The puzzle of trade union strength in Western Europe since 1980.

Kelly, John ; Hamann, Kerstin

Introduction

The period between 1945 and the early 1970s has been referred to as the 'Golden Age of Capitalism' (Marglin & Schor 1990). Most countries that comprised what was then known as the advanced capitalist world Western Europe, North America, Japan and Australasia recorded sustained and substantial rates of economic growth for over two decades. Living standards, measured by real incomes, ownership of consumer goods and health indices increased dramatically for millions of people (Marglin & Schor 1990). This period also represented a golden age of trade unionism in which membership, density, strike activity and economic and political bargaining power reached hitherto unprecedented levels (Shalev 1992, Western 1997). Some European governments in the 1970s in Ireland, Italy and the United Kingdom for example--responded to union power by creating corporatist policymaking committees designed to negotiate union wage restraint in exchange for improved legal rights and welfare provision (Schmitter & Lehmbruch 1979). Since the early 1980s however unions across the advanced capitalist world have witnessed prolonged membership decline, a reduction in bargaining power in the context of a globalizing economy and neo-liberal government policies, and a significant decline in strike rates. The world recession that began in 2008 appears to have dealt a further blow to unions already weakened by years of membership loss.

However, we argue that whilst recent economic and political developments have undoubtedly posed serious threats to trade unions, these can also be viewed as opportunities for trade union revitalization (Behrens, Hamann & Hurd 2004). We show that unions' weakening in organisational strength and economic bargaining power has in some cases been counterbalanced by the re-emergence of tripartite relations between governments, unions and employers and by the resurgence of general strikes against government policies. We interpret union involvement in tripartite talks and their continuing capacity to mobilize members and supporters in general strikes as evidence that some union movements have recognized the existence of opportunities in the midst of adversity. After a brief recap of the familiar dimensions of trade union decline we set out a number of theoretical propositions that enable us to identify contemporary sources of trade union power. We also present fresh evidence on recent patterns of union-government negotiations in Western Europe and on general strikes designed to influence government policy.

Parameters of Trade Union Decline

Evidence from 16 Western European countries as well as the USA, Canada, Australia and New Zealand shows that trade union density declined between 1980 and 2007 in all but two cases: Finland (up one percentage point from 69% to 70%) and Spain (up eight points from 7% to 15%) (Barratt 2009, Hamann & Kelly 2008). The mean, unweighted rate of density decline 1980-2007 is 15.4 percentage points. It is true that the decline has been far more pronounced in the private sector compared to the public sector and that the overall rate of decline has been quite modest in some countries, such as Belgium (- 1 point, from 54% to 53%), Norway (- 4 points, from 58% to 54%) and Canada (6 points, from 35% to 29%) but these national cases are unusual. It is also true that between 1980 and 2004 collective bargaining coverage has remained remarkably stable in most of Western Europe at approximately 80-90% of employees (Hamann & Kelly 2008). However, this institutional stability masks a significant erosion of trade union power, made evident by trends in national income distribution and in strike rates. Between 1960 and 1980 the wage share of value added in manufacturing industry in 15 OECD countries rose from 66% to 74% as union militancy shifted the distribution of national income away from profits and into wages and salaries. However between 1980 and 2000, as trade union membership declined, the distribution of national income shifted radically back in favour of owners of capital: wage share fell steadily and by 2000 had reached 65%, despite the economic growth and low unemployment of the mid-late 1990s (Glyn 2006: 7). Theory suggests that unions are able to push up wages and salaries at the expense of profits by the use (or threat) of collective action and it is therefore not surprising to observe that the redistribution of income from labour to capital over the past 20-30 years has coincided with a dramatic reduction in levels of strike activity, whether measured by days lost, numbers of strikes, or workers involved. Between 1970 and 1979 an average of 419 working days per 1000 employees were lost to strike action each year in the major OECD countries, but by 2000-2004 days lost to strikes had decreased by almost 90 per cent to just 51 days per 1000 employees (Piazza 2005: 290, Scheuer 2006: 144, 149, van der Velden et al. 2007).

There are substantial academic literatures on the causes of declining union membership and of falling strike rates. A wide range of causal variables have been linked to union membership, including changes in worker attitudes, away from collectivism and towards individualism; the limited effectiveness of trade unions caused by high unemployment and competitive, international product markets; the constraints placed on unions by deregulatory, neo-liberal economic policies; the opposition of employers to trade unionism, particularly in the private service firms that now dominate most Western economies and where traditions of trade unionism are weak; and the limited resources devoted by trade unions to organizing in the service sector of the advanced economies (see for example Fernie & Metcalf 2005, Martin & Ross 1999, Waddington & Hoffmann 2000, Western 1997). So far as strikes are concerned, the globalization of the economy has played a significant role in depressing overall levels of strike activity in Western Europe. Piazza's (2005) study of 15 OECD countries 1952-2001 found that two measures of globalization, increased international trade and reduced controls on capital mobility are strongly associated with a decline in days lost to strikes, especially in countries with low (below-median) union density. Declining trade union density since 1980 is also associated with strike decline over the same period although the relationship between levels of strikes and density (as distinct from changes) is more complex: the most strike-prone countries in Europe (2002-06) include high-density Finland, modest-density Italy and low-density Spain (Hale 2008).

Overall therefore, as is well known, the union movements in the advanced capitalist world are losing members and appear to be experiencing a decline in bargaining power and mobilizing capacity despite stable bargaining coverage.

Trade Unions, Electoral Volatility & Political Party Competition

However if we turn our attention from the roles of product and labour markets and employers to developments in governmental behaviour and electoral politics, it can be argued that a series of far-reaching changes in the attitudes and behaviour of voters and in the behavior of political parties has created new prospects for contemporary trade unionism. In the first place, since 1980 many West European governments have embarked on major reforms to their systems of welfare, labour markets and wage determination in order to enhance the competitiveness of their national economies. For example, we have witnessed attempts to reduce entitlements to unemployment benefits; raise the retirement age and increase levels of pension contributions; reduce legal protections for workers against dismissal or redundancy; and restrain the growth in real wages. Governments with a majority of seats in their legislature could in theory enact many of these reforms through legislation but in practice many of them, irrespective of party family, have chosen to engage trade unions in bipartite or tripartite (government, unions and employers) talks in an attempt to reach agreement (Hamann & Kelly 2010). Why would governing parties, either without strong ties to trade unions or with weakening ties, seek out agreements with trade unions, organizations whose membership base and bargaining power is supposedly eroding?

Broadly speaking, one main reason is that political parties are themselves facing a series of electoral problems because of the growth in party competition and because of the declining allegiance among voters to any particular party. Electoral volatility--the percentage of voters switching parties from one election to the next--has increased since the 1970s due to a wide range of factors including changing demographics, rising educational levels, the growing importance of mass media, the rise of new political issues, and softening class lines as an outgrowth of changing employment and production patterns. For example, while only 20% of Dutch respondents professed in 1971 they would sometimes vote for a different party, over two-thirds did so in 1991 (Dalton 2000). Parties therefore may see the need to reach out to voters that did not traditionally form their core support group. In this context, social pacts with trade unions and employers may thus present themselves as a useful strategy to seek broad support for potentially unpopular policies in an attempt to minimize the anticipated electoral costs of implementing such policies. Moreover if pact negotiations run into problems or the reforms prove to be especially unpopular, pacts may serve as a mechanism of 'blame avoidance', allowing the government to divert voter criticism onto trade unions (Hamann & Kelly 2007, 2010). We can therefore hypothesize that the more strongly governments are committed to reaching agreement with unions on a social pact, the more power unions may be able to wield in their relations with government. Moreover, in seeking to influence governments we would expect unions not only to deploy the kinds of arguments they would normally use with employers but to deploy various forms of pressure, including collective action.

Social Pacts & General Strikes

Social pacts almost disappeared with the onset of the world recession around 1980; yet, one of the most striking developments in the industrial relations and political economy of Western Europe has been their re-emergence after this brief hiatus (Hamann & Kelly 2010). Between 1980 and 1991, 29 social pacts were signed in a variety of Western European countries; during the years 199298, in the run-up to European Monetary Union (EMU), 40 social pacts were agreed on; yet, in the eight years after EMU, a further 41 social pacts were signed (Hamann & Kelly 2010). Although some commentators assumed that social pacts were strongly influenced by the fiscal and public spending requirements of EMU and dealt primarily with wages, the numerous pacts that both preceded and followed the EMU years 1992-98 suggests this 'political economy' approach is unconvincing (e.g. Hancke & Rhodes 2005). In terms of content, only 50% of the 110 agreed social pacts between 1980 and 2006 dealt with wages; the remainder covered labour market and welfare reforms.

In contrast, both quantitative and qualitative analyses of 16 West European countries between 1980 and 2006 show that electoral and party political factors played a significant role in shaping governmental offers of social pacts (Hamann & Kelly 2010). Such pacts were more likely to be offered by governing parties that had recently enjoyed electoral success and gained votes; by center or leftist governing parties rather than conservative parties, presumably because of their traditionally stronger ties to trade unions; by broad coalitions comprised very different party families (compared to single-party administrations); by governments that were relatively weak in the legislature, measured by seat share; and by governments working with legislatures that were highly fragmented, i.e. that included a large, effective number of political parties. Broad and weak coalitions in fragmented legislatures may experience considerable difficulty in legislating unpopular reforms. Social pacts with unions and employers may therefore assist such governments to overcome these problems as well as providing enhanced legitimacy for their policies. We also found that political parties responded to vote changes in different ways: vote losses, for example, were more likely to lead to pact offers from left parties (e.g. the Social Democrats in Sweden in the 1990s) but this was not the case for parties of the right (e.g. the Conservatives in Denmark in the 1980s). Legislation was more likely to be pursued by Conservative parties with a high proportion of seats, particularly where the legislature contained only a small number of effective parties (e.g. France since the 1990s).

Approximately 70% of social pact offers by West European governments have resulted in negotiated agreements on wages, labour market and welfare reforms. Whilst some unions have negotiated from positions of weakness, there is evidence in many cases that unions have secured significant concessions from governments. To illustrate, the most comprehensive study of pension reform in Western Europe provides many examples where unions have exerted significant influence on issues such as levels of contribution, reference periods for pension calculation and retirement age (Immergut et al 2007). Unions have therefore been able to exercise political influence on important issues despite membership losses and the erosion of bargaining power vis-a-vis employers.

However, many governments have shunned union inclusion in social pacts and have sought to legislate reforms, often in the face of union opposition. Although more often than not it is Conservative governments that have opted for trade union exclusion, as in France or Italy for example, we found that governments led by other party families also excluded unions from time to time: Social Democrats in Spain in the 1980s and Sweden in the 1990s, Centre Party governments in Finland in the 1990s and Christian Democratic governments in the Netherlands and Belgium at various times. Ongoing research suggests that union exclusion from government policymaking on wages, welfare systems and labour markets is strongly associated with a variety of union protests including general strikes. There is no generally agreed definition of the term 'general strike' or its various synonyms, such as 'political strike' or 'protest strike' (see for example Walsh 1983) but Hyman's (1989: 17) definition of a strike can be suitably revised, as follows:

'A general strike is a temporary, national stoppage of work by workers from many industries, directed against the executive or legislative arms of government, to enforce a demand or give voice to a grievance.'

Working days lost and workers involved per 1000 employees are the most commonly used measures of strike activity in preference to simple frequency counts (see for example Monger 2005, van der Velden et al 2007). In the case of general strikes, as with all large, multi-workplace strikes, data on days lost and workers involved are extremely unreliable (Lyddon 2007). Since the number of general strikes can be measured very accurately, the frequency measure is used here. Data were collected from a variety of sources for the EU 15 plus Norway (Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Portugal, Spain, Sweden and the UK) for the period 1980-2008. The frequency of general strikes to protest against government policy has risen since 1980. Fig. 1 shows that between 1980 and 1989, unions staged 18 general strikes against governments in Western Europe, a number that increased to 29 strikes in the following decade; and between 2000 and 2008, the number of general strikes grew again to 38 (Kelly & Hamann 2010). In addition, unions issued a total of 11 credible general strike threats, once during the 1980s, eight times in the 1990s, and twice after 2000. While Greece certainly contributed substantially to the large number of general strikes, the overall pattern of general strikes persists even if we exclude Greece. However a general strike in Greece (38 such strikes in 29 years) may have a different meaning and very different causes and consequences from a general strike in the Netherlands or Austria, where such actions are extremely rare. In terms of issues, it is reforms to welfare systems, pensions and labour market regulation that have precipitated the majority of general strikes.

[FIGURE 1 OMITTED]

Table 1 reveals the national distribution of general strikes and documents their concentration in the Southern European economies of France, Greece, Italy, Spain, and to a lesser degree Portugal. These five countries alone account for 82% (70) of the 85 strikes in this period. The remaining strikes (15) were organized in countries that have for many years recorded some of the lowest levels of industrial conflict in Europe: Austria, Belgium, Luxembourg, the Netherlands, and Norway, whilst several countries that have recently ranked high on economic strikes stand out for their absence of general strikes, e.g. Denmark and Ireland.

Theories developed to explain the existence, rise, or decline of economic strikes are poorly equipped to account for general strikes. For one, the empirical patterns displayed in Figure 1 demonstrate that patterns of general strikes diverge from those of economic strikes. In addition, general strikes differ fundamentally from economic strikes: general strikes are directed against governments and their (proposed) policies rather than employers; they are often organized around broad, rather than sectional or occupational, issues, of concern to large segments of the population beyond those employed in specific firms or sectors; the issues that motivate general strikes, such as welfare or labour market reform, are not generally those that are subject to regular collective bargaining processes; and general strike mobilizations may well extend beyond the unions' membership and activist base to include many non-union employees. Thus, it makes little sense to expect that explanations developed to account for workplace, company, or even industry-wide strikes will also be able to illuminate the causes of general strikes.

One of the most significant predictors of general strikes is government exclusion of unions from policymaking. The scatterplot in Fig. 2 reveals that the more often the government pursues legislative acts to address reforms in the areas of welfare, pensions, and labour markets (and to a lesser extent wages), the higher the propensity of unions to stage general strikes (Denmark, Germany, Ireland, Sweden and the UK are outliers because there were no general strikes in these countries during the period under study). Since governments led by Conservatives are most likely to exclude unions it is not surprising to find that the majority of general strikes has been launched against Conservative governments, 51% (43 out of 85). Yet a significant minority of strikes 23 out of 85 was also called to protest the policies of Social Democratic governments, in France, Greece, and Spain. Linked to union inclusion or exclusion are the formal channels available to unions for involvement in government policymaking, or 'opportunity structures' (e.g. Tarrow 1994). It could be argued that well developed structures for union involvement would reduce the propensity of unions to challenge government policies through general strikes. Using Siaroff's (1999) composite index of bargaining and tripartite institutions, it appears that there is a correlation between opportunity structures and the level of general strike action. Five of the seven countries with below-average scores on opportunity structures have the highest general strike totals: France, Greece, Italy, Portugal and Spain (Ireland and the UK are the two exceptions).

In addition to government policy and party family of the government, a third factor linked to general strike mobilizations is the structure of the trade union movement. Four of the six countries with the highest levels of general strikes have trade union confederations divided on ideological lines between 'communist' and 'socialist', i.e. France, Italy, Portugal and Spain. Although general strikes have often been called by all the union confederations in a particular country, some general strikes were staged only by the communist confederation: CGT (France), CGIL (Italy), CGTP (Portugal), and CC.OO (Spain). Since general strikes have occurred in six other countries with very different union structures, we know whilst the presence of leftist confederations is highly conducive to general strikes, such action does still occur even in their absence.

The recent general strikes have produced a range of outcomes for unions and raise interesting questions about the role of political action in union revitalization (Hamann & Kelly 2004). For example, the Spanish general strike of 2002 (with the threat of more to follow) forced the conservative (Popular Party) government to abandon almost all of its proposals for the reform of unemployment benefits. The Italian general strikes of 1994 against pension reforms led to the conservative government's resignation and its defeat at the next general election. On the other hand Austrian trade unions staged general strikes in 2000 and again in 2003 in protest at the unilateral implementation of pension reforms by the conservative-far right coalition government but extracted very few concessions. Even this limited evidence on variation in outcomes suggests that general strikes cannot simply be dismissed as futile protests by weak and declining union movements. Because of the weaknesses of governments, discussed earlier, some general strikes do appear to have elicited significant concessions from governments. Given our focus on governments' electoral considerations, this is perhaps not surprising as general strikes can present massive social and economic disruption and express widespread opposition of the electorate to governmental policies.

[FIGURE 2 OMITTED]

Conclusions

Union problems in bargaining with employers, particularly multinational employers in industry and financial services, have been well documented. Both this problem and the associated decline in economic strikes throughout most of Western Europe since 1980 have often been traced to a shift in the balance of power between increasingly mobile and powerful corporations and their unionized but vulnerable employees. Whilst the capacity and willingness of unions to strike against corporations has declined, relations between unions and governments look very different. Since the early 1980s there has been a re-emergence of union-government negotiations on issues of welfare and labour market reform and wage determination. The problems faced by political parties in gaining support from an increasingly volatile electorate and forming stable governing coalitions have provided opportunities for trade union movements in some countries to establish a key role as bargaining agents on behalf of the workforce as a whole. Governments that have sought to implement potentially unpopular reforms have sometimes found it helpful to secure the support of trade unions through negotiating a social pact. The dependency of government on such support provides unions with a degree of bargaining power outside of the regular workplace-related bargaining structures and many union movements in Western Europe have welcomed this opportunity to demonstrate their continuing relevance. At the same time government exclusion of unions from policymaking has proved to be a significant factor contributing to the resurgence of general strikes in Western Europe since 1980. From the vantage point of labour and product markets, unions' bargaining power in relation to employers often seems limited in its scale and impact. However, if we change our focus and examine union relations with governments in the context of party politics, a rather different picture of union power and vitality starts to emerge. Union strength is thus a multi-faceted and multi-dimensional concept that needs to be understood in a context that goes beyond the regular bargaining structures and instead also assesses unions' influence on government policies.

References

Barratt, Craig (2009), Trade Union Membership 2008, London: Department for Business, Enterprise and Regulatory Reform.

Behrens, Martin, Kerstin Hamann & Richard Hurd (2004), "Conceptualizing Labour Union Revitalization", In Carola Frege & John Kelly (eds), Varieties of Unionism: Strategies for Union Revitalization in a Globalizing Economy, Oxford: Oxford University Press.

Bird, Derek (1991), "International Comparisons of Labour Disputes in 1989 and 1990", Employment Gazette, 99(12): 653-58.

Dalton, Russell J. (2000), "The Decline of Party Identifications", In Russell J. Dalton & Martin P. Wattenberg (eds), Parties Without Partisans: Political Change Advanced Industrial Democracies. New York: Oxford University Press.

Fernie, Sue & David Metcalf (eds) (2005), Trade Unions: Resurgence or Demise? Abingdon: Routledge.

Glyn, Andrew (2006), Capitalism Unleashed: Finance, Globalization, & Welfare, Oxford University Press.

Hale, Dominic (2008), "International Comparisons of Labour Disputes in 2006", Economic and Labour Market Review, 2(4): 32-9.

Hamann, Kerstin & John Kelly (2010), Parties. Elections and Policy Reforms in Western Europe: Voting for Social Pacts, Abingdon: Routledge.

Hamann, Kerstin & John Kelly (2008), "Varieties of Capitalism and Industrial Relations". In Edmund Heery, Nick Bacon, Paul Blyton and Jack Fiorito (eds), The Sage Handbook of Industrial Relations, London: Sage.

Hamann, Kerstin & John Kelly (2007), "Party Politics and the Re-emergence of Social Pacts in Western Europe," Comparative Political Studies, 40(8): 971-94.

Hamann, Kerstin & John Kelly (2004), "Unions as Political Actors: A Recipe for Revitalization?" In Carola Frege & John Kelly (eds), Varieties of Unionism: Strategies for Union Revitalization in a Globalizing Economy, Oxford: Oxford University Press.

Hancke, Bob & Martin Rhodes (2005), "EMU and Labour Market Institutions in Europe: The Rise and Fall of National Social Pacts," Work and Occupations, 32(2): 196-228.

Hyman, Richard (1989), Strikes 4e, Basingstoke: Macmillan.

Immergut, Ellen M., Karen M. Anderson & Isabelle Schulze (eds.) (2007), The Handbook of West European Pension Politics, New York: Oxford University Press.

Kelly, John & Kerstin Hamann (2010), "General Strikes in Western Europe, 1980-2008", Paper to International Industrial Relations Association European Regional Congress, Copenhagen, June.

Lyddon, Dave (2007), "Strike Statistics and the Problems of International Comparison", In Sjaak van der Velden, Heiner Dribbusch, Dave Lyddon & Kurt Vandaele (eds) (2007), Strikes Around the World, 1968-2005: Case-Studies of 15 Countries, Amsterdam: Aksant.

Marglin, Stephen A. & Juliet B. Schor (eds.) (1990), The Golden Age of Capitalism: Reinterpreting the Postwar Experience, Oxford: Clarendon Press.

Martin, Andrew & George Ross (eds) (1999), The Brave New World of European Labour: European Trade Unions at the Millennium, New York: Berghahn Books.

Monger, Joanne (2005), "International Comparisons of Labour Disputes in 2003", Labour Market Trends, 113(4): 159-68.

Piazza, James (2005), "Globalizing Quiescence: Globalization, Union Density and Strikes in 15 Industrialized Countries", Economic and Industrial Democracy, 26(2): 289-314.

Pierson, Paul (1994), Dismantling the Welfare State? Reagan, Thatcher, and the Politics of Retrenchment, New York: Cambridge University Press.

Scheuer, Steen (2006), "A Novel Calculus? Institutional Change, Globalization and Industrial Conflict in Europe", European Journal of Industrial Relations, 12(2): 143-64.

Schmitter, Philippe & Gerhard Lehmbruch (eds) (1979), Trends Towards Corporatist Intermediation, Beverly Hills, CA: Sage.

Shalev, Michael (1992), "The Resurgence of Labour Quiescence", In Marino Regini (ed.), The Future of Labour Movements, London: Sage.

Siaroff, Alan (1999), "Corporatism in 24 Industrial Democracies: Meaning and Measurement". European Journal of Political Research, 36: 175-205.

Tarrow, Sidney (1994) Power in Movement: Social Movements, Collective Action and Politics. New York: Cambridge University Press.

Van Der Velden, Sjaak, Heiner Dribbusch, Dave Lyddon & Kurt Vandaele (eds) (2007), Strikes Around the World, 1968-2005: Case-Studies of 15 Countries, Amsterdam: Aksant.

Waddington, Jeremy and Reiner Hoffman (eds.) (2000), Trade Unions in Europe: Facing Challenges and Searching for Solutions, Brussels: European Trade Union Institute.

Walsh, Kenneth (1983), Strikes in Europe and the United States, London: Pinter.

Western, Bruce (1997), Between Class and Market: Postwar Unionization in the Capitalist Democracies, Princeton, NJ: Princeton University Press.

John Kelly is from School of Management, Birkbeck College, University of London. E-mail: j.kelly@bbk.ac.uk. Kerstin Hamann is from Department of Political Science, University of Central Florida, Orlando, USA. E-mail: khamann@mail.ucf.edu

Table 1: General 8 Economic Strikes by
Country

Country           General      Strike ranking
                  strikes     (days lost/1000)
                 1980-2006        1980-2008

Greece               38               1
Italy                13               3
France               10              11
Belgium              7               12
Spain                6                2
Austria              3               13
Netherlands          3               15
Portugal             3               10
Luxembourg           1               16
Norway               1                8
Denmark              0                6
Finland              0                4
Germany              0               14
Ireland              0                5
Sweden               0                9
UK                   0                6
TOTAL                85

Source: As for Fig. 1 plus Hale (2008), Monger
(2005), Bird (1991).