Policy Concertation in Europe: Understanding Government Choice Lucio Baccaro Institute for Work and Employment Research Massachusetts Institute of Technology baccaro@mit.edu Marco Simoni European Institute London School of Economics and Political Science m.simoni@lse.ac.uk Forthcoming in Comparative Political Studies, October 2008 Keywords: concertation, corporatism, trade unions, European politics, industrial relations Policy Concertation in Europe: Understanding Government Choice Abstract This paper focuses on the European governments’ decision to involve unions and employers in the design and implementation of public policy. Based on new measures of the phenomenon, it argues that between 1974 and 2004 no convergence on a pluralist model of policy formation is visible. The paper then uses these measures to identify and analyze the clearest cases of adoption or demise of concertation, namely the contrasting responses of the British and Irish governments to wage policy, and of the Austrian and Italian governments to pension reform. It argues that governments are willing to share their policy-making prerogatives when they are politically weak, and when unions, while still representing a credible threat to policy implementation, have been declining in the recent past. Additionally, a combination of partisanship and policy learning reinforces the push for change. 1 Policy Concertation in Europe: Understanding Government Choice The literature on corporatism and policy concertation has shown a tendency to conflate a number of questions that it would be best to keep separate: (1) Why are governments willing (or unwilling in some cases) to share their policy making prerogatives with trade unions and employer associations, not just informally, by incorporating their inputs, but, formally, by setting up a bargaining table and engaging in negotiations with them over the content of public policy? (2) In what conditions, contingent on government’s decision to in- (or de-)volve, does a negotiated agreement emerge? (3) In what circumstances does negotiated regulation cease to reflect the contingencies and power balance of the time in which the deal is struck and becomes a durable way of processing public policy? (4) In what respects, if any, is policy concertation more or less effective than alternative policy-making processes? 1 Despite its massive size, most of the literature has focused so far on questions two and four, while it has been much less vocal on the other two. Possibly due to an implicit assumption of corporatist convergence (see Streeck, 2004), much of the neo-corporatist literature of the 1970s and 1980s did not systematically distinguish between the reasons motivating governments and those motivating unions and employers. Policy concertation was assumed to be a functional necessity of advanced societies (Schmitter, 1974; Lehmbruch, 1979), such that all governments would, sooner or later, want to engage in it. From this vantage point, the interesting questions were not those about the conditions in which certain governments (and not others) would develop a demand for policy concertation, but those about the conditions in which policy concertation would be supplied by the interest group system (Schmitter and Lehmbruch, 1979; Lehmbruch and Schmitter, 1982; Berger, 1981; Goldthorpe, 1984; Molina and Rhodes 2002). Another strand of research (by now probably larger than the former) dealt with the 2 macroeconomic consequences of policy concertation, especially centralized or coordinated collective bargaining (for recent analyses, see Garrett, 1998; Iversen, 1999; Traxler et al., 2001; Kenworthy, 2003; Traxler, 2003; Mares, 2006). In this paper we are interested in question number one, which we regard as analytically prior to the others. Governments, accountable to national parliaments (or directly to the electorate), are the sole institutions with a clear mandate to take binding decisions. The constitutional standing and democratic legitimacy of mixed systems of policy-making granting private actors access to the public policy sphere have always been considered dubious (see Lowi, 1979; Habermas, 1989[1964]; 1996), including by theorists of neo-corporatism itself (see Schmitter, 1983, for an example). It needs to be explained why democratically elected governments are, in some circumstances but not in others, willing to share their policy-making prerogatives with the “social partners” rather than using them at full and proceeding unilaterally. To address this question, this paper begins by measuring the phenomenon of interest. We present two indicators of government willingness to engage in policy concertation in Europe, one pertaining to wage policy and the other to welfare policy, covering 15 European countries between 1974 and 2003. These indicators show that the propensity of governments to engage in negotiations with the social partners in at least one of the two policy areas has remained virtually stable over time, while the proportion of governments that are willing to bundle the two issues and deal with them simultaneously has grown. We then utilize these indicators as criteria for case selection in a small-n design aimed at contributing to an inductive theory of government choice. By selecting cases in which government’s willingness increased and decreased the most, according to our measures, while keeping background factors as much as possible constant (Mill, 1946: book 3, ch. 8; King et al., 3 1994), we identify four exploratory case studies: (1) the shift from decentralized wage bargaining to social partnership in Ireland; (2) the transition from the social contract to Thatcherism in Britain; (3) the move from policy inertia to negotiated pension reforms in Italy; and (4) the recent unilateral restructuring of public pensions in Austria. The stylized analysis of these cases identifies a particular configuration of factors, rather than single elements (Ragin 1987). Based on the case study evidence, we argue that the decision to involve private actors in policy bargains can be conceived of as the striking of an alliance between weak governments and weakened unions. In other words, governments are willing to share their policy-making prerogatives with unions when they are politically weak, and when the unions have seen their organizational fortunes decline in recent years, while still maintaining a capacity for oppositional mobilization. The Trajectory of Policy Concertation between 1974 and 2003 Both the early literature on corporatism/concertation and the more recent literature on social pacts often lack a measure of the object of analysis. To start obviating to this, we have designed an indicator of “government willingness to engage in concertation”, which captures government’s decision to involve formally the “social partners”, i.e. capital and labor, in policy making. The indicator does not capture whether or not there is a concertation agreement, but focuses on the initial intention of the government. 2 This indicator was created for two policy areas: wage policies (including income policies and centralized wage bargaining) and welfare policies (that is, spending policies aimed at providing social insurance of various kinds) for 15 European countries between 1974 and 2003. 3 The measures were based on information reported monthly in the European Industrial Relations 4 Review. By considering the whole period from the first oil shock to the present day, we do not assume that the conditions leading to policy concertation have changed over time (as it has been argued for the transition from “old” to “new” social pacts, see Negrelli, 2000; Hassel, 2003), but leave the issue to be determined empirically. For each country, year, and policy area, the government willingness index takes a value of 1 when: (a) there is textual evidence that, in the course of the year, the government publicly invites the social partners to negotiate a national agreement to set wage increases, or to design social security policies/welfare institutions; or that (b) the social partners are invited to design or implement autonomously, i.e. via bipartite agreements, any of the aforementioned policies. 4 If, instead, there is textual evidence that a government designs and/or implements policies without the formal involvement of social partners, the willingness score is 0. It must be emphasized that a government could rely on the implicit consensus of one or more of the social partners when designing policy, and still receive a 0 score, since our construct of interest, policy concertation, refers to formal involvement rather than informal influence. Graphs in Figure 1 plot the trajectory of government willingness over time. Figure 1(a) on wage policies shows a pronounced cyclical component, thus providing some support for the thesis articulated by Schmitter and Grote (1997) of a cyclical pattern of policy concertation over time. There appears to be a very mild downwards trend, however largely overshadowed by very pronounced cycles, which emerges clearly by comparing the values of this indicator in peaks and troughs. In 1977, 83 percent of European governments in our sample were willing to pursue concerted wage policies. 5 Only in France and Portugal, according to our measure, the governments were unwilling to do so. 6 This percentage dropped to 47 percent in 1985 (and 1986) when concerted wage policies were attempted only in Finland, 5 Greece, Sweden, Spain, Denmark, Austria, and Italy. A second trough is recorded in 1995, when only 40 percent of European countries pursued income policies or centralized wage negotiations, including Austria, Finland, Ireland, Norway, Greece, and Italy. Our data end in 2003 with another peak: 73 percent of European governments were willing to have some form of negotiated income policy, namely all except France, Denmark, Britain, and Italy. (Figure 1 about here) The trend in Figure 1(b) on welfare policy is markedly different. The graph shows a clear growing trend since 1974, peaking in 2000 when, according to our information, all governments, except in Austria and the United Kingdom, were willing to adopt a participatory approach. After 2000 the share of governments declined slightly. It was 60 percent in 2003: in that year, the governments of Austria, Britain, Greece, Italy, Denmark, and Portugal took a unilateral approach to welfare policy. The welfare policy graph shows a cyclical component which is much milder than in the wage policy graph. Some authors have proposed what we call a bundling hypothesis, namely that changes in international macroeconomic conditions in the 1990s have pushed governments to seek coordination in multiple areas simultaneously as opposed to single areas, as a way to increase national competitiveness (Ebbinghaus and Hassel, 2000; Rhodes, 2001). Figure 2 plots the number of governments that are willing to engage in concertation in at least one policy area (black line), and the number of governments that are willing to engage concertation in both policy areas at the same time (bars). The first curve is cyclical around a rather stable mean, suggesting no long-term growth or decline in government willingness to involve. On average, slightly more than 12 European countries per year (out of 15) have attempted some form of policy concertation. The second curve is instead growing over time at a decreasing rate. This 6 confirms the bundling hypothesis, even though, contrary to expectations, most of the growth in bundling seems to have taken place in the 1970s and 1980s, rather than in the 1990s. Indeed, in the 1990s, the tendency of governments to approach both wage and welfare issues through policy concertation seems to have reached a plateau. (Figure 2 about here) Linked to the bundling argument, scholars have also suggested that specific external pressures, coming from the run up to and the establishment of EMU, have increased the propensity of governments to seek explicit policy cooperation with the major interest groups, especially for those countries in which the established thresholds for qualification were more difficult to reach (Schmitter and Grote 1997; Crouch 2000; Rhodes 2001; Hassel 2003; Hancké and Rhodes 2004). To address this argument, Figure 3 distinguishes between EMU countries and non-EMU countries in the 1985-2003 period. On average, the former are more willing to engage in income policies than the latter throughout, but there is no appreciable growth trend in the 1990s compared with the beginning of the period. The picture changes slightly when one considers welfare policies. In the Euro group there is a fairly clear increase in willingness as the Euro approaches, and a decrease after qualification. However, this graph must be read against the evidence presented in Figure 1(b), which shows a growing trend of willingness in European countries in general. Against this backdrop, the EMU seems to have reinforced a tendency that was already present. 7 (Figure 3 about here) Overall, our measures suggest that government willingness to involve labor and capital in policy-making is not simply confined to the decades of the ‘Keynesian consensus’ (Goldthorpe 7 1984; Hall 1989), but is a rather stable feature of European politics in the last thirty years. European governments are not moving towards greater unilateralism in policy-making, to say the least. In this respect, theories predicting convergence towards a pluralist system of interest intermediation, and the demise of concertative/corporatist models of policy-making, do not find confirmation in our data (Streeck and Schmitter, 1991; Streeck, 1993). 8 The evidence illustrated above has nothing say, however, about the specific factors or combination of factors explaining government choice of cooperation vs. unilateralism. To explore these, the paper now moves to case study analysis. Case Selection We averaged our government willingness scores over five-year periods for each country (1974 to 1978; 1979 to 1983; 1984 to 1988; 1989 to 1993; 1994 to 1998; 1999 to 2003) and calculated the differences between the periods. By doing so, we created a quantitative measure of change in government willingness to engage in policy concertation and used it to identify the countries that had undergone the largest policy changes in either direction. We then selected one country at or near the top and one at or near bottom of the list. The specific choice attempted to match (and, hence, control for) background conditions in the two countries. In other words, we used an approximated “method of difference” research design (Mill, 1946: book 3, ch. 8), which allowed for maximum variation on the dependent variable, while controlling for as many background conditions as possible. The goal of the empirical analysis, based for the most part on secondary sources, was to identify the differences that could explain the stark divergence in outcomes. For wage policy, we decided to contrast Britain (1979-83/1974-78) – namely the transition from the “Social Contract” to Thatcherism – and Ireland (1989-93/1984-88) – 8 approximately, the emergence of social partnership in this country. For Britain, the choice was straightforward since this was the case which seemed to have undergone the greatest decline in government willingness to negotiate. Ireland, instead, had the second greatest increase in government willingness score after Germany 1999-2003/1994-98 (the Social Democratic-Green Party coalition’s unsuccessful attempt at producing a social pact), but was a good match to Britain in other respects. Indeed, the Irish economy is considered very similar to the British from various viewpoints, and especially from the point of view of the industrial relations system (see Murphy and Roche, 1997; Gunnigle et al., 1999). For welfare policy, we chose Austria (1999-2003/ 1994-98) as our negative case, corresponding to the Austrian government’s decision to abandon the traditional partnership approach and engage in unilateral pension restructuring, and Italy (1994-98/ 1989-93), approximately the transition from policy inertia to negotiated pension reforms, as our positive case. While Austria was at the bottom of our list, exhibiting the highest decrease in government willingness to engage in concertation over social policy issues, Italy was one of three countries at the top, together with Greece (1994-98/ 1989-93) and Norway (1989-93/ 1983-88). We chose Italy because in Italy (unlike other countries) the bone of contention was exactly the same as in Austria, namely pension reform, which the two countries tackled in dramatically different fashions. We provide stylized reconstructions of the four case studies. Ireland: from Free-For-All Collective Bargaining to Social Partnership During most of the 1980s, the Irish government was unwilling to regard collective actors, and especially trade unions, as more than “lobbying interests” (Hardiman, 1988: 225). In this period collective bargaining was decentralized and private sector negotiations took place at the 9 enterprise level. The government sought to influence wage settlements indirectly by taking a tougher bargaining stance towards public sector bargaining renewals than in the past. This approach was reversed in 1987, when the Irish government officially involved unions and employers in an ambitious and wide-ranging reform plan known as “Programme for National Recovery” (PNR). The PNR, a three-year agreement, committed the actors to seek stabilization of the exchange rate, reduction of public spending, and tax reform. By signing the agreement, the Irish Congress of Trade Unions (ICTU) accepted the principles of social peace and wage moderation, to be achieved, inter alia, through a ban on local bargaining. In exchange, government agreed to reform the tax system to increase take-home pay. The PNR was the first in a series of six successive three-year social partnership agreements, which last to date. In all these years, the basic structure of the agreements has remained more or less the same: against the backdrop of fiscal conservatism and monetary orthodoxy, these agreements have exchanged wage restraint for tax cuts, while targeting social inequalities and exclusion, first by keeping constant the real value of transfers, and then through more substantive measures (see Omitted). Three elements seem to have determined the adoption of social partnership: the electoral weakness of the government, the organizational weakening of unions, and the high problem load Ireland was facing at the time of change. First, the government that initiated social partnership was a one-party minority government by the Fianna Fail party, which held 48.8 percent of seats in the Irish lower chamber (Dáil). Its predecessor, a Fine Gael-Labor coalition, could instead count on 54.8 percent of the seats. The weakness of the Fianna Fail government was compounded by the party’s own interclass nature, which made it difficult for the leadership to pass policy decisions that penalized the party’s labor constituency (Hardiman 1988: 200-204). 10 Government weakness paired in Ireland with weakened unions, both in terms of membership and capacity to bring out workers on strike. Between 1983 and 1987, union membership declined from 57 to 50 percent. 9 However, they still maintained a considerable capacity to make life very difficult for government in case they decided to do so (Hardiman 1988: 215), which explains why government had an interest in involving them. These factors combined with an exceptional load of problems: all major actors in Ireland were aware that, with no growth, high unemployment, and public finance imbalances, the economy was out of control, and that something drastic needed to be done to bring it back on track. This is brought out clearly by the 1986 report produced by the National Economic and Social Council (NESC), a tripartite institution, which outlined clearly the gravity of the macroeconomic situation. Interestingly, the report was subscribed to unanimously by both government and the social partners, including the unions (Hardiman 1992; NESC 1986). Britain: The Shift to Neo-Liberal Orthodoxy A shift in the opposite direction – from concertation to unilateralism in wage policy – can be found in Britain. In 1973, when it was still in opposition, the Labour party signed a “social contract” with the Trade Unions Congress (TUC). The contract signaled that the party was ready to adopt a policy of partnership once in office, and this is exactly what it did. Since 1974, a number of labor-friendly measures – including price controls, lower income taxes, rent-controls, and various subsidies – were introduced, in exchange for the unions’ commitment to moderate wage claims (Pelling 1987: 291; Peden 1991: 203-204). However, the social contract did not deliver the expected economic results. Wage drift was so high in this period, that de facto average wage increases were twice or three time as high 11 as the agreed levels (Peden 1991: 204). Thus inflation was not brought down, and unemployment started to rise. Wildcat strikes dominated the industrial relations scene, culminating in the “winter of discontent” wave of protests between 1978 and 1979 (Scharpf 1991: 83-88; Mares 2006: 202). In retrospect, it is not especially surprising, given circumstances, that the Labour party lost the next elections (despite an attempt at rekindling the partnership with the unions through a new “social contract”) and Margaret Thatcher became the next PM in May 1979. The touchstone of the Tory Party’s industrial relations policy was the curtailment of trade union power (Letwin 1992: 130-158). To this purpose, various pieces of legislation were passed including the 1980, 1982, and 1988 Employment Acts, and the Trade Union Act in 1984. Instead of relying on unions’ voluntary restraint, the government centered its anti-inflation strategy on a monetarist approach (Peden 1991: 214-215), which it continued to pursue even when unemployment began to rise (Letwin 1992: 118-123). In contrast to the Irish case, the Thatcher government that terminated concertation in Britain was an exceptionally strong one, thanks to the largest electoral swing since WWII (Peden 1991: 214). Additionally, during its tenure, it faced an extremely fragmented and weakened opposition (Thorpe 2001: 190-193). British unions were at the peak of their organizational strength at the time of the policy shift. They were so strong that they were able to deadlock any negotiation with the government until all their requests were secured. Indeed, the excessive power of the unions featured prominently in the Conservative program and was one of the main arguments put forward by the Tories to support their requests for a demise of peak-level concertation as well as unions’ reform (Dale 2000: 268; Powell 1992: 138). However, in so doing, they reversed a long-standing tradition that considered policy concertation a crucial tool to 12 keep society stable and united (Letwin 1992: 130-132 Thomsen 1996: 96-98). The Thatcher government was convinced that the “country [was] faced with its most serious problems since the Second World War” (Dale 2000: 266), and that the unions were part of them. Austria: Unilateral Welfare Restructuring in a Quintessentially Corporatist Country The case of Austria is a striking example of government’s disengaging from policy concertation on welfare policies. From the end of WWII to 1999, Austria had been a model of corporatist governance. All policy issues, including welfare, were discussed with organized interests, including the trade unions confederation (ÖGB), the Chamber of Labor (BAK), the Economic Chamber (WKÖ), and the Chamber of Agriculture (PKLWK). In spite of a few divisive issues which had emerged in the late 1990s, the “forum for interest representation and [the] platform for consensus bargaining between the large societal groups ha[d] remained remarkably stable” (Tàlos and Kittel 2002: 48). In Austria, “virtually no decision of consequence [wa]s made that ha[d] not been intensively discussed by the social partners.” (Tomandl and Fuerboeck 1986: 12). Throughout the 1990s a number of concerted policies on issues of social security and welfare reform were agreed by the government and the social partners, including unemployment benefits (1994), 10 plans to cut public expenditure (1995), and a relatively minor pension reform (1997). 11 The situation became more difficult, however, in early 2000, when a new grand coalition government involving the SPÖ (the Social Democratic party) and the ÖVP (the People’s Party), tabled a proposal to reform the pension system by focusing on expenditure cuts. This proposal met with outright opposition from the ÖGB, which led to intense disputes between the union and the two coalition parties. As a consequence of this rift, the ÖVP pulled out of the 13 alliance with the SPÖ and formed a cabinet with the populist Freedom Party (FPÖ) (Fallend 2001: 242-246). Faced with growing deficit and increasing pension expenditures, the program of the new government contained a pension reform that was very similar to the one that had failed to be passed by the previous cabinet led by the SPÖ. Again, the ÖGB staunchly opposed the reform. This time, however, it was approved against the unions’ will in July 2000. 12 The demise of policy concertation was not just limited to pensions, but included other areas and institutions of the welfare state. In 2001, the governance of the Association of Social Security Providers (HSV), the central institution of the social insurance system, was altered in an unfavorable way for unions. For the first time in its history, the ÖGB called all its members to mobilize against the law. 13 In December 2002, the government presented another far-reaching pension reform, which included an increase in retirement age and benefit cuts. Again, the ÖGB refused to accept the new reform proposals and called for industrial action against them. 14 Strikes and demonstrations involved 1 million workers across the country. Mobilizations notwithstanding, the new reform was approved in July 2003. 15 Given the specificity of Austrian system of corporatist governance, whereby the social partners are institutionally embedded in policy-making and policy-implementation in virtually all areas, the demise of concertation in 2000 is particularly striking. The relative strength of the government seems again to play a role in this shift: in this case it pertains specifically to the issue on which concertation broke up. In the 1995-1999 legislature, over a fourth of SPÖ MPs were trade unions officials, while very few union officials were members of the FPÖ or ÖVP political groups. It was remarkably difficult for the grand coalition government to overcome the unions’ veto power on a key issue like pensions. The new coalition was instead free from direct union 14 influence and was much stronger vis-à-vis the major opponents, the trade unions, in this particular respect (Schludi 2003: 180). The Freedom Party – for the first time holding cabinet posts – had built its electoral fortunes, among other things, on a populistic critique of ‘union privileges’ in the corporatist system (Viebrok, 2003) and, unlike the socialist and the Christian democrats, did not have any special relationship with the interest representation bodies (Schludi 2003: 191-193; Tàlos and Kittel 2002: 39). Additionally, union density had been declining constantly in Austria from 60 percent in 1973 to 36 percent in 2000 – one of the biggest declines in advanced countries (OECD 2001; Golden, Lange and Wallerstein, 2002). The major problem, however, was not membership per se, which remained non-negligible, as mobilization capacity and strategic isolation. Unlike in Italy (and France), where union mobilizations were supported by the general public, in Austria citizens largely ignored the unions’ protests against pensions reform and the other unilateral welfare changes (Schludi 2003: 180, 191). Italy: From Inertia to Negotiated Pension Reform The Italian experience with pension reform provides a neat contrast to the Austrian, and presents a number of similarities with Ireland, the other case of increased governmental willingness to engage in concertation. In the late 1990s, Austria and Italy were the countries with the highest proportion of pension expenditures on GDP (see Schludi, 2001: Table 1, p. 7). Yet, unlike the Austrian, the Italian governments, for the most part, pursued a policy of negotiation with the trade unions during the 1990s, as opposed to unilateral reform. During the 1980s, the financial situation of the Italian pension system became increasingly difficult. Due to declining fertility rates and growing life expectancy, fewer young 15 people were being asked to support growing cohorts of retired. These phenomena were common to all advanced countries, but were particularly severe in Italy (Ferrera, 1996). In 1990 and 1991, two unsuccessful attempts were made by coalition governments, where Socialists occupied key cabinet posts, to reform the public pension system (Regonini 1996; Ferrera and Gualmini, 1999: Table 3.3., p. 115). The first comprehensive effort to tackle the pension problem was made in 1992, when the government, lead by a prominent socialist politician, faced with a difficult political and economic crisis, passed a drastic pension reform plan. This was not formally negotiated with the unions, but was tolerated by them because it took into account the most important union demand, name the safeguard of “seniority pensions” (Cazzola, 1995: 55). A second unilateral attempt at pension reform was made in 1994 by the center-right government of Silvio Berlusconi, which devised a far-reaching reform (Mascini, 2000: 16-77). In response, the major labor confederations organized the most impressive demonstration of union strength since the Hot Autumn wave of strikes in the late 1960s. Due, inter alia, to the unions’ massive mobilizations, the parliamentary coalition supporting the Berlusconi government began to fray and government was forced to scrap the reform plan. In 1995, a technocratic government, which lacked a clear parliamentary majority but was supported by the previous center-left opposition, engaged in negotiations over pension reform with the unions and the main employers’ association. While the latter withdrew early on, arguing that the proposed reform was not incisive enough, the unions and the government carried on and eventually agreed on a reformed pension system that contained several important innovations, including a “defined contribution” method for pension determination (in lieu of the previous “defined benefit” system) and the gradual phasing out of seniority pensions. A further negotiated reform to shorten the transition period to the new regime was passed in 1997. These 16 measures were key for Italy’s ability to qualify for the second phase of the European Monetary Union. The Cases Revisited The analysis of the Irish, British, Austrian, and Italian cases leads to interesting conclusions as regards both the particular combinations of factors that seem to explain both adoption and demise of policy concertation by governments, and those that, surprisingly, seem to hold no holding explanatory power at all. We begin with the latter. a) Plausible Hypotheses Ruled Out The four cases illustrated above suggest that the preferences of employer organisations and the macroeconomic regime (Keynesian vs. Monetarist) in which the government operates play little or no role in shaping governments’ approach to concertation vs. unilateralism. This is surprising because recent interpretations have instead contended that employer preferences and the macroeconomic regime are key to determining the interaction between state and organised interest groups. A simple version of the employer preference argument would state that the choice of governments reflects the preferences of organized employers. This hypothesis has not been explicitly formulated in exactly these terms, possibly due to its similarity to a coarse MarxistLeninist view of the state (Marx and Engels, 1978: 475). 16 However, it resonates with the recent emphasis in comparative political economy on the strategic choices of organized employers as the main determinants of policy outcomes (Hall and Soskice, 2001; Thelen, 2001, Swenson, 2002; Mares, 2003; Culpepper, 2007). 17 This hypothesis finds no corroboration in our case studies. In Britain, the monetarist shift implemented by the Thatcher government met with the opposition of the Confederation of British Industry, for it entailed a steep rise in interest rates. During the fall of 1980, “British industrialists had begun a sustained campaign against the government’s monetary policy” (Walsh 2000: 497). In Ireland, at the time of policy change, the Irish organized employers were relatively happy with the decentralization of collective bargaining they had been so keen to promote (Hardiman, 1988: 200, 221, 236). In December 1986 and then again in June 1987, the General Council of the major employer association “asserted that negotiations with the trade unions on pay and related matters should continue to take place at local level” (Hardiman, 1988: 236; see also Hardiman, 1992: 350). Finally, in Italy, not only did the employers withdraw from negotiations over pension reform in 1995, they also actively lobbied Berlusconi (who seemed uncertain) to take a draconian approach to reform in 1994 (Mascini, 2000: 181). 17 The second argument our case studies rule out states that the government’s interest in policy concertation on wage policy is a function of its macroeconomic stance. It has been argued that, in a Keynesian regime, the state has incentives to seek to involve the unions in negotiated wage moderation, while moderation is a strictly dominated strategy for the unions (namely irrational), since government is committed to full employment (Scharpf 1991). So, governments actively seek policy concertation in a Keynesian regime (even though they often fail to obtain it) but not in a monetarist one. In a monetarist regime inflation is exogenous to the collective bargaining system (for example because it is set by an independent central bank) and governments have little incentives to engage in negotiations with the unions over wage moderation, while the unions have incentives to moderate spontaneously their demands because of the negative effects of lower real money supply (in case the target inflation rate of the central 18 bank is exceeded) on employment (Sharpf, 1991; Streeck, 1994; Hassel, 2003). This hypothesis directly applies to government’s choice whether or not to engage in concerted wage policies. It also indirectly applies to choices concerning the welfare state because the latter is an important determinant of labor costs and wage inflation, especially in Continental welfare systems (Ebbinghaus and Hassel, 2000). We find no confirmation for this hypothesis in our sample. In Italy, progressively over the course of the 1980s, price stability took priority over unemployment and the increasingly independent central bank sought to mirror the behavior of the Bundesbank. Yet (some) of the governments of the 1990s explicitly pursued policy concertation on both wages (the 1992 and 1993 agreements) and pensions (1995 and 1997 agreements). The Austrian Central Bank (ÖNB) adopted a monetarist stance in the early 1980s, whose effect was a steady reduction of inflation levels and a temporary surge of unemployment. Under this macroeconomic regime there was no major disruption to the social partnership framework until 2000, when social partnership was cast aside by the right-wing government. The macroeconomic stance of the country seems to have had no influence, neither pro-concertation nor anti-concertation, on this particular policy choice (Kittel 2001: 125). In Ireland, the hypothesis seems to find outright disconfirmation, as a hard currency policy was accompanied by the staunch pursuit of a centralized wage agreement by government (Omitted). 18 b) Plausible Hypotheses Corroborated In our sample of countries, the governments that most dramatically increased their willingness to concert (Ireland and Italy) were all peculiarly weak from an electoral point of view. In Italy, the technocratic government of 1995 (similar to the government that initiated concertation on wage policy in 1993) lacked a clear popular mandate because it replaced the 19 Berlusconi’s government not after general elections, but thanks to a change in parliamentary alliances. The 1997 reform, in turn, was promoted by a very fragile coalition government, which was forced to resign less than one year later by a party in its own majority. In Ireland, the 1987 Fianna Fail government was also a minority government. Vice versa, the governments that moved away from concertation (Britain and Austria) were both holding comfortable majorities in parliament. This argument about strength needs to be qualified in the case of Austria. Here, the kind of strength at issue is not so much numerical as issue-specific. While the Social Democratic party was unable to overcome the opposition of an internal faction, which was able to block previous attempts at pension reform, the center-right coalition was free from direct influence from this group. The second factor which, in combination with the first, seems to explain a change in government willingness to engage in concertation is related to union strength and may be linked to the government’s expectations about the likely supply of partnership by the unions (Esping Andersen and Regini 2000). The analysis of our cases suggests a curvilinear relationship between union strength and government willingness to concert public policy. In other words, governments are interested in concertation when unions are neither too strong nor too weak. In Ireland and Italy, electorally weak governments were confronted with national union movements that had experienced losses in union density, even though they still maintained considerable mobilization capacities. Strong governments facing unions that are either so weakened that the government may not fear their reaction (as it may have been the case in Austria), or so strong that they are able to deadlock or highjack the negotiations, or worse undo the outcomes of negotiated regulation through collective action at the plant level (like in Britain), tend to abstain from concertation. 20 Overall, the four case studies both corroborate, in some respects, and qualify, in other respects, a particular line of argument in the literature on corporatism and concertation, which emphasizes the role of governments and regards the choice to engage in concerted policy-making with trade unions as essentially the response to a “governability” problem (Schmitter 1981; Lehmbruch, 1979). In a seminal contribution, Pizzorno (1978) pointed to the role that trade unions played in the mobilization of political consensus within democracies. He argued that because trade unions commanded the loyalty of thousands of workers, they had the potential to be valuable allies for governments, especially when these sought to pass controversial policies that might imply short-term costs in exchange for uncertain future long-term benefits (see also Regini, 1981). He referred to the ensuing entente between trade unions and governments as “political exchange,” an exchange in which the unions delivered consensus and were, in turn, repaid with access to the policy-making sphere and with the political/institutional resources that flowed from there. Pizzorno considered this type of arrangement as inherently unstable. The unions’ choice to enter the policy-making sphere led them to violate the representational mandate of at least a portion of their constituency. This created a representation gap, which, in turn, spurred a grass-root mobilization and the collapse of political exchange institutions. While the evidence presented here agrees with the basic thrust of this argument, it also suggests that concertation is not a general feature of all governments, independent of circumstances, but a peculiarity of electorally weak governments facing weakened unions. Concertation allows them to activate a non-parliamentary channel of consensus mobilization, In other words, sharing policy-making responsibility with the organizations representing those that are most likely to bear the brunt of policy changes, namely workers, protects weak governments from popular discontent they may be unable to handle otherwise. 19 Incidentally, in contrast to 21 Pizzorno’s argument, we find no evidence that concertative institutions are necessarily doomed to failure due to representation crises. Indeed, in Ireland social partnership has so far lasted for twenty years, and there are no signs that it may come to an end soon. From our analysis it also emerges that the combination of partnership and learning may lead to a change in policy regime. The Austrian and British were conservative governments and, in both cases, had come to the conclusion, based on past experience, that policy concertation had caused or compounded existing problems and needed to be disposed of. These two elements have to be considered in combination because both the Tory party in Britain and the People’s Party in Austria had, in the past, displayed completely different, and positive, attitudes towards policy concertation. The cases of adoption of concertation are, unlike the cases of demise, less clear cut. In Ireland, the government that engaged in concertation (Fianna Fail) and the one that preceded it and kept unions at arm’s lengths (a Fine Gael-Labor Party coalition) were probably equivalent in terms of represented interests (Mair 1992: 408). However, Ireland may be an outlier, also due to the difficulty of classifying the party system of this country along a left-right axis. Some corroboration for the partnership plus learning hypothesis comes from the Italian case, where governments that included socialist members did not attempt to concert policy solutions in the early 1990s, but did so later, when it became clear (especially after the Berlusconi government’s failure to pass pension reform) that a unilateral approach would be politically impractical. In other words, it seems that the element of partnership alone cannot explain the policy shift. It needs to be combined with a learning element, namely with the recognition that the past policy regime has been a failure. Concluding Remarks: Policy Concertation as a Political Phenomenon 22 This paper has sought to understand the European governments’ decisions to engage in policy concertation with unions and employers. To address this question, it has used both a large-N and small-n research strategy. The large-N results suggest there is no downward trend of government willingness to share policy prerogatives with the social partners. Between 1974 and 2003, the number of European governments that were willing to engage in at least one policy area between wage and welfare policy oscillated around a constant mean of 12 out of 15 countries. Additionally, there was a greater tendency over time to deal concertatively with both policy areas simultaneously. Government approach to concerted wage policies shows a strongly cyclical pattern, while the trend in welfare policies is markedly positive. The effects of EMU on government willingness to engage in policy negotiations seem less remarkable than is generally argued. The strongest impact is on the welfare policy domain, where the run-up to EMU seems to have provided further incentives for participating countries to engage in negotiations with the social partners. This conclusion needs to be qualified, however, because the overall 1974-2003 trend for government willingness to engage in concertative welfare policies is growing. Thus, the impact of EMU can at best be considered an accelerating element within a general growing trend. To understand the specific factors shaping government choice in favor or against policy concertation, the paper has then moved to a small-n study in which the measures of government willingness have been used to identify four extreme cases of policy change in both directions: Ireland, Britain, Italy, and Austria. The analysis of these four cases has aimed at contributing to a theory of government choice by identifying particular combinations of factors that could explain the adoption or the demise of concertation by national governments. 23 The results of this analysis suggest that the emergence of policy concertation can be likened to the building of an alliance between the “weak” and the “weakened.” Governments are willing to share their policy-making prerogatives with the social actors when they are politically weak, either because they lack strong and united parliamentary majorities, or because they are otherwise marred by internal struggles and legitimacy crises. In these circumstances associating trade unions to policy formation can be expedient for governments, as it allows them to activate supplementary channels of consensus mobilization. In particular, governments are willing to bring the trade unions on board when these still pose a credible threat for the smooth implementation of policy but their organizational fortunes have been declining in recent times, thus moderating their bargaining policies. Strong governments facing either unions that are too weak, so that the government does not fear their opposition, or that are too strong, so that they are able to deadlock any negotiation until their objectives are met, prefer a more unilateral approach to policy-making. In particular, strong governments are likely to change the policy regime, from concertative to unilateral and vice-versa, when their partisan orientation combines with a negative assessment of previous experiences. In other words, governments might move decisively away from concertation when they are of conservative orientation and when they have reached the conclusion that a system of policy codetermination has been a failure and has aggravated the problems it was intended to address. Conversely, centre-left governments might decide to adopt concertative approaches when unilateral policy-making has proved to trigger formidable social opposition. It should be emphasized that these are working hypotheses that have been inductively drawn from a structured comparison of the most dramatic cases of change in policy regime by governments. Not only do they need to be tested against cross-country evidence, but, even at a 24 purely theoretical level, they need to be refined and made more nuanced. For example, the argument that governments may be willing to engage in participatory policy-making when they are electorally weak seems acceptable only as a first pass. Clearly, not just minority governments may be interested in policy concertation, but also more electorally stable governments, when the opposition is well poised to benefit from unpopular policy choices (Kitschelt, 2001). Also, while the case study evidence suggests that there may be elective affinities, mediated by learning processes, between partisanship and particular modes of policymaking (concertative vs. unilateral), the conditions under which partisanship shapes the government’s choice of concertation vs. unilateralism do not seem entirely clear. 20 Even at this stage, however, one firm conclusion can be drawn from our study: policy concertation is an eminently political phenomenon that needs to be framed against the backdrop of the strategic configuration of the political system. Until now, most literature has relied on structural explanations, and has shown a tendency to read off presence or absence of concertation from structural grids of institutional and organizational preconditions. Yet, as witnessed by the case studies analyzed above, all cases of transition from one policy mode to another, concertation is not a stable characteristic of countries, but is more appropriately conceptualized as a set of possibilities that governments activate in some configurations of circumstances (potentially multiple, but not infinite), then drop, then perhaps reactivate again. Understanding exactly what these configurations might be is the task of future research, including our own. 25 References Berger, S. and Joint Committee on Western Europe. (1981). Organizing interests in Western Europe : pluralism, corporatism, and the transformation of politics. Cambridge [Eng.] ; New York: Cambridge University Press. Bonoli, G. (2001). Political Institutions, Veto Points, and the Process of Welfare State Adaptation. In P. Pierson, The New Politics of the Welfare State.(238-264) Oxford: Oxford University Press. Cazzola, G. (1995). Le nuove pensioni degli Italiani. Bologna: Il Mulino. Compston, H. (1997). Union Power, Policy Making and Unemployment in Western Europe, 1972-1993. Comparative Political Studies 30(6): 732-751. Compston, H. (2002). The Strange Persistance of Policy Concertation. In S. Berger and H. Compston, Policy concertation and social partnership in Western Europe : lessons for the 21st century.(1-16) New York, Oxford: Berghahn Books. Crouch, C. (2000). The Snakes and Ladders of Twenty-First-Century Trade Unionism. Oxford Review of Economic Policy 16(1): 70-83. Culpepper, P. (2007). The Politics of Common Knowledge: Ideas and Institutional Change in Wage Bargaining. Forthcoming in International Organization. Dale, I., Ed. (2000). Conservative Party General Election Manifestos, 1900-1997. London and New York: Routledge. Ebbinghaus, B. and A. Hassel (2000). Striking deals: concertation in the reform of continental European welfare states. Journal of European Public Policy 7(1): 44-62. 26 Ebbingghaus, B. and B. Kittel (2005). European Rigidity versus American Flexibility? The Institutional Adaptability of Collective Bargaining. Work and Occupations 32(2): 163195. Esping Andersen, G. and M. Regini, Eds. (2000). Why Deregulate Labour Markets?: Oxford University Press. Fallend, F. (2001). Austria. European Journal of Political Research (7-8): 238-253. Ferrera, M. (1996). The Southern Model of Welfare in Social Europe. Journal of European Social Policy 6(1): 17-33. Ferrera, M. and E. Gualmini (1999). Salvati dall'Europa? Bologna: Il Mulino. Garrett, G. (1998). Partisan Politics in the Global Economy. Cambridge: Cambridge University Press. Golden, M., P. Lange, et al. (2002). Union Centralization among Advanced Industrial Societies. Dataset available at /www.shelley.polisci.ucla.edu/data(Version dated July 28, 2004). Goldthorpe, J. H., Ed. (1984). Order and Conflict in Contemporary Capitalism. Oxford: Clarendon. Gunnigle, P., G. McMahon, et al., Eds. (1999). Industrial Relatons In Ireland: Theory and Practice. Dublin: Gill & Macmillan. Habermas, J. (1989[1964]). The Structural Transformation of the Public Sphere. Cambridge, Mass: MIT University Press. Hall, P. A., Ed. (1989). The Political Power of Economic Ideas : Keynesianism across Nations. Princeton, N.J.: Princeton University Press. Hall, P. A. and D. Soskice, Eds. (2001). Varieties of Capitalism: The Institutional Foundations of Comparative Advantage. New York: Oxford University Press. 27 Hancké, B. and M. Rhodes (2004). EMU and Labor Market Institutions in Europe. The Rise and Fall of National Social Pacts. Work and Occupations 32(2): 196-228. Hardiman, N. (1988). Pay, Politics, and Economic Performance in Ireland 1970-1987. Oxford: Clarendon Press. Hardiman, N. (1992). The State and Economic Interests: Ireland in Comparative Perspective. In J. H. Goldthorpe and C. T. Whelan, The State and Economic Interests: Ireland in Comparative Perspective Oxford: Oxford University Press. Hassel, A. (2003). The Politics of Social Pacts. British Journal of Industrial Relations 41(4): 707-726. Iversen, T. (1999). Constested Economic Institutions. The Politics of Macroeconomics and Wage Bargaining in Advanced Democracies. Cambridge: Cambridge University Press. Katzenstein, P. (1985). Small States in World Markets: Industrial Policy in Europe. Ithaca: Cornell Univeristy Press. Kenworthy, L. (2003). Quantitative Indicators of Corporatism. International Journal of Sociology 33(3): 10-44. King, G., R. O. Keohane, et al. (1994). Designing Social Inquiry : Scientific Inference in Qualitative Research. Princeton, N.J: Princeton University Press. Kitschelt, H. (2001). Partisan Competition and Welfare State Retrenchment: When Do Politicians Choose Unpopular Policies? In P. Pierson, The New Politics of the Welfare State.(265-302) Oxford: Oxford University Press. Kittel, B. (2001). Deaustrification? The Policy-Area-Specific Evolution of Austrian Social Partnership. West European Politics 23(1): 109-129. 28 Lehmbruch, G. (1979). Consociational Democracy, Class Conflict, and New Corporatism. In P. C. Schmitter and G. Lehmbruch, Trends toward corporatist intermediation.(53-61) London: Sage Publications. Lehmbruch, G. and P. C. Schmitter, Eds. (1982). Patterns of Corporatist Policy-Making. Sage modern politics series ; v.7. London: Sage. Letwin, S. R. (1992). The Anatomy of Thatcherism. Glasgow: HarperCollins. Lowi, T. (1979). The End of Liberalism: The Second Republic of the United States. New York: WW Norton. Mair, P. (1992). Explaining the Absence of Class Politics in Ireland. In J. H. Goldthorpe and C. T. Whelan, The development of industrial society in Ireland : the third joint meeting of the Royal Irish Academy and the British Academy.(383-410) Oxford: Published for the British Academy by Oxford University Press. Mares, I. (2003). The Politics of Social Risk: Business and Welfare State Development. Cambridge: Cambridge University Press. Mares, I. (2006). Taxation, Wage Bargaining, and Unemployment. Cambridge: Cambridge University Press. Marx, K. and F. Engels (1978). Manifesto of the Communist Party. In R. C. Tucker, The MarxEngels Reader New York and London: Norton & Company. Mascini, M. (2000). Profitti e salari. Bologna: Il Mulino. Mill, J. S. (1946). A System of Logic. New York: Harper &Brothers. Molina, O. and M. Rhodes (2002). Corporatism: the Past, Present and Future Of a Concept. Annual Review of Political Science 5: 305-331. 29 Murphy, T. and W. E. Roche, Eds. (1997). Irish Industrial Relations in Practice. Revised and Expanded Edition. Dublin: Oak Tree Press. Negrelli, S. (2000). Social Pacts in Italy and Europe: Similar Strategies and Structures; Different Models and National Stories. In G. Fajertag and P. Pochet, Social Pacts in Europe - New Dynamics.(85-113) Bruxelles: ETUI. NESC (1986). A Strategy for Development. 1986-90. Dublin: National Economic and Social Council. OECD (2001). Statistical Compendium, OECD. Peden, G. C. (1991). British Economic and Social Policy: Lloyd George to Margaret Thatcher. Oxford: Philip Allan. Pelling, H. (1987). A History of British Trade Unionism. London: MacMillan. Pierson, P. (1996). The New Politics of the Welfare State. World Politics 48(2): 143-179. Pizzorno, A. (1978). Political Exchange and Collective Identity in Industrial Conflict. In C. Crouch and A. Pizzorno, eds. (1978), The Resurgence of Class Conflict in Western Europe since 1968 London: Macmillan. Poulantzas, N. (1973). Political Power and Social Classes. New York: Verso. Powell, D. (1992). British Politics and the Labour Question, 1868-1990. New York: St. Martin's Press. Ragin, C. (1987). The Comparative Method: Moving Beyond Qualitative and Quantitative Strategies. Berkley: University of California Press. Regini, M. (1981). I dilemmi del sindacato. Bologna: Il Mulino. Regonini, G. (1996). Partiti e pensioni: legami mancanti. Un gigante dai piedi d’argilla. . In P. Isernia, Partiti, risorse e politiche nella crisi italiana.(73-139) Bologna: Il Mulino. 30 Rhodes, M. (2001). The Political Economy of Social Pacts: 'Competitive Corporatism' and European Welfare Reform. In P. Pierson, The New Politics of the Welfare State.(165194) Oxford: Oxford University Press. Scharpf, F. W. (1991). Crisis and choice in European social democracy. Ithaca, N.Y: Cornell University Press. Schludi, M. (2001). The Politics of Pensions in European Social Insurance Countries. Discussion Paper 01/11, Köln, Max-Planck-Institut für Gesellschaftsforschung. Schludi, M. (2003). The Reform of Bismarckian Pension Systems. A Comparison of Pension Politics in Austria, France, Germany, Italy, and Sweden. PhD Thesis. Humboldt University, Berlin. Schmitter, P. (1974). Still the Century of Corporatism? Review of Politics 36: 85-131. Schmitter, P. (1981). Interest Intermediation and Regime Governability in Contemporary Western Europe and North America. In S. Berger, Organizing interests in Western Europe : pluralism, corporatism, and the transformation of politics.(285-327) Cambridge [Eng.] ; New York: Cambridge University Press. Schmitter, P. (1983). Teoria della democrazia e pratica neo-corporativa. Stato e Mercato 9 (December). Schmitter, P. C. and J. R. Grote (1997). The Corporatist Sisyphus: Past, Present and Future. EUI Working Paper SPS. Florence: 22. Schmitter, P. C. and G. Lehmbruch, Eds. (1979). Trends Toward Corporatist Intermediation. Contemporary political sociology ; vol. 1. London: Sage Publications. 31 Streeck, W. (1993). The Rise and Decline of Neocorporatism. In L. Ulman, B. Eichengreen and W. T. Dickens, Labor and an Integrated Europe Washington, D.C.: The Brookings Institution. Streeck, W. (2005). The Study of Interest Groups: Before "The Century" and After. Philippe C. Schmitter's retirement from the European University Institute, Florence September 17-18. Streeck, W. and L. Kenworthy (2004). Theories and Practices of Neo-Corporatism. In Janoski, Alford, Hicks and Schwartz, A Handbook of Political Sociology Cambridge: Cambridge University Press. Streeck, W. and P. Schmitter (1991). From National Corporatism to Transnational Pluralism. Politics and Society 19: 133-164. Swenson, P. A. (2002). Capitalists Against Markets: The Making of Labor Markets and Welfare States in the United States and Sweden. Oxford: Oxford University Press. Tàlos, E. and B. Kittel (2002). Austria in the 1990s: The Routine of Social Partnership in Question? In H. Compston, Policy Concertation and Social Partnership in Western Europe Oxford: Berghahn Books. Thelen, K. A. (2001). Varieties of Labor Politics in the Developed Democracies. In P. A. Hall and D. Soskice, Varieties of Capitalism: The Institutional Foundations of Comparative Advantage New York: Oxford University Press. Thomsen, J. P. F. (1996). British Politics and Trade Unions in the 1980s: Governing Against Pressure. Aldershot: Dartmouth. Thorpe, A. (2001). A History of the British Labour Party. New York: Palgrave. Tomandl, T. and K. Fuerboeck (1986). Social Partnership: The Austrian System of Industrial Relations and Social Insurance. Ithaca, N.Y.: ILR Press. 32 Traxler, F. (2003). Bargaining (De)centralization, Macroeconomic Performance and Control over the Employment Relationship. British Journal of Industrial Relations 41(1): 1-27. Traxler, F., S. Blaschke, et al. (2001). National Labour Relations in Internationalized Markets. Oxford: Oxford University Press. Viebrok, E. (2003). Coping with Corporatism's Legitimation Deficit: Recent Attempts at Reforming the Austrian Chamber System. Unpublished Manuscript, Geneva, International Institute for Labour Studies. Walsh, J. I. (2000). When Do Ideas Matter? Explaining the Successes and Failures of Thatcherite Ideas. Comparative Political Studies 33(4): 483-516. 33 Figure 1 – Government Willingness To Engage in Concerted Policies Three-years moving averages Share of European Countries .8 .7 .6 .5 a) Wage Policy .4 1975 1980 1985 1990 1995 2000 2005 Share of European Countries .8 .7 .6 .5 .4 b) Welfare Policy 1975 1980 1985 1990 1995 2000 2005 34 Figure 2 – Bundling of Policy Issues Over Time Number of European Governments Yearly values 15 10 5 0 1970 1975 1980 1985 1990 1995 2000 2005 Concertation willingness in both income and welfare policies Concertation willingness in at least one policy area 35 Figure 3 – Government Willingness to Engage in Concerted Policies: EMU vs. non-EMU countries Yearly values Share of European Countries .9 .8 .7 .6 .5 .4 .3 .2 .1 a) Wage Policy 0 1985 1990 1995 2000 2003 Share of European Countries .9 .8 .7 .6 .5 .4 .3 .2 .1 b) Welfare Policy 0 1985 1990 1995 2000 2003 EMU Government Willingness Non-EMU Government Willingness 36 1 In this paper, the expression “policy concertation” is used as defined by Compston (Compston 2002), namely as a policy-making method whereby “employers and trade unions are involved in the making of decisions that are ultimately the exclusive province of the state, in particular decisions on the contents of legislation, regulations and administrative orders.” Policy concertation allows for the formal involvement of societal interests and is, in this sense, different form a pluralist model in which groups exercise informal influence on the governmental sphere. This definition is similar to the definition of corporatism as provided, for example, by Streeck and Kenworthy (2004: 11): “Corporatist theory and practice blur the boundary between state and society as the state shares authority with private interest associations, using the latter as agents of public policy by coordinating their behavior or delegating public functions and decisions to them. In a corporatist context, private interest representation thus shades into public governance.” 2 Our indicator shares some similarities with the Union Participation Index in Compston (1997), which is available for 13 OECD countries between 1970 and 1992. The goal of that index was to measure union influence in economic policy-making (except wage policy), ranging from no participation to narrow consultation to broad agreement (Compston, 1997: 737). Our goal is different: we focus on government’s propensity towards concertation independently of whether or not a concertation agreement materializes. This allows us to distinguish between the intentions of governments from the intentions of the social partners. We have also designed and measured an indicator of “social compacting” which captures weather or not, contingent on the government’s willingness to engage in concertation, a negotiated policy agreement is actually reached, and the extent to which the social partners underwrite it. We do not present it here because its reach is beyond the scope of this paper. 3 The 15 countries include the EU15 except Luxemburg and including Norway. 4 In the absence of information, we assume policy inertia. In other words, unless there is textual evidence that a change occurs in the process, the variable keeps the same score as in the previous year. 5 Notice that this does not transpire from the graphs in Figure 1 as these plot three-year averages and not annual data. 6 Notice, however, that we have no information on Spain, Finland, and Greece for that year. 7 We have also run a quick test of what we refer to as the “Katzenstein hypothesis” (1985), contending that greater exposure to international trade makes a government more willing to look for negotiated solutions to commonly 37 perceived problems. Our results (not displayed for reasons of space) suggest, at best, scant empirical confirmation for it. 8 Our data also show considerable within-country variation between concertation willingness and lack thereof, thus providing support for an argument emphasizing the institutional flexibility of several European countries (see Ebbinghaus and Kittel 2005). 9 Here and elsewhere union data are based on the OECD union membership database. 10 See EIRR 1994, 4 11 http://www.eiro.eurofound.eu.int/1997/11/feature/at9711144f.html; see also Schludi 2003: 186-190 12 http://www.eiro.eurofound.eu.int/2000/08/feature/at0008228f.html 13 http://www.eiro.eurofound.eu.int/2001/08/inbrief/at0108225n.html; http://www.eiro.eurofound.eu.int/2002/01/feature/at0201206f.html 14 http://www.eiro.eurofound.eu.int/2003/01/feature/at0301203f.html 15 http://www.eiro.eurofound.eu.int/2004/01/feature/at0401203f.html 16 A more refined Marxist theory of the state recognizes its relative autonomy and its capacity to pursue policies that may be in conflict which the immediate, short-term interests of the dominant class (Poulantzas, 1973) 17 For Austria we do not have sufficient information to pass a judgment. 18 This hypothesis is not testable in the case of Britain. Indeed, the shift to a monetarist regime – where inflation becomes the main priority and it is controlled through strict regulation of money supply – emerged simultaneously to the demise of policy concertation on wage policies. 19 In this regard, some of the recent literature on welfare state reform has proceeded on similar lines to classic corporatist theory, emphasizing the importance for governments of mustering the necessary popular consensus both through parliamentary and non-parliamentary means, but has at the same time made it very clear that only certain governments, and not all, take a participatory approach (see, for example, Kitschelt 2001; Schludi, 2001; 2003; Bonoli, 2001; Pierson, 1996). 20 See Omitted for a more elaborate theory of the decision of social democratic parties to engage with trade unions. The theory is based on the intuition that, at a time of electoral realignment, concertation allows social democratic parties to maintain strong links with organized working-class constituencies, while simultaneously pursuing policies that appeal to a shifting median voter. 38 39