Transnational trade union strategies in the field of European wage policyThe role of institutional power sources and institutional work in European

wage coordination

Susanne Pernicka1, Vera Glassner2

Abstract

In the context of market internationalization, the integration of European economic and monetary policies as well as managerial strategies of regime shopping and offshoring many academic observers expected national labor unions to compete rather than cooperate with foreign unions due to their conflicting national interests. Meanwhile there exist several well documented empirical evidences of cross-national collaboration of trade unions. Processes of transnational interaction and coordination can even be found in the field of wage policies, albeit largely confined to unilateral initiatives of trade unions. Our contribution poses the questions of why, when and how transnational trade union action occur. By adopting sociological Neo-Institutionalism we develop a framework of transnational union strategies and contend that both institutions (i.e., regulative, normative and cultural-cognitive institutions) and the strategies of business and national/European policy actors facilitate or constrain transnational union strategies Evidently, the predominant market logic at European and national levels stand in opposition to the logic of cooperation. In addition, transnational efforts of trade unions aiming at creating or maintaining institutions and practices of coordination does occur only if there are national and European ‘institutional entrepreneurs’, i.e. individual or collective actors who possess the legitimacy and resources at different levels (e.g., high sectoral union density). The paper concludes that 1) rising levels of international (wage) competition do not per se result in cross-border cooperation of trade unions. Rather, a minimum level of institutional support is necessary to open up room for strategic manoeuvre. 2) Business appears to be still reluctant to embark on bilateral wage bargaining coordination at transnational level, yet there is evidence that institutional pressures might induce business to enter negotiations with unions.

1 Institute of Sociology, Johannes Kepler University Linz, Austria (=corresponding author, Susanne.pernicka@jku.at, Altenberger Strass 69, A-4040 Linz)2 Institute of Sociology, Johannes Kepler University Linz, Austria

Introduction

The questions of why, when and how in the European Union (EU) transnational trade union action and solidarity occur have been preoccupying a growing number of industrial relations scholars. In the context of economic internationalisation, European economic and upcoming monetary integration as well as managerial strategies of regime shopping and offshoring many academic observers expected labor unions to compete rather than cooperate with foreign unions due to their conflicting national interests (Streeck and Schmitter 1991: 140; Altvater and Mahnkopf 1996; Streeck 1998) and to widely diverse national industrial relation

1

systems (Visser and Ebbinghaus 1992). According to this view, eastern enlargement of the European Union would have further decreased the likelihood of trade union cooperation across national borders. And wage policy seemed the least likely of labor relations fields where transnational trade union coordination would take place.

In contrast to these premises, several well documented empirical evidences exist of cross-national collaboration of trade unions (Turner 1996; Erne 2008; Gajewska 2009; Ales and Dufresne 2012), and processes of horizontal Europeanisation can even be found in the field of wage policies. However, due to the lack of interest on the employers’ side for transnational coordination in wage bargaining, the Europeanisation of wage policy is largely confined to unilateral initiatives of trade unions (Gollbach and Schulten 2000; Keller and Platzer 2003; Marginson and Sisson 2006; Traxler et al. 2008; Traxler and Brandl 2009, Glassner 2009; Platzer and Mueller 2011). In the wake of the European Monetary Union (EMU) early forms of transnational bargaining coordination were organised as bottom-up cooperation between national unions in neighbouring countries, e.g. the Doorn-initiative consisting of the Benelux countries and Germany (Gollbach and Schulten 2000). At a later stage the European Trade Union Confederation (ETUC) and its European Trade Union Federations (ETUFs) have institutionalised particular activities and norms for a top-down coordination of collective bargaining (e.g. information exchange, bargaining rules) (Schulten 2003, Marginson and Sisson 2006; 98).

How can these cross-national collaborations of trade unions in wage policy be explained? This question brings us to the problem of collective action (Olson 1965) that has been consistently addressed by industrial relations theory. The collective action problem suggests that rational, self-interested actors do not act in support of their group’s common interest or objective (e.g. higher wages or better working conditions) unless there is coercion or some other special incentive (ibidem;1f.). While the question of how trade unions might be able to overcome the collective action problem within national borders has been comprehensively addressed, there is no such analytical concept for exploring the determinants of transnational trade union action in wage bargaining. At national level distinct institutions of industrial relations provide trade unions with the ability for collective action. In Western Europe predominant multi-employer bargaining and practices of the legal extention of collective agreements stabilise organised industrial relations quite independently of union membership strength. And in most Western European countries where extention practices do not exist there are statutory incentives for unionisation, such as union-led unemployment systems (Traxler et al. 2001; Traxler and Brandl 2009).

At European level such supportive institutions of transnational wage bargaining coordination barely exist, and European umbrella organisations of labour (ETUC, ETUFs) are too weak in relation to their national affiliates to enforce transnational cooperation. deals with two dimensions of cross-border wage bargaining coordination: (1) the formation of more or less formalised transnational coordination initiatives by national unions and European trade union federations, and (2) decentralised forms of (latent) transnational coordination, i.e., pattern bargaining, where the following bargaining units orientate their settlement towards the pace setter such that they catch up with the pattern-setting agreement (Traxler

2

and Brandl 2009, 183). Because the standards derived from (1) are unilateral (union-based) and formally non-binding, Traxler et al. (2008: 222 and 232) argue that they can hardly be effective without being buttressed by (2).

While studies dealing with transnational wage bargaining coordination provide convincing evidences on the macroeconomic effects of cross-border pattern setting as well as of the pivotal role of certain sectors and countries in transnational wage coordination (Traxler et al. 2001 and 2008 ; Traxler and Brandl 2009), the question of why trade unions might be able and willing to cooperate across borders rather than pursuing competitive strategies that aim to undercut the pay increases of the latter (i.e., ‘inverse’ pattern bargaining), remains unresolved. Possible explanations for transnational trade union action are not at the centre of their analysis. Traxler and Brandl (2009, 197), for instance, stress the importance of ‘spill over’ processes of bargaining coordination to other sectors and countries, however, they do not specify the social mechanism that underlie the spread of such behaviour. This article develops a neo-institutional framework that might contribute to a better understanding of transnational trade union action by enriching institutionalist perspectives by a cultural dimension. In particular, the concepts of organisational field (DiMaggio and Powell 1983; Hoffman 1999), institutional logics and institutional worki (Lawrence and Suddaby 2006; Lawrence et al. 2009) are utilised to systematically take account of regulative, normative and cultural-cognitve institutions that might support or hamper transnational trade union action. In addition, we assume that trade unions themselves have contributed to the creation and maintainance of European normative and cultural-cognitive institutions that in turn might facilitate transnational collective action. Against this background, the article addresses the following questions:

1. Is there a European field of industrial relations in the metal industry, and if so, what roles have trade unions played in its construction?2. Why, how and when have trade unions engaged in strategies of creating, maintaining or disrupting European institutions of industrial relations?

In empirical terms we take a longitudinal perspective on the development of cross-national institutions of wage policy in the metalworking sector as an ‘extreme case’ (Ebbinghaus 2006; 394). The metal sector exhibits a high level of product market integration at international level, considerable trade union power, and holds a strong position as pattern setter in the multi-employer bargaining systems of most Western European countries. We expected that the history of European industrial relations in the metal sector would be richer and institutionally more developed than those in other industries, making it an ideal candidate for study. As such, the investigation of the metal industry is expected to contribute to our understanding of opportunities and constraints of trade unions’ institutional strategies at European level, and hence potential trajectories of horizontal Europeanisation.

The article is structured as follows. We first introduce the concepts of organisational field, institutional logics and institutional work. Determinants of actors’ institutional strategies in wage policy are presented in the following section. The question of whether an organisational field has emerged in the European metal sector is investigated by reviewing the extensive body of existing literature in the ensuing section. Conditions under which trade unions adopt institutional strategies leading

3

to (or not) the emergence of an organisational field are then discussed. Then we draw our conclusions.

Transnational strategies of trade unions: a neoinstitutional framework

The concept of organisational field applied to European wage policy

In order to understand and explain transnational collective action and their outcomes, i.e. the evolution and maintainence of institutions of transnational wage bargaining coordination, we adopt the neoinstitutional concept of ‘organisational field’. By organisational field DiMaggio and Powell (1983; 148) refer to ‘organisations that, in the aggregate, constitute a recognised area of institutional life ’. Organisational fields cannot be determined a priori but must be defined on the basis of empirical investigations. The authors distinguish between four parts of the process of institutional definition: ‘1. an increase in the extent of interaction among organisations in the field; 2. the emergence of sharply defined inter-organisational structures of domination and patterns of coalition; 3. an increase in the information load with which organisations in a field must contend, and 4. the development of a mutual awareness among participants in a set of organisations that they are involved in a common enterprise’ (ibid.; 148). Once a field becomes well established, processes of isomorphism are triggered.ii DiMaggio and Powell (1983; 150) distinguish between three mechanisms of institutional isomorphism (see below). In the (emerging) organisational field of European wage policy we can identify the following core populations of organisational actors that might fulfil the above characteristics: national and European trade unions, national and multinational business (associations), national and supranational policy actors. The individual populations of actors exhibit varying degrees of cross-national interaction and of mutual awareness that they are part of a common enterprise. In contrast to trade unions, MNCs and European business associations are largely reluctant to engage in bilateral wage bargaining (coordination) at European level. And EWCs are too weak to enforce company-level wage bargaining, especially since the European Trade Union Federations (ETUFs) have been reluctant to equip them with bargaining tasks (Traxler and Brandl 2009, 181).

Also with regard to trade unions one must distinguish between distinct sectors and countries. While most of national unions share a common interest in preventing ‘social dumping’ by the means of cross-border bargaining coordination, they differ in terms of their bargaining power and particular national interests. However, the focus on interest differences – adopted by rational-choice theory – tends to neclect that interests need to be recognised, defined and unified – which are particularly complex processes especially in the case of transnational collective interests and action (see Meardi 2011: 336). Thus, the collective action problem at transnational level will be investigated by adopting a conception of interests and behaviour that accounts for distinct rationalities (logics) and institutions within which they are embedded. Collective bargaining institutions differ widely across Europe. The most

4

fundamental distinction is between sytems of multi-employer bargaining and single employer bargainingiii.

Based on Scott (2008; 48) we perceive institutions as ‘… comprised of regulative, normative and cultural-cognitive elements that, together with associated activities and resources, provide stability and meaning to social life’. Two groups of organisational actors (supranational policy actors and trade unions) have intended to create particular institutions at European level that might lead to isomorphism or convergence of national bargaining structures and strategies and hence, to the evolution of an organisational field. However, organisational actors differ in terms of institutional power as well as normative-ideolgical objectives and interests. The EU Commission has aimed to strengthen the wages-productivity ratio by imposing European rules, in particular, on member countries which are hit by a worsening of their international (labor cost) competitiveness. Existing national institutions of multi-employer wage bargaining would be undermined and trade unions would be forced to accept restrictions on their bargaining autonomy which in turn might lead to a coercive isomorphism of union structures and strategies. In contrast, national and European trade unions have established and maintained European institutions (e.g., cross-border bargaining coordination rules) and provided the basis for institutional pressures on national wage bargaininers that might result in normative isomorphism. By adopting common bargaining targets formalised in coordination rules bagainers are expected to demand wages that are perceived from a transnational perspective as ‘fair’ and ‘appropriate’. And finally, the organisational field conception can be applied to the above mentioned form of pattern bargaining. The latter is perceived to induce trade union strategies to converge, i.e. mimetic isomorphism (DiMaggio and Powell 1983; 151, Marginson and Sisson 2006). Mimetic isomorphism rests on imitation, benchmarking and social learning and is mainly based on cultural-cognitive aspects of institutions that refer to shared conceptions and frames through which meaning is made (Scott 2008: 57).

The concept of institutional logics

Institutional logics are perceived as cultural beliefs, conceptions and practices that shape cognitions and behaviours and hence, organisational structures and strategies. As regards industrial relations and wage policy in particular, institutional logics – as ideal types – are assumed to differ along a continuum between market logic and the logic of coordination. According to the market logic, wages are best determined by competitive and efficient markets. The logic of coordination contains the view that bilateral wage bargaining secures interests of both parties to the collective agreement, labor and employers. Moreover, institutional logics might be further differentiated in terms of either conflicting or compatible sub-logics that prevail between different or even within populations of actors. For instance, German employers and trade unions used to share a common belief in collective bargaining cooperation. However, while employers have participated in multiple-employer wage bargaining due to their belief in its function to increase productivity by providing industrial peace, trade unions’ participation can be traced to the assumption that wage bargaining result in fairer and economically more sustainable wages than market competition.

5

Institutional strategies in the (emerging) organisational field of European wage policy

While earlier neoinstitutional accounts and most existing theorising in industrial relations focus on organisational behaviour and strategic action within (national) institutional settings, our main concern is on institutional strategies that aim at the creation of new institutions that facilitate collective action at European level. This brings us to the concept of institutional work or institutional strategies, a term we use synonymously (Lawrence and Suddaby 2006; Lawrence et al. 2009: 13). Institutional work is defined as work motivated significantly by its potential institutional effects. Institutional work can therefore also be understood as physical or mental effort done in order to achieve an effect on an institution or institutions (Lawrence et al. 2009; 15). For instance, trade unions’ efforts to create cross-border institutions which impact upon national processes and outcomes of wage bargaining are considered to involve ‘harder’ work than business activities aiming to enforce competitive wages. This is true at least under current conditions where EU policies of market liberalism prevail. The variation in required efforts between labor representatives and employers and their associations can be traced to at least two general explanations: 1) the existence of an asymmetrical power relationship between labor and capital in capitalist societies (Offe and Wiesenthal 1980; Traxler et al. 2001; 11) and in relation to this, 2) fundamentally contrasting interests. While structurally less powerful labor has an interest in building European and global institutions that might mitigate the effects of unlimited market competition such as downward pressures on wages, multi-national employers perceive unregulated markets as more appropriate to achieve their aims (profit maximising).

In focusing on institutional strategies of trade unions at European level, we arerequired to address the following question: is there room for strategic manoeuvre for trade unions that goes beyond existing institutions? While conventional industrial relations theory has tilted more toward structure (Dunlop 1958; Clegg 1976; Ferner and Hyman 1998; Traxler et al. 2001), since the late 1980s several industrial relations scholars have moved their focus toward ‘strategic choice’ of trade unions (Kochan et al. 1986; Voss and Sherman 2000; Frege and Kelly 2003; Hyman 2007; Pernicka et al. 2007). Frege and Kelly (2003, 2004) have developed the most elaborate framework of trade union strategies so far. Based on existing industrial relations and social movement literature (Kelly 1998; McAdam et al. 2001) they emphasise the importance of internal structures and framing processes that in combination with institutional differences, identities and differences in employer, political party or state strategies determine strategic action of unions (Frege and Kelly 2003: 12). Although their model remained within the borders of nation states their findings reveal an important predictor for institutional innovation and change: the existence of powerful individual and/or collective agents.

Sociological Institutionalism highlight the role of organised actors or institutional entrepreneurs – who possess sufficient resources (power, financial means, etc.) and have legitimacy – and see in them an opportunity to realise interests that they value high (DiMaggio 1988; Beckert 1999). Since institutional agents (in contrast to instrumental, rational actors) are perceived by definition as embedded in institutions one needs to address the issue of ‘enabling or hindering conditions’ for

6

institutional strategies. Until the early 1990s, trade unions have (with notable exceptions, see Erne 2008) rarely moved their wage-related strategies to the European or international level, not least because collective bargaining remained outside the purview of European legislation. Transnational market forces have neither led to a convergence of national institutions of wage setting nor to the internationalisation or Europeanisation of trade union strategies in response to functional requirements, i.e. economic pressures. This is largely due to trade unions dependence on legitimacy provided by both, their national membership and their counterparts on the labor market, business associations and the state, in order to secure necessary resources (financial resources, institutional power) for survival. In contrast, institutional changes at European level, such as the European Monetary Union or the adoption of the European Works Council Directive seem to entail some national trade unions to move their strategies to the European level. Thus, only in combination with changes in institutional environments, economic pressures seem to propel trade unions to engage in transnational strategies of institution building.

Unlike economic and regulative-normative environments, cultural determinants are rarely mentioned in explaining why local, national or European trade unions pursue transnational (i.e., Euro-technocratic or Euro-democratic) strategies to achieve their common targets (Erne 2008). Only very recently a fruitful discussion has evolved on the role of culture in explaining the behaviour of industrial relations actors, in particular European Works Councils (Klemm et al. 2011, Hürtgen 2011, Meardi 2011). Accordingly, culture is being framed in structural, power-tense national and transnational spaces (Hürtgen 2011, 301), and that its impact as independent variable to explain behaviour can barely be understood in isolation of other, socio-economic and institutional factors.

Hypotheses

Based on above premises we suggest the following hypotheses:

H 1: Changes in the institutional and economic environments increase the likelihood that trade unions contribute to the creation, maintenance (or disruption) of European institutions.

The larger trade unions’ power the greater the likelihood of unions to engage in the creation of European institutions. There are three sources of power, namely institutional power, structural power (e.g. a low unemployment rate) and organising power (i.e., composed of membership strength and cooperation with other social movements). Our focus is on institutional power.

H 2: Institutional power – derives from regulative, normative and cultural-cognitive elements and practices inherent in institutions that support trade union action. Moreover, institutions can also weaken trade union power. In such cases we refer to institutions as negative institutional power. Drawing on McAdam and Scott’s (2005) and Scott’s (1995, 2008) general framework of institutions, we differentiate between two components of institutional power, institutional logics (which relates to the cultural-cognitive dimension of institutions) and governance systems (which relates to the regulative and normative dimension):

7

H 2a: Institutional logics are defined as the belief systems, conceptions (including a common language)iv and associated practices that shape cognitions and behaviours and hence organisational structures and strategies. Dominant institutional logics are assumed to reflect the consensus of powerful actors, such as market logic that empowers multi-national business against labour in European fields of industrial relations. In contrast, secondary or repressed logics represent other, subordinated interests, such as the logic of cooperation of trade unions in organisational fields where policies of deregulation and marketisation prevail.

H 2b: Governance structures refer to those arrangements by which field-level power and authority are exercised in support (or to the detriment) of trade unions, involving regulative and normative mechanisms such as free collective bargaining, co-determination laws, trade unions’ political participation at macro-level (Sozialpartnerschaft) or existing strike laws in the U.K. as an example for negative institutional power.

Wage policy in the metalworking sector and beyond

The following subsection contains a brief summary of existing literature on the development of trade union bargaining coordination initiatives with a focus on the metal sector over time and against the background of the EMU. However, we also take account of earlier initiatives of cross-border bargaining coordination

The evolution of transnational collective bargaining coordination

Unions’ efforts in coordinating wage bargaining across borders took place against the background of developments in the economic and institutional sphere. In this respect, the implementation of the EMU is a key event; and although it originates in the external institutional environment of the field of European wage policy, it directly affects trade unions’ strategic behaviour at national and European level.

In view of the forthcoming EMU, trade unions in the metal sector begun to coordinate their bargaining policies and to adopt common principles for wage setting. Their main aim was to mitigate downward pressures on wages and working conditions deriving from the centralisation of monetary policy that ruled out currency devaluations as a strategy to improve a country’s international competitiveness (Traxler 2003; 86; Glassner 2009). Both ‘top-down’ and ‘bottom up’ strategies were adopted by unions. The former strategies refer to institutions initiated and implemented by the European Metalworkers’ Federation (EMF) while the latter ones refer to institutions established autonomously by national trade unions. The EMF had a strong and pioneering role in the cross-border coordination of collective bargaining policies. A ‘statement of principle on collective bargaining policy’ was adopted, in a ‘top-down’ fashion, at the EMF Collective Bargaining Conference in 1993; the EMF and its member organisations agreed on a ‘regular annual compensation for price increases in order to protect real wages, and to guarantee workers a share in productivity gains’ (EMF 1993). The ‘European coordination rule’, adopted in 1998, specifies quantitative and normative criteria for

8

wage bargaining but leaves the wage autonomy entirely to the national social partners, and is thus purely voluntary in nature.

The EMF’s coordination rule was adopted in the light of changes in the European legal-institutional framework and their impact on the national level. For instance, the enactment of the Belgian law on ‘the safeguarding of competitiveness’ in 1996 stipulating that wage growth should remain below wage increases in the neighbouring countries Germany, the Netherlands and France, is regarded as an important driving factor for enhanced coordination initiatives of central Western European trade unions (Marginson and Sisson 2006). In the late 1990s, interregional bargaining networks, aiming to support the implementation of the EMF coordination rule, developed rapidly through bottom up initiatives. For instance, the IG Metall, two Belgian metalworkers’ organisations, and the two sector-related Dutch unions established a cross-border network for exchange of collective bargaining information and trade union officials (Schulten 1998; Gollbach and Schulten 2000). Likewise, a coordination partnership between Austria, Bavaria, Czech Republic, Slovakia, Slovenia and Hungary, was established in 1999 by trade unions from these countries that make up the so-called ‘Vienna Memorandum Group’.

Unlike the bargaining coordination initiatives described above, there were a few but instructive examples of earlier trade union cooperation (see box below).

‘DACH’ initiative: Founded in the late 1960s, this multilateral trade-union network consisting of manufacturing and metal sector unions from Germany, Austria and Switzerland, provided a framework for the regular exchange of information on bargaining policies and other topics and aimed at the coordination of union policies across borders. DACH was formally included in the organisational structure of the International Metalworkers’ Federation (IMF) and, later on, the EMF.

‘Nordiska Metall’: Founded in 1970 by manufacturing sector unions from Denmark, Sweden, Norway, Finland and Iceland this cross-border cooperation aimed at the coordination of bargaining, industrial and health and safety policies. Since the 1990s the Nordic unions intensified their cooperation with the EMF, in particular in collective bargaining, transnational restructuring and other issues. With the accession of Sweden and Finland to the EU in the 1990s, Nordiska Metal became a regional body for the coordination of European policies within the EMF.

It is important to state that unions’ institutional strategies aiming at the cross-border coordination of wage bargaining are unilateral initiatives. However, for an organisational field to emerge, it requires that all groups of actors or field populations are at least affected in one or another way by institutions prevailing in

9

an organisational field. Evidence for interaction between employers and trade unions in the metal sector is scarce and inconclusive. Interestingly however, the rise of MNCs in the wake of market internationalisation did not per se result in a withdrawal of transnational companies from both national and European employers’ associations. MNCs are strongly represented in national employers’ associations (Marginson and Meardi 2010) and even more so in European associations. Assumedly, the topic of wages was on the agenda of both Business Europev and the Employers’ Organisation of Metal Trades in Europe (WEM). There is some evidence that in the metal sector, the EMF’s initiative has prompted WEM to exchange information on the development of collectively settled wages (Marginson and Sisson 2006; 204). The establishment of transnational bargaining networks by unions seems to have played an important role in employers’ attempts to coordinate information on wage-setting. For instance, the initiatives between the IG Metall Coastal district and the unions in Denmark include the involvement of employers’ associations in cross-border projects aiming at strengthening employment in the region (Andersen 2006). Moreover, the bargaining cooperation within the German-Belgian-Dutch network is quite active and strongly institutionalised. Thus, it can be expected that information exchange on wage bargaining occurs in some form or another among employers in the central-western metal sector.

Discussion

In this article we challenge the standard assumption that in increasingly internationalised markets trade unions from different countries compete over labor costs rather than cooperate transnationally to mitigate competitive wage pressures (Streeck and Schmitter 1991; Streeck 1998; Hoffmann 2002). We pose the questions of why, how and when trade unions have engaged in cross-border coordination and whether or not their institutional strategies have led to the emergence of a European organisational field in wage policy. By extending the theoretical definition of institutions to include regulative, normative and cultural-cognitive elements we contribute to a solution to the theoretical puzzle of voluntary collective action at transnational level.

Has a European field of wage policy emerged?

Growing levels of interaction and cross-border interdependencies of core populations of industrial relations actors would be an important indicator for the existence of an organisational field of European wage policy in the metal sector. However, as Marginson (2009; 64) points out, there is a striking asymmetry in the predominant level of activity between employers and trade unions: while the primary focus of employer activity is at the level of MNCs, for trade unions, the primary focus has been at the sector level.

The analytical consideration of the two basic institutional logics, i.e. belief systems and practices of wage bargaining cooperation or market competition, and how the field populations compete over their predominant form, might help to advance the question of whether or not a European field has emerged. Among all three field populations (i.e. European policy actors, trade unions and employers) the logic of

10

coordination tends to lose ground vis-à-vis the logic of competitionvi (see Table 1 below).

Field-external changes resulted in a change in the functions of multi-employer-bargaining (MEB) for employers and employees at national level. MEB is also considered as a key institutional precondition for the effective coordination of wage bargaining at the transnational level in two respects. First, from a functionalist perspective, the effective coordination of wage polices across borders presupposes that wage bargaining is coordinated effectively between different sectors, groups and workers and bargaining levels at the national level. Secondly, from an institutionalist perspective, we argue that firmly established MEB is perceived as an institution containing regulative, normative and cultural-cognitive elements, such as the legal extention of collective agreements or a belief in the benefits of sector collective bargaining among trade unions (see table 1, below). Hence, trade unions in MEB systems can draw on these institutional resources and are thus more than unions in SEB systems likely to engage into transnational wage policy initiatives.

Under what conditions do trade unions adopt institutional strategies?

As hypothesised, changes in both economic and (supportive) institutional environments are conducive for trade unions to carry out institutional work (H1). Competitive pressures on wages resulting from the centralisation of monetary policy in the Euro zone have intensified, in particular in the European metal sector that is largely exposed to global competition. However, it was institutional change introducted in the preparation of EMU rather than mere market forces that have induced national trade unions and the European Metalworkers’ Federation to embark upon a transnational approach to collective bargaining coordination.

Key determinants of institutional strategies of national unions are the following sources of power: institutional, structural and organising power (H2). Our focus is on institutional power deriving from supportive governance structures (i.e. regulative and normative mechanisms) and cultural-cognitive institutions. The latter are perceived as cultural beliefs, conceptions, and practices that shape cognitions and behaviours and hence organisational structures and strategies.

Trade unions that first attempted the cross-border coordination of collective bargaining policies from Germany, Switzerland and Austria as well as the Nordic region benefited from institutional power resources at national as well as at transnational level. In terms of goverance structures these include highly coordinated and inclusive MEB systems, high to medium-range union densities and a firmly institutionalised role of social partners in public policy-making.

As hypothesised, supportive regulative and normative institutions at national level facilitate transnational trade union action. In the course of the 1990s unions from Germany, Austria and the Benelux countries most actively promoted the establishment of institutions for wage policy coordination in the European metal sector. In the wake of EMU, unions from these countries had an active role, both autonomously and within the EMF, in strengthening and extending existing coordination initiatives to other European regions. Contrastingly, unions from countries that feature weakly coordinated MEB (most southern European countries)

11

or SEB systems (UK, Malta and most central and eastern European countries) and fragmented social partner organisations the involvement of which in public policy-making is weak, did not enter into institutional work with the aim to create institutions at European level.

However, it remains an open question why trade unions’ initiatives in the emerging field of European wage policy are clustering regionally and between trade unions from distinct industrial relations regimes (‘DACH’-initiative). Are there any transnational institutions that empower and shape the strategic behaviour of neighbouring trade unions? We have found evidences that both, in the Nordic as well as in the central-western European region, participating unions share similarities in cultural-cognitive characteristics. As Table 1 indicates, certain beliefs, conceptions and practices shared by powerful actors provide a resource for trade unions’ institutional strategies. For instance, a common language, continued experience of cross-country interaction and cooperation in the economic and social sphere as well as shared beliefs with regard to collective bargaining, wage, social and economic policy are conducive factors enhancing the adoption of joint transnational strategies by national unions. Socio-cultural closeness was a major determinant for the participation of the Swiss union from the German-speaking region in the DACH initiative in the 1970s when Switzerland was the only small country that maintained a completely independent monetary policy. Likewise, the EMF’s bargaining coordination rule was largely shaped according to the notion of stability- and productivity-based wage-setting common in Germany and Austria (and to a lesser extent also in the Nordic region) that became incorporated – via learning processes and common practice – into the belief systems of unions from other European regions.

Insert table 1 about here

In addition, structures and practices of transnational wage bargaining coordination that were established in the European metal sector in the course of the 1990s represent a self-created source of institutional power for trade unions. Institutions at European level, such as the EMF bargaining coordination rule, tend to correspond to common norms and values of unions from countries where MEB is firmly established and wage setting is highly coordinated to ensure both price stability and the participation of workers in productivity gains (in particular in Germany, Austria and, to a lesser extent, the Benelux countries). These basic principles, promoted via transnational peer review practices, begun to exert some degree of normative and moral pressure upon EMF members.

Conclusions

This paper aims to contribute to the explanation of transnational collective action in the field of European wage policy. Conventional theoretical accounts such as rational choice theory fall short in explaining cooperative behavior of unions in internationally highly competitive environments. Moreover, existing institutional

12

theorising in industrial relations has paid rather limited attention to the social mechanisms that facilitate or hinder cross-border collective action of unions. In order to overcome these weaknesses we have developed a conceptional framework for the determinants of trade union action that is either confined to national borders or transnational in scope. Transnational collective action in particular is assumed to be and cultural-cognitive commonalities and solidarity between trade unions.

By applying the concept of organisational fields to European wage policy we further aim to open up a new theoretical and analytical avenue to the analysis of processes of Europeanisation. Organisational fields are empirically defined and refer to a sphere of institutions, power relations and interactions between actors. Field actors representing the interests of labour, business and the state are endowed with various resources or power, and they are able to make use of power resources to a differing extent. By introducing the concept of institutional work, we aim to explain strategies of actors that are conscious and reciprocally related to each other. Transnational collective union action that is pursued to create, maintain (or disrupt) European institutions of wage bargaining is in the focus of our analysis. Theoretically, we draw on a conception of institutions that distinguishes between regulative, normative and cultural-cognitive pillars or institutional elements. Trade unions might be able to make use of these institutions that – besides socio-economic and associational power – determine their transnational collective action. In the (emerging) European field of industrial relations we have found two opposing logics, that is the dominant ‘market logic’ promoted by transnational companies, (parts of) the European Commission and some trade unions on the one hand and the suppressed ‘logic of cooperation’ which shapes the behaviour and strategies of those trade unions that aim at the creation of supportive institutions at transnational level to avoid wage competition.

However, empirical findings do not support the emergence of a European field of wage policy in the metal sector. Employers are reluctant to engage in Europe-wide pay negotiations with European trade unions. Therefore, some experts (not necessarily trade unionists) suggest the introduction of a legal regulation that replaces the current ‘optional’ framework for transnational company negotiations. However, European political actors such as the European Commmission do not actively support the creation of a legal framework for transnational collective bargaining. Thus, trade unions, in order to overcome competitive pressures on wages, have pursued institutional strategies of coordinating their bargainig policies transnationally. Such unilateral trade union initiatives have contributed to the creation of European institutions for the cross-border coordination of wage bargaining. Trade unions were able to establish common normative principles that might shape national wage negotiations and contribute to avoid wage competition.

Most interestingly, transnational collective action of unions is clustering regionally. Since this pattern can neither be traced to similarities in economic conditions nor to regulative-normative institutions (such as the level of collective bargaining) alone, we evaluated possible cultural-cognitive explanations. Besides similarities in industrial relations and welfare state traditions, we found indeed common cultural-cognitive institutions at transnational level, such as shared world views or beliefs in

13

cooperation rather than wage competition, linguistic closeness, and a common history of cross-border interation. This has facilitated a common understanding of trade unions from different countries which in turn has shaped their institutional strategies. Hence, transnational bargaining institutions feature a significant degree of persistence over time and were incorporated in the formal organisational structures of European trade union such as the EMB and the ETUC.

When it comes to assess these – though patchy and polycentric – processes of horizontal Europeanisation with regard to their potential to contribute to the emergence of a European field of wage policy, there is doubt that transnational pattern bargaining based on common wage norms will spill over to other regions and sectors. It is not realistic that unilateral trade union cooperation in wage bargaining will prompt employers to negotiate wage agreements at the transnational level. Only under conditions that increase the structural power of trade unions, such as shortages of skilled labour and companies’ interest in avoiding the poaching of qualified workers across borders, might induce employers to cooperate with trade unions, even without political support for the establishment of regulative transnational institutions for collective bargaining.

Literature

Ales E and Dufresne A (2012) Transnational collective bargaining: Another (problematic) fragment of the European multi-level industrial relations system. European Journal of Industrial Relations 18(2): 95-105.

Alvater E and Mahnkopf B. (1996) Grenzen der Globalisierung. Ökonomie, Ökologie und Politik in der Weltgesellschaft. Münster: Westfälisches Dampfboot.

Andersen SK (2006) Nordic Metal Trade Unions on the Move: Responses to Globalization and Europeanization. European Journal of Industrial Relations 12(1):29-47.

Beckert J (1999) Agency, entrepreneurs and institutional change: The role of strategic choice and institutionalized practices in organisations. Organisation Studies 20: 777–799.

Clegg HA (1976) Trade Unionism under Collective Bargaining. Oxford: Blackwell.

DiMaggio P (1988) Interest and agency in institutional theory. In : Zucker L (ed) Institutional patterns and organisations. Culture and environment. Cambridge: Ballinger, pp. 3–22.

DiMaggio P and Powell W (1983) The iron cage revisited: Institutional isomorphism and collective rationality in organisational fields. American Sociological Review 48(2): 147–60.

Dunlop JT (1958) Industrial Relations Systems. New York: Holt, Rinehart, and Winston.

Ebbinghaus B (2006) Qualitativer oder quantitativer Vergleich? Herausforderungen für die sozialwissenschaftliche Europaforschung. Zeitschrift für Staats- und

14

Europawissenschaften 4(3): 388–404.

Erne R 2008 European unions. Labor’s quest for a transnational democracy . Ithaca and London: Cornell University Press.

EMF (1993) Collective Bargaining Policy in a Changing Europe. Statement of principle on Collective Bargaining, adopted at the 1st EMF Collective Bargaining Conference. Luxembourg, March.

Esping-Andersen G (1990) Three Worlds of Welfare Capitalism, Princeton. New Jersey: Princeton University Press.

European Commission (2009) Industrial Relations in Europe. Luxembourg: Publications Office of the European Union.

Ferner A and Hyman R (eds) (1998) Changing Industrial Relations in Europe. Oxford: Blackwell.

Frege C and Kelly J (2003) Introduction: Union revitalization strategies in comparative perspective. European Journal of Industrial Relations 9(1): 7–24.

Frege C and Kelly J (2004) Varieties of unionism: Strategies for union revitalization in a globalizing economy. Oxford: Oxford University Press.

Gajewska K (2009) Transnational labour solidarity: Mechanisms of commitment to cooperation within the European trade union movement. Routledge: London.

Glassner V (2009)

Gollbach J and Schulten T (2000) Cross-border collective bargaining networks in Europe. European Journal of Industrial Relations 6(2): 161–179.

Hoffman AJ (1999) Institutional evolution and change: Environmentalism and the US chemical industry. Academy of Management Journal 42(4): 351–371.

Hoffmann J 2002 Beyond the myth: ‘international solidarity’ as a challenge to trade unions in the age of globalisation and europeanisation. In: Hoffmann J (ed) The Solidarity Dilemma: Globalisation, Europeanisation and the Trade Unions. Brussels: ETUI, pp. 119–144.

Hürtgen S (2011) Europäische Interessenvertretung – Eine Frage der nationalen Kultur? Industrielle Beziehungen, 18(4): 315-335.

Hyman R (2007) How can trade unions act strategically? Transfer. European Review of Labour and Research 13(2): 193–210.

Keller B and Platzer H-W (eds) (2003) Industrial relations and European integration: Trans- and supranational developments and prospects. Aldershot: Ashgate.

Kelly J (1998) Rethinking Industrial Relations. Mobilization, Collectivism and Long Waves, Routledge: London.

Kochan TA, Katz HC and McKersie RB (1986) The transformation of American industrial relations. New York: Basic books.

15

Klemm M, Kraetsch C and Weyand J (2011) ‚Bei ihnen ist das Umfeld völlig anders‘. Kulturelle Grundlagen der europäischen betrieblichen Mitbestimmung. Berlin: edition sigma.

Lawrence TB and Suddaby R (2006) Institutions and institutional work. In: Clegg S, Steward C H, Lawrence TB and Nord R W (eds) Handbook of Organisation Studies, 2nd.edition. London: Sage, pp. 215–254.

Lawrence TB, Suddaby R and Leca B (eds) (2009) Institutional work. Actors and agency in institutional studies of organisations. Cambridge: Cambridge University Press.

Marginson P and Sisson K (2006) European integration and industrial relations. Multi-level governance in the making. Houndmills: Palgrave.

Marginson P and Meardi G (2010) Multinational companies: transforming national industrial rela-tions? In: Colling, T. and Terry, M. (eds) Industrial relations: theory and practice, pp. 207-230.

McAdam D, Tarrow S and Tilly C (2001) Dynamics of Contention. Cambridge: Cambridge University Press.

McAdam D and Scott (2005) Social Movements and Organization Theory. New York: Cambridge University Press.

Meardi G (2011) Understanding Trade Union Cultures. Industrielle Beziehungen 18(4): 336-345.

Offe C and Wiesenthal H (1980) Two Logics of Collective Action: Theoretical Notes on Social Class and Organisational Form. Political Power and Social Theory 1: 67-115.

Olson M (1965) The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge: Harvard University Press.

Pernicka S and Aust A. (eds) (2007), Die Unorganisierten gewinnen. Gewerkschaftliche Rekrutierung und Interessenvertretung atypisch Beschäftigter – ein deutsch-österreichischer Vergleich. Berlin : edition sigma.

Platzer H-W and Mueller T (2011) Global and European Trade Union Federations. Oxford: Peter Lang.

Schulten T (1998) Collective bargaining in the metal industry under the condition of the European Monetary Union. In : Kauppinen T (ed) The impact of EMU on Industrial Relations in European Union. Helsinki : Finnish Labour Association, pp. 207-224.

Schulten T (2003)Europeanization of collective bargaining. In: Keller B and Platzer H-W (eds) Industrial relations and European integration: trans- and supranational developments and prospects. Aldershot: Ashgate, pp. 112–136.

Scott R (1995) Institutions and organisations. Thousand Oaks: Sage.

Scott R (2008) Institutions and organisations: ideas and interests. 3rd edition. New York: Sage.

16

Streeck W and Schmitter P (1991) From National Corporatism to Transnational Pluralism. Organized Interests in the Single European Market. Politics and Society 19: 133–164.

Streeck W (1998) The internationalization of industrial relations in Europe: Prospects and problems. Politics and Society 26(4): 429–459.

Trampusch C (2010) The welfare state and trade unions in Switzerland: an historical reconstruction of the shift from a liberal to a post-liberal welfare regime. Journal of European Social Policy 20(1): 58–73.

Traxler F (2003) European Monetary Union and collective bargaining. In: Keller B and Platzer H-W (eds) Industrial relations and European integration: trans- and supranational developments and prospects. Aldershot: Ashgate, pp. 85–111.

Traxler F, Blaschke S and Kittel B (2001) National labour relations in internationalized markets, Oxford: Oxford University Press.

Traxler F, Brandl B and Glassner V (2008) Pattern bargaining: An investigation into its agency, context and evidence. British Journal of Industrial Relations 46(1): 33–58.

Traxler F and Brandl B (2009) Towards Europeanization of Wage Policy: Germany and the Nordic Countries. European Union Politics 10(2): 177-201.

Turner L (1996) The Europeanization of labor: structure before action. European Journal of Industrial Relations 2(3): 325–344.

Visser J and Ebbinghaus B (1992) Making the Most of Diversity? European Integration and Transnational Organisation of Labour. In: Greenwood J, Grote JR and Ronit K (eds) Organized Interests and the European Community. London: Sage, pp. 206–237.

Voss K and Sherman R (2000) Breaking the iron law of oligarchy: Union revitalization in the American labor movement. American Journal of Sociology 106(2): 303-349.

17

Table 1: Determinants of transnational trade union initiatives

Initiatives: Founding year and countriesa of participating unions

Late 1960s: ‚DACH‘ initiative (DE, AT, CH)

1970: ‚Nordiska Metall‘ (SE, FI, DK, NO, IC)

1993, 1998: EMF Collective bargaining coordination guideline

1997: Transnational bargaining network; Strongest in Central-Western Europe (BE, NL, DE) and Nordic Europe (SE, FI, DK, NO and IC)

Determinants

(Hypotheses)

Changes in economic and institutional environments

Crises of international monetary regime (Bretton Woods)

Increasing internationalisation of financial and product markets

Coordination of monetary policies (ERM, currency pegs to Deutschmark, except CH)

Crises of international monetary regime (Bretton Woods)

Oil Shocks Increasing

internationalisation of financial and product markets

Coordination of monetary policies (ERM, temporary currency pegs to Deutschmark, except FI, IC)

Market internationalisation and completion of EU single market

European monetary union (EMU, 1998): Centralisation of monetary policy, enhanced fiscal and macroeconomic policy coordination

Development of EU ‚social dimension‘ (e.g. European social dialogues, EWCs)

‚Eastern‘ enlargement of EU (2004, 2007)

Sources of institutional power:

Supportive governance structures (i.e. regulative and normative institutions) are largely confined to national level

Coordinated MEB systems (predominantly at sector level; CH sector and company level); High to medium-range union densities, practice to extend collective agreements (CAs), high employer density; high collective bargaining coverage;

Coordinated MEB systems (sector level); High union and employer densities; unions in DK, SE and FI administrate unemployment funds (‘Ghent system’), high bargaining coverage, strong role of social partners in public policy making.

(Diversity of welfare state and IR regimes)

Coordinated MEB systems (Nordic and central-western Europe, including Slovenia)

Coordinated MEB systems (Nordic and central-western Europe); Central-western Europe: Medium-range union densities (high in BE), practice to extend CAs (lower in DE), medium-range employer density; high CB coverage (lower in DE);

Nordic Europe: high union and employer densities;

18

- National industrial relations (IR) regimes

- Welfare state and IR traditions Institutionalisation of

class conflict to ensure social peace; High-trust relations between unions and employers‘

associations and enterprise-level social partners.

Conservative continental welfare state tradition based on political exchange and participation of social partners in public policy-making (exception CH).

CH: Mix of liberal and continental traditions with a shift to more centralized tripartite social policy concertation in the 1970s and 1980s.

Institutionalisation of class conflict to ensure social peace; Cooperative relations between unions and employers‘ associations; ‚productivity-coalitions‘ between employers and employees at enterprise- level.

Nordic social-democratic welfare state tradition (general and inclusive welfare state and maintenance of full employment).

Cooperative CB and universalistic welfare state tradition in Nordic Europe

Cooperative CB and conservative corporatist welfare state tradition in Central-Western Europe (incl. Slovenia)

unions in DK, SE and FI administrate unemployment funds (‘Ghent system’), high bargaining coverage;

Continental welfare state model in BE, NL, DE: highly institutionalised role of social partners in public policy making;

Nordic social-democratic welfare state model: highly institutionalised role of social partners in public policy-making;

Cooperative CB and universalistic welfare state tradition in Nordic Europe;

Cooperative CB and conservative corporatist welfare state tradition in Central-Western Europe

19

Supportive institutional logics

- Cultural beliefs, conceptions and practices at (trans-) national level

- The logic of coordination at transnational level

Common language of Germanophonic countries, common conceptions of industrial relations institutions and practices (derived from cross-country interaction and cooperation in economic and social sphere)

Strong belief in stability-oriented and solidaristic wage policy

Cultural and linguistic closeness of Scandinavia, long experience of cross-country interaction and cooperation in economic and social sphere, similarities in CB practices.

Strong belief in stability-oriented and solidaristic wage policy

(Diversity of cultures and practices)

Dominance of Germanophonic and Benelux unions within EMF

that promote orientation towards price stability and productivity-orientation of wage settlements as a common principle for European coordination of CB (‚best practice‘, ‘social learning’).

Cultural and linguistic closeness of Scandinavia (partly BE and NL); (derived from long experience of cross-country interaction and cooperation in economic and social sphere)

Strong belief in stability-oriented and productivity-based wage policy in DE, NL (metalworking);

Avoidance of transnational wage competition in BE (due to ‚Law on competitiveness’).

Stability-oriented and productivity-based wage policy in Nordic Europe;

a Country abbreviations: AT (Austria), BE (Belgium), CH (Switzerland), DE (Germany), DK (Denmark), FI (Finland), IC (Iceland), NL (Netherlands), NO (Norway); SE Sweden

20

Sources: Andersen 2006, European Commission 2009; Esping-Andersen 1990, Ferner and Hyman 1998; Trampusch 2010; Traxler et al. 2008, Traxler and Brandl 2009;

21

22

iEndnotes

We use the terms institutional work and institutional strategies synonymously. ii Isomorphism is a concept that captures the process of homogenisation. According to Haley (1968, cited in DiMaggio & Powell 1983; 149), ‘…isomorphism is a constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions’.iii Multi-employer bargaining refers to negotiations between trade unions and employers‘ associations while in systems of single-employer bargaining (SEB) trade unions negotiate with individual employers, i.e. company’s management.iv Conceptions contain structured world views that are bound to cultural and linguistic contexts (e.g., ‘German worker participation’ and ‘French co-determination’ refer to different conceptions of reality due to differences in institutional and cultural contexts). v Business Europe is the European peak-level organisation of national employers’ associations.

vi Different degrees or sub-logics of the respective two main logics are not considered here.