European Monetary Union and collective barganing

EUROPEAN MONETARY UNION AND COLLECTIVE BARGAINING Franz Traxler University of Vienna WP núm. 207 Institut de Ciències Polítiques i Socials Barcelona, 2002 2 The Institute of Political and Social Sciences (ICPS) is a consortium created in 1998 by the Barcelona Provincial Council and the Universitat Autònoma de Barcelona, the institution to which the ICPS is officially linked for academic purposes. “Working Papers” is one of the collections published by the ICPS, specialising in the publication of work currently being undertaken by social researchers -in the author’s original languagewith the aim of facilitating their subsequent scientific debate. Inclusion of work within this collection does not prohibit its future publication by the author, who maintains full rights over the article. The work in question cannot be reproduced without the permission of the author. Edición: Institut de Ciències Polítiques i Socials (ICPS) Mallorca, 244, pral. 08008 Barcelona (España) http://www.icps.es © Franz Traxler Design: Toni Viaplana Printer: a.bís Travessera de les Corts, 251, entr. 4a. 08014 Barcelona ISSN: 1133-8962 DL: B-43.407-02 3 INTRODUCTION Collective bargaining has remained within the realm of the nation state even more than most other areas of industrial relations. These national bargaining systems still vary considerably in their institutions and practices across the EU member states. At the same time, economic integration has made substantial progress in particular due to the single market and European Monetary Union (EMU), which have created strictly supranational markets within the EU. This coincidence of deepening economic integration and persisting national hegemony over collective bargaining has given rise to concerns in two respects. First, there is the question of how economic integration will affect collective bargaining. Regardless of their manifold differences in institutions and practices, the national bargaining systems under the umbrella of the EU share one essential property in that they are more ‘organized’ than their counterparts are in most other (namely non-European) OECD countries: this is manifested in the predominance of multi-employer bargaining, covering the vast majority of employees (Traxler et al., 2001). In this respect, one may well speak of a EU model of bargaining. The notable exception to this model is the United Kingdom, where single-employer bargaining supplanted multi-employer settlements already in the 1980s, resulting in a decline in collective bargaining coverage from about 70 per cent in the early 1980s to little more than 20 per cent in the mid-1990s. Furthermore, the reform countries of Central and Eastern Europe are also characterized by rather unorganized bargaining systems. Against this background, three scenarios of the future of bargaining have been discussed. The pessimistic scenario assumes that EMU will prompt a general process of disorganization, as Europe’s supranational markets will tend to undermine bargaining systems the scope of which is still limited to the nation state (e.g. Mahnkopf and Altvater, 1995). Such prospects of disorganization may have even been one important goal of the neoliberal policy community, when advocating EMU (Crouch, 2000; Dølvik, 1999). 4 Under these circumstances, Europeanization would occur in a very restricted way, giving rise to highly fragmented, company-based practices of transnational bargaining. These practices are supposed to develop at the expense of the national level, since the Eurocompanies when adopting these practices will tend to opt out of the national bargaining structures (Visser and Ebbinghaus, 1992). The optimistic scenario suggests that economic integration may spill over to industrial relations, leading to a multi-level bargaining system that includes European level collective agreements (e.g. Jacobi, 1998). While both the pessimistic and optimistic scenario imply significant changes in the bargaining system, a third line of reasoning stresses the continuity and path dependency of industrial relations developments (Ferner and Hyman, 1998a, Traxler et al., 2001). The key argument underlying this position is that economic pressures affect industrial relations only indirectly, in that these pressures are processed and filtered by the established institutions. Notably, this does not rule out the possibility of growing divergence in terms of substantive bargaining outcomes within the national systems (Marginson and Sisson, 1998), provided that flexibility rather than rigidity is the institutional code that guides the system’s response to the economic challenge. At any rate, the path dependency means that the national bargaining systems will not easily change even under the aegis of EMU, and Europeanization of bargaining can take place only in a way which is compatible with the given institutional diversity of bargaining across the EU (Traxler, 2002b). The second concern refers to the macroeconomic performance of collective bargaining under EMU. It is widely assumed that EMU places the main burden on the labor market, when it comes to adjusting to asymmetrical shocks (i.e. economic imbalances within the eurozone). This is because EMU deprives the eurozone countries of the exchange rate as a means of compensating for imbalances. Since transnational labor mobility is low in Europe, collective bargaining, namely bargaining over wages becomes the key parameter of labor market adjustments. As is in the case of the future development of bargaining, so the problem with performance 5 ensues from the coincidence of economic integration and national fragmentation of bargaining. Again, one can differentiate between three scenarios. The first one contends that this coincidence will provoke inflationary wage policies (e.g. Soskice and Iversen, 1998). As a consequence of national fragmentation, each single bargaining unit is too small to have a noticeable impact on macroeconomic development in the eurozone. Hence, there is no incentive to moderate wages for the sake of price stability and employment. The counterposition is that there is a deflationary risk in that the distinct nationally differentiated bargaining units embark on competitive wage policies aimed at lowering the labor costs of one’s own country in relation to the country’s trading partners (Busch, 1992). Finally, the national fragmentation of bargaining is supposed to reinforce economic imbalances in the eurozone, since the national bargaining systems strongly differ in their ability to align their wage policy with macroeconomic requirements (Traxler, 1999b). This paper will address the development and performance of bargaining under EMU in line with the following methodological considerations: scholarly debates on the development and performance of institutions tend to treat these questions as one and the same problem. The assumption underlying such reasoning is that institutional change is driven by performance pressures bringing about a natural selection that enables only well performing institutions to survive. However, there is the possibility that inefficient institutions can survive due to market imperfections that fail to enforce efficiency (North, 1990). Furthermore, functionally equivalent institutional arrangements may exist, such that institutional diversity across Europe may last even when EMU magnifies the scale of performance pressures. At any rate, it is reasonable to consider the development and performance of bargaining as distinct questions. A second methodological problem is that a sound empirical basis for predicting the development and performance of bargaining under EMU is lacking, since the pre-EMU phase (beginning with the Maastricht treaty) and EMU as such have covered a rather short time period. Hence, we 6 adopt a long-term perspective on these issues, including 1970 to 1996 for 15 European countries (Table 1). As regards developments, this enables analysis to examine how the national bargaining systems responded to essential changes in the economic policy regime which are comparable to the establishment of EMU: this refers to the shift from Keynesian to neoliberal policies in the late 1970s/early 1980s, and the preparation and implementation of the single market during the second half of the 1980s. In the case of performance, a long-term perspective is preferable, since institutions can have only a structural (i.e. long-term) impact on performance due to their high degree of stability. As far as European economic integration primarily means the rise of a supranational market, it is almost impossible to isolate its effect from more encompassing processes of ‘globalization’ of markets. This is important since the main challenge posed by EMU to bargaining ensues from growing market competition. Under these circumstances, a long-term perspective is appropriate, because this offers the opportunity to study how the bargaining systems tend to respond to growing market pressures. For the reasons outlined above, we have to consider both the national and the European level, when analyzing the interplay of bargaining and EMU. Finally, we will limit our comparative analysis to those dimensions of collective bargaining which have most hotly been debated, as far as issues of development and performance are concerned: that is, centralization and coordination of bargaining. The degree of centralization refers to the level at which collective agreements are formally fixed. Given multi-level bargaining in all countries under consideration, country scores are based on a country’s most important bargaining level. Coordination of bargaining means that there is an economy-wide synchronization of bargaining. 7 Table 1 Forms of Wage Setting in Western Europe* COORDINATION 1970–1973 % 1974–1976 % 1977—1979 % 1980–1982 % 1983–1985 % 1986–1990 % 1991–1993 % 1994–1996 % State-sponsored Coordination DKx, FINx B, Dx, IRL 38.5 FINx, Nx, Sx B, Dx, I, IRL, UK 57.1 FINx, Sx B, E, IRL, UK 40.0 FINx, IRL 13.3 FINx, DKx, E, I, NLx, Sx 40.0 FINx,Nx, B, DKx, IRL, NLx, Sx 46.7 FINx, Nx, B, DKx, I, IRL, NLx, P, Sx 60.0 DKx, I, IRL, Nx, NL 33.3 Inter-associational Coordination Nx, Sx 15.4 0.0 0.0 Sx, E 13.3 Nx 6.7 I 6.7 0.0 0.0 Intra-associational Coordination Ax,CHx,I, NLx 30.8 Ax,CHx 14.3 Ax,CHx,I, NLx 26.7 Ax,CHx,I 20.0 CHx,IRL+,P 20.0 CHx,E,P 20.0 CHx, E 13.3 CHx, E, P 20.0 Pattern bargaining 0.0 0.0 Dx 6.7 Dx, DKx, 13.3 Ax, Dx 13.3 Ax, Dx 13.3 Ax, Dx 13.3 Ax, Dx 13.3 State-imposed coordination F 7.7 DKx, F, NL, P 28.6 DKx, F, Nx, P 26.7 B,F,Nx,NLx, P 33.3 B, F 13.3 F 6.7 F 6.7 B, F 13.3 No coordination UK 7.7 0.0 0.0 UK 6.7 UK+ 6.7 UK+ 6.7 UK+ 6.7 FINx, Sx,UK+ 20.0 Total 13 100.0 14 100.0 15 100.0 15 100.0 15 100.0 15 100.0 15 100.0 15 100.0 BCEN 5.38 5.21 5.90 5.90 6.77 6.43 6.43 7.46 * Modal values of annual observations per period. Portugal since 1975, Spain since 1977. a DK: state-sponsored in 1997, state-imposed in 1998 b S: no coordination in 1997, pattern bargaining in 1998 c N: central-level bargaining in 1997 Central-level bargaining put in italics. x High (i.e. state-licensed) bargaining governability, + predominance of single-employer bargaining. Country codes: A = Austria, B = Belgium, CH = Switzerland, D = Germany, DK = Denmark, E = Spain, FIN = Finland, F = France, IRL = Ireland, I = Italy, NL = Netherlands, N = Norway, P = Portugal, S = Sweden, UK = United Kingdom BCEN = Degree of bargaining centralization; centralization decreases with score (For definition of BCEN, see Traxler et al. 2001). 8 THE DEVELOPMENT OF COLLECTIVE BARGAINING The development of collective bargaining in terms of centralization and coordination at the national level is summarized in Table 1. The three decades covered by Table 1 are differentiated into eight subperiods. Country scores for both variables refer to collective bargaining over wages. This is because centralization as well as coordination of bargaining varies with bargaining issues in some countries. For instance in Spain, several central-level accords on non-wage issues were struck during the 1990s, whereas the level most important to wage formation was the industry. As regards the centralization variable (i.e. BCEN), Table 1 shows a clear trend towards decentralization of bargaining. As the average score on BCEN for all countries under consideration indicates, that bargaining has significantly become decentralized from the first period to the last one (paired t-test; t= -2.60, p= 0.03, n= 12). On closer examination of the subperiods, one finds distinct movements within the general trend. During the first half of the 1970s (i.e. from the first to the second subperiod) collective bargaining experienced a process of (insignificant) centralization (t-test; t= -1.00, p= 0.34, n= 13). From this subperiod until 1980-82, a significant decline in the bargaining level took place (t-test; t= -2.05, p= 0.06, n= 14). Further decentralization in the following subperiods occurred although the scale of this change in relation to 1980-82 remained statistically insignificant (t-test; t= -0.79, p= 0.45, n= 14). One should note that the scores on BCEN somewhat downplay the actual scale of decentralization in that they report only changes in the most important bargaining level. Hence, the rise of shop floor bargaining is not registered as long as this did not become prevalent in a country’s bargaining system. For the 1980s and 1990s, Table 1 does not document what is often called organized decentralization (Traxler, 1995). This means that certain bargaining issues are delegated by the higher-level bargaining parties to regulation at lower level within a binding framework set by the higher-level settlement. Since agenda-setting and control over the lower 9 level thus remains with the higher bargaining level within a framework of organized decentralization, even delegation processes flowing from the most important level to lower levels do not constitute a change in the prevalent (i.e. most important) level. Organized decentralization, mainly in the form of a delegation of issues from the industry level to the company, has become widespread in Europe (Ferner and Hyman, 1998a; Traxler et al., 2001). This happened in two waves. From the mid-1980s onwards, working time has become the subject of organized decentralization, while wages have become so since the early 1990s. Both waves were initiated by the employers for the sake of more flexible arrangements. What follows from these observations is that the bargaining level underwent gradual modifications rather than radical transformations in terms of the predominant bargaining level over the three decades. Most essentially, only two countries (i.e. Ireland and the UK) saw a move from multito single employer bargaining (Table 1). In the case of Ireland, this remained a short-term episode. In contrast to this, single employer bargaining consolidated in the UK which thus moved away from the European model. In all the other countries, multi-employer bargaining continued to be prevalent in the overall bargaining system. This implies that the long-term trend towards decentralization, as documented by the scores on BCEN, mainly result from a more or less lasting shift from central-level to industry-level bargaining in several countries (namely Denmark, Sweden and Spain). Given all these tendencies towards decentralization within the national bargaining systems, it is no surprise that European-level bargaining has not developed so far. Notably, there are European-level negotiations within the framework as introduced by the Social Protocol of the Maastricht treaty (Keller and Bansbach, 2001). However, these negotiations do not meet the properties of free collective bargaining for two main reasons. Within this framework, there is no possibility of advancing one’s interests by means of industrial conflict. Furthermore, wages are explicitly excluded from the range issues that can be negotiated. 16 Not surprisingly, either of these competing positions has come up with empirical studies that support their reasoning (e.g. Cameron, 1984; Calmfors and Driffill, 1988). However, their explanatory power is strongly questioned by more recent studies (e.g. OECD 1994, OECD 1997, Traxler and Kittel, 2000; Traxler et al., 2001) which could not detect any kind of a robust association between bargaining centralization and macroeconomic performance. Above all, bargaining centralization has turned out to be a very poor predictor of labor cost growth. This is essential because labor costs are generally seen in the related debate as the causal link between the bargaining institutions and macroeconomic performance other than labor costs. If convincing evidence of a significant labor cost effect is lacking, then any empirical finding on a significant effect of centralization on such performance criteria as inflation and employment becomes inconclusive by standards of that causality assumption underlying both hypotheses. This lack of compelling empirical evidence can be traced to several conceptual weaknesses which are common to any reasoning regarding bargaining centralization as the key to explaining structural performance effects caused by the wage-setting institutions. The main weaknesses can be summarized as follows: -The debate does not systematically differentiate between centralization and coordination. As mentioned above, the premise of the debate is that alternative bargaining institutions differ in their ability to internalize wage externalities. This includes the point that bargaining units whose domain of activities is so encompassing that it achieves macroeconomic relevance are unable to externalize the costs of wage increases. A concept that focuses only on centralization neglects the fact that a situation of macroeconomic relevance also applies to bargaining systems composed of a larger number of formally independent bargaining units which coordinate their wage policy across the economy (Soskice, 1990). Hence, what really matters with regard to macroeconomic weight of bargaining is coordination, while centralization is just a special 17 manifestation of coordination. Studies centering on centralization thus ignore the existence of decentralized forms of wage coordination. The magnitude of this shortcoming can be seen from the fact that in the vast majority of countries characterized by economy-wide wage coordination this coordination takes place in a non-centralized context (Table 1). -As an implication of the tendency to confuse centralization with coordination, the debate does not fully capture the scale of cooperation problems arising from economy-wide coordination. Mainstream reasoning on the structural ability to internalize wage externalities refers only to what may be called the horizontal problem of cooperation: that is, the relationship between the distinct sectoral and occupational groups of employers and employees. These groups are argued to cooperate or free ride, depending on how the domain of the bargaining units are demarcated (i.e. the degree of bargaining centralization). Poor performance, implying the ability to externalize the costs of wage increases, will occur when the domains of the bargaining units coincide with the scope of powerful, particularistic interest groups which enable these groups to take a free ride to the disadvantage of the other groups. This focus on horizontal cooperation neglects another vertical problem of cooperation. This ensues from the fact that the rank and file does not directly participate in collective bargaining but is represented by certain, nominated agents in this process. Manifold conflicts of interest between the rank and file and its representatives may emerge, such that the former’s compliance with the latter’s decisions cannot be taken for granted. Conventional reasoning suggests that central-level bargaining is capable of overcoming the horizontal as well as the vertical cooperation problem. While it is evident that central-level bargaining resolves the horizontal problem, it tends to aggravate the vertical problem of cooperation. The reason for this is that decision-making processes face increasing legitimation problems with growing centralization. In line with this, recent empirical research shows that labor costs even significantly increase with growing centralization of bargaining, when a supportive framework devised to overcome this vertical 18 cooperation problem is lacking (Traxler, 2001; 2002a). This evidence of a highly contingent effect of bargaining centralization helps understand why there is no robust, general effect of the bargaining level. At any rate, attempts at horizontal coordination –regardless of whether relying on a centralized or a decentralized setting– do not guarantee that the bargaining units’ members comply with these coordination efforts. On the contrary, any such effort brings about problems of vertical cooperation. -The focus on bargaining centralization fails to take account of the manifold forms of state regulation. There is evidence from other crossnational research that differences in how wage formation is regulated by the state affect performance (Traxler, 1999a). -The above points have another important methodological consequence for empirical studies in that they rule out the possibility of constructing one single, ordinal or parametric scale for measuring the wage setting institutions. The only dimension that meets this scale type is bargaining centralization which as such has turned out as no powerful predictor of performance. All the other dimensions discussed above differ not simply in quantitative but also in qualitative terms. These caveats include coordination (Traxler and Kittel, 2000). There is no theoretical argument that can show that decentralized coordination forms are more/better coordinated than centralized bargaining or vice versa. Therefore, any kind of one single, composite measure based on ordinal or parametric ranking of bargaining coordination is pointless3. Due to the multi-dimensionality of wage setting the scaling process must address each single dimension separately. The classification of bargaining coordination used in Table 1 captures all the above points, as far as the problem of horizontal coordination is concerned. To overcome the problem of vertical cooperation, state regulation is essential, since employer organizations and unions are voluntary associations which can hardly bind their rank and file merely by fiat when conducting collective bargaining (Crouch, 1993; Golden, 1993). This difficulty with assuring compliance is evidenced by the 19 above mentioned finding that wage growth significantly increases with bargaining centralization, when such supportive legal framework is lacking. Its most important provisions are the legal enforceability (i.e. the ‘lawmaking’ effect) of collective agreements and the peace obligation during the validity of a collective agreement (Traxler et al., 2001). For the purpose of a crossnational comparison, this results in a dichotomous categorization. A national system of wage setting is classified as being characterized by high bargaining governability if legal enforceability and the peace obligation are established. Otherwise, a country’s bargaining governability is low. It should be noted that this differentiation between horizontal and vertical coordination corresponds with the distinction between to qualitatively different forms of state interference in wage setting. Stateimposed and state-sponsored coordination are the two possible forms of substantive state regulation in that the state actively participates in the wage setting process and influences its outcome. Bargaining governability is a matter of procedural state regulation, since it forms part of the legal framework for bargaining. Table 2 The General Relation Between Bargaining and Wage Performance Wage moderation Strong Pattern bargaining Voluntary peak-level coordination* with high bargaining governability Medium Uncoordinated bargaining State-imposed coordination Weak Voluntary peak-level coordination* with low bargaining governability Low Medium High Vertical coordination * State-sponsored coordination, inter-associational coordination, intra-associational coordination. Source: Traxler et al. (2001: 247) Empirical studies which used these categorizations of bargaining coordination as predictors of economic performance for differing country samples and time periods (Traxler and Kittel, 2000; Traxler et al., 2001) 20 could show that the institutional differences in bargaining coordination across countries indeed affect labor cost growth significantly. The institutional effects on inflation and unemployment echo those on labor costs. However, the impact on unemployment is much weaker than on labor costs and inflation. The main findings of these studies are summarized in Table 2. State-sponsored, inter-associational and intraassociational coordination are subsumed under peak-level coordination in Table 2. This is because they are similar in terms of both their performance profile as well as their mode of coordination in that the peak associations have the key role in the coordination process in the case of all three categories. The most striking result is that peak-level coordination delivers contrasting performance, depending on whether bargaining governability is high or low. When embedded in a context of high governability, all forms of peak-level coordination perform very well. Yet in combination with low governability, they perform worse than any other kind of bargaining. This is consistent with the above-cited finding on the contingent effect of bargaining centralization. As any form of peak-level coordination is relatively centralized compared to alternative forms of bargaining, it is also burdened with more severe problems of compliance and legitimacy in relation to the rank and file. As a consequence, there is a much stronger need for a legal framework capable of making the rank and file comply than is the case of more decentralized forms of voluntary coordination. As regards the other forms of bargaining, pattern bargaining also records a high capacity for internalizing wage externalities. In comparison, stateimposed coordination and uncoordinated bargaining achieve average labor cost increases, with non-coordination performing somewhat better than state-imposed coordination. One reason for this comparatively inferior performance of state-imposed coordination is that the state usually resorts to authoritative wage control as a means of ultima ratio in a situation of extremely poor performance of free collective bargaining. The upshot of these findings is that institutional differences in bargaining cause indeed significant and robust differences in 21 macroeconomic performance, in particular as far as (unit) labor costs and inflation are concerned. This relationship, however, is far more complex than orthodox reasoning suggests. It is not simply a matter of gradual differences in the degree of bargaining centralization. In contrast to this, there are qualitatively different properties of wage setting, leading to potential substitutes (i.e. functionally equivalent arrangements) in terms of effective wage moderation (i.e. pattern bargaining and the distinct forms of peak-level coordination backed by high governability). Conversely, one and the same form of (horizontal) coordination produces contrasting performance effects, depending on whether legal provisions for governability are established or not. EMU becomes important with regard to the performance of bargaining, since differences in monetary policy affect the scope for bargaining. A monetary policy that does not accommodate to what is seen as an inflationary policy makes it more difficult for the bargainers to externalize the costs of wage increases, as compared to an accommodating monetary regime. This means that a non-accommodating monetary policy may work as a substitute for the structural ability of the bargaining institutions to internalize negative externalities, when it comes to moderating wages. However, one cannot rule out the possibility that the need to enforce wage moderation by means of a tough monetary policy brings about real costs (e.g. rising unemployment) which an effective voluntary wage moderation does not create. In any case, the power of the monetary authorities to sanction and discipline the bargainers and their wage policy works as an intervening variable that modifies the above relationship between the bargaining institutions and economic performance, in a way depending on whether these authorities adopt a non-accommodating or an accommodating policy line. Aside from these general interaction effects, EMU creates a specific and new situation in that it tends to change the conditions for the national bargainers in the eurozone in two main respects. First, the ECB is modelled on the Bundesbank the high institutional independence of which 22 is argued to assure far more effective commitment to price stability than is the case of central banks which are less independent of the other political actors (e.g. Cukierman, 1992). This means that the bargainers of several countries (e.g. Italy, Spain) which had experienced a long record of accommodating monetary policy will have to face a nonaccommodating regime under EMU. Second, and even more importantly, EMU alters the structure of interaction between monetary policy and wage policy. On the one hand, monetary policy in the eurozone cannot target each single national bargaining system, since the ECB will orient its policy towards European money supply and inflation. On the other hand, the economic weight of each single national bargaining system –even when effectively coordinated across the domestic economy– does not suffice to have a noticeable macroeconomic impact on the eurozone as a whole. This as well as the European orientation of the ECB improves the opportunity of the national bargainers to externalize wage externalities. For all these reasons, the interaction between monetary policy and wage bargaining has attracted growing attention. While there are both analytical and empirical studies in this issue (Franzese, 2000), we again concentrate on empirical research here. Needless to say, related studies differ widely in how they operationalize wage bargaining and monetary policy. Since monetary responses to wage movements are just one certain kind of wage externality, it is reasonable to disregard work on centralization and follow the above categorization of bargaining coordination instead, when conceptualizing the institutional setting of wage policy. As regards monetary policy, institutional analysis focuses on the independence of the central bank in relation to other political actors, as noted above. In line with this, most interaction analyses examine the interplay of bargaining and central bank independence. However, this is not the most appropriate approach, as far as our problem is concerned. In this context, the question of whether the hypothesized positive association between central bank independence and price stability holds is of secondary importance. What actually interests us, is the responsiveness of alternative wage setting 23 institutions to monetary policy as such, regardless of how this policy relates to central bank independence. Hence, it is preferable to capture differences in monetary policy as such, namely differences in its restrictiveness, all the more since a standard measure of central bank independence is not available (Iversen, 2000). The empirical findings on the interaction between monetary policy and the bargaining institutions (operationalized in terms of horizontal and vertical coordination) are documented in Traxler et al., (2001). Hence, we again refrain from going into technical details of this analysis which is based on data for 20 OECD countries from 1970 to 1996. Monetary policy is operationalized as growth rates of monetary supply of M1. The main results of this analysis, as far as voluntary forms of wage setting are concerned, are summarized in Table 3. Table 3 The Interaction Between the Bargaining Institutions and the Monetary Regime Bargaining institutions The conditional effect of monetary policy caused by a shift from accommodation to non-accommodation Labour costs+ Inflation Change in unemployment Uncoordinated bargaining Insignificant Insignificant Significantly increasing* Pattern bargaining Significantly dampening** Significantly dampening** Insignificant Peak-level coordination, high governability Significantly dampening* Significantly dampening** Significantly dampening** Peak-level coordination, low governability Significantly dampening** Significantly dampening** Insignificant * p ≤ 0.05 ** p ≤ 0.001 + Increase of unit labour costs Source: Traxler et al. (2001: 270). Most importantly, there is only one category of wage bargaining which is not responsive to differences in monetary policy: uncoordinated bargaining. In contrast to this, any form of voluntary coordination of bargaining is sensitive to monetary policy in the way one should expect: that is, a significantly dampening effect of a non-accommodating monetary policy on labor costs. This dampening effect also applies to peak-level coordination under low governability which records the lowest structural capacity for internalizing externalities. This indicates that a non- 24 accommodating monetary policy can indeed work as a substitute for a lack of effective vertical cooperation. Although this tends to increase unemployment, this effect remains insignificant by conventional standards according to the econometric analysis (Traxler et al., 2001). The shift from an accommodating to a non-accommodating monetary policy thus causes significant real costs (in terms of rising unemployment) only in a situation of uncoordinated bargaining. Under these circumstances, given peak-level coordination backed by high governability, such a change in the monetary regime even combines with real benefits in the form of a decline in unemployment. These findings plausibly suggest that the main divide in terms of responsiveness to monetary policy is between coordinated and uncoordinated wage bargaining. The explanation for this observation lies in the fact that any kind of attempt at horizontal (i.e. economy-wide) coordination is of macroeconomic relevance, such that the bargainers involved have to take account of the responses of monetary authorities to their bargaining outcomes. Conversely, the monetary authorities can deliberately target the bargainers and influence their behaviour even in advance of wage accords, by means of signalling the monetary consequences of a certain wage agreement. Put otherwise, coordinated bargaining enables the actors to embark on concertation of wage policy and monetary policy. The real benefits arising from the coincidence of a non-accommodating monetary policy and highly governable peak-level coordination indicate that this combination can most effectively make use of concertation. A possible explanation for this is that peak-level coordination is better prepared than pattern bargaining to comprehensively concert all relevant, interdependent policy fields, such that the benefits of price stability and wage moderation can more effectively translated into positive employment effects. The Dutch ‘miracle’ (Visser and Hemerijck, 1997) is probably the most obvious example of such successful setting. This kind of intentional interaction between wage policy and monetary policy is impossible in the case of uncoordinated bargaining. On the one hand, it is 25 not rational for the bargainers to internalize monetary policy effects because none of the numerous bargaining units is so comprehensive that it exerts a notable macroeconomic impact. On the other hand, the monetary authorities cannot deliberately address the bargainers due to the fragmentation of the bargaining process. As a consequence, a wage policy which is regarded as inflationary by the monetary authorities can be disciplined only ex post in that growing unemployment that ensues from the move to monetary nonaccommodation restricts the scope for further wage increases. This explains the observed increase in unemployment in the wake of a shift to monetary nonaccommodation that coincides with uncoordinated bargaining (Table 3). These findings also help explain why there has been a convergence of labor cost growth across Europe despite the significant differences of the national bargaining systems in their structural capacity for internalizing wage externalities (Table 2). One important determinant of this convergence has been the general shift to nonaccommodation which has been able to have such a notable impact on wage movements, since virtually all European countries have sought to coordinate their wage policy (Table 1). As Table 1 also shows, these efforts have often taken the form of tripartite, state-sponsored coordination which –often captured as ‘social pacts’ in the literature (e.g. Fajertag and Pochet, 1997)– have explicitly aimed at improving one country’s competitiveness by lowering comparative labor costs. As a result, wage setting has converged towards decreasing growth rates of pay to an extent that real wages have remained far behind productivity growth in Europe (Schulten and Stueckler, 2000; Schulten, 2001). Hence, the coexistence of national fragmentation of bargaining and economic integration has created a risk of deflationary wage bargaining rather than inflationary wage policies. Transnational coordination of bargaining, initiated by the unions, is a means of overcoming this risk. As delineated above, however, this project is still in its infancy, such it is too early to empirically examine its performance. 32 Regardless of this, transnational coordination is feasible in the eurozone, since it is the essence of pattern bargaining that its effectiveness does not hinge on an encompassing and complete accord among all the bargaining units of the overall system. What is actually needed is a critical mass capable of setting the pace, while the other bargaining units simply follow. In all the OECD countries recording a long tradition of pattern bargaining (i.e. Japan, Germany and Austria) the critical mass of the pattern setter (i.e. the metal industry) represents less than 15 per cent of the total number of employees (Traxler et al., 2001). Put otherwise, the effect of pattern bargaining clearly transcends the formal scope of the initiating bargaining unit. Likewise, complete participation of all national bargaining systems across the eurozone is not needed for coordination in the context of EMU. In the case of EMU the critical mass for effective coordination across the eurozone would be reached if the coordination process covers a range of bargaining units which are so economically important that their coordinated policies secure the achievement of macroeconomic goals in the eurozone. NOTES 1. One finds less pronounced resistence among employers in a few sectors like construction and road haulage, where incoming workers from low-standard countries outside the EU tend to distort inter-firm competition (Marginson and Sisson, 2001). 2. For instance, one standard assumption is that there exists only one single (i.e. monopoly) union per country. 3. Such a scaling becomes possible only by means of ranking the distinct coordination modes according to their comparative performance. Such an approach, however, results in tautology when using this measure as a predictor of performance. 33 BIBLIOGRAPHY ARROWSMITH, J.; SISSON, K.: “Americanisation or Europeanisation? The Message of Pay Developments”, Paper prepared for the Workshop “Industrial Relations–an Anachronism?”, University of Tuebingen, 15-16 June 2001. BAHNMÜLLER, R.: “The Short-lived ‘End of Modesty’ Versus the Long Arm of the ‘Alliance for Jobs’” in BISPINCK, R.; SCHULTEN, T. (eds.): Collective Bargaining Under the Euro–Experiences from the European Metal Industry. Brussels, ETUI and EMF, 2001, p. 163-198. BAUMGARTNER, J.; URL, T.: “Gemeinsame Geldpolitik–Unterschiedliche Auswirkungen?”, Monatsberichte, vol. 72, 1999, p. 75-88. BUSCH, K.: “Die Wirtschafts-und Währungsunion in Europa und die Konsequenzen für die Tarifpolitik der Gewerkschaften”, WSI Mitteilungen, vol. 45, 1992, p. 267-274. CALMFORS, L.; DRIFFILL, J.: “Bargaining Structure, Corporatism and Macroeconomic Performance”, Economic Policy, vol. 6, 1988, p. 13-61. CAMERON, D.: “Social Democracy, Corporatism, Labour Quiescence and the Representation of Economic Interests in Advanced Capitalist Countries” in GOLDTHORPE, J.H. (ed.): Order and Conflict in Contemporary Capitalism: Studies in the Political Economy of Western European Nations. Oxford, Clarendon Press, 1984, p. 143-178. CROUCH, C.: Industrial Relations and European State Traditions. Oxford, Clarendon Press, 1993. CROUCH, C.: “National Wage Determination and European Monetary Union” in CROUCH, C. (ed.): After the Euro. Shaping Institutions for Governance in the Wake of European Monetary Union. Oxford, Oxford University Press, 2000, p. 203-226. CUKIERMAN, A.: Central Bank Strategy, Credibility, and Independence. Theory and Evidence. Cambridge, MIT Press, 1992. DØLVIK, J.E.: “Economic and Monetary Union: Implications for Industrial Relations and Collective Bargaining in Europe”, Typescript. ECONOMIC POLICY COMMITTEE: “European Employment Pact: The Implementation of the Macro-economic Dialogue”, EPC/II/443/99-Final. Brussels, Economic Policy Committee, 1999. FAJERTAG, G.; POCHET, P. (eds.): Social Pacts in Europe. Brussels, ETUI and OSE, 1997. 34 FAJERTAG, G.; POCHET, P. (eds.): Social Pacts in Europe–New Dynamics. Brussels, ETUI, 2000. FERNER, A.; HYMAN, R. (eds.): Industrial Relations in the New Europe. Oxford, Blackwell, 1992. FERNER, A.; HYMAN, R.: “Introduction: Towards European Industrial Relations?” in FERNER, A.; HYMAN, R. (eds.): Changing Industrial Relations in Europe. Oxford, Blackwell, 1998a, p. xi-xxvi. FERNER, A.; HYMAN, R. (eds.): Changing Industrial Relations in Europe. Oxford, Blackwell, 1998b. FRANZESE, R.: “Credibly Conservative Monetary Policy and Wage/PriceBargaining Organization”, Paper presented at the Workshop ‘Wage Bargaining within EMU’, Vienna, 9 October 2000. GOLDEN, M.: “The Dynamics of Trade Unionism and National Economic Performance”, American Political Science Review, vol. 87, 1993, p. 439-454. GOLLBACH, J.; SCHULTEN, T.: “Cross-Border Collective Bargaining Networks in Europe”, European Journal of Industrial Relations, vol. 6, 2000, p. 161-179. HANCKE, B.: “European Works Councils and Industrial Restructuring in the European Motor Industry”, European Journal of Industrial Relations, vol. 6, 2000, p. 35-59. HEADEY, B.W.: “Trade Unions and National Wage Politics”, Journal of Politics, vol. 32, 1970, p. 407-439. HENLEY, A.; TSAKALOTOS, E.: “Corporatism and the European Labour Market after 1992”, British Journal of Industrial Relations, vol. 30, 1992, p. 567-586. IVERSEN, T.: “Decentralization, Monetarism and the Social Democratic Welfare State” in IVERSEN, T.; PONTUSSON, J.; SOSKICE, D. (eds.): Unions, Employers, and Central Banks. Cambridge, Cambridge University Press, 2000. JACOBI, O.: “Contours of a European Collective Bargaining System” in KAUPPINEN, T. (ed.): The Impact of EMU on Industrial Relations in European Union. Helsinki, Finnish Industrial Relations Association, 1998. KELLER, B.; BANSBACH, M.: (2001), “Social Dialogues: An Interim Report on Recent Results and Prospects”, European Industrial Relations Journal, Annual Review 1999/2000, p. 65-85. 35 MAHNKOPF, B.; and ALTVATER, E.: “Transmission Belts of Transnational Competition?”, European Journal of Industrial Relations, vol. 1, 1995, p. 101117. MARGINSON, P.; SCHULTEN, T.: “The Europeanisation of Collective Bargaining”, http://www.eiro.eurofound.ie/1999/07/study/TN9907201S.html., 1999. MARGINSON, P.; SISSON, K.: “European Collective Bargaining: A Virtual Prospect?”, Journal of Common Market Studies, vol. 36, 1998, p. 505-528. MARGINSON, P.; SISSON, K.: “European Dimensions to Collective Bargaining: New Symmetries within an Asymmetric Process?” Paper prepared for the IIRA 6th European Congress, Oslo, 25-29 June 2001. MARSDEN, D.: “Incomes Policy for Europe? Or will Pay Bargaining Destroy the Single European Market”, British Journal of Industrial Relations, vol. 30, 1992, p. 587-604. NORTH, D.C.: Institutions, Institutional Change and Economic Performance. Cambridge, Cambridge University Press, 1990. OECD: The OECD Jobs Study, Part II. Paris, OECD, 1994. OECD: “Economic Performance and the Structure of Collective Bargaining” in OECD (ed.): Employment Outlook. Paris, OECD, 1997, p. 64-92. PETERS, T.: “European Monetary Union and Labour Markets. What to Expect?”, International Labour Review, vol. 134, 1995, p. 315-332. POCHET, P. (ed.): Monetary Union and Collective Bargaining in Europe. Brussels, P.I.E.-Peter Lang, 1999. SCHULTEN, T.: “A European Solidaristic Wage Policy?”, Paper prepared for the IIRA 6th European Congress, Oslo, 25-29 June 2001. SCHULTEN, T.; STUECKLER, A.: “Wage Policy and EMU”, http://www.eiro.eurofound.ie/2000/07/study/TN0007402S.html., 2000. SOSKICE, D.: “Wage Determination: The Changing Role of Institutions in Advanced Industrialized Countries”, Oxford Review of Economic Policy, vol. 6, 1990, p. 36-61. SOSKICE, D.: “Macroeconomic Analysis and the Political Economy of Unemployment” in IVERSEN, T.; PONTUSSON, J.; SOSKICE, D. (eds.): Unions, Employers, and Central Banks. Cambridge, Cambridge University Press, 2000, p. 38-74. 36 SOSKICE, D.; IVERSEN, T.: “Multiple Wage-Bargaining Systems in the Single Currency Area”, Oxford Review of Economic Policy, vol. 14, 1998, p. 110-124. TRAXLER, F.: “Farewell to Labour Market Associations? Organized versus Disorganized Decentralization as a Map for Industrial Relations” in CROUCH, C.; TRAXLER, F. (eds.): Organized Industrial Relations in Europe: What Future? Aldershot, Avebury, 1995, p. 3-19. TRAXLER, F.: “The State in Industrial Relations: A Cross-national Analysis of Developments and Socioeconomic Effects”, European Journal of Political Research, vol. 36, 1999a, p. 55-85. TRAXLER, F.: “Wage-Setting Institutions and European Monetary Union” in HUEMER, G.; MESCH, M.; TRAXLER, F. (eds.): The Role of Employer Associations and Labour Unions in the EMU. Aldershot, Ashgate, 1999b, p. 115-135. TRAXLER, F.: “National Pacts and Wage Regulation in Europe” in FAJERTAG, G.; POCHET, P. (eds.): Social Pacts in Europe–New Dynamics. Brussels, ETUI, 2000, p. 401-417. TRAXLER, F.: “Die Metamorphosen des Korporatismus”, Politische Vierteljahresschrift, vol. 42, 2001, p. 590-623. TRAXLER, F.: “Bargaining (De)Centralization, Macroeconomic Performance and Control over the Employment Relationship”, Typescript, 2002a. TRAXLER, F.: “Wage Regulation between Industrial Democracy and Market Pressures”, European Sociological Review, vol. 18, 2002b, p. 1-16. TRAXLER, F.; KITTEL, B.: “The Bargaining System and Performance: A Comparison of 18 OECD Countries”, Comparative Political Studies, vol. 33, 2000, p. 1154-1190. TRAXLER, F.; BLASCHKE, S.; KITTEL, B.: National Labour Relations in Internationalized Markets. A Comparative Study of Institutions, Change and Performance. Oxford, Oxford University Press, 2001. VISSER, J.; EBBINGHAUS, B.: “Making the Most of Diversity?” in GREENWOOD, J.; GROTE, J.R.; RONIT, K. (eds.): Organized Interests and the European Community. London, Sage, 1992, p. 206-237. VISSER, J.; HEMERIJCK, A.: “A Dutch Miracle”: Job Growth, Welfare Reform and Corporatism in the Netherlands. Amsterdam, Amsterdam University Press, 1997.