Sources: Brown et al. 1997; Ebbinghaus and Visser 1997; Jordana 1996; Kauppinen and Köykkä 1991; Visser 1989; Waddington 1992.

A range of factors contribute to the pattern of decline, which include; the nature of management organization and practices (Kochan et al. 1986), country size (Wallerstein 1989) and inflation rates (Crouch 1993). However, two factors are generally seen as most influential in western Europe; employment restructuring and unemployment (Visser 1994; Galenson 1994). In many member states of the EU there has been a shift in the composition of the labour force away from manufacturing towards private sector services. Between 1975 and 1995 the proportion of the labour force employed in services rose from 49 per cent to 64.5 per cent in the EU as a whole. In several states this shift has been accompanied by a decline in public sector employment, as privatisation and restrictions on public sector expenditure take effect. While there is considerable variation in the extent of these shifts, their overall effect is a movement of employment away from areas of traditional union strength to areas of relative union weakness. In most European countries unionisation is highest in the public sector, and is higher in manufacturing than in private sector services. Associated with these employment shifts are increases in the proportion of employment found at small and medium sized enterprises (SMEs - 66.2 per cent of the total EU workforce employed in enterprises of less than 250 employees in 1992), and increases in the number of part-time and contingent workers (in 1985 part-timers accounted for 12.5 per cent of the EU workforce and by 1995 this figure had risen to 16 per cent; in 1985 9.1 per cent of the workforce were employed on fixed-term contracts and by 1995 the figure was 11.5 per cent). Both of these trends also mitigate against trade union membership, as union organization traditionally tends to concentrate on larger sites and full-time workers. These shifts have contributed to membership decline, but cannot be regarded as the sole reason for the decline as many of these shifts in employment were underway during the late-1960s and early-1970s when trade union membership tended to increase.

High levels of unemployment have compounded the impact of employment shifts. With the exception of member states where unions retain a formal role in the administration of unemployment and welfare benefits, unions are poor at retaining unemployed workers in membership. Membership declines arising from high levels of unemployment were particularly sharp in Ireland, the Netherlands, Portugal, Spain and the UK, where union benefits are almost exclusively directed towards the workplace, rather than supporting the unemployed worker back into employment. 

These two reasons account for a large proportion of union decline during the 1980s and 1990s. A consequence of the decline in union density is that trade unions are less representative of entire workforces than they were during the late 1970s or early 1980s. If unions are to reverse this decline, union membership must be strengthened in weakly organized sectors of the economy. A further consequence of the decline is that public sector unions now tend to organize relatively greater shares of total membership than hitherto, resulting in shifts in the balance of power within union confederations. Membership decline and labour market restructuring raise many issues for European trade unions, three of which are considered here; recruitment, trade union structural reform, and union articulation and cohesion. In selecting these three issues we do not infer that these are the most important, but we argue that these issues will at the centre of the European trade union agenda in the foreseeable future.

A new emphasis on recruitment

Changes in the composition of the labour force have reduced the full-time male worker employed in manufacturing to a minority in most labour forces. The decline in the number of these workers, the traditional bedrock of union organization, has required unions to extend the scope of their recruitment. In practice, this means deepening recruitment in private sector services, among women workers, among part-time and contingent workers, at SMEs, and among young workers. A wide range of measures have been introduced in order to deepen recruitment in these areas.

Packages of financial services, such as credit cards and insurance discounts, have been offered as an inducement to membership by unions in Belgium, Denmark, Italy, The Netherlands and the UK. Implicit in their introduction is a changed relationship between union and member, based on individual consumer interest rather than collective industrial interests. In practice, unions offer discounted financial services on the basis of allowing exclusive access to the membership to specific companies offering such services. It is far from clear that these measures have been effective in attracting new groups of workers into membership. For example, survey evidence in the UK indicates that only 3 per cent of new members regard such services as being among their first or second reasons for joining a union (Waddington and Whitston 1997). Similarly, in Belgium and Denmark such services are of little appeal (Jørgensen et al. 1992). However, in the Netherlands individual financial services are claimed to have contributed to reversing membership decline during the 1990s (IRS 1994). It is argued that packages of financial services represent an adjustment by unions to a 'new individualism'. In this issue of Transfer Armingeon considers whether this development is associated with a new relationship between union membership and political behaviour in his study of voting patterns.

Improvements in the 'core' business of unions is on the agenda for most union movements. Unions have traditionally undertaken a social insurance role in representing individual members in their workplaces. This purpose has regularly been ranked as the prime reason for membership by workers. In The Netherlands joining a union to 'avoid personal problems in the work environment' is the primary reason for joining for both manual and white-collar workers (van de Vall 1970; Klandermans 1986). In the UK no fewer than 72 per cent of new members cite the provision of 'support if I have a problem at work' as being one of the two main reasons for joining a union (Waddington and Whitston 1997). Danish results are comparable as joining a union 'to ensure that my interests are protected' is ranked second after joining in order 'to become a member of the unemployment scheme' (Jørgensen et al. 1992). These results are consistent with evidence which shows union membership to have been more stable where local union organization is B (Hancké 1993). The legal underpinning of workplace organization is particularly influential on the establishment of workplace strength. However, it is also possible for unions to introduce some improvements to their workplace organization in order to arrest rates of membership decline. In several countries, for example, attention is now being directed towards examining how the quality of workplace union organization may be improved through the provision of additional training to union representatives and by making available a wider range of support and advisory services from the union centres.

Integral to the improvement of workplace organization is the participation of larger numbers of workers from under-represented groups. There is no agreement on how this can be achieved. Take, for example, the case of women. Some argue that organizations directed exclusively towards representing the interests of women are most successful in raising a range of issues of central concern to women (Trebilcock 1991; Cockburn 1983 1991). The long-standing independence, and recent membership growth, of exclusive unions for women in Denmark and The Netherlands supports this view. In recognition of this point many unions have established committee structures to facilitate the emergence of bargaining agendas which include issues of concern to women. The contrary argument is that these issues may be marginalised, rather than integrated, by the presence of exclusive organizations (Briskin and McDermott 1993, Crain 1994). Furthermore, items generated by such organizations for inclusion on the bargaining agenda are among the first to be dropped during negotiations (Colling and Dickens 1989; Cyba and Papouschek 1996). Furthermore, women are inadequately represented in the decision-making structure of most unions at national and European levels (Braithwaite and Byrne, n.d.). It is apparent that further reforms must ensure that, in addition to defining targets for representation and participation, the activities of women become integral to the decision-making and social processes of trade unions.

A range of other measures have also been introduced to encourage the membership and participation of under-represented groups. Discounted membership contributions for part-time and young workers are now widespread. Furthermore, a range of additional institutions to encourage participation have been introduced by many unions. Committee structures for young workers and ethnic minority workers are now common throughout European trade unions. However, the establishment of a union specifically for young workers by the Federatie Nederlandse Vakbeweging (FNV) in The Netherlands failed to attract large numbers of young members. Efforts to attract young workers into membership also encompass union sponsorship of music festivals and concerts. The GMB union in the UK sponsors Fulham Football Club and Metal in Denmark sponsors the national womens' handball team, thereby ensuring that union logos appear on the players' shirts. Several unions have also used advertisements on television in order to attract potential members. It is unclear, however, how many new members have been recruited by these means. However, IG Metall has recorded membership increases by offering incentives, such as radios or CD players, to existing members who recruit a minimum of ten new members.

While these measures may have slowed the rate of decline, they have yet to reverse it in most countries. Private sector services, in particular, are proving resistant to unionisation. Even in Sweden and Denmark, where union density is relatively high, it took long and concerted union campaigns to secure collective representation for workers employed by the American retailer Toys 'R Us, by the McDonalds burger chain, and at restaurants on Nyhavn in Copenhagen. Legal support for workplace organization is not sufficient to ensure the unionisation of private sector services, as the relatively low density in Austrian and German private sector services illustrate.

In the UK the Trades Union Congress (TUC) proposes to establish an 'Organizing Academy' at which young recruiting officers will be trained before being sent out to recruit in private sector services. This approach follows similar initiatives launched by the Australian Council of Trade Unions (ACTU) and the American Federation of Labor and Congress of Industrial Organizations (AFL-CIO). The rationale underpinning this approach is that existing full-time officers of trade unions have insufficient time to devote to recruitment activities, therefore, it is necessary to train additional officers who will be concerned exclusively with recruitment in the first instance. Although it remains to be seen whether this approach will be successful, it indicates the extent of the measures necessary to secure recruitment footholds in private sector services. Leisink's article in this issue of Transfer examines the issues associated with the recruitment of white-collar and professional workers.

Apart from unions which have a role in administering unemployment insurance, European unions have failed to recruit sufficient members in private sector services. Although a range of new recruitment tools have been introduced, they have proved inadequate to replace the members lost from manufacturing. Similarly, representative structures intended to promote the participation of under-represented groups have yet to yield their intended results in terms of union practices and policies.

Trade union structural reform

Arising from the changing composition of the labour force and membership decline are a range of policies intended to reform trade union structures. Mergers are central to such structural reform. Although the character of merger activity in each country is dependent upon existing forms of union organization, a number of features characterise many recent developments.

First, many of the unions that organize in declining industries or occupations have been acquired by larger unions. For example, unions of leather workers, woodworkers and agricultural workers have been acquired by their larger counterparts in Germany, The Netherlands, Sweden and the UK. Furthermore, in several other countries, merger discussions are underway involving unions with similar declining recruitment bases. The point here is that many sectors of traditional union strength are now not of sufficient size to support an independent union. Only by merging can members of these unions be retained within the trade union movement.

Second, the extent of merger activity is altering the role and internal politics of union confederations, and, in particular circumstances, is bringing into question the future of union confederations. Historically, union confederations tended to devote more of their resources to supporting smaller unions. As small unions were unable to fund extensive research, legal and technical services, support from union confederations enhanced the quality of services provided to members. On acquisition by larger unions these support services from union confederations are no longer required by the smaller unions, as they are available through the post-merger organization. The support role of union confederations is thus changing. Similarly, the internal politics of union confederations are in flux. Where membership size determines voting strength and the number of delegates, mergers have led to altered power relations. For example, between 1922 and 1993 the Transport and General Workers' Union (TGWU) was the largest union in the UK. However, the merger to form UNISON#1# resulted in a larger union than the TGWU within the TUC. Similarly, the sharp decline in the membership of the National Union of Mineworkers (NUM) has resulted in a loss of influence.

In particular circumstances, merger activity is bringing into question the future of union confederations. IG Metall (IGM) in Germany, for example, will organize around 40 per cent of the total membership affiliated to the Deutscher Gewerkschaftsbund (DGB) when it completes the acquisition of unions of textile workers and woodworkers. The enormous influence of IGM arising from its voting strength has led several other large unions to question whether affiliation to the DGB is still appropriate. To put this in practical terms: why affiliate to a confederation when little influence can be exerted on the policy of the confederation? An additional feature of merger activity in Germany is that other large unions are more actively engaged in merger discussions than hitherto, in an attempt to forestall IGM from achieving an overall majority of DGB members. Several mergers have been completed by IG Chemie Papier Keramik (IG CPK) and IG Bau-Steine-Erden, and merger discussions are underway involving Öffentliche Dienste, Transport und Verkehr (ÖTV), Handel, Banken und Versicherungen (HBV) and Deutsche Angestelltengewerkschaft (DAG). If these latter merger discussions are successful, the industrial organization that has characterised union structure in Germany since 1946 will be transformed. Furthermore, the membership of DAG, which is not affiliated to the DGB, will be incorporated within a DGB-affiliated union, thereby influencing issues of relative membership size.

A third feature of recent trade union structural reforms is the influence of changes in bargaining structure on the outcome of union structural change. This influence is no more marked than in Denmark. In addition to merger activity, which has reduced the number of unions affiliated to the Landsorganization i Danmark (LO) by about 50 per cent since 1970, Danish unions have established five industry-based cartels which are responsible for the conduct of bargaining in different sectors. In other words, structural reform was introduced in the form of cartels to facilitate adjustments to changes in bargaining structure from central agreements to industrial agreements. The decentralisation of bargaining structure has also influenced structural reform. In the UK, for example, bargaining decentralisation accelerated the development of conglomerate unions (Waddington 1995a). With the demise of industrial bargaining the political and industrial advantages of mergers with other unions in the same industry tended to diminish. As a result, many unions have merged with unions which offer better deals on support and membership servicing, rather than those with similar recruitment bases.

Finally, it should be noted that trade union structural reform has concentrated organization within nation states, rather than crossed national boundaries. Although separate discussions involving unions of print and postal workers from several member states have been mooted, no agreements have yet been reached. In the Deutsche Post Gewerkschaft (DPG) there is a debate between those who advocate merger with other unions organizing service workers in Germany and those proposing an international merger comprising unions of postal workers in other EU member states. Apart from supranational affiliations, trade union organization remains firmly based on the nation state. However, closer working relations have certainly been encouraged by European involvement, particularly in transport industries. A further form of closer working relationship has been established by IG CPK from Germany and the GMB from the UK. The agreement concluded by the two unions requires them to represent members working in Germany and the UK. IG CPK would thus have to represent a GMB member who was working in Germany. It remains to be seen how this agreement will operate in practice. For example, the GMB recruits engineering and construction workers, yet IG CPK does not have representation rights in these two industries in Germany. It does not seem likely that unions in engineering and construction in Germany will readily cede representation rights to the IG CPK in order for it to represent GMB members. However, senior officers from the two unions welcomed the initiative as representing the first step towards greater co-operation and the creation of a European trade union. A more wide-ranging initiative is proposed by the European Metal Workers' Federation, which suggests that a common membership for metalworkers throughout the EU be made available.

Articulation and cohesion

Initiatives to recruit the new workforce and trade union structural reform have led to increases in membership heterogeneity. In many unions heterogeneity is rising as overall membership is falling. Increasing membership heterogeneity has highlighted issues of trade union articulation, cohesion and interest aggregation. In short, can new forms of solidarity be generated within trade unions? As Richard Hyman's paper in this issue of Transfer addresses the issue of interest aggregation, we restrict our comments to identifying the different forms taken in the breakdown of union articulation.

Following Crouch, an articulated trade union movement is 'one in which B relations of interdependence bind different vertical levels' (1992:54). Wide-ranging evidence indicates that in countries where union structures are highly articulated, such as in Belgium or Sweden, union membership is more stable (Hancké 1993; Kjellberg 1983). Similarly, highly articulated union movements are able to co-ordinate policy formulation and implementation. Since 1980, however, very few union movements in western Europe have been able to maintain articulation, although there is considerable variation in the form of its decline. Key declines in articulation are noted regarded the isolation of workplace organization, increasing competition between federations, and changing relations between federations and affiliated unions. Where unions are weakly articulated it is difficult to prevent the exploitation of weaker local union organizations. Similarly, bringing influence to bear at national level is more problematic in circumstances where unions are unable to aggregate the divergent interests of their memberships.

Italy and the UK illustrate different forms of failure to articulate workplace organization. The rapid growth of workplace-based unions in Italy (Comitati della base or Cobas), autonomous from established unions and federations markedly reduced opportunities for articulated trade unionism during the 1980s. Local union organization attempted to win control over the bargaining agenda as managements sought to settle locally an increasing range of issues (Locke 1990). However, the reform of the system of workplace representation during the 1990s, in part prompted by national union policy, partially restored union-workplace organization relationship (Regalia and Regini, forthcoming). Similarly, the decentralisation of collective bargaining in the UK has made articulation between union organization in the workplace and regional and national centres more difficult to sustain. This issue is particularly acute in manufacturing, where bargaining decentralisation is more pronounced, compared to the public sector where many of the features of national bargaining are still in place.

Denmark and Sweden illustrate a second form of decline in articulation; the rise of competing confederations. In Denmark and Sweden the growth of white-collar unionism, coupled to the organization of such workers into confederations separate from manual workers, has led to inter-confederal disputes. In Denmark the Funktionärernes og Tjenestemändenes Fällesrad (FTF) and the Akademikernes Centralorganization (AC) now organize over half a million members, compared to the 1.4 million members of the LO. In Sweden a similar pattern is in evidence, as the combined memberships of the Tjänstemännens Centralorganization (TCO) and the Sveriges Akademikers Centralorganization (SACO), respectively the white-collar and the professional union confederations, approach that of the Landsorganizationen i Sverige (LO), traditionally the leading trade union confederation of primarily manual workers. As the LO lost its dominant position, a myriad of alliances between competing union blocks emerged which contributed to the 'great conflict' of 1980, prohibited the development of agreed union positions and, in turn, allowed employers to exploit the differences between unions (Kjellberg 1992; Hammarström 1994).

A third way in which articulation has diminished is in the weakening of links between union federations and their affiliated unions. Several union federations have few constitutional powers over the activities and policies of affiliates. For example, with the exception of the power to expel an affiliate, the position of the TUC and the DGB relative to their affiliated unions rests on the capacity to generate broad internal alliances. During the 1970s the TUC was able to strengthen its position relative to affiliated unions through two mechanisms. First, the TUC adjudicated in inter-union disputes. Second, the TUC acted as a conduit to government. Legislation enacted by the Conservative Government weakened the TUC's role in resolving inter-union disputes, and, more significantly, the dismantling of tripartite institutions eliminated the TUC's role as a conduit to government. These changes weakened the position of the TUC relative to its affiliated unions and prompted a major internal review of its role (Waddington 1995b). Similarly, disputes between unions affiliated to the DGB, changes in relative membership levels, and mergers between affiliates, have made the DGB's task of securing broad union alliances on policy matters more difficult. In consequence, the influence of the DGB has diminished and individual affiliated unions have assumed greater prominence.

Employers' relations with unions and workers

Since the 1970s the emphasis of European industrial relations has shifted away from concertation, neo-corporatism and politics to the industrial and, more specifically, to the firm (Regini 1992; Locke and Kochan 1995). These shifts co-incided with the decline in worker militancy, and employers taking a more significant role in driving through change to adapt to new forms of international competition. This shift towards the firm has been facilitated, albeit to different degrees, by state deregulation of the terms on which labour is used. Such deregulation has encouraged the growth of part-time and temporary employment, particularly in Italy, The Netherlands, Spain and the UK. Privatisation has also reinforced the emphasis on the firm and, in specific circumstances, has disturbed union organization, thereby further reducing membership levels. Although adaptation to international competition is neither universal nor even, three features directly impinge on union organization and on relations between unions and their members, bargaining decentralisation and coverage; the introduction of new forms of workplace participation; and the forms of workers' workplace representation, such as works councils. As Ida Regalia addresses inter alia workplace representation in her article in this issue of Transfer, the first two of these features are discussed below. Although varying markedly in institutional form, they raise issues for the key relationships identified in this paper between union and member and between union and workplace organization. The final part of this section assesses the pressures on this relationship in different institutional circumstances and identifies some of the responses developed by unions to employer-led change.

Decentralisation and the coverage of bargaining

Institutions of collective bargaining are intended to facilitate economic growth by encouraging ceteris paribus employee commitment. Collective bargaining set a floor below which wages would not be undercut. However, neo-liberals argue against this institutionalist orthodoxy and propose that institutions of collective bargaining should be dismantled, as they restrict the operation of the free market. Institutionalists and neo-liberals thus debate whether collective bargaining has a role in a modern economy. There can be little doubt that the influence of neo-liberalism has grown in recent years. Many aspects of decision-making have been decentralised to the level of the firm. This has involved shifts away from industry-wide or neo-corporatist bargaining arrangements towards local levels of settlement. Associated with decentralisation in several countries is the emergence of weaknesses in the coverage of collective bargaining. Katz (1993:13-16) identifies three specific reasons propelling decentralisation.

- Shifts in bargaining power. Whereas centralisation was a success for unions in that it took wages out of competition, decentralisation is a reversal.

- Changes in work organization. Workforce restructuring around new technologies and forms of work organization have required local settlements.

- Diversification of corporate structure or worker interests. Decentralisation is a product of corporate decentralisation and facilitates the independence of business units or profit centres. Associated with corporate restructuring, it is argued, is a breaking down of workers' solidarity, thereby reducing the attraction of centralised bargaining to labour.

The forms of decentralisation vary across countries, reflecting the different prevailing political and institutional conditions when decentralisation was initiated. These are considered below, together with an assessment of weaknesses in the coverage of collective bargaining.

Many of the decentralisation initiatives, however, have taken place against a backdrop of tripartite social pacts. During the mid-1990s neo-corporatist social pacts or incomes policy agreements were implemented in Ireland (1987), Italy (1993), Finland (1995) and Portugal (1996) in order to establish a stable macro-economic environment with the objective of reducing inflation and unemployment. Of course, such developments assume some centralisation of collective bargaining and union involvement at national level (Fajertag and Pochet 1997). Trends towards decentralisation have also required many unions to act locally. 

In Sweden highly stable neo-corporatist arrangements prevailed until the 1980s, in Finland they prevailed until the early 1990s, and in Denmark they remain largely in place. However, recent developments are characterised as 'centralised decentralisation' (Due et al. 1994), in which the central employers' associations retain a strategic role in the conduct of decentralisation. The central employers' associations in Denmark and Sweden retained this key strategic role by organizing several cartels of employers associations through which agreements affecting small and medium sized enterprises are concluded, and also by setting up a range of support functions that affiliated employers call upon when conducting bargaining. Decentralisation in the Danish and Swedish public sectors is associated with the introduction of differentiated or individualised wage rates, bargained at the local level within a framework set by national agreements. Integral to this pattern of centralised decentralisation are relatively high levels of bargaining coverage. Bargaining coverage is estimated to be 83 per cent of employees in Sweden, 95 per cent in Finland, and 74 per cent in Denmark (Traxler 1996b). However, a recent Danish survey shows that bargaining coverage is much lower than the national average among salaried employees and among workers in private sector services, primarily due to the reluctance of employers to join employers' associations (Scheuer 1997).

In contrast to Denmark and Sweden, institutional stability characterises the bargaining arrangements in Austria and Germany. Although under considerable pressure from small and medium-sized companies, employers' associations have remained largely intact and retained a co-ordinating function in bargaining. In practice, decentralisation is taking the form of a wider range of issues being handled through works councils than hitherto. In Germany, following reunification, the use of 'opening clauses' (Öffnungsklauseln) has burgeoned in order that company-specific terms can be settled within a framework of industry bargaining. During 1997 the first opening clause on wages was agreed. This covered the chemicals industry and allowed companies to reduce pay by up to 10 per cent for a limited period either to save jobs or to improve competitiveness. This was followed by similar developments in other sectors, including the influential engineering industry. In both countries the coverage of collective bargaining is high (Austria, 98 per cent; West Germany, 82 per cent: Traxler 1996b). However, works councils are more sparse in private sector services and among white-collar workers, a situation replicated in terms of the workplace organization of unions (Jacobi and Müller-Jentsch 1990; Jacobi et al. 1992). In other words, in the areas of likely employment growth, workplace representation in Germany is at its weakest.

In Belgium, The Netherlands and Luxembourg moves towards the firm were achieved without widespread disturbance to existing institutions; 'the deep embeddedness of industrial relations institutions in national life, and the fact the employers can usually achieve flexibility at company level when they want it, inhibit changes to the inherited mechanisms of regulation' (Crouch 1994:217). Although some national institutions were withdrawn in Belgium during the early 1980s, most were reinstated shortly afterwards. There are marked differences in the coverage of bargaining; Belgium, 90 per cent and The Netherlands 71 per cent in 1990, having fallen from 76 per cent in 1980 (Traxler 1996b). The Dutch case replicates the countries mentioned above, however, in so far as it is private sector services that tend not to be covered by bargaining (Valkenburg 1995).

In Italy and Ireland there is no universal trend apparent in bargaining structure. If there is a pattern, it is one of oscillation. During the 1980s some bargaining institutions were dismantled, but not on a consistent basis between industries. New attempts to find neo-corporatist solutions were instigated during the early 1990s in Ireland (von Prondzynski 1992) and Italy from the mid-1990s (Regalia and Regini, forthcoming). Similarly, it is difficult to categorise the direction of change in France, Portugal and Spain, as attempts to institutionalise bargaining arrangements have featured Bly since 1980. As a consequence, bargaining institutions have been strengthened at company, sectoral and national levels, albeit with mixed results (Goetschy and Rozenblatt 1992; Martinez-Lucio 1992; Barreto 1992). In France and Portugal institutionalisation resulted in marked increases in the coverage of collective bargaining during the 1980s to 92 per cent and 79 per cent respectively (Traxler 1996b). The rise in bargaining coverage in Portugal also coincided with the implementation of a form of neo-corporatist social pact. As part of the expansion in France the Confédération Française Démocratique du Travail (CFDT) secured union rights at McDonalds, a company that has resisted unionisation in many EU member states (IRS 1996a). Weaknesses in bargaining coverage in Portugal and Spain, however, remain in private sector services where employers resist unionisation and involvement in bargaining.

In the UK decentralisation was wide-ranging and was accompanied by the dismantling of long-standing bargaining institutions. Employers' associations collapsed in many industries and unions were excluded from any role in macro-economic policy formulation. The settlement of pay and conditions within many larger companies was decentralised to business unit or establishment level, often within limits determined at company head offices. The UK thus represents an extreme point on the centralisation-decentralisation continuum in Europe, marked by its 'disorganized' character. Furthermore, the coverage of collective bargaining declined sharply from 72 per cent in 1972 to 47 per cent by 1990 (Brown 1993). Apart from some areas of retailing and in banking and insurance, unions and bargaining are almost non-existent in private sector services. Furthermore, employers have been most actively derecognising unions for white-collar staff (Claydon 1996).

The implications of decentralisation and weaknesses in the coverage of bargaining for trade union organization are as wide-ranging as the developments themselves. Trade unions are now required to provide collective bargaining services on a larger range of levels than before. However, three features are of specific concern.

The institutional form of decentralisation differs from country to country, making it difficult for unions to establish 'best practice'. Several of the changes to trade union structure mentioned above reflect the range of available options. Common to most of these options is the need to review relationships between central and workplace union organization.

As more issues are handled within the workplace, trade unions will have to assess the quality of support they provide to members at the workplace. Where trade union membership is declining this may require taking some difficult decisions in allocating scarce resources.

The coverage of bargaining tends to be weaker where employment growth is likely to be Best, in private sector services and among white-collar workers. It will be necessary to extend bargaining and union organization into these areas in order to exert an influence over terms and conditions. 

Participation and employee commitment

Associated with decentralisation are measures introduced by employers to increase flexibility in work organization and the deployment of labour. Approaches to flexibility have been different among employers. At one end of the continuum, employers emphasise the removal of the constraints of the post-war settlement as the key to achieving flexibility. This may include the exclusion of unions or resistance to a union presence at the workplace. These employers are also most likely to seek improvements in competitiveness by reducing costs rather than introducing new methods of work. Employers taking this approach to flexibility are close to the neo-liberal ideal. At the other end of the continuum are employers endeavouring to introduce new workplace practices, forms of internal labour market and to generate employee commitment. The recent Green Paper of the European Commission reaffirms its commitment to this view, based on 'high skill, high productivity, high quality and good real wages' (Commission of the European Communities 1997). While this approach does not directly threaten a union presence, it brings into question many traditional union practices and often requires innovative union responses. The majority of employers are to be found at different points between these extremes. Our purpose is to identify some of the most common measures introduced by employers and suggest their implications for union organization.

Forms of participation are central to many approaches intended to achieve greater workplace flexibility. Two forms of direct participation have been introduced on a more widespread basis in recent years, financial participation and employee involvement in problem-solving; for example, by means of team-working, quality circles and total quality management. In the United States where participation schemes have been implemented at non-union sites, union recruitment campaigns have tended to fail due to the commitment of workers to the company generated by the schemes (Kochan and Wever 1991).

During the 1970s many labour movements engaged in campaigns to secure forms of economic democracy. Although some long-lasting successes were achieved in Sweden, most of these campaigns dissipated during the 1980s and were followed by management attempts to encourage financial participation, thereby linking pay to the performance of the enterprise and widening employee share ownership arrangements. These schemes are intended to promote employee commitment to the enterprise and have increased in coverage throughout most EU member states since 1980, although national differences remain (Kester and Pinaud 1996). Traditionally most unions have been sceptical, if not opposed, to financial participation schemes, as they tended to be unilaterally introduced by management, widened workplace inequalities and reduced the influence of collective bargaining. However, in countries such as France, where financial participation schemes are regulated by legislation which obliges managements to settle the terms of profit-sharing schemes with worker representatives, trade unions have recently adopted more pragmatic approaches, based on securing agreement on the terms of implementation. The contrast between the French approach and that in Belgium and the UK, goes some way to explaining the views of unions towards such schemes. In Belgium and the UK financial participation schemes tend to be introduced unilaterally by management, which have often used them as part of a strategy to exclude unions (Vaughan-Whitehead 1996)#2#. However, some local unions are now attempting to influence the implementation of such schemes, due to the interest expressed in them by members. In other words, opposition from unions at national level is being undermined by developments within the localities.

A second direct form of participation that has increased in coverage over recent years is participation in workplace decision-making. A variety of participation schemes have been introduced by employers to generate workplace flexibility through commitment. These range from autonomous work groups, where workers have some autonomy in the control and conduct of their work, to the increased use of suggestion schemes, sometimes linked to a cash reward if a suggestion is implemented to cost-saving effect.

Unions from Denmark and Sweden have traditionally pursued an interest in workplace decision-making, particularly in the areas of job design, work environment and work organization (Sandberg 1995). Elsewhere, until the late-1970s, unions tended to oppose direct participation on the basis that it might lead to the marginalisation of the union at the workplace. Much of this principled opposition dissipated during the 1980s under the influence of three factors; state promotion, acceptance by employees, and the number of schemes unilaterally imposed by employers. France, Germany and the UK illustrate this range of influences.

During the early 1980s in France the Socialist government formalised some existing informal arrangements in terms of rights to direct workplace participation (le droit d'expression). These measures were principally concerned with the working environment. However, the momentum behind this initiative fell away in the face of high levels of unemployment during the mid-1980s and divisions among the trade unions toward direct participation (Pinaud 1996). In other words, economic and political circumstances undermined state-prompted initiatives, leading the field clear for employer-imposed schemes.

In Germany the combination of well-articulated union organization and legally supported works councils enabled the major unions to shift their position from one of opposition to direct workplace participation schemes to one of support, conditional on agreement being reached before implementation. This shift was initially conditioned by the acceptance of direct participation by employees (Müller-Jentsch and Sperling 1995). By insisting that works agreements on direct participation be settled before schemes were implemented, unions, in co-operation with works councils, have been able to extend the bargaining agenda. In particular, a higher priority has been placed on training issues, to ensure that all members within a team are able to perform the duties required of the team (Streeck 1993). The pursuit of a training agenda has thus promoted functional rather than numerical forms of flexibility.

Although many unions in the UK remain formally opposed to direct 

participation schemes, the large number of such schemes unilaterally imposed by employers brings this formal position into question in practice. In consequence union organization in many workplaces has been left to fend for itself. Where workplace organization is B, it has been able to place representatives in key positions within the direct participation scheme, for example as team leaders, thus retaining some traditional workplace controls (Heaton and Linn 1989). However, this development is the exception. Where less B union organization is present, many direct participation schemes have contributed to union exclusion (Smith and Morton 1993). In other words, the adversarial bargaining characteristic of the UK, remains in place and influences the policy of the parties and the outcome when direct participation schemes are introduced. Where unions have not established a presence, direct participation schemes contribute towards the commitment of the employee to the company, but do not explain the failure to unionise (McLoughlin and Gourlay 1994).

Forms of union influence at the workplace

Evidence from several countries suggests that managements of unionised workplaces are more likely to introduce innovations in participation and work organization (Eaton and Voos 1992; Sorge and Streeck 1988). Furthermore, in the absence of unionisation, managements appear less likely to implement change, with the result that Dickensian 'Bleak Houses' remain, characterised by low pay, low skill and low trust (Sisson 1993). The central issue is the extent of influence unions can bring to bear on the nature and form of workplace restructuring. Different institutional configurations clearly effect the influence unions can exert. We have identified three institutional variants to illustrate the pattern of influence. The first of these is labelled adversarial bargaining as characterised by the UK during the 1980s and 1990s. The second variant is that of weak institutional regulation, as associated with France, Portugal and Spain. The third variant is that of B institutional regulation where the legal framework promotes social bargaining and the integration of workplace organization into managerial decision-making, as in Austria, Germany and Sweden.

In the UK many employers have restructured without attempting to establish a consensus with unions. This adversarial approach has been fostered by the neo-liberal economic and political agenda of Conservative governments since 1979, integral to which was the exclusion of union influence. 'Them and us' attitudes remain firmly entrenched at the workplace (Kelly and Kelly 1991), and many participation schemes are intended to secure the commitment of workers to the company at the expense of that to the union. However, evidence suggests that at highly unionised workplaces it is possible for unionists to reverse some of the losses sustained on the introduction of workplace restructuring by positioning shop stewards as team leaders, by establishing networking arrangements to ensure links between plants are maintained, and by negotiating for the introduction of worker-oriented benchmarks (Heaton and Linn 1989; Beaumont 1992; Martínez-Lucio and Weston 1995). The rapid rate of bargaining decentralisation also adversely affected union articulation and cohesion. In an attempt to mitigate the impact of these adverse developments, the TUC and several affiliated unions have advocated 'social partnership' at the workplace, in which issues of common interest to employers and unionists - such as, training, job security, and health and safety - are addressed in partnership. This national initiative has yet to be endorsed by the Confederation of British Industry (CBI), the largest employers' federation, and has yet to filter down to most workplaces, where efforts to intensify work have resulted in management attitudes becoming the principal source of workers' grievances (Waddington and Whitston 1995). Furthermore, it appears that employers are now acting strategically in their preparations to derecognise the unions of relatively well organized groups of workers (Claydon 1996, Higgs 1994). In other words, if social partnership is to become widespread in the UK, employers as well as unionists must move away from an adversarial approach to bargaining. As many institutions through which social partnership may have been fostered were dismantled after 1979, it is difficult to imagine how this transformation will be brought about in the short term.

In contrast to the deregulatory impetus that characterises developments in the UK, in France, Portugal and Spain initiatives were launched to regulate relations between employers and unions, thereby integrating workplace union organization with managerial decision-making. The most notable example of this shift towards legal underpinning were the French Auroux reforms of 1982 which extended legal protection to local union organization in all workplaces, irrespective of size, and obliged employers to conduct annual bargaining rounds over pay, working time and other terms of employment. Facility time for local activists was improved, as were disclosure of information provisions. Although not as wide-ranging as the Auroux reforms, legislative underpinning in Portugal granted the comissões de trabalhadores (workers' commissions) consultation rights on restructuring and on decisions concerning working hours and grading (Barreto 1992), while in Spain the Ley Orgánica de Libertad (Law of Trade Union Freedom) provides statutory support for workplace union branches and specifies workplace facilities for local union activists (Martinez-Lucio 1992).

For unions these reforms have generated contradictory pressures. It was anticipated that these reforms would lead to a new relationship between employers and unions in France (Erbčs-Seguin 1984). However, participation in workplace institutions remains muted (Wilson 1991), and it is argued that their introduction has accelerated membership decline (Goetschy and Rozenblatt 1992). While French unionists in the localities have been able to conclude an increasing number of local agreements and have extended the scope of the bargaining agenda (Boulin 1996), their links with national organizations remain perilous. In some circumstances local bargaining has replaced engagement in industrial bargaining, thereby further jeopardising union articulation and making it more difficult to maintain a common union position across localities. In Portugal and Spain similar pressures are reported, as weak, central union organization is unable to support the large number of local union initiatives (Barreto 1992; Escobar 1995). Unions in France, Portugal and Spain remain heavily reliant on their capacity for national mobilisation. At the workplace, they are relatively weak although institutions are in place through which they might exert influence over workplace change.

Where unions are both articulated and integrated into management decision-making, they are best placed to exert influence over workplace change (Turner 1991:241). Certainly in countries where national legal frameworks promoted social dialogue, unions have adapted more readily to workplace change and have been able to develop, if not always implement, an alternative agenda. The case of IG Metall in Germany is illustrative in two regards. First the union developed a framework proposal for involvement in direct participation (Turner 1992:229). Second, it has attempted to implement a training programme in conjunction with workplace change from which members directly benefit (Mahnkopf 1992). In practice union involvement in workplace restructuring highlighted functional rather than numerical flexibility. The high levels of training in Germany facilitated this strategy. Such a strategy would be difficult to implement in large parts of the UK economy due to the emphasis on numerical flexibility and low skill. However, similar programmes have been introduced by the LO in Denmark and, under B influence from Metallarbetereförbundet, by the LO in Sweden, where there is also an emphasis on training and skills.

The Europeanisation of labour: some foundations in place

Supranational developments contrast with the decentralisation and the accentuation of the workplace that characterises national trajectories. Two features of the supranational environment have had particularly wide-ranging effects on organized labour. First, heightened international competition, coupled to the growth of trans-national corporations, has weakened the position of unions, which steadfastly remain organized on a national basis. Indeed, if supranational developments lead to a decline in the influence of the nation state, as envisaged in early neo-functionalist analysis (Haas 1957), labour is Best at a point of declining influence. Second, the rejuvenation of European integration in the form of the Single European Act, the Maastricht Treaty and proposals for European Monetary Union (EMU), necessitates supranational organization and activity in order to influence policy formulation. Assessments of the impact of these developments are as wide-ranging as the developments themselves. On an optimistic note, Jacobi (1996) sees EMU as likely to accelerate the introduction of European-wide collective bargaining within which a union influence would be secure. This prospect is explored by Hege in her article in this issue of Transfer. A more pessimistic outcome is foreseen by Streeck (1994; 1995). He views the process of European integration as structurally biased towards capital and, more particularly, towards neo-liberalism. The shift to setting minimum standards in the Action Programme of 1990 and away from moving towards upward harmonisation, which informed social policy initiatives during the 1970s, is noted as indicative of the contradictory position taken by unions in both attempting to secure greater protection for labour within a political order that they are simultaneously trying to transform.

There is no doubt that structural impediments restrict the labour agenda. The absence of suitable employers' associations and the weaknesses in the competencies of the different institutions of the EU in labour relations are often cited in this regard (for example, Streeck 1994; Dølvik 1996). The political, social and economic diversity of member states compounds these difficulties. The argument advanced here, however, is that a broad range of institutions are in place through which labour may develop policy objectives and has secured policy gains. We do not argue that labour is Europeanised, nor do we argue that institutional development is complete. We argue that the rudiments of structures which might form the basis from which more wide-ranging policy objectives will be sought are emerging. The outcome of this ongoing development is intertwined in the tension between pressures towards deregulated competition and pressures towards regulation or re-regulation. As there is no predetermined outcome of this tension, the issue for trade unionists is to struggle to resolve it in their favour. To illustrate this argument we examine the evolution in the politics of the European Trade Union Confederation (ETUC), developments in European collective bargaining, and the emerging role of European Works Councils.

The ETUC and evolving policy objectives

Since the 1970s the ETUC has enlarged its coverage by accepting new organizations into affiliation, irrespective of their ideological tradition, and is now a unified European trade union confederation. It represents about 52 million trade unionists throughout Europe. Following the ETUC Congress in 1991, the European Industry Committees (from 1995 called European Industry Federations) became full members of the ETUC, thereby consolidating a bifurcated geographical and sectoral structure. Whether the confederal or the sectoral element of the Confederation should be pre-eminent in trade union integration remains an open question and a source of internal tension. In addition, in 1995 and 1996 the ETUC Executive approved the membership of a total of twelve trade union confederations from six central and eastern European countries, while observer status has been granted to a further four confederations#3#. The French Confédération Générale du Travail (CGT) is the most significant organization that remains outside of the ETUC, although it has applied to join and seems likely to do so within the next two years. In addition, organizations of civil servants, and professional and managerial workers retain separate minor confederations.

The sheer breadth of coverage, heterogeneity of membership and the embeddedness of member organizations in diverse political and economic institutional environments, combine to ensure that co-operation is always contentious, necessitating consideration of competing interests. The ETUC has no direct constitutional power over member organizations. In consequence, the ETUC is reliant on the authority affiliates are prepared to delegate and on generating broad alliances among member organizations while protecting minority interests. During the 1970s national union federations were not prepared to permit the ETUC to initiate campaigns on behalf of European trade unionists. However, this position changed in 1995, since when the ETUC is mandated to negotiate at the European level by member unions.

Throughout much of the 1970s these limitations combined to restrict the ETUC to principally an administrative role, focused on lobbying European institutions and the co-ordination of national union policies together with, very occasionally, activities. With the acceleration of the integration process during the 1980s, the ETUC sought to integrate itself more closely into the emergent European polity and, in particular, within the 'social dimension complex' (Dølvik 1993), comprising the European Parliament, the Economic and Social Committee and the European Commission. This integration was encouraged by the Delors presidency as part of a strategy to develop a wider-ranging social dimension (Ross 1995: 150-151). Operating within this network, the ETUC has evolved as a political actor and as an efficient lobbying organization (Goetschy 1996). The structural reforms introduced in 1991 involving the European Industry Committees were part of the process of adaptation to this new role.

There are three principal limitations to this pattern of ETUC development. First, affiliated unions remain embedded within national perspectives and continue to pursue national objectives. This is exacerbated by the failure of affiliated unions to rotate staff through their European offices. The sovereignty that member states have yielded to the EU level has yet to be matched by similar developments among affiliated unions. Second, in the absence of adequate financial support from affiliates, the ETUC is heavily reliant on support from the Commission, making it vulnerable to political shifts within the Commission. This reliance may tie the ETUC to the strategy adopted for European political integration by the Commission even though it may damage the interests of unions and the workers they represent (Martin 1996). For example, the transition process to EMU chosen by the Commission does not assist in the generation of full employment. Although the ETUC raised reservations about this process (Ramos Yuste and Foden 1995; Foden 1996), the point remains that the reliance on superstructural linkages creates further ambiguities in developing an 'independent' ETUC policy. A third and associated limitation of the ETUC's pattern of development is that it has not fostered a transnational workers' identity. In consequence, the diverse interests and identities that comprise the ETUC are largely unaltered. Indeed, transnational workers' identity generated through action may result from opposition to integration rather than in its support (Tarrow 1995).

Although these limitations have restricted the pursuit of some wider policy options, the ETUC has been successful in institution building, in campaigns to influence the content of the social dimension, and in influencing the development of a social policy regime which accommodates aspects of national diversity. Through its involvement in the emerging European polity, the ETUC has campaigned to some effect during the Val Duchesse discussions, launched in 1985, and on the Social Charter (1987-1989), the Action Programme (1990-1993) and the Maastricht Treaty (1991). Furthermore, the gains secured by the ETUC through this involvement has influenced the development of the Union of Industrial and Employers' Confederations of Europe (UNICE), which, in turn, may promote further opportunities for the ETUC (Pochet and Arcq 1997).

The ETUC has also influenced the development of a social policy regime comprising the elements of social dialogue and legislation. This regime has been established against the wishes of employers and may allow the ETUC to develop a role more independent of superstructural linkages. Furthermore, both the social dialogue and legislative elements of the social policy regime meet key ETUC requirements; namely, that negotiations are restricted to some employment and social policies, and that any workers' rights are the subject of legislation. Negotiations within the Social Policy Agreement attached to the Maastricht Treaty, now integrated in the Amsterdam Treaty, have resulted in European-level framework agreements covering parental leave and part-time work. These negotiations have involved the ETUC, UNICE and the European Centre of Enterprises with Public Participation and of Enterprises of General Economic Interest (CEEP). The ETUC was successful in these areas only because it initially lobbied for legislation. UNICE calculated that it would be preferable to conclude agreements rather than submit to legislation. In other words, the ETUC was able to bring the employers' associations to the negotiating table and secure agreement because of the strategy it adopted. By way of the legislative route, the ETUC was influential in promoting measures on equality between men and women and on collective redundancies. Furthermore, a range of health and safety measures were implemented from the mid-1980s. Employers had blocked earlier broader initiatives. However, more recently the Directive on European Works Councils (EWC) has been implemented. The subject and content of this Directive reflects issues of long-standing ETUC concern, and the extent to which the ETUC able to implement a European agenda when opposed by the more defensive employers' associations (Goetschy 1996).

Since the mid-1980s the ETUC has established an institutional framework and a mode of operation through which policy objectives are being sought and achieved. The approach adopted by the ETUC has linked it closely with the emerging European polity and has not yet involved it in the generation of a transnational workers' identity. While there are limitations arising from this approach, they have not prevented the realisation of some long-standing ETUC policy objectives. It is to one of these that we now turn.

Towards European collective bargaining?

A central objective of the ETUC has been the development of some form of European collective bargaining, to protect working conditions in the face of the gathering momentum of global competition. Several commentators view European Monetary Union as a key stimulus to such a development, as it is likely to result in greater transparency for comparative analyses of employment terms and conditions, and of national variation in the costs of living (Coldrick 1991; Keller 1995). Some see such a development as necessary in the short term if trade unions are to prevent employers from 'regime shopping' in search of lower wage and social costs for production, as occurred in the often-cited case of Hoover (Jacobi 1996; IRS 1993).

Less optimistic analysts point to the durability and diversity of national systems of collective bargaining; the absence of employers' associations coupled to widespread opposition to such a development from employers; the absence of overt support from the Commission; and the unwillingness of unions to cede authority on bargaining issues to European Industry Federations (Due et al. 1991; Traxler 1996a; Streeck 1995). In combination, these features are viewed as likely to undermine movements towards European collective bargaining. This view is particularly Bly held by commentators who conceive of European collective bargaining in terms of pan-European sectoral bargaining, and by those who accentuate the homogeneity of national systems.

While we acknowledge that pan-European sectoral bargaining is a long way off, there are signs that suggest a rather different pattern of development is underway. National collective bargaining policies are influenced by The Maastricht Treaty convergence criteria. This influence is particularly B in the public sector, which is being increasingly squeezed by restrictive budgetary policies implemented by member states. A single currency and the resulting disappearance of exchange rate adjustments will also have wide-ranging effects on incomes policies and social policies,

'if exchange rates no longer serve as a sluice gate to even out the economic levels of different countries, then pay and labour costs take on this function and the role of sluice-gate keeper calls to the trade unions' (Altvater and Mahnkopf 1993: 246).

Noé advances a further aspect of a similar argument in stating,

'In the absence of a centrally co-ordinated (but not necessarily centralised) fiscal policy, and without an equally institutionally safeguarded wage and incomes policy, a common Europe of money and capital, rather than one of work and employment will come into being' (1996:43).

The urgency of tackling the question of Ber coordination of collective bargaining policy at European level emerges clearly from the approach of the Belgian government after the collapse of negotiations on an employment pact. The government decreed that pay increases must not exceed average pay increases in the neighbouring countries of Germany, France, the Netherlands and Luxembourg. This, of course, is state-decreed coordination of collective bargaining. Even before the implementation of the third stage of EMU there is thus an acute need for trade union action.

However, trade unions are still some way distant from a convergent and coordinated collective bargaining strategy. The considerable diversity of national bargaining and pay systems inhibit movement towards a coordinated European bargaining policy. The ETUC proposed closer cooperation on realising that Maastricht and EMU left little room for manoeuvre (ETUC 1993). The amendments to the statutes adopted at the 8th Congress of the ETUC held in May 1995 in Brussels, introduced a concrete negotiating mandate (Article 13), which served to clear some formal internal hurdles, but little more. At the European level, the framework agreements on parental leave and part-time work are the outcome of negotiations between the social partners. At the European sectoral level developments are more limited. Although there are numerous joint texts and opinions on Commission proposals, these do not mandate national affiliates and, hence, are not necessarily diffused to the national level. At the company level there are an increasing number of European Works Councils (EWCs), discussed in more detail below. These components, comprising 'virtual collective bargaining' (Marginson and Sisson 1996), represent the development of a hybrid form of European collective bargaining in which features of company-level, national, and supranational systems may be integrated. 

The challenges for trade unions arising from these developments are wide-ranging, particularly for those organizations with limited resources. The Europeanisation of bargaining also requires the institutional development of employers' organizations empowered to conclude agreements. The statutes of UNICE contain a right of veto which makes it extremely difficult for employers to commit themselves to binding terms required for the negotiation and conclusion of agreements. The institutional difficulties of European employers can be illustrated by reference to engineering where no fewer than eighty different organizations exist, all of which point out that they are not empowered to deal with social policy issues. In short, there are no employers' counterparts to the European Industry Federations. The issue that this shortfall raises for trade unions in no new. During industrialisation in most member states it was trade unions that encouraged the development of employers' organizations. However, during recent years there have been some small steps made toward Europeanisation among employers. In construction and distribution, for example, where representative employers' organizations are established, joint opinions have taken on a somewhat wider significance than elsewhere (Baumann et al. 1996; IRA 1996b). Furthermore, some employers have implicitly afforded additional status to European Industry Federations by approaching them to assist in establishing European Works Councils.

The crossing of frontiers and the opening up of new prospects for a Europeanisation of industrial relations will depend on how far the instruments already to hand (social dialogue, in particular at sectoral level, EWCs and the coordination of collective bargaining policy) can be developed; how trade unions can break down their national identification; and how a legal framework is created or developed at European level. It is to one of these developments, EWCs, to which we now turn.

European Works Councils

The European Works Council (EWC) directive, adopted September 22 1994, represents a further stage in the institution-building process of European unions. The Directive is a significant step in an ETUC campaign of more than twenty years duration to improve workers' information and consultation rights. As the Directive requires workers from different member states to meet at least once per year, it may encourage the development of a transnational workers' identity and new forms of transnational union collaboration. Furthermore, implementing EWCs has strengthened the position of several European trade union federations, as managements have preferred federal involvement in the negotiation process leading to the establishment of EWCs.

Employers opposed the Directive. UNICE (1991) found the legislative component of the draft directive 'totally unacceptable' and viewed it as adding a bureaucratic tier which would be inconsistent with the wide range of company-specific practices it would overlay. For example, Peter Reid of the Engineering Employers' Federation in the UK states,

'We saw the directive as being too rigid, and bearing no relation to the real shape of industrial relations. Every company has a different culture, different shareholders, different employees and the sectoral approach simply cannot work' (quoted in ETUC/ETUI 1996).

Furthermore, employers argue that the introduction of EWCs runs counter to the general tendency towards decentralisation and that consultation is best undertaken at workplace level rather than supranationally. Despite this opposition, ETUC lobbying ensured inclusion in the Directive of several key union prerequisites, notably, that legislation supported the establishment of EWCs. Although the ETUC expressed disappointment at the weakness of the consultation rights and the high workforce-size thresholds, the Directive is viewed as the foundation onto which a range of other measures can be added, either through further ETUC lobbying or by negotiation within EWCs (Coldrick 1991; Hoffmann 1997).

Responses from ETUC member unions varied, reflecting the institutional situation in which they were embedded and the structure of capital within their country of origin. In the UK the TUC welcomed the Directive as a means by which collectivism could be reinforced, although a neo-liberal Conservative Government was in office. Furthermore, the TUC also saw in the terms of the Directive possibilities of initiating a debate about the future of workplace representation under of Labour Government. From a position of relative strength, the DGB welcomed the Directive, but expressed reservations about the high thresholds of 1000 and 150 employees, the inadequacy of the rights to information and consultation, the inability of workforce representatives to exert influence within a framework allowing a minimum provision of one meeting per year, the capacity of employers to undercut minimum provisions by concluding voluntary agreements, and the length of the negotiating period (see Lecher 1995). In The Netherlands FNV emphasised a third perspective, in arguing that EWCs should be seen as a second system of representation which complements the domestic system (Stoop 1994). A foremost issue for FNV is thus ensuring that the relationship between domestic and supranational systems of representation does not develop into an issue of regime competition, whereby managements are able to undercut one system of representation by reference to the other.

At the time of writing about four hundred EWCs have been set up. These are primarily, but not exclusively, in manufacturing, reflecting the weaknesses of labour organization in private sector services mentioned in the previous section. Critics of EWCs have labelled them as 'neo-voluntarist' (Streeck and Vitols 1993) or as indicative of 'enforced voluntarism' (Krieger and Benneton 1995). Certainly recent events at Renault have afforded this view some credence (for details, see the News and Background Information section of this issue of Transfer). Even though the Renault Group EWC agreement states:

'The Renault Group has for several years been conducting a policy based on total quality and the constant improvement in competitiveness. ... Given this, the enterprising spirit and motivation of the Renault Group's workforce and the continuing development of a good quality of social dialogue are decisive factors for ensuring the success of the undertaking' (quoted in Hoffmann 1997), 

the company did not feel obliged to consult or notify EWC worker representatives when announcing the closure of its Vilvoorde plant in Belgium with the loss of 3,500 jobs. The absence of consultation and the co-incidence of the announcement of job losses with Renault's application for EU funding to expand its Spanish operations, have led the Commission, the European Parliament and the ETUC to call for a review of the EWC and the Collective Redundancy Directives. The response to these developments have involved workers from Renault operations throughout Europe, rather than just workers from the affected Belgian plant, suggesting that elements of a European workers' identity may be emerging. Furthermore, the Renault Group EWC was influential in co-ordinating the workers' response. In other words, the institutional presence of the EWC was central to the generation of transnational workers' activity.

Events such as those at Renault are unusual. For example, there was little reaction in Europe to Ford's plans to cut the number of jobs at its Halewood plant. However, existing practice suggests that extensive networks of workers' representatives are developing and benchmarking practices are becoming more commonplace. It seems likely that, as a minimum, EWCs will make aspects of international comparison more transparent for those involved (Marginson and Sisson 1994). Opportunities for management to misinform workers regarding the nature of working practices at other plants will be reduced. Managements' ability to withhold information from worker representatives, as reported by Martinez-Lucio and Weston (1994), will also be jeopardised by direct contact between worker representatives from different sites. Furthermore, these developments in labour organization are financed by the transnational companies.

Conclusions

During the twentieth century trade unions in Europe continually adjusted to changed circumstances with the result that their development was cyclical, characterised by peaks and troughs in membership levels, institutional stability, and influence. Furthermore, each period of adjustment was marked by the adoption of new policies and priorities. Undoubtedly the 1980s and 1990s were difficult times for trade unions, due the extent of employer restructuring, the greater influence of neo-liberal policies, and the failure of many unions to formulate and enact coherent counter-measures. In practice, unions in Europe are simultaneously required to act locally, nationally and internationally. At each of these levels of activity outcomes are not predetermined, but may be influenced by trade union interventions. The key question is whether unions can develop appropriate structures and policies to make successful interventions. 

We have shown that many reforms were introduced by trade unionists in order to modernise their organizations during the 1980s and 1990s. While many issues still require attention, it is clear that policy questions such as recruitment, structure and institutional articulation are afforded a higher profile than hitherto. It is also apparent that further research is required to identify 'best practice', and to explain why similar policies introduced by unions have resulted in a range of different outcomes. In addressing these questions broader exchanges between unions in Europe are required in order to facilitate the search for appropriate policies.

Throughout this review we have highlighted the changing nature of the relationships between unions and members and between unions and workplace organization. It is these relationships that are central to securing influence for trade unions at all levels of activity. Only by improving and modernising these relationships can trade unions expect to expand organization into poorly unionised sectors, to exert influence over employers, and to extend international co-operation.

References

• Altvater, E. and B. Mahnkopf. 1993. Gewerkschaften vor der europäischen Herausforderung, Münster: Westfälisches Dampfboot
• Baglioni, G. 1990. 'Industrial Relations in Europe in the 1980s' in G. Baglioni and C. Crouch. (eds.). European Industrial Relations. London: Sage.
• Barreto, J. 1992. 'Portugal: Industrial Relations under Democracy', in A. Ferner and R. Hyman. (eds.). Industrial Relations in the New Europe, Oxford: Blackwell.
• Baumann, H., E-L Laux and M. Schnepf. 1996. 'Collective Bargaining in the European Building Industry - European Collective Bargaining?', Transfer, Vol. 2, No. 2, pp. 321-333.
• Beaumont, P. 1992. 'Trade Unions and Human Resource Management', Industrial Relations Journal, Vol. 22, No. 4, pp. 300-308.
• Boulin, J-Y. 1996. 'Trade Union Modernisation in France: Is There Still Time?', Transfer, Vol. 2, No. 2, pp. 129-145.
• Braithwaite, M. and C. Byrne. n.d. Women in Decision-making in Trade Unions. Brussels: ETUC.
• Briskin, L. and P. McDermott. 1993. (eds.). Women Challenging Unions: Feminism, Democracy and Militancy. Toronto: University of Toronto Press.
• Brown, W. 1993. 'The Contraction of Collective Bargaining in Britain', British Journal of Industrial Relations, Vol. 31, No. 3, pp. 189-200.
• Brown, W., S. Deakin and P. Ryan. 1997. 'The Effects of British Industrial Relations Legislation', National Institute Economic Review, No. 161, July, pp. 69-83.
• Claydon, T. 1996. 'Union De-recognition: A Re-Examination', in I. Beardwell. (ed.). Contemporary Industrial Relations. Oxford: Oxford University Press.
• Cockburn, C. 1983. Brothers. London: Pluto.
• Cockburn, C. 1991. In the Way of Women: Men's Resistance to Sex Equality in Organizations. London: MacMillan
• Coldrick, P. 1991. 'Collective Bargaining in the New Europe', Personnel Management. Vol. 22, pp. 58-61.
• Colling, T. and L. Dickens. 1989. Equality Bargaining - Why Not? Equal Opportunities Commission. London: HMSO.
• Commission of the European Communities. 1997. Green Paper. Partnership for a New Organization of Work. COM (97) 128\9. Brussels: European Commission..
• Crain, M. 1994. 'Gender and Union Organizing', Industrial and Labor Relations Review. Vol. 47, No. 2, pp. 227-248.
• Crouch, C. 1993. Industrial Relations and European State Traditions. Oxford: Clarendon.
• Crouch, C. 1994. 'Beyond Corporatism: The Impact of Company Strategy' in R. Hyman and A. Ferner. (eds.). New Frontiers in European Industrial Relations, Oxford: Blackwell.
• Cyba, E. and U. Papouschek. 1996. 'Women's Interests in the Workplace: Between Delegation and Self-Representation', Transfer, Vol. 2, No. 1, pp. 61-81.
• Dølvik, J-E. 1993. Nordic Trade Unions and the Social Dimension: Towards Europeanisation or Renationalisation of Trade Union Strategies?, FAFO: Oslo.
• Dølvik, J-E. 1996. Europeanisation of Trade Unions - Dynamics and Constraints, paper presented at the IREC Conference, Industrial Relations in Europe: convergence or Diversification?, Copenhagen.
• Due, J., J-S. Madsen, and C-S. Jensen. 1991. 'The Social Dimension: Convergence or Diversification of Industrial Relations in the Single European Market?', Industrial Relations Journal, Vol. 22, No. 2, pp. 85-102.
• Due, J., J-S. Madsen, C-S. Jensen and L-K. Petersen. 1994. The Survival of the Danish Model. Copenhagen: DJØF Publishing.
• Eaton, A. and P. Voos. 1992. 'Unions and Contemporary Innovations in Work Organization, Compensation and Employee Participation', in L. Mishel and P. Voos. (eds.). Unions and Economic Competitiveness, Armonk, New York: Economic Policy Institute.
• Ebbinghaus, B. and J. Visser. 1997. 'When Labour Institutions Matter: Union Growth and Decline in Western Europe 1950 -1990', paper presented at ESA Conference, University of Essex. 
• Erbčs-Seguin, S. 1984. 'Trade Unions, Employers, and the State: Toward a New Relationship?' in M. Kesselman. (ed.). The French Workers' Movement. London: Allen and Unwin.
• Escobar, M. 1995. 'Spain: Works Councils or Unions?' in J. Rogers and W. Streeck. (eds.). Works Councils. Chicago: University of Chicago Press.
• ETUC/ETUI. 1996. European Works Councils and the Europeanisation of Industrial Relations. Conference Report. Brussels: ETUI.
• Fajertag, G. and P. Pochet. 1997. (eds.). Social Pacts in Europe. Brussels: ETUI/OSE.
• Foden, D. 1996. 'EMU, Employment and Social Cohesion', Transfer, Vol. 2, No. 2, pp. 273-286.
• Galenson, W. 1994. Trade Union Growth and Decline. Westport: Praeger.
• Goetschy, J. 1996. 'The European Trade Union Confederation and the Construction of European Unionism' in P. Leisink, J. Van Leemput and J. Vilrokx. (eds.). The Challenges to Trade Unions in Europe. Cheltenham: Edward Elgar.
• Goetschy, J. and P. Rozenblatt. 1992. 'France: The Industrial Relations System at a Turning Point', in A. Ferner and R. Hyman. (eds.). Industrial Relations in the New Europe, Oxford: Blackwell.
• Haas, E. 1968. The Uniting of Europe: Political, Social and Economic Forces 1950-1957. Stanford: Stanford University Press. 2nd Edition.
• Hammarström, O. 1994. 'Local and Global: Trade Unions in the Future', in J. Niland, R. Lansbury and C. Verevis. (eds.). The Future of Industrial Relations. London: Sage.
• Hancké, B. 1993. 'Trade Union Membership in Europe 1960-90: Rediscovering Local Unions', British Journal of Industrial Relations, Vol. 31, No. 4, pp. 593-614.
• Heaton, N. and I. Linn. 1989. Fighting Back: A Report on the Shop Steward Response to New Management Techniques in TGWU Region 10. Barnsley: Northern College and TGWU Region 10.
• Higgs, F. 1994. 'Briefing Report on Shell Haven UK', The Future of Trade Unions. House of Commons, Employment Committee, Third Report, Volume II, HC 676-II. London: HMSO. pp. 154-155.
• Hoffmann, R. 1997. 'European Works Councils: A Building Block in the Europeanisation of Industrial Relations', in E. Gabaglio and R. Hoffmann. (eds.). European Trade Union Yearbook. Brussels: ETUI.
• IRS. 1993. 'The Hoover Affair and Social Dumping', European Industrial Relations Review, No. 230, march, p. 14-19.
• IRS. 1994. 'Debate on the Future of Trade Unions', European Industrial Relations Review, No. 246, July, p. 24-25.
• IRS. 1996a. 'McDonalds Signs Union Rights Agreement with CFDT', European Industrial Relations Review, No. 274, November, p. 6.
• IRS. 1996b. 'Commerce Social Dialogue', European Industrial Relations Review, No. 266, November, p. 24-27.
• Jacobi, O. 1996. 'European Monetary Union: A Quantum Leap?', Transfer, Vol. 2, No. 2, pp. 233-244.
• Jacobi, O., B. Keller and W. Müller-Jentsch. 1992. 'Codetermining the Future' in A. Ferner and R. Hyman. (eds.). Industrial Relations in the New Europe, Oxford: Blackwell.
• Jacobi, O. and W. Müller-Jentsch. 1990. 'West Germany: Continuity and Structural Change', in G. Baglioni and C. Crouch. (eds.). European Industrial Relations. London: Sage.
• Jordana, J. 1996. 'Reconsidering Union Membership in Spain 1977-1994: Halting Decline in a Context of Democratic Consolidation', Industrial Relations Journal, Vol. 27, No. 3, pp. 211-224.
• Jørgensen, H., M. Lassen, J. Lind and M. Madsen. 1992. Medlemmer og Meninger. Copenhagen: LO.
• Katz, H. 1993. 'The Decentralisation of Collective Bargaining: A Literature Review and Comparative Analysis', Industrial and Labor Relations Review, Vol. 47, No. 1, pp. 3-22.
• Kauppinen, T. and V. Köykkä. 1991. Union Membership and Density in Finland 1989. Helsinki: Ministry of Labour.
• Keller, B. 1995. 'Towards a European System of Collective Bargaining?', in R. Hoffmann, O. Jacobi, B. Keller and M. Weiss. (eds.). German Industrial Relations under the Impact of Structural change, Unification and European Integration. Düsseldorf: Hans-Böckler Stiftung.
• Kelly, J. and C. Kelly. 1991. 'Them and Us: Social Psychology and the New Industrial Relations', British Journal of Industrial Relations, Vol. 29, No. 1, pp. 25-48.
• Kester, G. and H. Pinaud. 1996. 'The Decline, Consolidation or Growth of Democratic Participation?', in G. Kester and H. Pinaud. (eds.). Trade Unions and Democratic Participation in Europe, Aldershot: Avebury.
• Kjellberg, A. 1983. Facklig Organizering i Tolv länder. Lund: Arkiv.
• Kjellberg, A. 1992. 'Sweden: Can the Model Survive?', in A. Ferner and R. Hyman. (eds.). Industrial Relations in the New Europe, Oxford: Blackwell.
• Klandermans, B. 1986. 'Psychology and Trade Union Participation: Joining, Acting, Quitting', Journal of Occupational Psychology, Vol. 59, pp. 189-204.
• Kochan, T., H. Katz and R. McKersie. 1986. The Transformation of American Industrial Relations. New York: Basic Books.
• Kochan, T. and K. Wever. 1991. 'American Unions and the Future of Worker Representation' in G. Strauss, D. Gallagher and J. Fiorito. (eds.). The State of the Unions, Industrial Relations Research Association Series, Madison: IRRA.
• Krieger, H. and P. Bonneton. 1995. 'Analysis of Existing Voluntary agreements on Information and Consultation in European Multinationals', Transfer, Vol. 1, No. 2, pp. 188-206.
• Lecher, W. 1995. 'European Works Councils and Direct Participation: The Dual Shift as a Challenge to the German Industrial Relations System' in R. Hoffmann, O. Jacobi, B. Keller and M. Weiss. (eds.). German Industrial Relations under the Impact of Structural Change, Unification and European Integration, Düsseldorf: Hans-Böckler Stiftung.
• Locke, R. 1990. 'The Resurgence of the Local Union: Industrial Restructuring and Industrial Relations in Italy', Politics and Society, Vol 18, No. 3, pp. 347-379.
• Locke, R. and T. Kochan. 1995. 'Conclusion: The Transformation of Industrial Relations? A Cross-National Review of the Evidence', in R. Locke, T. Kochan and M. Piore. Employment Relations in a Changing World Economy. Cambridge: MIT Press.
• Mahnkopf, B. 1992. 'The Skill-oriented Strategies of German Trade Unions: Their Impact on Efficiency and Equality Objectives', British Journal of Industrial Relations, Vol. 30, No. 1, pp. 61-81.
• Marginson, P. and K. Sisson. 1994. 'The Structure of Transnational Capital in Europe: The Emerging Euro-Company and its Implications for Industrial Relations', in R. Hyman and A. Ferner. (eds.). New Frontiers in European Industrial Relations. Oxford: Blackwell.
• Marginson, P. and K. Sisson. 1996. 'European Collective Bargaining: A Virtual Prospect?', ETUI Discussion and Working Paper, Brussels: ETUI.
• Martin, A. 1996. 'European Institutions and the Europeanisation of Trade Unions: Support or Seduction?' ETUI Discussion and Working Paper, Brussels: ETUI.
• Martínez-Lucio, M. 1992. 'Spain: Constructing Institutions and Actors', in A. Ferner and R. Hyman. (eds.). Industrial Relations in the New Europe, Oxford: Blackwell.
• Martínez-Lucio, M. and S. Weston. 1994. 'New Management Practices in a Multinational Corporation: The Restructuring of Worker Representation and Rights', Industrial Relations Journal, Vol. 25, No. 2, pp. 110-121.
• Martínez-Lucio, M. and S. Weston. 1995. 'Trade Unions and Networking in the Context of Change: Evaluating the Outcomes of Decentralisation in Industrial Relations', Economic and Industrial Democracy, Vol. 16, No. 2, pp. 233-251.
• McLoughlin, I. and S. Gourlay. 1994. Enterprise Without Unions. Buckingham: Open University Press.
• Müller-Jentsch, W. and H-J Sperling. 1995. 'Towards a Flexible Triple System?: Continuity and Structural Changes in German Industrial Relations' in R. Hoffmann, O. Jacobi, B. Keller and M. Weiss. (eds.). German Industrial Relations under the Impact of Structural Change, Unification and European Integration, Düsseldorf: Hans-Böckler Stiftung.
• Noé, C. 1996. 'The Maastricht Process is Marginalising the Unions and Employers', in ETUI (ed.) The Future of the European Union: The Trade Unions' Demands and Expectations for the Intergovernmental Conference 1996. Brussels: ETUI.
• Pendleton, A., J. McDonald, A. Robinson and N. Wilson. 1995. 'The Impact of Employee Share Ownership Schemes on Employee Participation and Industrial Democracy', Human Resource Management Journal, Vol. 5, No. 4. pp. 44-60.
• Pinaud, H. 1996. 'Participation in France' in G. Kester and H. Pinaud. (eds.). Trade Unions and Democratic Participation in Europe, Aldershot: Avebury.
• Pochet, P. and E. Arcq. 1997. 'UNICE in 1996: Becoming a Real Social Partner', in E. Gabaglio and R. Hoffmann. (eds.). European Trade Union Yearbook. Brussels: ETUI.
• Ramos Yuste, A. and D. Foden. 1995. 'Labour Market, Unemployment and Employment Policy: Aspects of the European Debate', Transfer, Vol. 1, No. 4, pp. 499-519.
• Regalia, I. and M. Regini. forthcoming. 'Italy: The Dual Character of Industrial Relations' in A. Ferner and R. Hyman. (eds.). Changing Industrial Relations in Europe. Oxford: Blackwell
• Regini, M. 1992. 'Introduction: The Past and Future of Social Studies of Labour Movements', in M. Regini. (ed.). The Future of Labour Movements. London: Sage.
• Ross, G. 1995. Jacques Delors and European Integration. London: Polity Press.
• Sandberg, Å. 1995. (ed.). Enriching Production. Aldershot: Avebury.
• Scheuer, S. 1997. 'Collective Bargaining Coverage and the Status Divide: Denmark, Norway and the United Kingdom Compared', European Journal of Industrial Relations, Vol. 3, No. 1, pp. 39-57.
• Sisson, K. 1993. 'In Search of HRM', British Journal of Industrial Relations. Vol. 31, No. 2, pp. 210-210.
• Smith, P. and G. Morton. 1993. 'Union Exclusion and the Decollectivisation of Industrial Relations in Contemporary Britain', British Journal of Industrial Relations, Vol. 31, No. 1, pp. 97-114.
• Sorge, A. and W. Streeck 1988. 'Industrial Relations and Technical Change: the Case for an Extended Perspective', in R. Hyman and W. Streeck. (eds.). New Technology and Industrial Relations, Oxford: Blackwell.
• Stoop, S. 1994. (ed.). European Consultations Scenario: A Practical Guide for Employees' Representatives to Arrive at a European Consultation Structure within one's own Company. Amsterdam: FNV.
• Streeck, W. 1993. 'Training and the New Industrial Relations' in S. Sleigh, (ed.). Economic Restructuring and Emerging Patterns of Industrial Relations, Kalamazoo: Upjohn Institute.
• Streeck, W. 1994. 'European Social Policy after Maastricht: The Social Dialogue and Subsidiarity', Economic and Industrial Democracy, Vol. 15, No. 2, pp. 151-177.
• Streeck, W. 1995. 'From Market Making to State Building? Reflections on the Political Economy of European Social Policy' in S. Leibfried and P. Pierson. (eds.). European Social Policy. Washington, D. C.: The Brookings Institution.
• Streeck, W. and S. Vitols. 1993. European Works Councils: Between Statutory Enactment and Voluntary Adoption. Discussion Paper. Berlin: Wissenschaftszentrum für Sozialforschung.
• Tarrow, S. 1995. 'The Europeanisation of Conflict: Reflections from a Social Movement Perspective', West European Politics, Vol. 18, No. 2, pp. 223-251.
• Traxler, F. 1996a. 'European Trade Union Policy and Collective Bargaining: Mechanisms and Levels of Labour Market Regulation in Comparison', Transfer, Vol. 2, No. 2, pp. 287-297.
• Traxler, F. 1996b. 'Collective Bargaining and Industrial Change: A Case of Disorganization? A Comparative Analysis of Eighteen Countries', European Sociological Review, Vol. 12, No. 3, pp. 271-287.
• Trebilcock, A. 1991. 'Strategies for Strengthening Women's Participation in Trade Union Leadership', International Labour Review, Vol. 130, No. 4, pp. 407-426.
• Turner, L. 1991. Democracy at Work. Ithaca: Cornell University Press.
• Turner, L. 1992. 'Industrial Relations and the Reorganization of Work in West Germany: Lessons for the U.S.' in L. Mishel and P. Voos. (eds.). Unions and Economic Competitiveness, Armonk, New York: Economic Policy Institute.
• Turner, L. 1996. 'The Europeanisation of Labour: Structure before Action', European Journal of Industrial Relations, Vol. 2, No. 3, pp. 325-344.
• UNICE. 1991. Position Paper on Proposed European Works Councils Directive. Brussels: UNICE.
• Valkenburg, B. 1995. 'Trade Unionism in The Netherlands: Back to the Future', Transfer, Vol. 1, No. 1, pp. 64-79.
• van de Vall, M. 1970. Labor Organizations. Cambridge: Cambridge University Press.
• Vaughan-Whitehead, D. 1996. 'Financial Participation: A New Challenge for Trade Unions', in G. Kester and H. Pinaud. (eds.). Trade Unions and Democratic Participation in Europe, Aldershot: Avebury.
• von Prondzynski, F. 1992. 'Ireland: Between Centralism and the Market' in A. Ferner and R. Hyman. (eds.). Industrial Relations in the New Europe, Oxford: Blackwell.
• Visser, J. 1989. European Trade Unions in Figures. Deventer: Kluwer.
• Visser, J. 1994. 'European Trade Unions: The Transition Years' in R. Hyman and A. Ferner. (eds.). New Frontiers in European Industrial Relations, Oxford: Blackwell.
• Waddington, J. 1992. 'Trade Union Membership in Britain 1980-1987: Unemployment and Restructuring', British Journal of Industrial Relations, vol. 30, No. 2, pp. 287-324.
• Waddington, J. 1995a. The Politics of Bargaining. London: Mansell.
• Waddington, J. 1995b. 'UK Unions: Searching for a New Agenda', Transfer, Vol. 1, No. 1, pp. 31-43.
• Waddington, J. and C. Whitston. 1996. 'Empowerment versus Intensification: Union Perspectives of Change at the Workplace' in P. Ackers, C. Smith and P. Smith. (eds.). The New Workplace and Trade Unionism. London: Routledge.
• Waddington, J. and C. Whitston. 1997. 'Why Do People Join Unions in a Period of Membership Decline?', British Journal of Industrial Relations, Vol. 35, No. 4.
• Wallerstein, M. 1989. 'Union Organization in Advanced Industrial Democracies' American Political Science Review, Vol. 82, No. 2, pp. 481-502.
• Wilson, F. 1991. 'Democracy in the Workplace: The French Experience', Politics and Society, Vol. 19, No. 4, pp. 439-462.

*Senior Research Fellow and Senior Lecturer, Industrial Relations Research Unit, University of Warwick, UK.

** Director, European Trade Union Institute, Brussels, Belgium.

*** Senior Research Fellow, Institut for Sociale Forhold og Organization, Aalborg Universitet, Denmark.

1 UNISON was formed in 1993 from the National and Local Government Officers' Association (NALGO), the National Union of Public Employees (NUPE) and the Confederation of Health Service Employees (COHSE).

2 In the case of the UK, Pendleton et al distinguish between technical, paternalist and representative employee share ownership schemes, arguing that only the latter offer some opportunities for the extension of 'industrial democracy'. However, these authors show that these schemes have tended to disintegrate in the UK as workers opt for financial rewards, and they have made little difference to the working life of employees (1995: 57-58).

3 A total of twelve confederations were accepted into membership from Bulgaria, the Czech Republic, Hungary, Poland, Romania and Slovakia; four from Hungary and Slovenia, have observer status.

Jürgen Hoffmann, Reiner Hoffmann

1. Trade unions in Europe - merely victims of globalization?

"The Globalization Trap", "The Limits of Globalization", "On the Treadmill of Global Competition", "Policy Regarding the Location of Production Sites", "Eurosclerosis". These are all the (translated) titles of books and articles, as well as being "merely" provocative terms that refer to an ongoing debate on the restructuring of the world market. They also highlight the apparent or actual change in the situation facing Europe's economies and societies, a change which is again apparently or actually subjecting present and future trade union policy in all the European countries to constant blackmail. However one assesses this debate, one thing is certain: in the economic and political processes of change currently under way, the trade unions are finding themselves confronted with a new economic structure, one in which the boundaries of individual sectors - by which we mean their boundaries where their (union) organization is concerned - are increasingly becoming "europeanized" under pressure from the European single market. At the same time they are being internationalized as a result of companies' decentralization strategies (outsourcing, global sourcing, etc.). But the fact remains that they are not being sufficiently prepared for all this in organizational terms despite all the progress made in the area of policy at the level of the European Trade Union Confederation (ETUC) (see below). At the same time, the unions are confronted with a development in which the importance of national and sectoral links is declining, while global options - as indeed options at the regional and company levels - are gaining in prominence, and the significance of the "classical" form of the social closure of labor markets, the nation state, is waning. In other words, as the market boundaries defined by capital are expanding, the various trade union organizations are finding themselves faced with the risky dilemma of opening up the previously established, familiar forms of "social closure" which had a stabilizing effect on organization within the framework of the national labor market, and doing so both outwardly (with a view to Europe) and inwardly (under pressure from the growing differences within as well as between the various industrial sectors). As we all know, such "opening up socially" is "one of the most difficult decisions facing any organization, and something that normally comes into question only during a crisis or when all other alternatives have been ruled out" (Ebbinghaus/Visser 1994, p. 232 - with reference to Brunsson/Olsen).

Behind these challenges lies not only the European single market, but also, and increasingly, the dominance of international financial markets, which are forcing companies to utilize short-term, efficiency-oriented reasoning and arguments in their business policy - and this, in turn, is defining the social discourse and challenging the traditional, received forms of regulation. Today's business culture in Western Europe means not only the traditional "Fordist" approach as a potential model of productivity, but also - as Michel Albert termed it - "Rhenish capitalism" (Albert 1992), designed as a cooperative, social system to function in the long term. And unions are finding themselves in the curious - and to them surely sometimes uneasy - situation of having to defend this form of capitalism, while at the same time being obliged to contain it as well as develop it further in social and environmental terms - in the face of Anglo-American deregulation which is geared towards a radical form of market-oriented capitalism, which in the short run is apparently more successful on the world market. It is a form of radical market-oriented capitalism that has gained an increasingly higher profile as a social prospect for the future, both through the collapse of "real forms of previously existing socialism" and as a result of the pressure exerted by the labor markets in the East. Moreover, in its present form it leaves the trade unions little room for manoeuvre.

Trade union policy today appears to have few ways of combating these global trends. Even pro-union, critical analyses label politicians generally - and the trade unions in particular -as mere actors caught up in the "treadmill" of globalization, i.e. primarily as victims. "You have no chance, but make sure you use it!" is the fatalistic slogan; and the listed alternatives without actors thus seem like "whistling in the wind." Politically, this feeling of being victimized has thus far only very rarely served as a catalyst for action; as a rule, it leads the actors involved into isolation, resignation and fatalism, or into unconscious revolt. So, the question is whether this really is the only prospect remaining in Europe when the globalization process is analysed.

However, to be in a position to answer this question, we must seriously address the thesis of the globalization of the economy, looking beyond the sensational, circulation-boosting headlines and spectacular isolated cases. At the analytical level, this does not o