Spain: From Tripartite to Bipartite Pacts

Transcription

1 Time:12:07:48 Filepath:d:/womat-filecopy/ D 8 Oscar Molina, and Martin Rhodes 8.1. Introduction The history of social pacts in Spain is one of success in the early 1980s dramatic failures ten years later and the consolidation of peak concertation and bipartite pacts after the late 1990s. This changing and irregular pattern poses numerous challenges to scholars interested in social pact emergence and institutionalization. Among explanations in the literature on Spain, functionalist and neoinstitutionalist arguments have predominated. According to the functionalist approach, social pacts respond to the time-specific needs of the economy and/ or political context. Hence, the economic and political crisis that accompanied Spain s transition to democracy was for many a sufficient explanation for its negotiated character (Estefanía and Serrano, 1990; Zaragoza and Varela, 1990; Roca, 1993; Trullén, 1993; Heywood, 1995). But this argument cannot explain on its own why there was no social pact in the first half of the 1990s, when unemployment reached record levels and economic imbalances jeopardized Spain s membership of Economic and Monetary Union (EMU). Nor does it fully account for the revival of concertation in the second half of the 1990s. The second approach neo-institutionalism focuses on interest representation and intermediation, and attributes pacts in the transition years to a neo-corporatist pattern of policymaking, notwithstanding the weakness of such institutions in Spain. According to this argument, neo-corporatist innovations including the centralization of decision-making and tripartite bargaining occurred under the pressure of socio-economic circumstances (Pérez-Diaz, 1984; Espina, 1999). Others contest this view, claiming that the abandonment of pacts after the mid-1980s was precisely due to social partner weakness and the absence of supporting institutions (Foweraker, 1987; Martínez Lucio, 1989; Maravall, 1997; Pérez, 2000; Royo, 2002). 174

2 Time:12:07:48 Filepath:d:/womat-filecopy/ D This chapter presents a different account. In line with the general argument of this book, we reject the explanatory capacity of single-variable analyses. While taking political and economic contexts and institutions into account, we focus on changing perceptions of relative power and interaction between social actors and governments. Problem loads are important in shaping those perceptions and clearly affect the dynamics and outcomes of negotiations. Political and organizational factors are also important, and one of our more robust conclusions is that government weakness is the best predictor of successful pacting, alongside relations between and within the Spanish unions. The success of unions as social and political actors depends on their ability to strike a balance between their logics of influence and membership or representation (Streeck and Schmitter, 1999). Their context-specific perception of how concertation will affect that balance helps determine whether a pact is signed. We argue similarly that no single variable can explain the evolution of Spanish social pacts. Broad or encompassing tripartite social pacts were abandoned after the mid-1980s, with the single exception of the 1997 April Agreements. But concertation has remained alive and well, and has become to some extent institutionalized, via policy-specific bipartite agreements, notably the annual inter-confederal wage agreements, in place since Functionalist arguments contribute much to our understanding at critical junctures, especially regarding the problem load of high unemployment and the massive expansion of fixed-term labour contracts. But utilitarian, power-distributional, and normative theories provide valuable insights into the timing, character, successes, and failures of Spain s social pact responses to economic challenges The Emergence of Social Pacts Political and Organizational Variables Social pacts in Spain have been signed under very different circumstances. While providing rich evidence for analysis, this diversity of contexts makes it difficult to single out sufficient or necessary conditions. Pacts have been signed in periods of recession and declining competitiveness as well as in periods of growth and employment creation. It could be argued that political institutions have been an obstacle to pacting in Spain. There are no strong, institutionalized mechanisms for social partner participation in policymaking, except for the Economic and Social Council (ESC), created in 1990, in which unions and employers have a purely advisory role. Indeed, the Spanish political system has little tradition of consensual politics, apart from the years of democratic transition, and relations between the two main political parties the PSOE (Partido Socialista Obrero Español) and the PP (Partido Popular) are confrontational. Nevertheless, the Moncloa 175

3 Time:12:07:48 Filepath:d:/womat-filecopy/ D Social Pacts in Europe Pact of 1978 on economic management and the Toledo Pact of 1995 on pensions between the political parties, both supported by the social partners, provided the basis for significant consensus-based policymaking in the years that followed; and the period since 1997 has witnessed a strengthening of the ESC, as well as the institutionalization of union employer commissions on employment and collective bargaining issues. The colour of government a precondition of social pacts in neo-corporatist theory appears to explain little: the years of majority Socialist rule ( ) saw a high level of social unrest and conflict between the unions and the government (Astudillo, 2001), while the resurgence of tripartite concertation after 1996 occurred under the centre-right PP. But although Spain has almost perfect party-system bipolarism, regionalist (e.g. the Catalan and Basque) parties play an important role in helping stabilize governments without ample majorities a rather frequent occurrence and the first clue as to why Spanish governments turn to pacts. Thus, social pacts have most often been signed when governments are weak and dependent on support from regionalist parties, even if this does not guarantee success, as revealed by the failure of pacting under a Socialist-led minority government in 1993 and Governments also tend to rely on pacts more so during their first terms in office. The first mandates of both major parties the PSOE ( and ) and the PP ( ) were periods of intense tripartite bargaining and pacting. Those years except for the PSOE s incumbency were followed by similarly intense periods of social and political conflict, coinciding with their consolidation or attainment of parliamentary majorities (the PSOE in ; the PP in ). This suggests that typically social pacts have been used strategically and instrumentally by governments, rather than as a dependable tool of socio-economic governance, although, as we discuss below, that may have begun to change in the 2000s. As for organizational factors, only unity of action between the two major unions the UGT (Unión General de Trabajadores) and the CC.OO (Comisiones Obreras) seems to increase the probability that a pact will be signed. When inter-confederal unity was achieved in the early 1990s, social pacts re-emerged; and enduring unity of action after the mid-1990s has underpinned their subsequent stability and gradual institutionalization. Union centralization and membership, by contrast, explain little. Both have remained almost unchanged since the early 1980s, the first at high levels, the second very low. But the low density levels of Spanish unions have never impeded their participation in pacts, for they derive their strength and legitimacy primarily from workplace elections rather than membership. Also important is the power conferred on unions by the state through official most representative status and the erga omnes extension clause of legally regulated collective agreements that increases their bargaining coverage to high levels (Nonell et al., 2006). 176

4 Time:12:07:48 Filepath:d:/womat-filecopy/ D The unions political roles and capacity for mobilization therefore far exceed their meagre organizational resources Negotiation Processes and Actor Preferences In the following, four episodes of pacting are subject to close scrutiny: the early 1980s pacts, the failed negotiations of the early 1990s, the social pact of 1997, and pacting in the 2000s (see Appendix A for the full range of social pacts signed in Spain between 1976 and 2010) THE TRANSITION TO DEMOCRACY AND SOCIAL PACTS ( ) The negotiation of social pacts in the early 1980s occurred against a backdrop of democratic transition, a deep economic crisis, and the consolidation of unions and employers organizations. Balancing the demands of different actors at a time when political and social institutions were also being reconfigured required self-restraint and a search for consensus. Fears of democratic reversal, especially after the failed coup d état of February 1981, reinforced incentives for cooperation. The weakness of the Union de Centre Democrático (UCD) coalition government led by Adolfo Suárez ( ) which combined Christian Democrats, Liberals, and former Spanish National Movement leaders helped bolster the appeal of the political left to voters demanding a stronger departure from Francoism. A critical legacy of forty years of dictatorship was the weakness of civil society and the social partners. The different ideological and associational traditions of the two main union confederations, the CC.OO and the UGT, produced divergent strategies. The CC.OO enjoyed much stronger roots at the company level, a more decentralized organization and a relatively strong capacity for mobilization. The UGT, by contrast, looked to state support for building organizational strength, making it more supportive of pacting than its rival, at least until the late 1980s. Its close links with the Socialist Party also made it more inclined to cooperate via pacts, whereas the most powerful influence within the CC.OO, although declining after the late 1980s, was the Spanish Communist Party (PCE). Inter-union competition for members and votes in workplace elections prevented strong labour movement unity until the mid-1990s. Organizational weakness was even more pronounced among employers. The Confederación Española de Organizaciones Empresariales (CEOE), created in June 1977, had a highly fragmented membership and lacked disciplinary control over its constituency. Its organizational structure has always been highly complex, with 114 sectoral associations and 50 territorial associations in 1985 (firms are usually affiliated to these organizations rather than the peak CEOE), both growing in number in subsequent years (EIRO, 1999). As a result, the CEOE has had an ambivalent view of concertation. If organizational weakness 177

5 Social Pacts in Europe has obliged it to take a critical position towards pacts, from which employers expect to obtain less than the unions, participation can also provide it with influence. Several months after the failed coup d état of February 1981, the UCD government initiated pact negotiations with the aim of building political and economic confidence. Even though all actors agreed on the need for a pact to strengthen democratic stability, and wage moderation as necessary for relaunching the economy, the communist CC.OO opposed any agreement that would moderate its industrial militancy. Initial contacts between the unions, the government, and the CEOE were accordingly difficult. Both unions feared that participation would legitimize government policies at the expense of workers. Because of the political context, the pact signed in June 1981 the ANE (Acuerdo Nacional sobre el Empleo) is popularly known as the Pacto del miedo ( Pact of Fear ). In order to secure the commitment of a reluctant labour movement, the government struck a number of parallel pacts, providing organizational and financial compensation to break down union resistance. Thus, the participation of the CC.OO was secured in part by the new roles given to the social partners in regulatory bodies, the National Social Security Institute (INSS), the National Employment Institute (INEM), the National Health Institute (INSALUD), and the National Institute of Social Services (INSERSO), as well as by a government commitment to create 350,000 public sector jobs (Estefanía and Serrano, 1990; Encarnación, 1999). But CC.OO agreement was also a response to competition from the UGT. The AMI Acuerdo Marco Interconfederal (National Multi-Industry Framework Agreement) signed by the UGT and the CEOE in January 1980, and renewed in 1981 rapidly increased the number of workers covered by national bargaining agreements and drove a wedge between the two unions. 1 Opposed by the CC.OO, these agreements strengthened the UGT which made important gains in workplace elections as a reward for its participation, and convinced the CC.OO that it was better off joining the ANE than opposing it. Both unions saw their bargaining position as strong vis-à-vis the fragile centre-right government and the fragmented CEOE, whose leaders regarded any pact as better than no pact at all, or the social conflict that failure might foment. But as predicted by the bargaining model, gaining union consent to a moderation of wage demands and militancy required the active involvement of a third partner the state. The second important social pact of the period was the 1984 AES (Acuerdo 1 The AMI established guidelines for collective agreements with the aim of restraining wages, reducing working hours, and raising productivity. It also created a framework for the new industrial relations system by regulating collective agreements and the role of trade unions in the workplace. 178

6 Económico y Social). The context for negotiations was now quite different. The 1982 elections gave the new Socialist government a comfortable majority, and fears of democratic instability were subsiding. The economy, however, was in crisis: inflation was at 14 per cent, domestic demand contracted by 1 per cent in real terms in 1984, and the unemployment rate was rising, reaching 21.7 per cent by the end of that year (Figure 8.1). This situation, plus the prospect of European Community membership, and a commitment to join the European Monetary System (EMS) in June 1989, increased the government s desire for an agreement. The AES and the wage moderation it secured, combined with a tight monetary policy, would allow the government to bring inflation down to 8 per cent in 1985 and 6 per cent by In this context, the government initiated a new round of talks. The unions reacted positively, although they imposed some a priori conditions: they would only accept wage moderation in exchange for an increase in social spending and an extension of welfare entitlements. But as both CC.OO and UGT officials 25 Success: ANE Success: AES Failure: Solidarity Pact Failure: Social Pact For Employment Success: April Agreements 20 Unemployment Rate Inflation 0 5 Fiscal Deficit Centre-Right- UCD Government -Relative- Majority Left PSOE Government -Absolute Majority Left PSOE Government -Absolute Majority Left PSOE Government -Absolute Majority Left PSOE Government -Relative Majority (coalition) Right PP Government -Relative Majority (coalition) Right PP Government -Absolute Majority Left PSOE Government -Relative Majority Figure 8.1. Social Pacts, Governments, and the Macro-Economy in Spain, Note: The unemployment data in this figure are the national data current at the time, rather than those in the new OECD historical series which have been revised downwards due to a methodological discrepancy between national data and OECD/Eurostat data that produced an overestimation of Spanish unemployment figures hitherto. 179

7 Social Pacts in Europe point out, they did not know if the government was willing to accept significant changes to its economic policy, or how much compensation they would be awarded in return for their support (El Pais, 17 July 1984). This uncertainty made the initial meetings difficult, as the unions struggled to articulate their bargaining strategies. The unions responses differed significantly. While the weaker UGT kept negotiations alive by making parallel deals with the government, the CC.OO s leaders abandoned the talks when they realized that the government s commitment to economic austerity might alienate their base. According to a leading CC.OO official: (... ) contrary to the practice today of signing wage agreements that establish guidelines for negotiations at lower levels, the early 1980s were real bi-partite or tri-partite incomes policy pacts. These pacts established bargaining limits for the actors. Accordingly, they were a permanent source of conflict. But the most worrying aspect of these pacts was that they limited the possibilities for the participation of the grass-roots in bargaining processes. This weakened the links between grass-roots and peak officials, thereby potentially affecting the representative capacity of the union. 2 The pact was therefore negotiated by the UGT, the government, and the CEOE. The employers, who feared a strong alliance between the Socialist government and the UGT, pursued a strategy of desgaste ( wear and tear ) and threatened several times to abandon negotiations if their demands were not met. This strategy was particularly threatening to the UGT whose reliance on concertation left it vulnerable to an employer-biased pact entailing too many costs for its constituency. The outcomes of the pact are in line with the bargaining model s predictions. The government perceived itself to be in a strong bargaining position thanks to its solid electoral mandate and the external constraint of securing EMS membership. By contrast, the CC.OO perceived itself as weaker than the Socialist government and its trade union ally, which explains its exit from the talks. The UGT had become increasingly critical of some of the government s policies, but believed that participation would deliver positive results. Securing the UGT s agreement to wage moderation required government commitment to a new social-democratic demand- and supply-side policy. This meant considerable policy innovation, including a loosening of the tight monetary policy the previous government had introduced to bring down inflation in 1982, and an agreement to increase spending on public investment, public employment, education and training, and unemployment benefits, as well as making the tax system more progressive. The support of the employers was secured by allowing for an easier use of temporary contracts (a concession that would lead to a profound segmentation of the Spanish labour market), part- 2 Interview with Ignacio Fernandez Toxo, Madrid, 16 April

8 time and seasonal employment, and by a modification of minimum wage standards for workers under 18. The 1984 pact would be the last for over a decade. The demise of tripartite concertation and pacting can be attributed to several factors: the disappointment of the unions with the results of the ANE and AES, the neo-liberal turn of the Socialist government, and its embrace of economic austerity, and the view of the employers that wage moderation could now be secured by high unemployment and the expansion of low-paid temporary work contracts. Internal dissent and contestation within each union was weak in the early 1980s, but increased after the 1984 pact, especially in the UGT whose members began to link their union to Socialist government policies and blamed their leaders for the strong moderation of wage increases, the continuing job crisis (unemployment reached 21 per cent in ), as well as for the weaker regulation of temporary contracts due to the AES and a revision of the Workers Statute in Some CC.OO workers also criticized their officials for not cooperating. Eventually, these developments would lead to a convergence of the UGT and CC.OO on a common model of trade union action, based on a revitalization of collective bargaining (Molina, 2006). Initially, however, the unions converged on a joint rejection of concertation as such. The UGT joined the CC.OO in a new phase of industrial militancy to recoup some of the concessions made in the 1980s pacts and to re-establish its reputation with workers, having lost some 40 per cent of its members between 1978 and The unions were not alone in rejecting concertation. The Socialist government had taken a firm neo-liberal turn under Miguel Boyer and Carlos Solchaga, the ministers of finance and industry respectively, who oversaw the introduction of fiscal austerity and a new commitment to labour market flexibility (Recio and Roca, 1998). The government continued to solicit pacts in the late 1980s (especially after the anti-government UGT CC.OO-led strike of 1988) but without much conviction: it walked away from negotiations with the UGT in 1987 after the union requested an upper wage band two points above the government s inflation forecast (Pérez, 1999: 673), and a growing budget deficit (at 7 per cent of GDP in 1985) ruled out further compensation for wage moderation through public spending. In any case, the government now believed it had an alternative to incomes policy: the rapid spread of temporary contracts moderated wage costs (in 1988 temporary workers wages were barely more than half those of permanent contract workers), while from 1987 the exchange rate of the peseta in the EMS began, for a while at least, to work as an anchor for the government s anti-inflation policy (Ferreiro and Gomez, 2005). 3 The outcomes of those pacts had been well below union expectations. Although in both public spending and social spending increased (the latter from 21.6 to 25.2 per cent of GDP), real wages fell by 9 percentage points between 1977 and 1986, and inflation targets were only met in one year out of seven (Ferreiro and Gomez, 2005). 181

9 Social Pacts in Europe Spain s employers, who had never been keen on concertation, also believed they could now avoid commitments and concessions to labour given the role of high unemployment and the expansion of fixed-term contracts, not just in moderating wage demands (real wages would rise by a marginal 0.9 per cent between 1984 and 1989) but also in weakening the influence of the unions (Rhodes, 1997; Encarnación, 1999; Ferreiro and Serrano, 2001) FAILED NEGOTIATIONS UNDER THE THREAT OF MAASTRICHT (1990 6) After six years of sustained growth, the 1990s began with a deep economic recession. The Socialist government s strategy of substituting a restrictive monetary policy and an overvalued currency for bargained wage moderation had failed to control wage growth and inflation (Pérez, 1999; Royo, 2001). In 1992, a large and increasing budget deficit, together with high inflation and external disequilibria, led to speculative attacks against the Peseta, and its devaluation within the ERM. The real costs of the crisis were skyrocketing unemployment (which rose to 23 per cent in 1993) and a further increase in the budget deficit. The political situation also deteriorated significantly for the Socialists. Although the 1989 elections allowed them to stay in government, their performance was far from the substantial victory achieved in 1986: the PSOE government won half the seats and only the absence from parliament of the Herri Batasuna deputies (the political arm of ETA) gave it a working majority (Lancaster, 1994). The severe recession, and a series of corruption scandals, further weakened the government over the next few years. In the 1993 elections, the PSOE remained Spain s largest party, but was forced to form a minority government when coalition talks with regionalist parties broke down. The economic and political circumstances should have been conducive to a pact, especially given the problem load of high unemployment and government infirmity. But perceptions of mutual weakness on the part of the unions, employers, and the government worked against it. After a decade of organizational development and consolidation, the UGT, CC.OO, and CEOE were ostensibly in a stronger position. In December 1988, the UGT and CC.OO led a successful 24-hour general strike against government plans for new weakly regulated contracts to boost youth employment, which led to the withdrawal of the proposals and government agreement to boost spending on social services. But union representation remained weak at the firm level and was mostly limited to large companies. Concertation, and its associated benefits, had strengthened them as political actors, but not in terms of membership or bargaining power. As CEOE President José María Cuevas put it: notwithstanding the enormous influence the unions have on public opinion and political decisions, the presence of unions at firm level has been 182

10 weakening in recent years. Why? Because their strategy has been focused on lobbying governments. 4 The unions had become increasingly aware of this problem. Consequently, they shifted away from building sociopolitical strength towards improving their own internal channels of communication and extending their presence in the business system. When unity of action became stronger in the early 1990s assisted by the collapse of support for the PCE, and the diminution of its influence over the CC.OO inter-confederal competition was replaced by intraorganizational concerns. But the unions still perceived themselves to be weaker vis-à-vis employers and the government than a decade earlier, the success of the 1988 general strike notwithstanding. As both the government and employers were aware, the deterioration of the economy, rising unemployment (at 21 per cent in 1997), and an increasingly large proportion of employees under fixed-term contracts had all reduced their capacity to attract and mobilize members. Economic and political conditions in 1993 favoured new attempts at pact negotiation: GDP fell by 1 per cent that year, caused by falling public investment, an adverse reaction of private investment to higher interest rates, and a substantial decline in private consumption as unemployment rose. This created something of an emergency and bolstered the perceived need for social partner cooperation. Employment conditions not only reduced the unions bargaining power but also made the abandonment of negotiations more costly for them. Given the prospect of worsening electoral performance, the government proposed a Solidarity Pact consisting of wage moderation and labour market reform. The unions declared for the first time in eight years that the inclusion of wage moderation in the agenda for negotiations would not in itself constitute an obstacle in the path towards a social pact. 5 The UGT s willingness to negotiate stemmed in part from a financial scandal, and it sought the support and financial assistance of the government to hold off bankruptcy. Unity of action with the UGT made the CC.OO more inclined to negotiate certain policies. And both unions reacted pragmatically to the government s call for a catch-all pact, although they called for a revision of its economic policy and for key issues to be negotiated at separate tables. But the government strongly defended its Plan de convergencia for economic adjustment to EMU, involving a tight fiscal policy, a reduction in inflation and interest rates, a control of nominal wage increases, and the deregulation and flexibilization of capital and labour markets. This stance, and the employer s refusal to help the troubled Socialist party, sunk the negotiations. 4 Diario de Sesiones del Senado, Comisión de Trabajo y Seguridad Social, no. 95, 19 April 1994, p European Industrial Relations Review and Report, 1993, 235:

11 Social Pacts in Europe This episode reveals how under perceptions of mutual weakness the successful negotiation of a social pact becomes more difficult, as predicted by our bargaining model. In 1993, the unions and the government were both negotiating from weak positions that made a pact more costly for all involved. The government was in a minority in parliament and under pressure from other parties to stick with austerity. The unions were dealing with internal problems and disagreements, and the employers had no intention of boosting the ailing Socialist government. All parties had negative expectations as to what they could obtain. Politics and perceptions overrode the problem load. The same occurred in 1994 when the government initiated further talks on a Social Pact for Employment. Asked about the government s strategy, Prime Minister Felipe González replied that it had the greatest interest in a pact but little scope for providing an acceptable quid pro quo. The fact that the Socialists could only govern with the tacit support of the centre-right Catalan CiU (Convergència i Unió) diminished the unions expectations as to adequate compensation. This perception, together with the stronger bargaining position of the employers, made failure even more likely. For the general secretary of the UGT, the government was more interested in reaching a political pact with other parties than in signing one with the unions and employers. 6 For the CC.OO, however, other factors were important, in particular, the bad relationship between the leaders of the PSOE and UGT 7 (Royo, 2001). The CEOE blamed the government s take-it or leave-it strategy 8 and the instability in union government relations created by an excessive use of strike tactics. 9 As the bargaining model suggests, had the government been in a position to compensate union involvement, these impediments might have been overcome. Some months after their exit from the talks, the unions agreed to return to negotiations on two issues: employment regulation and incomes policy. The unions rejected the government s plans for mandatory wage moderation, and both they and the employers rejected the proposed labour market reform. In response, the government implemented the changes (which allowed for a more flexible use of working hours and greater flexibility in company pay scales) unilaterally, provoking a general strike in January Looking back at their 1988 success, the unions expected to improve their bargaining position and force the government to reconsider. But the new strike had nothing like the same impact either on government policies, which sought to foster a greater use of part-time contracts, allow for private employment agencies, and decen- 6 Comisión de Economía y Hacienda, testimony of Nicolás Redondo (UGT), no. 57, 29 October 1993, p Interview with Salvador Bangueses, Madrid, 14 April Comisión de Economía y Hacienda, testimony of José María Cuevas (CEOE), no. 57, 29 October 1993, p Interview with CEOE leader Gabriela Uriarte, Madrid, 13 April

12 tralize collective bargaining, or on public opinion which favoured a negotiated solution to the crisis THE REVITALIZATION OF CONCERTATION IN 1997 By 1997, the political context had changed significantly: the centre-right PP led by José María Aznar won the 1996 elections and formed a coalition with three right-wing regionalist parties the Basque Nationalists, the Catalan CiU, and the Canary Coalition. But the economy was still sick and perilous for the new minority government: inflation and the fiscal deficit were improving, but unemployment had peaked at 24 per cent in 1994, and almost all new jobs created in 1995 and 1996 were on temporary contracts. The government was also highly apprehensive of the unions reaction to its policies the PSOE, in opposition for the first time since 1982, warned that social rights were now under attack and sought a dialogue with them and the employers to build a consensus on its proposed agenda. The unions and employers had shifted to a pro-pact position. Both believed that the change of government opened up new opportunities for concertation but at a lower cost than hitherto, and both were moving towards a common understanding of the country s growing predicament (if not the solution), as a low value-added, low-productivity economy, one over-reliant on cheap and insecure employment contracts and price-based competition. Pressure on the unions was especially strong. A majority of their members now supported dialogue and concertation, especially given the failure of the 1994 strike, as did their leaders whose reputation had been damaged by the rapid increase in unemployment and their experiment with wage militancy (Royo, 1996; Pérez, 2010). Not only had they been losing the battle for public opinion, but critically, support had haemorrhaged in workplace elections, their main source of strength. In the elections, both unions lost votes to independent candidates and company representatives, especially in large firms. And for the first time since 1982, the UGT took second place to the CC.OO, now led by moderates after the defeat of its more militant wing in 1991 (Hamann, 2001; Royo, 2006). The strength of the unions had been sapped by the segmentation of the labour market, and they had failed to recruit temporary workers in significant numbers: in 1994, the membership rate of permanent workers was 19 per cent against only 8 per cent for temporary workers (Polavieja, 2001; Llorente Sanchez, 2007). Concertation to restore political influence now seemed much more attractive than when they rejected it in the mid-1980s, precisely to prevent a loss of worker support. A month after his election, Prime Minister Aznar held a meeting with the unions. As they remained opposed to catch-all pacts spanning numerous policy areas, the social partners and the government agreed to seven separate negotiating tables. Even though the government respected the social partners 185

13 Social Pacts in Europe bargaining autonomy, it threatened on several occasions to act unilaterally if they failed to reach an agreement. The first step was the negotiation of social security reforms, as outlined in the 1996 Toledo Pact among the political parties, 10 and an agreement on the Consolidation and Rationalization of the Social Security System was signed in October. The government then launched talks on labour market reform. An agreement was signed in 1997, consisting of two collective bargaining reforms the AINC (Acuerdo Interconfederal sobre Negotiation Colectiva) and the AICV (Acuerdo Interconfederal de Cobertura de Vacios) and a reform of the labour market, the AIEE (Acuerdo Interconfederal para la Estabilidad del Empleo). 11 A quid pro quo secured the pact. In the labour market reform, the unions presenting themselves for the first time as defenders of outsider as well as insider workers obtained the introduction of new permanent employment promotion contracts to help counter the expansion of fixed-term work, with lower dismissal costs than for other such contracts. Employers received a reduction in social security charges as their part of the deal. In the collective bargaining agreement, the unions achieved a degree of re-centralization in collective bargaining. The AINC was especially important in this regard, in aiming to reduce the number of collective agreements and prioritize national industriallevel agreements. Although bipartite deals, endorsed by the government, and then passed into law, the 1997 April Agreements have nonetheless had the significance and effect of a fully fledged social pact. 12 Not only did they create a new paradigm for peak negotiations and social dialogue going forward, but they also established the quality of employment as a priority, and sought to enhance the articulation of collective bargaining with concertation. How do we explain the successful return to pacting? The problem load, combined with critical organizational and political factors, provides the answer. The economy was beginning to improve, but still very high unemployment (at 21.5 per cent in 1997) and a massive expansion of atypical contracts helped 10 The Toledo Pact involved a commitment by all political parties to maintain and reform the public pension system on a consensual basis. In 1996, the trade unions gave it their support, thereby ensuring that in principle pension reform would be subject to social dialogue. The consensual spirit of the pact was reinforced by the creation of a monitoring commission to secure the input of social and economic actors. 11 The AINC clarified the roles of different levels of collective bargaining and expanded the issues to be dealt with through collective bargaining, which enhanced the influence of the social partners in the policy debate on labour market reforms; the AICV aimed to extend the coverage of collective bargaining to those sectors lacking collective agreements; and the AIEE introduced incentives for employers to transform temporary contracts into indefinite ones, as well as financial incentives to hire on permanent basis. The reform also removed some of the protections for employees with indefinite contracts. 12 See Diario de Sesiones del Congreso de los Diputados, Comisión de Política Social y Empleo, no , May 1997, for parliamentary testimony by union and employer leaders regarding the success of the 1997 Agreements. 186

14 induce all three actors to sign. In terms of the bargaining model, they all saw themselves as relatively weak, but all were under pressure to reach agreement quickly. This was especially true for the unions, given declining workplace support and accusations of defending permanent workers while neglecting the now very large temporary workforce, comprised mainly of younger people and women (Richards and Polavieja, 1997). 13 The employers had also come to acknowledge the problems created by the very high level of fixed-term work: in the view of manufacturing companies, it had begun to damage productivity while failing to weaken the bargaining power of permanent employees 14 (Serrano et al., 1998; Royo, 2006). As for the government, its weak parliamentary position made it the perfect partner in a positive-sum pact SOCIAL PACTING IN THE 2000S The Shift to Bipartite Pacting, Developments in the 2000s reveal the vitality and vulnerability of Spanish concertation as well as the strengths and weaknesses of its institutionalization. When the PP was re-elected with an absolute majority in 2000, it agreed again to dialogue with the social partners over the four years of its mandate. At the top of its agenda for reform were the labour market, collective bargaining, pensions, and unemployment benefits. The government preferred to respect the autonomy of unions and employers, and it asked them to follow the pattern of the 1997 agreements and negotiate a pact as the basis for legislation. It also had a strong interest in securing agreements to legitimize its policies. But given its new parliamentary majority, the Aznar government could afford to play a double game sponsoring agreements when the social partners could achieve them, but making policy unilaterally, and largely ignoring them, when they could not. The unions and employers had quite different objectives, even if their diagnosis of the country s economic ills was similar. The unions Common Proposal for Social Dialogue focused on job creation and a reduction in the use of temporary contracts. The 1997 agreements had delivered some positive results 13 The segmentation of the Spanish labour market, which had commenced in the mid- 1980s, had become fully entrenched by 1997: that year, 34 per cent of the active labour force were employed on temporary contracts, which also accounted for 85 per cent of flows into and out of employment (the annual transition between the two segments was only 1 per cent). Moreover, the average tenure of a temporary worker was twelve months, compared with twelve years for workers on permanent contracts, and 34 per cent of temporary workers were unemployed, compared to only 6 per cent of permanent ones. All evidence suggests that temporary employment in Spain is predominantly involuntary and non-transitional (Amuedo-Dorantes, 2000; Polavieja, 2003). 14 Between 1992 and 1995, years of deep recession, temporary workers lost 25 per cent of their purchasing power, while the earnings of permanent workers increased by 4 per cent in real terms (Ferreiro and Serrano, 2001; Ferreiro and Gomez, 2005). 187

15 Social Pacts in Europe (mainly due to government incentives) in increasing the percentage of permanent contracts. But while unemployment fell to around 13 per cent in 2000 from 21 per cent three years earlier, the proportion of temporary contracts remained unchanged at around one-third. The employers Approaches to the New Stage of Social Dialogue prioritized the reform of collective bargaining, a reduction in dismissals costs, greater flexibility for part-time contracts, and a reduction in employers social charges. Common to both proposals was a preference for more flexible forms of social dialogue and concertation over allencompassing social pacts. In fact, talks at separate tables had already been institutionalized, and consisted of three negotiating commissions : on the social protection of the unemployed, on increasing secure employment, and on part-time employment. In the late 1990s, negotiations in the first two were deadlocked, with employers and the government seeking more resources for active employment policy against union calls for higher unemployment benefit and job promotion via working-hours reduction. This discord did not prevent success, however, in the third commission where an agreement on improving the social security entitlements of part-time workers was reached (the employers abstained) in November 1998 (EIRO, 1998a, 1998b). This more flexible strategy produced further results, regardless of a surge in industrial conflict in , and the government s predilection for breaking with concertation on many critical reforms. In March 2001, following union employer failure to reach agreement on replacing the expiring 1997 agreements, the government unilaterally reformed the labour market, extending the categories of worker who could be hired on open-ended contracts with employer social security reductions, and providing limited compensation for the dismissal of temporary workers. In May 2002, the government also unilaterally reformed unemployment benefits, removing certain categories of workers for example, those on seasonal work contracts from entitlement, and abolishing the interim wages employers had to pay workers waiting for rulings on unfair dismissals claims. The breakdown in social dialogue and union opposition to these changes led to a general strike in June 2002, which, alongside a precipitous decline in government opinion poll support, forced the resignation of the labour minister and a repeal of most of the benefit reform package. Surprisingly, none of this prevented a pensions agreement in April 2001, which was supported by the CC.OO but not the UGT, 15 or government union agreement in November 2002 on the modernization and improvement of the 15 The reform allowed workers to take early retirement if they had not paid contributions before 1967, created a reserve fund, improved widows pensions, allowed flexible work and retirement beyond 65, and ensured the ongoing participation of unions in the management of the social security system. 188

16 public administration. 16 Indeed, the UGT claimed that the 2002 strike had strengthened the unions and forced the government to concede more than it wanted to secure their support. What we see, then, across this period is a process in which both unions and employers engaged in concertation to preserve an influence over social and labour market policy (using this strategy, as in the past, to compensate for associational weaknesses), while the government used the social dialogue instrumentally, exerting a strong shadow of hierarchy via the threat (and reality) of unilateral policymaking. A government with a strong majority could afford to embrace a take-it or leave-it strategy, more often than not to the disadvantage of the unions, for whom an abandonment of social dialogue promised marginalization from the policy process, with no guarantee that militancy and social movement opposition would deliver better results. Perhaps the most important innovation in this period was the return in late 2001 to an incomes policy agreement (the first since 1984) on a union employer bipartite basis, but once again under the shadow of state hierarchy. It would be renewed every year until 2009, when it broke down amidst the financial and economic crisis, but was reprieved in early 2010 a development made all the more remarkable by open conflict at the time between the unions and the government over its anti-crisis measures. 17 The pattern for the rest of the decade was set in December 2001, when the unions and employers concluded an Acuerdo para la Negociación Colectiva, or ANC, for The ANC established guidelines and set out criteria for lower level bargaining, linked pay rises to inflation and productivity gains, and also included a general commitment to employment stability and quality. It even recommended the creation of national sectoral observatories to analyse economic and employment trends (EIRO, 2002), and a monitoring commission was created in 2003 to promote their implementation. Subsequent ANCs have often been beset by stormy negotiations, as well as by opposition from critics in both employer and union camps. But the system has now endured for nearly a decade, making it the most persistent of reiterated wages pacts in the EU, apart from Ireland s incomes policy (see Chapter 5). It is worth considering briefly the reasons for the institutional success of this implicit incomes policy pact in terms of the bargaining model, prior to an extended discussion of its institutionalization in the next section. 16 This reform sought to rationalize the public workforce and increase its efficiency in return for above-inflation pay increases, a 35-hour week, and measures to improve employment stability. 17 Pérez (2000) presents evidence that the unions began unilaterally to restrain their wage demands from the mid-1990s onwards, once they had prioritized lowering unemployment and reducing the number of non-standard contracts. According to Ferreiro and Gomez (2005), the UGT and the CC.OO signed joint documents laying out joint wage-setting criteria every year between 1997 and

17 Social Pacts in Europe A review of almost a decade of ANC agreements reveals the following features. The central concern has been wage moderation on the part of both employers and unions, driven in part by their shared view that containing real unit labour costs will promote employment (and a diminution of temporary employment) and enhance the competitiveness of Spanish firms. The results have been decidedly mixed. Job creation proceeded apace in the 2000s on the back of a booming economy (only to collapse precipitously in the recent crisis) and while the expansion of temporary contract workers in the private sector fell significantly, they began to grow in the public sector. Nominal wage growth has been contained and real unit labour costs have fallen although wage moderation only accounts for part of this trend; immigration and an ongoing expansion of low-cost labour have also contributed as has declining purchasing power (wage drift has been negative for most of the period) in a tighter labour market. But productivity growth has been low and relative unit labour costs have risen well above the Euro average: wage moderation alone cannot cure Spain s multifactor productivity problem. If we think of the ANCs as an implicit incomes policy pact instead of a series of annual ad hoc wage agreements (for they have depoliticized pacting and created a stable system for trade-offs and exchange between employers and unions), then we can more readily understand the power balance underpinning this structure. The unions entered the ANC process from a rather weak position given the failure of either pacting in the 1980s or militancy in the 1990s to counter an erosion of workers rights and security, or provide them with greater influence over collective bargaining and the workforce. By the mid-1990s, even the power derived from workplace elections was under threat. Their political influence was restored to some extent by the 1997 Agreements, and their adoption of job promotion as a primary goal led them to internalize wage moderation even before the ANC formally committed them to it. At the same time, both Socialist and centre-right governments had sought to flexibilize the labour market in response to the employment crisis and persistent demands from employers, and experience had proven that the unions only real means of slowing or blocking that process was concertation. Although a second-best option for the unions especially after 2000 when a growing economy, much lower unemployment, and rapid job creation might have backed a more militant stance concertation on wages and work-related issues under the ANCs have allowed for a mutually reinforcing relationship between the exercise of political influence and a strengthening of the unions organizational powers. The Problems of Pacting under the New Socialist Government The minority government led by Prime Minister José Luis Rodriguez Zapatero elected in March 2004 and supported in parliament by two small left-wing parties the Republican Left of Catalonia and United Left revealed a strong 190