushion its effects in a way that allowed increased security for both. When the long-term character of the crisis became clear, more extensive measures were required to safe- guard employment (Eurofound 2012). Workplace actors, for instance, bargained on wage reductions or freezes, cutting back benefits, or paying lower starting wages (Carley and Marginson 2011). When the effects of the crisis got harsher and unemployment rose, employers found it easier to impose higher flexibility, while unions were able in some cases to ne- gotiate employment guarantees in exchange (Marginson et al. 2014).

Concession bargaining has important implications for company-level employment relations. Firstly, by linking labour costs to job security, concession bargaining puts pressure on workers and contributes to un- dermining wages and working conditions by enabling deviations from higher-level agreements through bargaining decentralization. Margin- son (2014) sees accelerated bargaining decentralization as one of the main outcomes of the recent crisis. Secondly, by implying a trade-off between flexibility and job security, it links company-level bargaining to a company’s economic situation. This creates a dilemma for unions be- cause they need to choose between safeguarding employment or main- taining wage levels and benefits and thereby running the risk of the plant being closed (Pulignano et al. 2013).

Although a lot is known about collective bargaining in crisis times, the effects produced for different groups of workers have hardly been in- vestigated. Labour markets have changed from a system of formerly strong internal labour markets protecting core employees towards a more fragmented system in which employment regulatory institutions have become weaker and organizations utilize the wide range of avail- able contractual arrangements (Grimshaw and Rubery 1998). Following Grimshaw et al. (2001:29), ‘deregulation has allowed managers, even within a single organization, to consider and adopt different employ- ment policies and practices as solutions for different groups of workers’. Hence, we can expect that workplace-level actors engaged in bargaining in crisis times may use different approaches for different groups of work- ers, with different effects on their jobs and working conditions.

3.    Anti-crisis measures adopted in Germany and Belgium

Various collectively bargained measures cushioning the social effects of the crisis were available at workplace level in Belgium and Germany, two countries characterized by multi-employer bargaining systems. Howev- er, differences regarding the linkage between sector and company levels can be observed (Marginson and Galetto 2014). Because of the limited use of opening clauses and the restricted capacity for company-level de- viations countries characterized by ‘delegation’ (Belgium) could react to the crisis in a way preserving the encompassing nature of the bargain- ing institutions, and thereby better able to resist downward pressure on wages and working conditions. By contrast, in countries featuring ‘derogation’ (Germany) from sector-level provisions through opening clauses, it was more difficult to counteract employer pressure for col- lective bargaining decentralization and for wage moderation and more flexible working conditions (Glassner and Keune 2012). This difference is reflected in the specific way collective bargaining was used in both Germany and Belgium as a mechanism through which the state, trade unions and employers were able to mitigate the effects of the crisis at different levels.

Generally speaking, higher-level agreements defined the framework for further company-level negotiations. In Belgium, a two-year inter- sector agreement was concluded in December 2008, aimed at simulta- neously balancing companies’ competitiveness, maintaining high levels

of employment and supporting workers’ purchasing power (Eurofound 2009). It increased financial benefits for those in temporary unemploy- ment and thus stabilized their income without increasing labour costs for companies. Moreover, many sectoral agreements specified modest wage increases (Glassner et al. 2011). In Germany, sector-level agree- ments were frequently further detailed at company level. Most impor- tantly, previously guaranteed wage increases in sectoral agreements could be suspended at company level on crisis grounds, leading to wage moderation. Furthermore, a one-year employment guarantee for those working short-time was implemented in the metal sector (Glassner et al. 2011).

Different statutory measures were available to prevent mass layoffs. In Germany, short-time working (Kurzarbeit) can be used in the case of a substantial drop in production volumes, subject to management and works council agreement. Weekly working hours can be reduced to zero for a limited period for part of or the entire workforce. The intention of short-time working is thus to prevent layoffs and offer companies the possibility to retain workers with critical skills in crisis times. The work- ers concerned are paid for the hours worked in the company, while the German Federal Employment Agency pays 60/67 per cent (without/ with children) of the difference between the normal and actual wage level. The mechanism can normally be used for six months, though this timeframe was extended several times during the crisis to safeguard em- ployment over longer periods. Finally, short-time working did not apply to temporary agency workers prior to the crisis, but was temporarily in- troduced for them from February 2009 until the end of December 2010. The question of extending the scheme indefinitely to agency workers has since been under discussion in Germany.

In Belgium, statutory short-time work was introduced and could be ap- plied after concluding company-level agreements independent of the sector. Moreover, temporary unemployment (tijdelijke werkloosheid/ chômage économique) provisions were enhanced for companies in eco- nomic difficulties. Employment contracts can be partly or wholly sus- pended for limited periods, with the affected workers entitled to unem- ployment benefits paid by the National Employment Office (70/75 per cent of the last income without/with family). Before the crisis, tempo- rary unemployment only applied to blue-collar workers, but the sys- tem was extended to white-collar employees in 2012. In addition, the system of time credits (tijdskrediet/credit-temps) allowing time-off (for

example, for further education or family life) was used to keep workers employed (Vandaele 2009; Eurofound 2009). Finally, in 2008, unions and employers agreed to extend temporary unemployment to workers on fixed-term and agency work contracts.

4.    Features of the investigated MNCs and their subsidiaries

Auto is a US-American manufacturing MNC employing 130,000 staff (mainly low- to medium-skilled blue-collar workers). It uses a low level of technology to supply standardized components mainly to the auto- motive industry. Although Auto has a leading market position, it faces global competition and thus needs to be flexible and cost-competitive. Consequently, the use of different contractual arrangements has in- creased. The German plant uses its agency workforce mainly for simple production tasks, with agency and permanent workers working side by side. In the Belgian plant, fixed-term workers work mainly in production alongside the permanent workforce (see Table 1). Auto was particularly hit by the crisis in 2009 after all pre-crisis orders had been fulfilled, and the lack of new orders led to a declining work volume. The European and Northern American operations were most affected, while developing markets were hardly hit at all. Overall, Auto remained profitable during the crisis, though profits were unevenly distributed.

Machine is a European MNC active in power generation and transport with about 92,000 employees, half of whom are blue-collars and for the most part highly skilled. The investigated transport division develops and manufactures high-tech, customized products. Though these differ across sites, the workforces’ skill profile is similar. Because of growing competition, internal adaptability had to be strengthened and labour costs reduced through contractual flexibility. Machine uses agency workers to perform relatively simple production-related tasks, and ex- ternal contractors (mostly high-skilled engineers) are actively involved in product development and R&D (see Table 1). Machine entered the crisis in 2010 with order intake dropping by almost 40% in Europe and North America. Although its operations in emerging markets were prof- itable, substantial restructuring was carried out in Europe.

With both companies featuring different workforce skill levels related to the technology used, this allows us to assess the extent to which such

Table 1 Plant characteristics

company-level differences are reflected in the outcomes of crisis-related collective bargaining.

While both multinationals operate in the manufacturing sector and have a European Works Council (EWC), Machine is the only one to have a Eu- ropean Framework Agreement (EFA) on training and mobility practices, negotiated within the framework of transnational restructuring. The four plants feature high levels of unionization and use about 20% flex- ible labour, enabling comparisons across contractual groups. At Auto, the permanent and temporary staff worked side by side, performing the same tasks, while at Machine the two tended to be segmented. Hence,

the differences between the two companies – both of which were hit hard by the crisis – allow us to examine whether company-level factors played a role in the way Auto and Machine dealt with the crisis. For instance, high-tech companies like Machine may be more interested in skill re- tention and providing an attractive work environment even in times of crisis, since the specific skills of its workforce are not readily available on the labour market. By contrast, companies like Auto employing a pre- dominantly low-skilled workforce may tend to lay off employees to cope with the short-term effects of the crisis because it is relatively easy for them to find new staff on the labour market when new orders come in.

5.    Crisis-related collective bargaining and its effects on different groups of workers

5.1                      The case of Auto

Auto’s German plant had faced severe economic difficulties in 2004, and as a consequence several flexibility practices had been negotiated to help cope with the initial crisis phase. A working-time corridor of 28- 42 hours depending on production volume was introduced based on the sector-level agreement provision that derogations from the 35-hour week were possible if management and the local works council agreed. Furthermore, there was a 20 per cent agency work quota to react more flexibly to changing volumes. As a result of these measures, Auto’s Ger- man plant was profitable. However the lack of incoming orders drasti- cally decreased the workload in late 2010. Management and the works council immediately agreed to reduce weekly hours for all blue-collar and half of the white-collar staff to the negotiated minimum of 28 hours. However, three months later, the two parties had to negotiate further ‘adjustment mechanisms’ to counteract incurred losses. Short-time working – which allowed jobs and skills to be retained simultaneously – was implemented for permanent workers in most production-related and some administrative departments. As a consequence, the workers concerned faced income losses since unemployment insurance only provided partial compensation. Furthermore, management sent almost all agency workers back to their agency and did not renew fixed-term contracts because of the lack of work. They were able to take this deci- sion without the works council, since the latter only has codetermination rights when temporary workers are hired, not when they leave. Thus, the works council could not formally intervene.

‘The situation in the crisis was indeed difficult. Apart from our ordi- nary employees, we had also 500 temporary workers – agency and fixed-term – about 350 in production. Only few of them were kept on since their skills were required. The others were made redundant.’ (Works councillor, Auto Germany)

Hence, temporary workers – the first to leave – were instrumental in safeguarding the core workforce’s jobs. Only the few agency workers with critical skills remained in the plant’s workforce, though they were still employed via agencies and thus remained in a vulnerable position.

Auto’s Belgian plant had to cope with a 70 per cent decline in produc- tion volume in 2009. Management wanted to react immediately through temporary unemployment for almost all permanently employed blue- collar workers. However, as temporarily suspending these workers’ con- tracts would have meant a loss of income, the unions refused to give their necessary consent.

‘We cannot accept our company demanding workers to make conces- sions with regard to their wages when they are doing okay overall – that’s impossible, that’s not conceivable.’ (Unionist, Auto Belgium)

The unions demanded that the wages of the workers concerned be topped up so as not to incur any financial penalty through temporary unem- ployment and asked for a training budget to use the ‘time-off’ to invest in the workforce’s skills. Management – in need of the union’s consent – quickly agreed. As a result, the unemployment insurance paid most of the wages of those concerned, with Auto topping them up to maintain ini- tial levels. Additionally, various training courses were offered, giving the permanent workforce the possibility to upgrade skills. Thus, the perma- nent workforce’s working conditions remained stable in the initial crisis phase. Management also intended not to renew any fixed-term contracts to react to the declining workload and rising cost pressure. However, a workplace-level agreement on ‘employment paths’ for temporary work- ers negotiated between management and the unions ruled that fixed- term contracts were the mode of entering the company and were to be upgraded to permanent contracts after one/three years (white-collar/ blue-collar workers). The agreement could be temporarily suspended under special circumstances, but the unions refused to accept the imme- diate interruption of its application. As a result, management and unions reached a compromised, with those 40 employees who had been there

long enough receiving a permanent contract and thus becoming part of the core workforce. The other fixed-term workers, however, had to leave when their contracts expired and were thus the first victims of the crisis.

‘We lost 20% of our workers in the crisis. They were all on fixed- term contracts that were not extended. There was hope because Auto brought people back, but this didn’t last and today there is a hiring freeze.’ (Unionist, Auto Belgium)

These initial measures proved insufficient to cope with the scope of the continuing crisis. In Germany, additional measures were negotiated in mid-2011, while short-time working was still in place. Management de- manded concessions from the workforce to safeguard the plant, and the works council – under pressure from management, fearing the loss of jobs as in the 2004 crisis and lacking the security offered by a strong sector-level agreement – felt it had to give in to safeguard jobs. Agree- ments were thus concluded in three areas. Firstly, internal mobility was fostered by redeploying employees in other departments. For instance, production workers with computer skills were sent to administrative units to give support there. Likewise, management established coop- eration agreements with other companies in the region to send its staff there for some time. This ‘leasing’ measure was encouraged by the sec- toral agreement, but its success was limited since the whole region was affected by the crisis. Generally, the higher internal mobility decreased employees’ level of discretion regarding their job content as they could be moved from one position to another based on company needs. Sec- ondly, working time for production workers became more flexible. The shift system for instance became directly linked to production volumes and could thus change within days. This also meant that workers had to cope with more weekend and night work in times of high workloads, leading to a decreased level of employee control over their working time. Finally, an extensive early retirement scheme offered workers the pos- sibility of leaving the company at the age of 55 on relatively favourable conditions, granting financial stability to those leaving. This put pres- sure on the older core workers, though only 5% of them left. In sum, the works council agreed to higher levels of flexibility and conceded on working conditions to safeguard the core workforce’s jobs.

In Belgium, problems were caused because temporary unemployment provided protection exclusively to blue-collar workers. Hence, solutions to overcome the absence of a protective scheme for white-collar staff

had to be found. In mid-2009, management and unions jointly decided to encourage white-collar workers to use their time credit (as recom- mended by the state), a scheme available under Belgian law. Employees can reduce working hours, and the state pays an allowance to partially compensate for the loss of salary. To encourage uptake, unions demand- ed that an additional allowance be paid to partly close the income gap. Management – in need of the unions’ support to promote the uptake of time credit – agreed. Many white-collar workers took time credit, as it allowed them to take time off without big cuts to their salary. Finally, internal mobility was strengthened, meaning that employees could be temporarily moved to other departments when they had the appropriate skills. This was also viewed as on-the-job training and thus as a way of multi-skilling the workforce. Compared to the German plant, the crisis measures adopted in Belgium were less drastic, since unions success- fully worked to safeguard the permanent workforce’s jobs and working conditions.

Altogether, the aforementioned measures helped Auto to overcome the crisis. The German plant’s situation improved as new orders increased in early 2013, leading to the termination of short-time working. The ability to use the scheme for 24 months helped to substantially limit the plant’s losses and thus prevented plant closures or forced redundancies.

‘Flexibility is certainly a crucial factor for the survival of the plant. An agency work level of 15-20% seems to be necessary for the well-being of the company and the permanent staff. Of course it is the agency workers who suffer from that situation. In the end I think that both the instruments of our government and the ones that we agreed on were crucial for securing jobs.’ (Works councillor, Auto Germany)

The Belgian plant recovered faster, since new orders came in a year after the onset of the crisis. This ended temporary unemployment, but re- vealed an immediate lack of staff, meaning that the company had to seek to rehire workers whose contracts had expired.

‘In 2009, we terminated hundreds of temporary contracts, month after month, but in 2010, business increased very sharply, and we needed staff. We managed to hire back less than 300 people – 300 temporary contracts that we had just terminated.’ (HR Manager, Auto Belgium)

5.2                         The case of Machine

Machine’s German plant entered the crisis in 2010, when order volumes plummeted. Management sent half of the plant’s agency workers back to their agencies, and the remaining half left within weeks. Hence, they were the first to be affected by the crisis. Those agency workers were low- to medium-skilled, performing auxiliary functions in production and ad- ministration. The works council – lacking codetermination rights on this issue – was informed, but not consulted. From a management perspective, letting agency workers leave was an easy way to rapidly reduce costs. The works council could not formally oppose, but at the same time, letting them leave potentially assured the jobs of Machine’s core workforce.

‘In the crisis most works councils were quite happy that only agency workers were made redundant. There is nothing worse for a works council than for the core workforce to be targeted by management.’ (Works councillor, Machine Germany)

In 2010, several large-scale orders were cancelled at Machine Belgium, leading to a drop in workload. Compared to the more production-orient- ed German plant, the Belgian branch was focused on research and prod- uct development, i.e. with a smaller share of production activity (mostly prototyping) and a lower number of (medium- to high-skilled) blue- collar workers, mostly hired via work agencies. As in Germany, man- agement wanted to reduce its agency workforce, but there were workers with critical skills among them. Hence, management offered permanent contracts to those ten high-skilled agency workers, while the other medi- um-skilled ones had to leave. Unions were not in favour of this, but could not formally intervene due to the absence of a corresponding company- level agreement. Thus, agency workers were the first ones to leave due to the crisis.

‘When things got bad, the company told us ‘we have to dismiss peo- ple’. And the first ones to be dismissed are those on temporary con- tracts. Most temps haven’t got any career perspective at all here. This workforce is there when needed. If not needed, it’s over.’ (Works councillor, Machine Belgium)

These initial measures were insufficient to cope with the intensifying crisis. In Germany, management did not renew fixed-term contracts, and only a few of these workers – those with scarce skills – were offered

permanent contracts. The works council did not have to be consulted be- cause of the absence of co-determination rights in this respect. Termi- nating the contracts of external contractors, however, was not considered an option by management even though these high-skilled engineers were relatively costly. Similar to agency workers, they were not directly hired by Machine, and therefore, it would have been relatively easy to termi- nate their contracts. However, their skills distinguished them from the agency workforce. Meanwhile, the works council demanded employ- ment guarantees for the plant’s remaining workforce while management wanted something in exchange. This led to the two parties agreeing to use up hours in the staff’s working-time accounts as a precondition to introduce short-time working. The consequence was that the core work- force had to sacrifice hours saved for instance to go on sabbatical or take early retirement, and to cope with a loss of income due to short-time working. Management also demanded greater internal mobility and the works council felt it had to consent in order to safeguard jobs. At that time, the work councils were able to use the European Framework Agreement (EFA), negotiated between Machine’s European headquarters and the European Metalworkers’ Federation (EMF) in 2010. Under it, workers were sent to other German or even foreign plants for short periods when their profiles fitted. Hence, the EFA made national best practices on mo- bility and training-related issues available for use at Machine’s European subsidiaries. The EFA was used by the works council as an institutional resource during the crisis to promote job security in the German subsidi- ary (Pulignano 2014). Inter-plant mobility was voluntary, but employees were afraid of losing their jobs if they did not agree to it. Hence, func- tional flexibility and mobility were demanded from all employees, nega- tively impacting discretion on job content and work location.

‘We have no problems with flexibility at all as long as the jobs of the employees are secure. We know that companies have to be able to react to developments on the markets.’ (Works councillor, Machine Germany)

However, further headcount reductions were demanded by manage- ment to cut labour costs, leading to the negotiation of generous volun- tary redundancy and early retirement schemes. More than 300 workers accepted the offers within six months. Since the plant was located in an economically strong region, the prospects of finding work again were good. Almost two years later, new orders came in and soon after, new agency workers were hired.

In Belgium, further measures to react to the declining workload had to be implemented. Using temporary unemployment was no option, as it only applied to blue-collar workers. As in Germany, management and unions decided to strengthen cross-border internal mobility in accordance with the previously negotiated EFA, sending employees to other plants in the neighbouring countries on short-term assignments, and thereby safeguarding jobs. This measure was considered as inter- cultural on-the-job-training, though it only applied to a specific group of employees with good language skills. Furthermore, the unions pro- posed using the time credit scheme as encouraged by the state, but management refused to grant extra financial incentives since companies cannot force employees to take it up. Management demanded a head- count reduction and the unions agreed based on the promise that it be done on a voluntary basis with generous severance payments, and most importantly that it stabilized the working conditions of the remaining workforce. Hence, those leaving the company enjoyed favourable condi- tions, while the remaining workforce did not experience any change in their working conditions. Furthermore, the classroom training budget for the lower-skilled workforce was cut and more emphasis put on on- the-job training. This hardly affected the permanent workforce, since low-skill tasks were mostly performed by agency workers. Finally, a few external contractors had to leave, though, as in Germany, the majority of the high-skilled and high-paid contractors were kept on since their skills were indispensable. When orders increased in 2012, the situation improved markedly and 100 agency workers were hired for the produc- tion department.

‘It has become common to have a “safety buffer”. As work decreases, we simply reduce our “extra” temporary staff. Of course, when the company fires someone, it doesn’t want to give severance payments, etc. It’s easy to fire a temp.’ (Works councillor, Machine Belgium)

6.    What are the effects of crisis-related collective bargaining on different groups of workers?

The interaction between institutional factors at national and European levels (i.e. collective bargaining and social dialogue structures) with company-level features (i.e. skill levels) explains the differences in the nature of crisis-related collective bargaining and its effects on con- tractual groups in different companies across and within countries. As

Table 2 The effects of crisis-related collective bargaining on different groups of workers

illustrated in the previous section and in Table 2, temporary workers – who were already vulnerable before the crisis – were hit hard. The measures adopted to counteract the crisis in Auto and Machine point to safeguarding permanent workforces, using flexible staff as a buffer to achieve this. Our comparison of the two companies also reveals dif- ferences in both the nature and the outcomes of crisis-related collective bargaining. Specifically, Auto with its medium- to low-skilled workforce was characterised by more concession-oriented bargaining patterns while negotiations at Machine, employing predominantly high-skilled staff, were more consensus-driven.

In the initial crisis phase, the local Belgian unions succeeded in keeping on a number of temporary workers, who were given permanent contracts and became part of the core workforce, while the German works coun- cils could only stand and watch as most temporary workers left, as they

only have co-determination rights when temporary workers are hired.3 Cross-country differences persisted when the crisis endured. While jobs and working conditions were maintained for the permanent Belgian workforces, their German counterparts had to concede on working con- ditions to safeguard employment. Different collective bargaining struc- tures in Belgium and Germany as well as the differences in institutional mechanisms used to retain jobs in crisis periods explain cross-national differences.

Although Germany and Belgium both feature multi-employer bargain- ing systems, Belgium is characterized by ‘delegation’, while the German system allows for ‘derogation’ (Marginson and Galetto 2014), i.e. ena- bling company-level deviations if foreseen in the sector-level agreement. Such ‘opening clauses’ exist regarding working time and wages in the German metalworking sector. The resulting high degree of discretion was used by local management at Auto’s and Machine’s German subsidi- aries to push for concessions on working conditions to safeguard jobs. The works councils could not make use of the procedural security offered by binding sector-level agreements to oppose such concessions, and also lacked ‘local’ institutional resources linked to co-determination rights. The fact that these rights did not cover the dismissal of temporary work- ers made it impossible for works councils to formally oppose manage- ment decisions. Hence, they were in a defensive position, unable to keep on temporary workers and stabilize the permanent workforce’s work- ing conditions. Conversely, since deviations from sectoral agreements are hardly possible in Belgium, employers had no leeway to undermine standards set at higher levels. Therefore, Belgian company-level man- agement was limited in the practices they could adopt in comparison to Germany (Pulignano et al. 2013), i.e. the unions could use sector-level agreements as a power resource to protect the regular workforces’ jobs and working conditions.

Nevertheless, the findings also illustrate inter-company differences. For example, in contrast to Auto’s German plant where all temporary workers had to leave, a few fixed-term workers and nearly all external contractors were retained at Machine Germany. Similarly, in Belgium

3. Works council rights to intervene may be strengthened in the future, since the 2012 metal sector agreement encourages them to negotiate on temporary agency work at work-

place level. Further information: http://www.igmetall.de/tarifabschluss-sichert-faire- leiharbeit-10140.htm

the local unions at Auto leveraged their consent to temporary unemploy- ment to enforce stable working conditions for the permanent workforce, demonstrating the more concession-driven bargaining pattern. The un- ions used an earlier negotiated agreement on ‘employments paths’ to ne- gotiate concessions to save some temporary workers’ jobs. Such conces- sions were absent at Machine’s Belgian subsidiary.

Looking at Machine, the findings illustrate the importance of the linkage between national and European information and consultation structures in association with the conclusion of an EFA on mobility and training in situations of transnational restructuring in 2010 as a relevant factor explaining the empirical outcomes. The EMF played a crucial role in developing the transnational bargaining coordination and in ensuring communication between employee representatives and trade unions at Machine. In this way, the conditions for signing the EFA at sector level were created. Distinctive workplace-level agreements adopted during the crisis illustrate the EFA’s positive effects on local negotiation. In both countries, local employee representatives negotiated on internal mobil- ity to keep workforce capacity in line with expected demand, relying on the transfer and rotation of expertise across different plants when pro- duction peaks occurred (Pulignano 2014). In short, none of the meas- ures adopted at Machine’s subsidiaries were the outcome of concessions, but rather the result of management and employee representatives com- promising to find consensus-based solutions balancing flexibility and employment security. However, we observed cross-national differences within Machine. In Belgium, the EFAs contributed to empowering local unions to negotiate on retaining a number of temporary workers and to safeguarding the core workforces’ jobs and working conditions because of the more encompassing collective bargaining system. By contrast, the German works councils suffered from bargaining decentralization and felt they had to agree to management requests to safeguard employment. Hence, a trade-off between retaining jobs and conceding on working conditions evolved as an outcome for the German permanent workforces overall, implying that the job stability of the core workforce and the in- stability of temporary workers were complementary, increasing the gap between these groups of staff (Hassel 2014).

The institutional system, including European and national employment regulation, defined the framework under which crisis-related collective bargaining took place, though the outcomes of such negotiations were also shaped by the interaction with company-level contexts. Auto and

Machine differed mainly in terms of workforce skill levels. The speci- ficity of skills could be leveraged by unionists at Machine, helping to keep on some temporary workers in both countries and also entailing a consensus-driven bargaining pattern. Consensus was also stimulated by the linkage between European and national levels, leading to greater autonomy for local unions to work with management to shape flexibility in a way fostering long-term employment security for different groups of workers. This was different at Auto, where unions used local institution- al resources where they existed (Belgium) to enforce a degree of protec- tion for the workers via concessions. Overall, the replaceability of skills at Auto in contrast to the scarcity of skills at Machine, within a context characterized by European and national employment regulation, medi- ated the effects of crisis-related collective bargaining, explaining the dif- ferent bargaining patterns observed in both companies and their effects on different groups of workers.

7.    Conclusion

The study’s main objective was to investigate the effects of crisis-related collective bargaining on different contractual groups of workers in MNC subsidiaries in Germany and Belgium. It identified the conditions un- der which crisis-related collective bargaining led to more positive or less positive outcomes for different groups of workers. We explain the ob- served diversity in outcomes by referring to differences in the extent of bargaining decentralization and company-level specificities. Specifically, differences in employment regulatory systems – and collective bargain- ing in particular (i.e. derogation vs. delegation) – and company-level differences related to workforce skill levels (i.e. low- vs. high-skilled) as well as aspects of MNC employment governance at transnational level (i.e. the existence of a European Framework Agreement) explain the observed outcomes. The aforementioned factors mediated the effects of crisis-related collective bargaining, with decentralized bargaining prov- ing to be particularly problematic in Germany where company-level de- viations from sector-level agreements limited union capacity to main- tain working conditions.

When the crisis hit the plants, the initial reaction was to let temporary workers leave. While the German works councils lacked codetermina- tion rights to prevent this, their Belgian counterparts were able to get a number of temporary workers given permanent contracts. As the crisis

persisted, the Belgian unions were able to stabilize both the jobs and the working conditions of the permanent workforces, while the Ger- man works councils conceded on working conditions to safeguard those workers’ jobs. In Belgium, the encompassing multi-level bargaining sys- tem hardly allowed for company-level deviations, creating a favourable environment for unions to stabilize jobs and working conditions alike. In this context, Auto’s unionists engaged in concession bargaining to keep on a number of temporary workers on the basis of a previously negoti- ated agreement that served as a power resource. As the crisis persisted, they were able to stabilize the core workforces’ jobs and working condi- tions in exchange for them agreeing to temporary unemployment. Con- versely, at Machine, both temporary workers with indispensable skills and highly-qualified external contractors were kept on. This confirms the argument at the core of dualization literature, claiming that the ef- fects of liberalization are filtered by the skill level (Emmenegger et al. 2012). However, our study adds to this by illustrating how, under spe- cific conditions, unions can mediate the effects of dualization for both skilled and unskilled workers. In particular, it demonstrates that unions and management at Machine consented to several measures in accord- ance with the previously negotiated EFA and without engaging in con- cession bargaining. Overall, within a framework respecting the existence of systems of employment governance at transnational level, skills me- diated crisis-related collective bargaining, leading to a more consensus- oriented pattern at Machine. However, because of the progressive use of company-level deviations in Germany, power was shifted from the sec- tor- to the company-level, resulting in a weakening of works councils. This had worse consequences at Auto, where all temporary workers had to leave, and works councils had to endorse concessions on the perma- nent workforce’s working conditions to safeguard their jobs. Conversely, at Machine’s German plant, the high skill level and the presence of the previously mentioned EFA mediated the effects of crisis-related collec- tive bargaining, with several fixed-term workers and nearly all external contractors being kept on and with negotiations more consensus-driven.

Overall, the chapter has demonstrated that the effects of crisis-related collective bargaining on groups of workers differed. Generally, perma- nent workers in all investigated workplaces enjoyed a certain level of protection, while most temporary workers had to leave. Findings also demonstrate that negotiated practices are not simply a function of the regulatory context, but also depend on the skill level of the workforce as an important company-level contextual variable, together with the

degree of coordination between information-consultation at European and local level. The interaction between the regulatory context and com- pany-level factors also explains the observed variation in crisis-related collective bargaining – concession-driven at Auto and consensus-driven at Machine. In other words, even in crisis times, negotiations do not nec- essarily have to result in concessions. The study emphasizes that there is scope for unions to avoid concessions under certain conditions, even in economically difficult times. At the same time, the findings show that concessions can contribute to producing poor working conditions for different groups of workers in the long run. Therefore, they might not be the best path for unions to follow.

Acknowledgement

The research was funded by grants from KU Leuven (OT/10/015) and the Flemish Research Council (FWO) (ZKC2575/G.0773.11).

References

Cappelli P. (1985) Plant-level concession bargaining, Industrial and Labor Rela- tions Review, 39 (1), 90-104.

Carley M. and Marginson P. (2010) Negotiating the crisis: social partners re- sponses, in European Commission (ed.) Industrial relations in Europe 2010, Luxembourg, Publications Office of the European Union, 55-84.

Crouch C. and Keune M. (2012) The governance of economic uncertainty: beyond the ‘new social risks’ analysis, in Bonoli G. and Natali D. (eds) The politics of the news welfare state, Oxford, Oxford University Press, 45-67.

ETUC (2010) Undermining the bargaining position of European labour, ETUC Austerity Watch 3. http://www.etuc.org/sites/www.etuc.org/files/NR_3_ wekaening_coll_bargaining_plusjob_protection.pdf

Eurofound (2009) Temporary unemployment is buffer to economic crisis. http:// www.eurofound.europa.eu/eiro/2009/04/articles/be0904019i.htm

Eurofound (2012) Social dialogue in times of global economic crisis, Luxembourg, Publications Office of the European Union.

Glassner V. and Keune M. (2010) Negotiating the crisis? Collective bargaining in Europe during the economic downturn, DIALOGUE Working Paper 10, Geneva, International Labour Office.

Glassner V. and Keune M. (2012) The crisis and social policy: the role of collective agreements, International Labour Review, 151 (4), 351-375.

Glassner V., Keune M. and Marginson P. (2011) Collective bargaining in a time of

crisis: developments in the private sector in Europe, Transfer; European Review of Labour Research, 17 (3), 303-322.

Grimshaw D. and Rubery J. (1998) Integrating the internal and external labour market, Cambridge Journal of Economics, 22 (2), 199-220.

Grimshaw D. et al. (2001) Organisations and the transformation of the internal labour market, Work, Employment and Society, 15 (1), 25-54.

Holst H. (2014) ‘Commodifying institutions’: vertical disintegration and institu- tional change in the German labour relations, Work, Employment and Society, 28 (1), 3-20.

Marginson P. (2014) Coordinated bargaining in Europe: from incremental corrosion to frontal assault?, European Journal of Industrial Relations. doi:10.1177/0959680114530241

Marginson P. and Galetto M. (2014) Engaging with flexibility and security: redis- covering the role of collective bargaining, Economic and Industrial Democracy. doi: 10.1177/0143831X14538850

Marginson P., Keune M. and Bohle D. (2014) Negotiating the effects of uncertain- ty? The governance capacity of collective bargaining under pressure, Transfer; European Review of Labour Research, 20 (1), 37-51.

McKersie R. and Cappelli P. (1982) Concession bargaining, Working Paper 1322- 82, Cambridge, Massachusetts Institute of Technology.

Pulignano V. (2014) Negotiating flexibility and security in multinationals in Europe: the case for extending European Framework Agreements, Policy Brief 1/2014, Brussels, ETUI.

Pulignano V. (2015) Transnational flexibility and ‘shareholder capitalism’: a case for extending EWC competences to cover flexibility, Policy Brief 2/2015, Brussels, ETUI.

Pulignano V., De Franceschi F. and Doerflinger N. (2013) Local bargaining over flexibility and security: varieties of outcomes and diversity of ‘dualisation’ effects for the European workforce, 10th European Conference of ILERA, Amsterdam, 20-22 June, 2013.

Vandaele K. (2009) The Ghent-system, temporary unemployment and the Belgian trade unions since the economic downturn, Transfer; European Review of Labour Research, 16 (3-4), 589-596.

.

Part 4

Addressing the outsourcing challenge: statements from practitioners in trade unions

Goodbye TINA!

Rudi Wagner

GPA-djp (Union of Private Sector Employees, Graphical Workers and Journalists),

Austria

‘There is no alternative’ (abbreviated as TINA) was a slogan often used by British Prime Minister Margaret Thatcher. In economics, politics and political economy, it has come to mean that ‘there is no alternative’ to economic liberalism — that free markets, free trade and capitalist glo- balization are the best or only way for modern societies to develop.

Outsourcing is part of this game and this book challenges its nature. For instance, Anna Mori´s overview of outsourcing in local government con- stitutes extremely valuable input to the discussions on the roots of out- sourcing and what it involves for the workers concerned.

‘Savings indeed may be lower than expected and declining over the long term’ was a key sentence and an everyday experience for me as a union- ist.

One example from an Austrian hospital which wanted to outsource its cleaning department: a business consultant calculated cost savings of

€ 7,419.31 per year – I am sure the cost of the consultant was nearly the same.

A second example concerns the outsourcing of a hospital canteen. Now that a private provider is doing the cooking for patients and hospital workers, the canteen is empty because the quality of the meals has got so bad.

It would be very interesting to know the total expenditure on consultants and lawyers, spending their time on counselling outsourcing processes (for example). It would seem to be a very profitable business model – for the consultants and lawyers.

A few decades after Margaret Thatcher’s TINA, the opposite is true: TATA – There are thousands of alternatives.

Creating a toolbox for unionists and works councils, these are the prior- ity tools:

1. TUPE – The Transfers of Undertakings Directive 2001/23/EC is a European Union directive protecting the employment contracts of people working in businesses transferred between owners. It re- placed and updated the directive previously known as the Acquired Rights Directive 77/187/EC.
2. Fighting for social clauses in calls for Achieving better con- ditions in outsourcing contracts is a major field of action.
3. Making use of the OECD Guidelines. In most cases works councils are informed about planned outsourcing moves at the last A comprehensive, non-binding code of conduct, the Guidelines state that companies should respect human and labour rights in the areas in which they operate. So just publicize breaches – companies don’t like bad media coverage.

So let’s say goodbye to TINA. And give TATA a very warm welcome – ‘There are thousands of alternatives!’

Outsourcing in the public sector:

how do we respond and move forward?

Clare Harrisson

UNISON, United Kingdom

UNISON is a UK-wide public services trade union. With over 1.2 million members we have extensive experience of organizing across the public sector and our membership also includes people working in the private or charitable sector to deliver public services.

Our experience with outsourcing directly relates to Anna Mori’s chapter ‘Low road convergence? Implications for labour following outsourcing in local government. A comparative perspective’. The article raises many important issues for trade unions and it is certainly true that we rec- ognize many of the patterns described. We have seen how outsourcing is pushing down wages, in particular for those who do manual roles or roles viewed as being lower-skilled. These jobs are often performed by low paid-women. Outsourcing is designed to do this as often a contract’s biggest cost factor is labour: the only way to deliver services at a cheaper cost is to pay people less or reduce staff numbers. In addition, this frag- mentation of delivery by numerous different organizations across the private and charitable sector has made it increasingly difficult to bargain on behalf of these workers or to organize them. As mentioned in the ar- ticle, the erosion of TUPE over time means that many transferred work- ers are seeing their terms and conditions downgraded, or more simply these protected ex-public sector employees are gradually being replaced by cheaper workers paid at the lowest rate. In many workplaces we are seeing the development of two- or even three-tier workforces. This is played out not just in terms of pay, but also a segregation of terms and conditions and other benefits such as pensions. Low-paid outsourced workers also find it more difficult to move up in an organization or seek new opportunities, trapping many in a cycle of low-paid work. This is of particular concern for women, as in the private sector it is much harder to hold employers to account on their equal pay obligations, i.e. further embedding the segregation of pay by gender.

The article makes some very interesting points on convergence; in par- ticular the impact that depressed pay for outsourced/agency workers can have on core workers. This is something we have recognized for some time; however there is still some way to go in convincing core workers that it is in their interest to focus on organizing for better terms for periphery workers. And we are seeing this not only at a local level. Low wages for outsourced workers push down the ‘market rate’ for par- ticular types of work, in turn making it harder to negotiate and win the argument for decent wages for directly employed (or core) staff. Work in public services is effectively being devalued by the downward pressure on wages throughout the sector and across different types of employer. The language of ‘crisis’ and ‘austerity’, coupled with market pressure has been a pay disaster for our members.

However the impact and response to outsourcing is complicated. The im- portance of tendering bodies in setting the terms, and particular the cost envelope, for contracts to be delivered cannot be underestimated. In real- ity, the profit organizations make from public sector contracts is often relatively marginal; especially when examined on a local level. A good ex- ample of this complexity can be seen in how unions need to approach cam- paigns such as for a living wage in the outsourced sector. These campaigns are unlikely to succeed without appropriate pressure on the tendering body to fund through the increase needed. It is also a very varied landscape in the type and make-up of the companies taking on these contracts. We work with organizations of such varying size, from huge multinationals with hundreds of thousands of employees to tiny organizations and individu- als employed on personal care budgets. Size is just one of the factors that can have a huge impact not only on a company’s approach to pay and negotiation, but also their capacity to deal effectively with HR issues. In fact this increasing diversification in the market has caused some con- cern amongst some of the larger outsourcing companies themselves who have called for a better code of conduct in the sector. Some worry that the proliferation of unregulated, small service providers is pushing down the bid level of contracts to a point where they are effectively undeliver- able; undercutting ‘more reputable’ companies in the process. On a prac- tical level this variety provides us as a union with a rage of organizing and bargaining challenges, meaning our strategy for meeting these must take into account a number of different approaches and tactics.

It is also worth looking at how the issues and problems with outsourcing have changed over time, in particular the last five years of a Conservative-

Outsourcing in the public sector

led government. Although outsourcing has been a growing problem for a much longer period, the rate of change has accelerated over recent years. Not only has this government championed a pro-privatization agenda, passing numerous new pieces of legislation that further open up the pub- lic services to market competition (e.g. the Health and Social Care Act). They have also reduced the amount of money going to public sector bod- ies like local councils and the NHS. This has pushed many public bodies to see outsourcing as their only option, the only way they can start to cope with dramatically reduced budgets. However plenty of research has shown this model to be a bit of a false economy, with cost levels already pushed as far to the floor as they can go. Over this period we have also seen many of the protections previously in place to try and protect out- sourced workers undermined and TUPE has also been chipped away at over the years.

So how do we move forward as a movement and respond to some of these challenges? We can for instance start putting pressure on tender- ing bodies during the procurement phase. We are often relatively well organized in many of these organizations, whether NHS trusts of local authorities. We can organize to put pressure on these bodies to, for ex- ample, be ‘living wage employers’, specifying (and costing) such in the tenders they write. We can push for TUPE-plus and challenge them to write procurement policies that take into account social value as well as baseline cost. However as a note on TUPE, we need to be careful that we don’t over-rely on what is a legal protection only at the point of transfer. Although a vital tool in protecting outsourced workers, the examples in the article show that as time goes on it proves less and less of a shield to hide behind. For people employed by the private company after the point of transfer TUPE offers very little.

Pan-workplace solidarity is important but so is building up a sense of employer-based identity for our private sector members; getting the bal- ance right is key if they are not to become marginalized either in the workplace or our union. However at the same time we need to equip them with the tools and capacity to bargain and organize in their own right. We also need to find a way to balance our local relationships with a more central approach that deals with some of these huge national com- panies within a holistic national strategy. These interplays between the small and the large, the local and the national (or international), and between contractor and tendering body will be central to our ability to protect these outsourced workers.

Overcoming challenges in organizing contact centre workers

Alan Tate

UNI Global Union, UNI ICTS, Switzerland

UNI ICTS is the Information & Communication Technology Services sector of UNI Global Union and represents over 3 million ICTS workers worldwide.

We fight to ensure workers’ rights globally, in particular the right to or- ganize and the right to collective bargaining.

As a global organization, UNI is particularly concerned not just with out- sourcing but also offshoring – relocating work to countries where even lower wages and more intense working conditions are possible. Offshor- ing or the threat of offshoring puts even more downward pressure on pay and conditions as workers in both outsourced and in-house contact centres are ‘benchmarked’ against workers in the offshored countries.

In February 2014 UNI held a Global Call Center Conference in Orlando, Florida, together with our US affiliate, the CWA. The conference drew close to 300 participants, half from CWA locals across the US and half from countries across the globe. The conference heard presentations from activists and union leaders from countries as diverse as Brazil, In- dia, Belgium, Tunisia, Germany, Portugal, France and the US. The as- tounding thing was how similar the working conditions were in contact centres in each of these different situations, confirming the findings out- lined in the papers that subcontracting has led – not just in Europe but worldwide – to increazed downward pressure on pay and conditions and to further work intensification and precarity. The presentations and dis- cussions in Orlando also clearly found that the one factor protecting the conditions of contact centre workers was the presence of strong unions and a strong tradition of collective bargaining.

UNI ICTS is working with our affiliates in Morocco, the Philippines, In- dia and Latin America to organize workers in contact centres there to raise standards for all contact centre workers.

There are huge challenges in organising contact centre workers – the in- creased use of temporary agency workers and part-time and contingent workers, a huge turnover of staff and the fact that contact centres can be closed or moved relatively easily.

To face these challenges unions need to be very strategic about how we organize contact centres – how can we leverage our bargaining power and relationships with the clients – major telecom, finance, retail and other companies? How can we reach agreements with the key outsourc- ing companies on workers’ rights? How can we develop innovative or- ganising and bargaining strategies able to overcome the challenges we face? How can we campaign for improved legislation to protect and en- hance decent work in contact centres? The papers presented in this vol- ume help tackle these important questions.

Employee voice in the metal sector: trade unions play a key role

Fidel Gavilan

ABVV Metaal/FGTB Métal, Belgium

My comments come from the perspective of a practitioner involved in the metal sector, both in Belgium and on the international level. I have been working as an economist for the Belgian socialist metalworking un- ion ABVV since 2002, supporting works council members in their an- nual analyses of business and financial information and on restructuring measures. I have also advised members of European works councils and is himself a member of a number of EWCs in his capacity as IndustriALL Europe Coordinator.

The article written by Ramioul and Van Hooteghem is particularly in- teresting even though no companies from the metalworking sector are concerned. Like Ramioul and Van Hooteghem we are also seeing work- loads increasing in many companies after restructuring. In times of cri- sis workers are prepared to accept sacrifices – both financially and ma- terially – in exchange for jobs. Any restructuring is not only difficult for those losing their jobs, but also for those who remain.

Benassi’s article relates to agency workers in Germany, and their dis- crimination vis-à-vis the core workforce. This is a problem that we are also seeing in the UK. Agency workers are exerting downward pressure on the pay and working conditions of employees on open-ended and fixed-term contracts. The increasing precariousness of the German la- bour market has been putting heavy pressure on German workers, as evidenced by the fact that their real wages stagnated after 2001.

On the other hand, we should note that this discrimination – based on our experience in European works councils – (or should I weaken this some- what, because in the meantime labour laws have come under pressure from the International Monetary Fund and the European Union and have been changed in several EU countries) – is no great problem in the other Western European countries, where agency work is strictly regulated and where agency workers enjoy the same conditions as the core workforce.

This does not detract from the text’s special interest. Benassi describes how the German IG Metall colleagues are gradually improving the sit- uation of agency workers first at a company level and then at sectoral level. She rightly points out that the European Union and the European Central Bank are spotlighting the German system as an example for the whole European Union. Yet the European Commission and the ECB are forgetting that total hours worked in Germany in the last decade have remained constant. Germany is creating jobs by replacing full-time jobs with part-time ones. And here’s another case of EC and ECB forgetful- ness: Germany is the EU country where poverty has risen most since 2000.

The article by Doerflinger and Pulignano upholds the – from a union per- spective – strength of the Belgian social governance model. While nego- tiations are conducted in Germany by works councils, in Belgian this is the job of sector-level union delegations and officers. On the employee side, members from just the three recognized unions, the socialist ABVV, the Christian ACV and/or the liberal ACLVB, are responsible for social dialogue in Belgium. In Germany, anybody may be a candidate in the social elections, from pro-employer ‘unions’ to radical left-wing lists.

Doerflinger and Pulignano show that German works councils are pre- pared to make great sacrifices in an attempt to save jobs – sacrifices often undermining existing wages and working conditions. In Belgium, social dialogue between employers and unions takes place in a consensus-ori- ented manner. One discusses the consequences of this or that sacrifice. Can a company be rescued through such measures? Moreover, the exist- ing (collective) agreements are filed with the Ministry of Employment, where a further check takes place. Union leaders also have a whole ar- senal of instruments at their disposal in negotiations, something that is not always the case with German works councils. They can possibly push for legislative changes to allow collective agreements to be approved as in the case of VW Vorst. Last but not least, Belgian sectoral collective agreements are generally binding, in other words, they apply to a whole sector, making any (unfair) competition between companies in the same sector more difficult. This is not the case in Germany, where collective agreements are concluded between the sectoral union and the sectoral employer federation, with the result that only members of the federation are bound by the collective agreements. Moreover, opt-out clauses allow companies to circumvent the agreements.

Employee voice in the metal sector

The Doerflinger and Pulignano study needs to be read with reservations. The study looked at companies with trade union presence, i.e. DGB un- ions and three recognized Belgian unions. The result could well be com- pletely different if smaller German companies without DGB unions were studied!

List of contributors

Carina Altreiter is University Assistant (prae doc) at the Department of Sociology at the University of Vienna, Austria.

Chiara Benassi is Lecturer in Human Resource Management at Royal Holloway, University of London, United Kingdom (UK).

Jan Drahokoupil is Senior Researcher at the European Trade Union Institute (ETUI) in Brussels, Belgium.

Sonila Danaj is a doctoral student in Sociology at the Department of Social Sciences and Philosophy at the University of Jyväskylä, Finland.

Virginia Doellgast is Associate Professor of Comparative Employment Relations at Cornell University, USA.

Nadja Doerflinger is a PhD student at the Centre for Sociological Research (CeSO) at KU Leuven, Belgium.

Jörg Flecker is Professor at the Institute of Sociology at the University of Vienna and Chairman of the Working Life Research Centre (FORBA), Austria.

Theresa Fibich is University Assistant (prae doc) at the Department of Sociology at the University of Vienna, Austria.

Fidel Gavilan is Advisor at ABVV-Metaal, the Belgian socialist metalworkers federation, Belgium.

Luca Giustiniano is Associate Professor of Business Organization at LUISS Guido Carli University, Rome, Italy.

Bettina Haidinger is Senior Researcher at the Working Life Research Centre (FORBA) in Vienna and lecturer at the University of Vienna, Austria.

Clare Harrisson is National Officer at Private Contractors (Outreach) Bargaining Unit, UNISON Centre, UK.

Stefan Kirchner is Post-Doc Researcher at the Institute of Sociology at the University of Hamburg, Germany.

Nathan Lillie is University Lecturer in Social Policy at the Department of Social Sciences and Philosophy at the University of Jyväskylä, Finland.

Lucia Marchegiani is Assistant Professor of Business Organization at Roma Tre University, Rome, Italy.

Anna Mori is Ph.D. Fellow in Labour Studies at the University of Milan, Italy.

Elisa Pannini is PhD Candidate in Employment Relations & Organisational Behaviour at The London School of Economics and Political Science, UK.

Enzo Peruffo is Assistant Professor of Management at LUISS Guido Carli University, Rome, Italy.

Luca Pirolo is Assistant Professor of Management at LUISS Guido Carli University, Rome, Italy.

Valeria Pulignano is Professor of Sociology of Labour and Industrial Relations at the Center for Sociological Research (CeSO) at KU Leuven, Belgium, and Associate Fellow at the Industrial Relations Research Unit (IRRU) at Warwick University, UK. She also coordinates the CeSO.

Monique Ramioul is Head of the research group Work and Organisation at Research Institute for Work and Society – HIVA-KU Leuven and Associate Profes- sor at the Center for Sociological Research (CeSO) at KU Leuven, Belgium.

Katja Sarmiento-Mirwaldt is Lecturer in Politics at Brunel University London, UK.

Markku Sippola is University Researcher at the Department of Social Sciences and Philosophy, University of Jyväskylä, Finland.

Alan Tate is Head of Department UNI ICTS at UNI Global Union, Switzerland.

Geert Van Hootegem is Professor at the Center for Sociological Research (CeSO), KU Leuven, Belgium.

List of contributors

Ines Wagner is Post-Doc Researcher at the Institute for Work and Qualification at the University of Duisburg-Essen.

Rudi Wagner is Union Secretary at the collective bargaining department of GPA-djp, Austria.