Social Democratic Parties and Trade Unions in Deflationist Times

Transcription

1 Social Democratic Parties and Trade Unions in Deflationist Times Marco Simoni London School of Economics Paper prepared for the 2004 Conference of Europeanists: "Europe and the World: Integration, Interdependence, Exceptionalism?" Chicago, March 11-13, 2004 This paper is part of an ongoing PhD project. Please do not cite or quote without permission. Comments welcome.

2 Abstract This paper aims at challenging conventional theory according to which the abandonment of Keynesianism by social democratic governments entailed a divorce between once very close parties and trade unions. During the 1980s and 1990s, Social Democratic parties endorsed the new economic consensus grounded on restrictive monetary policies. Nonetheless, a collaborative relationship between social democratic governments and trade unions is apparent in a significant number of European countries. Meanwhile, in other countries a new kind of adversarial relationship apparently supports the divorce theory, as in late 1990s Britain. Contrary to any expectation, however, this adversarial relationship paired with redistributive outcomes favourable to wage earners. Arguably, more must be at stake than a general political economy stance, to be able to explain different types of party/unions relationships throughout Europe. The paper concludes by drawing an alternative explanation of the causal factors of the party/unions relationship to be tested through further research. Reversing the causal direction of the conventional explanation, it argues that the nature of the relationship depends on the features of trade unions movement. Parties and unions will benefit from a collaborative relationship, only if the latter embed the substance of an encompassing union in their programs and attitudes. 2

3 Introduction The relationship between Social Democratic parties and Trade Unions has been and continues to be a pivotal juncture of the European left in particular, and European politics in general 1. The nature, features, and rationale of this relationship will be at the centre of this paper, with particular reference to the period that followed the end of the Keynesian consensus (Hall 1989). De-industrialization and internationalization of developed economies have had a deep impact on Social Democracy and trade unionism. The consequences for party politics and industrial relations have been analysed thoroughly by two distinct streams of comparative political science and political economy (see for example Baglioni and Crouch 1990; Kitschelt 1994; Boix 1998; Martin and Ross 1999; Notermans 2000). The first objective that this paper endeavours to accomplish is therefore to bring together two adjacent topics by focusing the analysis on the point they share. The nature and, more indirectly, the causal factors of the party/unions relationship are part of all the studies on late twentieth century Social Democracy and organized labour. However, there has been comparatively little research focusing on their juncture in spite of the importance such relationship holds for the fortunes of both the actors (Bergounioux 1983; Ebbinghaus 1995; Astudillo Ruiz 2001; Howell 2001; Piazza 2001). The core of this paper aims at challenging conventional wisdom according to which the abandonment of traditional Keynesianism by Social Democratic parties in the 1980s and 1990s entailed a divorce between once very collaborative actors. This view stems from the literature on industrial relations and party politics fin de siècle. However, the observation of late 1990s social democratic governments will highlight that this theory falls short of explaining either renewed party/unions collaborations, or blatant divorces. The paper will argue that factors other than social democrats stance on political economy intervene to shape the characters and features of the party/unions relationship, and it draws a working hypothesis for further research based on the analysis of the new collaborative relationship. Reversing the causal direction of the conventional explanation, it will be argued that the nature of the relationship depends on the ability of trade unions to be valuable and reliable partners for social democratic cabinets. 1 This paper is concerned with the reformist parties of the Left, and uses the descriptor labour, socialist, or social democratic in a loose manner. For a similar loose use of the descriptor social democratic see Kitschelt, H. (1994). The transformation of European social democracy. Cambridge ; New York, Cambridge University Press.; or Ladrech, R. and P. Marlière, Eds. (1999). Social democratic parties in the European Union : history, organization, politics. New York, St. Martin's Press. 3

4 The remainder of this paper is organized into five sections. The first sets the broad research question, focusing on the occurrences that throughout the 1970s undermined the traditional party/unions relationship. It then reviews the most common answer to the problem, according to which the end of socialist demand-side policies entailed a de-linkage or divorce between the two actors. The paper then presents empirical material to contradict the conventional theory. The emergence of Social Partnership in the late 1990s, grounded on close collaboration between parties and unions, contrasts fundamentally with the de-linkage prediction. Moreover, contrary to expectations arising from received wisdom, in countries where a divorce is apparent, as in 1990s Britain, wages perform well and income distribution is favourable to salary earners. The fourth section draws an hypothesis for an alternative explanation, which reverses the usual causal direction, and according to which the nature of the relationship depends primarily on unions structures and policies. The fifth section concludes. The Background: Crisis in a Marriage In the aftermath of the Second World War, Social Democratic parties and trade unions in northern and central European countries developed a distinctive kind of relationship, whose causes and effects have been widely studied. The most immediate reason for their collaboration was their primary reference to the industrial working class. Socialists ceased to be class parties as soon as they endorsed the liberal democratic method (Przeworski 1985), nonetheless they continued to build social majorities around a working class core for decades. The emergence of an organized mass party implied that the membership of the two actors overlapped to a large extent (Marlière 1999). This class foundation of the party/union relationship progressively faded during the 1970s, being undermined by the process of de-industrialization. While the industrial working class shrunk, personal identities as formed in the workplace multiplied (Crouch 1999). Unions membership decreased as collective representation of new occupational strata became harder to achieve (Ebbingghaus 2002). Meanwhile, the variety of societal interests that the party needed to manage became more complex (Kitschelt 1994). The party s structure turned into a lighter cadre party, so that the overlapping membership progressively reduced its salience (Inglehart 1990; Mair and van Biezen 2001; Ebbingghaus 2002). In sum, the class foundation of the party/union relationship came to an end while post-industrial society gained terrain as a new social and economic order. Meanwhile, the conditions for the political exchange carried out between the two actors during social democratic governance periods were also fading. The political exchange (Pizzorno 1978; Goldthorpe 1984) was a deal between demand-side policies and wage 4

5 moderation and represented the core of the social democratic project 2. In fact, as early as the late 1960s, this started showing worrying weaknesses. Buoyant economic growth triggered a period of wage militancy across European countries. When, later on, wage militancy was joined by two oil shocks, Europe entered the worst period of raising inflation since the 1920s (Maddison 1991; Notermans 2000). Aimed at counteracting this phenomenon, a new international consensus grounded on restrictive monetary policy took the place of the postwar Keynesian consensus, that was also seriously under attack by the monetarist critique (Friedman 1968). Inflation control took the place of full employment as the main economic policy objective, and from the mid-1970s, deflationist monetary stances spread across the continent (Notermans 2000). Capital market liberalization and the establishment of the European Monetary System also forced reluctant European countries such as the social democratic Sweden to converge (Notermans 2000: 196). In spite of some resistance, traditional political exchange ceased to be a viable political option since 1981 U-turn of Mitterand s government. In that year the French socialists engaged an austerity programme to counteract the economic crisis giving up Keynesianism in one country (on the U-turn see Hall 1987). This change in the political economic framework occurred just when in a number of countries such as France, Spain, or Greece, Social Democratic parties were winning elections for the first time (Gallagher and Williams 1989). In other words, while the bases for a stable and successful socialist government were fading, the family of (potentially) governing parties was enlarging, with new socialist rule coming to power across Europe 3. Paradoxically, a genuinely continental European Left, constituting the Party of European Socialists (Hix 1999), was emerging while the traditional bases for a collaborative relationship with their sister trade unions were fading. If the traditional foundations have vanished, what is driving the party/union relationship in Europe during the 1980s and 1990s? Which forms does this relationship actually take? 2 Under the political exchange framework, social democratic governments used to pursue demandside policies aimed at full employment and income redistribution. In exchange for this, concentrated and centralized trade unions would have provided an institutional framework for inflation control, moderating their wage claims (cf. Schmitter, P. C. and G. Lehmbruch (1979). Trends toward corporatist intermediation. London, Sage Publications.). The two elements of the deal were complementary: where inflation was not under control any more, eventually unemployment would have risen (cf. Cameron, D. (1984). Social Democracy, Corporatism, Labour Quiescence, and the Respresentation of Economic Interest in Advanced Capitalist Society. Order and conflict in contemporary capitalism. J. H. Goldthorpe. Oxford, Clarendon: ) 3 The Italian PDS has been the last new European Social Democratic party to join the family of the Socialist International in It won election as the mainstream reformist party in

6 The De-linkage Answer: Announcing Divorces The most common answer that established literature gives to this question can be summarized as follows. With the end of the overlapping of parties and unions memberships, the rationale for any party/unions relationship has to be found within social democratic policies. However, the theory continues, during periods of social democratic governance, a collaborative relationship between Social Democratic parties and trade unions can run if and only if the former adopts a traditional Keynesian stance in political economy, able to benefit unions and union members. In other words, without some sort of income redistribution or expenditure geared towards the working class, or more generally salaried employees, a divorce between the two actors will take place causing an antagonistic relationship. Since socio-economic conditions in the 1980s and 1990s have induced Social Democratic parties to abandon their traditional Keynesian stances, the relationship between these and their old trade unions allies will end in divorce. James Piazza wrote the article that probably systematizes this de-linkage argument in the most rigorous manner, using multivariate analysis on a sample of 16 countries over 45 years (Piazza 2001). Comparing two different periods, and , he finds that globalisation, defined as both trade and capital account liberalization, was negatively correlated to increase in trade union density. Globalization, therefore, ended in reducing the political relevance of organized labour. Hence, he argues, trade unions ceased to be pivotal partners for social democrats. Indeed, he shows that the parties electoral success in the second period was not positively correlated (as it used to be) to growth in union density. Therefore, infers Piazza, Social Democratic parties have abandoned their traditional social democratic policies in order to capture a wider array of interest groups and forge their electoral basis on a new socio-economic coalition. This is demonstrated through a positive correlation between the abandonment of traditional Keynesian policies by social democrats and globalisation. Interestingly, Piazza s regression analysis is the ultimate test able to correlate the tight monetary discipline pursued by social democrats to the new structural conditions of the economy as they have emerged since the 1980s. Piazza concludes his study by adding a further consequence to its findings: Centre-left parties are not as closely committed to symbiotic friendship with the unions and adherence to generous social spending, nationalized industry or centralized bargaining (Piazza 2001: 426). 6

7 By inducing a drift in social democratic political economy, he argues, globalisation has undermined the party/unions relationship, and a de-linkage between the two actors has occurred. The assumption underpinning Piazza s conclusions stems also from another perspective focusing on Social Democratic parties. In his seminal study, Herbert Kitschelt (1994) explored the determinants of electoral fortunes of Social Democratic parties in post-industrial Europe. According to Kitschelt, in the new context a close relationship with trade unions is no longer an asset for Social Democratic parties, rather it is a liability. Parties still entrenched in a collaborative relationship with trade unions, which may be reinforced by formal links (as used to be the case in Sweden and the UK), were doomed to pay a high electoral price (Kitschelt 1994: 225). This conclusion derives from his conceptualization of the electoral arena in post-industrial society. Kitschelt argues that electoral competition is no longer fought only on a socialist/capitalist divide. The latter is cross-cut by a new liberty/authority cleavage in which post-materialist values play a crucial role 4. In the new context, if they are to win elections, Social Democratic parties have to manage a greater variety of interests and beliefs rather than relying on a single well-defined bloc of interests. Even if one fully endorses Kitschelt s conceptualization of post-industrial electoral arena, his conclusions still depend on the assumption that a collaborative relationship between trade unions and Social Democratic parties will only bring about traditional Keynesian-type of policies. Indeed, only in this case would the relationship hinder the possibility for social democratic governments to pursue the management of a greater variety of interests and values. In other words, according to Kitschelt the party/unions relationship becomes a liability because it limits the party s room for manoeuvre to establish more complex societal coalitions around its programs. Again, the assumption behind this conclusion is that the party/unions relationship has to be founded on redistributive economic policies. The literature on trade unions and industrial relations is perhaps the most direct in identifying the causes of the party/unions divorce. While focusing on the traditional dichotomy between capital s and labour s interests, Ferner and Hyman underline that the 1980s and 1990s have witnessed a dominance of the former over the latter. In particular, they observe that the economic context has induced Social Democratic parties to retreat and to give up their Keynesian redistributive stances. As a consequence, in countries like France and Spain trade unions have remained orphans of their political ally. As empirical evidence of the 4 Kitschelt builds on Inglehart, R. (1977). The silent revolution : changing values and political styles among western publics. Princeton,, Inglehart, R. (1987). "Value Change in Industrial Societies." American Political Science Review 81(4):

8 divorce, in these countries the relationship has ended in very antagonistic attitudes, with trade unions calling for anti-governmental strikes (Ferner and Hyman 1992: xix). Ross and Martin (1999) are similarly clear-cut. Since in many countries social democratic governments took the lead in implementing liberalizing policy shift, the relationship with the unions is doomed to end. The logic is always the same: Social Democratic parties need more leverage over their economic policies in order to forge stronger electoral blocs, and the traditional alliance with the unions, which included redistributive policies, had no hope of survival any more (Ross and Martin 1999: 15). Finally, and turning again to one of the few studies focusing on the relationship as such, similar considerations and analyses stem from a class-based approach to the party/unions relationship. Howell (2001) argues that the end of pro-workers spending policies leads to the divorce of the two actors, because it entails political isolation for trade union movement (Howell 2001: 9). Howell insists in particular on the social democratic industrial relations approach. He links the divorce of the two actors to a departure of the former from a clear idea of the role that labour should play in the regulation of socio-economic relations. This brief review of the literature could be widened to encompass more studies on trade unions or Social Democratic parties, but the result would not change (among others: Baglioni and Crouch 1990; Marlière 1999). Scholars from different fields and traditions, applying different methods, all draw the same conclusions. Indeed, they assume that a non-keynesian, non-redistributive social democratic political economy entails a divorce between organized labour and its long-standing political ally. Or, vice-versa, that the successful implementation of a new leftist political economy needs some sort of confrontational stance in the relationship with unions. In the following section I will offer empirical material in order to challenge this vision and the conclusions it entails. Observing More Complex Feelings Social Partnership, or Competitive Corporatism The conventional approach to the party/unions relationship entails a clear prediction. The process of European convergence towards restrictive monetary stances went ahead without interruption from the mid-1970s, and it was endorsed by governing Social Democratic parties. Therefore, the more we depart from the beginning of this process, the more we should encounter antagonistic party/unions relationships across Europe. 8

9 Figure 1 indicates the number of European countries with at least one Social Democratic party in government for each year between 1980 and At the end of the 1990s, a real social democratic boom occurred in western Europe. For a biennium, namely in 1998 and 1999, only Spain was ruled by a conservative government. In the mid-to-late 1990s, tight monetary policy was combined with budget restraints and fiscal consolidations geared at complying with the Maastricht criteria for the monetary union. Politically, the scene was dominated by a new generation of social democrats. If the divorce argument was to hold, this period should therefore have been the theatre of widespread antagonistic relationships between parties and unions Figure 1 Social democratic Governments in Europe, The graph indicates the number of governments in which Social Democratic parties had a role, as single ruling party or in coalitions with others, in the EU-15 member states barring Luxembourg and Ireland. Source: author s own compilation. On the contrary, however, the second half of the 1990s was characterized by the reemergence of a policy method centred on peak consultations between governments, the trade unions and the employment associations. Further, in countries such as Ireland or Italy where the preconditions for corporatist agreements were absent, Social Partnership became the new buzzword of economic governance (Goldthorpe 1984; Baccaro 2003). According to Fajertag and Pochet, virtually all European countries in the late 1990s, with the significant exception of Britain, endorsed this method of shared policy-making (Fajertag and Pochet 1997; Fajertag and Pochet 2000). In countries as different as Italy, Finland, Germany, Denmark and others, Social Partnership was associated with social democratic governance 9

10 periods. A close collaborative relationship between the party (in government) and their closest trade unions confederation remained pivotal for this policy project. In countries such as Italy, Portugal and Finland, this collaborative relationship led to fullyfledged social pacts, i.e. tripartite agreements covering a broad range of different socioeconomic policies and often including norms on wage increases. In other countries, such as Germany, Denmark and France, the intense consultation of social parties did not produce encompassing, broad agreements, but rather a number of issue-specific understandings (Fajertag and Pochet 1997; Fajertag and Pochet 2000). However, the substance of this new party/unions relationship was quite homogeneous. The political economic framework was dominated by an orthodox approach based on supplyside policies, budget restraints, and inflation control. The unions delivered their consensus over policies, often engaging the body of their organization (Baccaro 2002). Moreover, they often secured wage moderation, which would have enhanced cost competitiveness for trading sectors 5. It should be emphasised that in this case wage restraints played a very different role than that in the old political exchange. In the new macroeconomic framework, inflation is curbed through tight monetary stances. Thus, in order to pursue their new political economy aimed at combining low unemployment with a relatively equitable income distribution socialist policymakers only need a solid electoral majority. 6 (Boix 1998:35). In exchange for consensus and social peace, unions were gaining a significant political role, which, starting from labour-related issues, achieved a wider scope (Hyman 2001). In their role as Social Partners, trade unions had a say in most socio-economic policy areas whilst keeping a strong grip over labour market dynamics. In many countries, labour markets were being restructured since traditional institutions were blamed for leading European countries to persisting unemployment (OECD 1997). In countries where a collaborative party/unions relationship was established, trade unions were able to lead the restructuring process, retaining a role in the management of flexibilized labour markets, or reducing the impact of flexibilization on their own members (Esping Andersen and Regini 2000). The conventional theory which links the new social democratic political economy to the end of the party/unions relationship is not compatible with this evidence from a bird s eye perspective: social democrats do not need unions and unions are not gaining from traditional Keynesian policies, nonetheless the two collaborate over an extremely wide array of policies. 5 Interestingly enough, the Social Partnership approach has been labelled as competitive corporatism so as to mark both continuity and discontinuity with earlier kinds of agreements (cf. Rhodes, M. (1996). Globalisation, labour markets and welfare states :a future of' "competitive corporatism"? Florence, European University Institute.). 6 According to Boix (1998) the new policies would have been centred on active labour market policies and education. 10

11 On the Causes of Blatant Divorces As already mentioned, at least one country seemed to verify the divorce theory in late 1990s Europe. In 1997, after 18 years spent in opposition, the British Labour party was elected to government and a new era for pro-labour industrial relations was forecast (Taylor 1998). In reality however, the relationship between the two actors turned less than friendly. The Labour government made it clear from the beginning that the supply side of the economy was to rest at the core of its concerns. In this respect it plainly stated that: The extent of trade union growth and organization is dependent on [ ] how much help they can bring to the success of the enterprise (DTI 1998). This orthodox approach to economic management clashed with the unions expectations, and throughout the governance period, the relationship between these and the Labour government was much less collaborative than predicted, instead becoming confrontational and adversarial. Trade unions considered that the employers got the favours that the government was denying to trade unions, whereas employees were not getting the fairness they were promised (Morris 2003). The Labour government was hence singled out as a pro-business one. According to its discontents, the Labour s program in industrial relations required nothing less than the marginalization of unions as independent forces (Smith and Morton 2001). While being harshly blamed, the government was pleading for Partnership in the workplace (Undy 1999), clearly echoing experiences from abroad and urging unions to modernize if they were to keep pace with Britain. It is arguable that these frictions should mirror low trends in wages and in income distribution. According to conventional theory, the reason for the divorce resides in the policy shift that ended redistributive policies. Therefore one should expect indicators on wages and income distribution in Britain to fair comparatively poorly. In order to verify this, I compared British trends in real compensation per employee and in wage share of GDP with both the EU average and Italy. This country is easily comparable, being ruled during roughly the same period by a coalition led by social democrats 7. In contrast with Britain, however, Italian centre-left coalition has had very collaborative relationships with the trade unions (Baccaro 2001). 7 Colin Crouch (2000) makes a strong point for the high level of comparability between Italy and the United Kingdom. 11

12 The comparison shown in Figure 2 and 3 is quite surprising. In Great Britain, trends in real compensation per employee and in wage share of GDP between 1995 and 2002 compare favourably both with the EU average and with Italy. In particular wage share of GDP in Britain has a reversed trend. Between 1997 and 2001, it was decreasing from 61.3 percent to 58.3 percent in Italy, while in Britain was increasing from 61.9 percent to 63.5 percent (European Commission 2003). These aggregate data, one could argue, might however be biased by the disproportionate impact of high-level wages. In other words, these trends might have been tricked by highprofile employees in the financial or legal services earning very high salaries and not representative of the average unionized employee United Kingdom Italy EU-15 Figure 2 Real compensation per employee, selected countries , 1995 = 100 Source: European Commission (2003) United Kingdom Italy EU-15 Figure 3 - Wage share of GDP, selected countries, Source: European Commission (2003) 12

13 Data shown in Tables 1 and 2 offer a long-run perspective over income distribution in Britain, and rule out this objection. The economic gains achieved during the second half of the 1990s were distributed evenly between different income classes of British households. The lowest percentiles actually increased their share slightly more than the highest. The figures are even more significant if compared to the previous 18 years of Conservative rule. Between 1978 and 1996, real disposable income of households in the 75 th percentile increased by 42 percent, while the same figure for households in the 25 th percentile increased merely by 18 percent. Between 1996 and 2001, i.e. during the Labour government, the picture is reversed with the 25 th percentile increasing its real disposable income by 19.5 percent, and the 75 th percentile by 17.7 percent. In sum, the trends observed in aggregate data on wages and wage share of GDP is reflected on trends in income distribution across different strata. In this respect, compared to previous decades the Labour government marked a high degree of discontinuity. Table 1 Distribution of real household disposable income in the UK during Conservatives Gov t 1978 = th 10 th percentile percentile Median 75th percentile 90th percentile / Table 2 Distribution of real household disposable income in the UK during Labour Gov t 1995/96 = th percentile 25th percentile Median 75th percentile 90th percentile 1995/ / / / / / Source: National Statistics website: 13

14 Renewing Vows or Painful Divorces? Instead of being the theatre of widespread divorces between parties and unions, in the late 1990s the majority of European countries saw a renewed collaboration between the two actors. Moreover, when confronting the conventional theory with data on Britain and Italy the puzzle becomes more intriguing. An adversarial party/unions relationship in the first country was coupled with net gains for salaried employees, while in Italy the case was reversed. Against this background, the driving forces of the party/unions relationship remain obscure. However, drawing from the features of the new collaborative party/unions relationship, it is possible to set an hypothesis to be tested though further research. The old political exchange was founded on a clear deal: wage moderation versus public spending. The two elements were complementary and mutually reinforcing (Cameron 1984). The new collaborative relationship is different both in content and in form. In order to pursue their new political economy, Social Democratic parties do not need the collaboration of unions as they used to during the post-war decades (Boix 1998). This does not imply, however, that they cannot benefit from the collaboration of unions. In many European countries, financial and demographic reasons pushed towards urgent reforms of welfare state institutions. Some scholars underline that the collaboration of a large trade unions confederation can significantly ease the task (Baccaro 2002; Streeck and Hassel 2003). More generally, given the exposure of European economies to international competition, the willingness of trade unions to coordinate wage increases can improve economic growth, hence benefit the party electorally. Consequently, it seems far from sensible for the party simply to dismiss such a powerful ally. The unions also have much to gain from a collaborative relationship, so that their agreement on disinflation is not so strange as it seems otherwise (as in Hyman 2001: 56). As mentioned earlier, in a period characterized by restructuring and flexibilization of labour market institutions (OECD 1997), collaborative trade unions were able to keep their grip on those dynamics, and reduce the negative impact that the multiplication of labour patterns inevitably would have had on them (Ross and Martin 1999; Esping Andersen and Regini 2000). From a renewed role in their core business, unions were then able to depart from their original boundaries. They took the lead in representing wider societal concerns such as the environment, social exclusion, or international conflicts. In this way in the larger arena of civil society trade unions re-gained the terrain they had lost in the working place from the effect of market internationalization (on 'civil society': Baccaro 2001; on unions: Hyman 2001: 56-68). Seen from another perspective, trade unions renewed their old quest for citizenship rights (Ebbinghaus, 1995) in their engagement with citizenship concerns (Streeck and Hassel 2003). 14

15 Against this background, the research question becomes the following: under which conditions were unions and parties able to benefit from a collaborative relationship? The conventional explanation implies a clear causal direction, i.e. the parties abandoned Keynesian stances, hence trade unions were left as political orphans, and the divorce took place. If the theory is fundamentally challenged by empirical observations, the causal direction should also be reconsidered. The new political economy of the social democrats, aimed at combining market efficiency with their distinctive quest for equality, has been grounded on employability, education, and, more generally, equal access to opportunities (Bobbio 1995; Boix 1998). By their own nature, these new leftist supply-side policies are not attributable to single occupational groups, but are directed towards the society as a whole. To use economic terminology, they appear as non-rival and non-excludable public goods. This program does not clash in principle and in practice with a collaboration with trade unions confederations, as long as the latter does not seek for particularistic, sectoral, short-term gains which would undermine the overall policy project. In other words, social democratic governments will benefit from collaborative relationship with trade unions, only if the latter internalize the systemic consequences of their actions, anticipating the effects of their claims. Olson (1982) first introduced this concept referring to the internal structure of what he called encompassing unions. He argued that unions organizing across different sectors internalize the systemic consequences of their actions because their universe approximates the whole of society. Other literatures on peak agreements between social partners over wage increases used to make a distinction between unions in exposed, trading sectors, and unions in sheltered sectors. The former will internalize (anticipate) the impact that wage militancy can have on competitiveness, because their firms are price takers, therefore increasing costs could result in reducing employment (Crouch 1988; Franzese 2001). Therefore, similarly to Olsonian encompassing unions, confederations dominated by unions in trading sectors will internalise systemic consequences of their claims. Borrowing from these theories it is arguable that in countries where the process of industrial restructuring did not lead public sector trade unions to dominate the confederation, encompassing-type attitudes would prevail. Indeed, the completion of the European Common Market which occurred in the late 1990s played a crucial role to this outcome. Given that competence for fiscal policies rested within individual member states, the single market blurred the distinction between exposed and sheltered industries, so that only the public sector remained genuinely non-trading. Moreover, it is likely that tradable (productive) sectors remained more salient in countries where unions had a greater role during the industrial restructuring process in the 1970s and 1980s, whereas if the unions suffered from a top-down process of de-industrialization, 15

16 unionised private sectors remained subdued. In this latter scenario besieged unions retrenched and adopted defensive militant attitudes. Fragmented and weakened unions, aiming at maximizing short-term benefits for their members, are not viable partners for modern Social Democratic parties. Conversely, trade unions internalising the consequences of their policies, albeit radical in their quest for social justice and firm in their grip on labour market dynamics, will be precious allies for a nonclass-based leftist political project (Streek and Hassel 2003). In turn, collaboration with governments is not a viable option for militant trade unions. Credibility at the workplace level can be hindered by concessive attitudes, and in a fragmented environment these might trigger fierce competition at the workplace level for workers representation (Astudillo Ruiz 2001). Conclusions Based on the analysis of primary and secondary sources, as well as trends in wages and income distribution, this paper has argued that conventional theory predicting a divorce between European Social Democratic parties and trade unions falls short of explaining most of the development observable in 1990s Europe. With the end of the Keynesian consensus and the emergence of post-industrial society, the long-lasting relationship between the two actors has been called into question. Social democratic governments which seized government after 1980 widely endorsed the new consensus grounded on inflation control, giving up their traditional stances. Most scholars in party politics and industrial relations argued that this policy drift entailed a divorce between the two actors. At the end of the 1990s, the great majority of European countries was ruled by Social Democratic parties that continued to pursue restrictive monetary policies. Unexpectedly, in the majority of cases, budget constraints and supply-side policies came together with strong policy coordination with trade unions, which endorsed their aims and objectives. In other countries like Britain, orthodox economics was matched with antagonistic party/unions relationship, which seemed to track the conventional divorce theory. However, data on trends in wages and income distribution show that something more than the political economy must be at stake in order to explain the factors driving the relationship. In fact, trends in wages and income distribution during Labour government compare favourably both to previous periods in Britain and to other European countries. Reversing the causal direction of the conventional explanation, this paper has argued that the party/unions relationship might be explained by the nature and features of the trade unions 16

17 confederations. Indeed, only if trade unions adopted encompassing policy stances would a collaborative relationship have benefited both the actors. If, conversely, trade unions were confined to militant attitudes by harsh de-industrialization processes, the best option for both the actors would have been to stress their respective independence, ending in antagonistic relationships. This hypothesis seems plausible to explain both European trends during the 1980s and 1990s, and the British experience in the late 1990s. However, further empirical research would be needed to test its validity. 17

18 References Astudillo Ruiz, J. (2001). "Without Unions, but Socialist: The Spanish Socialist Party and Its Divorce from Its Unions Confederation ( )." Politics & Society 29(2): Baccaro, L. (2001). Civil Society, NGOs, and Decent Work Policies: Sorting out the Issues. Discussion Papers. International Institute for Labour Studies. Geneva, International Labour Organization. Baccaro, L. (2001). The Construction of "Democratic" Corporatism in Italy. Baccaro, L. (2002). "Negotiating the Italian pension reform with the unions: lessons for corporatist theory." Industrial and labor relations review 55(3): Baccaro, L. (2003). "What is Dead and What is Alive in the Theory of Corporatism." British Journal of Industrial Relations(Forthcoming). Baglioni, G. and C. Crouch, Eds. (1990). European industrial relations : the challenge of flexibility. London, Sage. Bergounioux, A. (1983). "Typologie des rapport syndicats-partis en Europe occidentale." Pouvoirs 26: Bobbio, N. (1995). Left and right : the significance of a political distinction. Cambridge, Polity Press. Boix, C. (1998). Political parties, growth and equality : conservative and social democratic economic strategies in the world economy. Cambridge, U.K ; New York, N.Y., Cambridge University Press. Cameron, D. (1984). Social Democracy, Corporatism, Labour Quiescence, and the Respresentation of Economic Interest in Advanced Capitalist Society. Order and conflict in contemporary capitalism. J. H. Goldthorpe. Oxford, Clarendon: Crouch, C. (1988). Trade Unions in the Exposed Sectors: Their Influence on Neo-Corporatist Behaviour. Labour Relations and Economic Performance. R. Brunetta and C. Dell'Aringa. London, Macmillan: Crouch, C. (1999). Social change in Western Europe. New York, Oxford University Press. DTI (1998). Fairness at Work. London, Stationery Office. Ebbingghaus, B. (2002). "Trade unions' changing role: membership erosion, organizational reform, and social partnership in Europe." Industrial Relations Journal 33(5): Ebbinghaus, B. (1995). "The Siamese Twins: Citizenship Rights, Cleavage Formation, and Party- Union Relations in Western Europe." International Review of Social History 40(Supplement 3): Esping Andersen, G. and M. Regini, Eds. (2000). Why Deregulate Labour Markets?, Oxford University Press. European Commission (2003). AMECO Database. Directorate General for Economic and Financial Affairs. Bruxelles, European Commission. Fajertag, G. and P. Pochet, Eds. (1997). Social Pacts in Europe. Bruxelles, ETUI. Fajertag, G. and P. Pochet, Eds. (2000). Social Pacts in Europe. New Dynamics. Bruxelles, ETUI. 18

19 Ferner, A. and R. Hyman (1992). Introduction. Industrial relations in the new Europe: Seventeen Types of Ambiguity. Industrial relations in the new Europe. A. Ferner and R. Hyman. Oxford ; Cambridge, Mass., Blackwell: xvi-xlix. Franzese, R. J. (2001). Institutional and Structural Interactions in Monetary Policy and Wage/Price- Bargaining. Varieties of Capitalism: The Institutional Foundations of Comparative Advantage. P. A. Hall and D. Soskice. New York, Oxford University Press: Friedman, M. (1968). "The Role of Monetary Policy." American Economic Review 58: Gallagher, T. and A. M. Williams (1989). Southern European socialism : parties, elections and the challenge of government. Manchester, Manchester University Press. Goldthorpe, J. H., Ed. (1984). Order and conflict in contemporary capitalism. Oxford, Clarendon. Hall, P. A. (1987). The Evolution of Economic Policy under Mitterand. The Mitterrand experiment : continuity and change in modern France. G. Ross, S. Hoffmann and S. Malzacher. Oxford, Polity: Hall, P. A., Ed. (1989). The political power of economic ideas : Keynesianism across nations. Princeton, N.J., Princeton University Press. Hix, S. (1999). The Party of the European Socialists. Social democratic parties in the European Union : history, organization, politics. R. Ladrech and P. Marlière. New York, St. Martin's Press: Howell, C. (2001). "The End Of the Relationship Between Social Democratic Parties and Trade Unions?" Studies in Political Economy 65(Summer): Hyman, R. (2001). Understanding European trade unionism : between market, class and society. London, Sage. Inglehart, R. (1977). The silent revolution : changing values and political styles among western publics. Princeton,. Inglehart, R. (1987). "Value Change in Industrial Societies." American Political Science Review 81(4): Inglehart, R. (1990). From Class-Based to Value-Based Politics. Oxford readings in politics and government. P. Mair. Oxford, Oxford University Press: Kitschelt, H. (1994). The transformation of European social democracy. Cambridge ; New York, Cambridge University Press. Ladrech, R. and P. Marlière, Eds. (1999). Social democratic parties in the European Union : history, organization, politics. New York, St. Martin's Press. Maddison, A. (1991). Dynamic forces in capitalist development : a long-run comparative view. Oxford, Oxford University Press. Mair, P. and I. van Biezen (2001). "Party Membership in Twenty European Democracies, " Party Politics 7(1): Marlière, P. (1999). Introduction: European Social Democracy, in Situ. Social democratic parties in the European Union : history, organization, politics. R. Ladrech and P. Marlière. New York, MacMillan: Martin, A. and G. Ross, Eds. (1999). The brave new world of European labor : European trade unions at the millennium. New York, Berghahn Books. 19

20 Morris, B. (2003). Deliver on fairness not favours in ERA review, says Morris, T&G Press Release. Notermans, A. (2000). Money, markets, and the state : social democratic economic policies since Cambridge, Cambridge University Press. OECD (1997). Implementing the OECD Job Strategy: Member Countries' Experience. Paris, OECD. Piazza, J. (2001). "De-linking labour. Labour Unions and Social Democratic Parties under Globalization." Party Politics 7(4): Pizzorno, A. (1978). Political Exchange and Collective Identity. The resurgence of class conflict in Western Europe since C. Crouch and A. Pizzorno. London, Macmillan.: xxiv,349p. Przeworski, A. (1985). Capitalism and social democracy. Cambridge, Cambridge University Press. Rhodes, M. (1996). Globalisation, labour markets and welfare states :a future of' "competitive corporatism"? Florence, European University Institute. Ross, G. and A. Martin (1999). European Unions Face the Millennium. The brave new world of European labor : European trade unions at the millennium. New York, Berghahn Books: Schmitter, P. C. and G. Lehmbruch (1979). Trends toward corporatist intermediation. London, Sage Publications. Smith, P. and G. Morton (2001). "New Labour's Reform of Britain's Employment Law: The Devil is not only in the Detail but in the Values and Policy Too." British Journal of Industrial Relations 39(1): Streeck, W. and A. Hassel (2003). Trade Unions as Political Actors. International Handbook of Trade Unions. J. T. Addison and C. Schnabel. London, Edward Elgar. Taylor, R. (1998). "Annual Review Article." British Journal of Industrial Relations 36(2): Undy, R. (1999). "Annual Review Article: New Labour's 'Industrial Relations Settlement': The Third Way?" British Journal of Industrial Relations 37(2):