Conference on Organized labour
Responses to the Conference Paper
Professor Mario Regini
University of Milan
18 September 1998.
This is also intended as my contribution to the debate that you want to organize on 'Organized Labour in the 21st Century'. Unfortunately, I did not
have the time to answer your specific questions. You may, however, be interested in a paper that Ida Regalia and I have contributed to a forthcoming publication of the ETUI and which covers some of the concerns of your network. I attach this paper as well.
DIFFERENT TRAJECTORIES IN 1990S EUROPE: DE-REGULATION VS. CONCERTATION
(Paper presented at the SASE 10th International Conference on Socio-Economics, Session on 'Old and New Forms of Class Compromise', Vienna, 13-16 July, 1998 ROUGH AND INCOMPLETE VERSION).
In the 1990s, the challenges raised against the traditional forms of regulation of the European economies, which were based on varying mixes of market, state and associational regulation, have dramatically increased and grow ever more threatening. According to some scholars (see e.g. Dahrendorf 1995), and even more so the mass media, the greatest threat seems to be raised by the globalization of markets - that is, the intensification of international competition - which obliges national economies to rapidly adjust prices, products, technologies and labour forces. The process of European monetary unification, too, imposes similar exigencies of increased competitiveness on national economies, which are no longer able to rely on many of the traditional economic policy instruments of the past, like currency devaluation and various forms of covert protectionism. No less disruptive is the threat arising from the structurally high levels of unemployment which signal the diminished efficiency of traditional forms of economic regulation (Esping-Andersen & Regini 1998). Finally, also the spread of new 'production regimes' (Soskice 1994) and of forms of tertiary and post-Fordist employment (Esping-Andersen 1990) call the conventional modes of regulating labour markets and systems of social protection into question.
These processes imply largely convergent pressures for change, ones that are substantially the same for all European countries. But have the responses to these pressures been uniform as well (or at least are they bound to become so in order to be effective)? Or are the European countries responding (and will presumably continue to do so) to the common challenges in different ways? In the latter case, the crucial question becomes: how can we account for the emergence of different responses in the European economies?
1. The trends to convergence
Obviously supporting the view that national responses are growing more uniform, through mechanisms of either explicit imitation or organizational learning via trial and error, are the numerous versions of a 'theory of convergence' in the advanced economies. This theory, which periodically reappears in the social sciences although it is almost invariably gainsaid by history, considers the role of the processes of change affecting all the advanced economies to be so powerful and decisive that they drastically reduce the possibility of alternative responses to these universal processes or tendencies. At the basis of all convergence theories is the idea that the modernization of advanced economies and societies must follow established paths, essentially dictated by exogenous factors; however much pre-existing institutions and groups of 'losers' may resist them, they can only delay the course of history.
Numerous old versions of this theory - ranging from the emergence of a standardized 'industrial man' (Kerr et al. 1960) through theories of the 'embourgeoisement of the working class' (Goldthorpe et al. 1968) or, alternatively, of the 'degradation of work' (Braverman 1974), predictions of the 'institutionalization of conflict' (Dahrendorf 1959) or indeed its 'withering away' (Ross & Hartman 1960), to theories of the 'end of ideology' (Bell, 196 ) versus the radicalization of a 'new working class' (Mallet 1963, Gorz 196 ) - have all been confuted in their empirical evidence or the logic of their arguments, or else they have simply been contradicted by subsequent events. Nevertheless, the 1990s have seen the accumulation of such a striking series of universal trends and processes, like those mentioned at the outset, and which seemingly enjoin common responses, that convergence theories have regained credibility among scholars and policy-makers.
In particular, the tendencies towards the globalization of markets, on the one hand, and persistently higher levels of unemployment in Europe compared with North American ones, on the other, are at the origin of the widely perceived need for greater de-regulation of labour markets, of industrial relations and of welfare systems; a perception which easily turns into a generalized prescription, dispensed by economists, central banks and supranational bodies to the European governments on an almost daily basis.
(16)
Since this prescription has already been amply applied in the Anglo-American model of capitalism, the implication is that all European economies are expected to converge on this model. The view widely shared by mainstream economists and the international institutions on this issue is, at bottom, nothing but a rather crude version of a theory of convergence. But this theory has regained the lost hegemony in more indirect and sophisticated forms as well. An example is provided by the debate on the decline of neo-corporatism - that is, the "decline of institutional arrangements for collaborative or tripartite governance of labor markets by representatives of capital, labor and the state" (Pontusson & Swenson 1996, p. 224); or by analyses - highly pessimistic regarding the future of the German model of capitalism - of the consequences on this country's regulatory system of the 'regime competition' triggered by the process of European economic unification (Streeck 1997).
However, the main problem with a new version of convergence theories is that, in numerous countries, certain recent changes seem to move in a direction opposite to - or at least at odds with - the phenomena of deregulation, decentralization and unilateral initiative in labour markets, industrial relations and welfare systems considered to be universal trends. It is above all the increasing reliance on centralized social pacts to cope with the new challenges, and therefore to moderate wages, flexibilize the labour market and reform the welfare state (a practice which in the 1990s has spread with greater or lesser success among several continental European countries) that seems to empirically gainsay the inevitability of common responses. Some scholars are even tempted to advance the alternative hypothesis, namely that the European countries may be now converging on a revival of concertation, rather than on the deregulation of their economies (Fajertag & Pochet 1997, Dore 1994).
Aside from the differing popularity of these two views, they commit a symmetrical error of perspective. Both of them underestimate the profound differences among the European countries in whichever dimension that they believe is bound to prevail; or else they explain these differences simply in terms of delays or of different stages in the same process. Moreover, both focus their attention on changes in some areas of socio-economic regulation most exposed to the new challenges, while they neglect other and equally decisive areas.
Thus, a good part of the economic and political science literature on the decline of neo-corporatism focuses on trends towards bargaining decentralization and the flexibilization of labour markets while neglecting the differing degrees of cooperation among organized interests in these processes, as well as in those of welfare reform. Moreover, the uniformity of these tendencies is - as we shall see - clearly overestimated. The differences among European countries are often conceptualized as a contrast between radical ruptures and changes made slower and more difficult by institutional stickiness (see e.g. Silvia 199 regarding attempts to decentralize wage bargaining in Germany, which have been less successful than in Sweden), rather than as the outcome of alternative trajectories. Conversely, the sociological literature on the revival of concertation tends to focus on the search for social pacts for employment and competitiveness, which naturally give priority to consensual incomes and welfare reform policies, but it neglects the tendencies - present everywhere but of greatly varying magnitudes - towards collective bargaining decentralization and labour market deregulation. Since at least some debate on the desirability of a social pact in order to meet the new challenges has emerged in almost all the European countries, implicit in this perspective is a simple distinction between successful pacts and (temporarily) unsuccessful ones, which is once again related to a difference in gradation, rather than to alternative trajectories.
Hence, although both perspectives claim to describe (or to prescribe) universal trends, the empirical evidence they provide is based solely on some countries and on certain areas of socio-economic regulation, while it is apparently contradicted by other countries or regulatory areas. These shortcomings indirectly provide grist for the mill of 'neo-institutionalist' theories, which are the most determined and best equipped to counter the thesis of the inevitable convergence among advanced economies. Yet I shall seek to show that nor do these theories offer a satisfactory explanation of the different alternatives pursued by the European countries in response to common challenges.
2. The role of institutions
Neo-institutionalism traditionally enjoys greater prestige in political science, especially among scholars of political economy and industrial relations. And this theory too comes in numerous versions (for an excellent discussion, see Hall 199 ). The basic idea, however, is that pre-existing institutions play a key role in shaping the responses to exogenous factors, by acting as a filter, or an intervening variable, between external pressures and the responses to them. The institutional context, in fact, provides actors with a set of resources and constraints which they must necessarily take into account when choosing among different alternatives, and which consequently shapes their action. Since institutional contexts vary from one country to another, given that they are rooted in their histories, the neo-institutionalist perspective has no difficulty in recognizing and explaining the divergence among responses to common challenges: different alternatives are pursued because the pre-existing institutions, and their role in mediating the impact and direction of change, are different.
Traditionally, the objective of neo-institutionalist theories has been to draw up typologies which capture the crucial characteristics of institutional arrangements from the point of view of their impact on the underlying economic-social variables. Various examples can be cited of typologies which have sought to interpret the divergences in the areas discussed in this article - that is, in the regulation of industrial relations, of the labour market, and of welfare systems.
Among the best known, mention should first be made of the modes of interest intermediation called respectively neocorporatist and pluralist (Schmitter 1974, Lehmbruch 1977); or of the types of welfare systems, which were originally distinguished into institutional-redistributive, residual and meritocratic-particularistic (Titmuss 1974), but then repeatedly redefined to give rise to the currently most widely accepted typology comprising the Scandinavian, Anglo-Saxon, and Continental European models (Esping-Andersen 1990). Also the organization of production - which labour economists and sociologists often only study as a consequence of technological change and managerial strategies - is viewed in this perspective as a choice closely conditioned by the institutional context. This has given rise to the concept of 'production regimes', which were simply distinguished between Fordist and post-Fordist in early studies (Piore & Sabel 1984, Boyer 1986), while the subsequent literature has shown that post-Fordist production may follow different paths according to the institutional environment in which it is located: German 'diversified quality production' (Sorge & Streeck 1991) is different from Japanese 'lean production' (Dore 1990), and both are different from the British 'low road' based on a 'low wage - low skill - low quality equilibrium' (Soskice 199 ). Finally, the distinction between different 'models of capitalism' - the Anglo-American and the Rhenish, to which the Japanese one is sometimes added (Albert 1991, Berger & Dore 1997, Crouch & Streeck 1997) - covers most aspects and institutional variables taken into account by the previous typologies.
Also the two polar types of response to the common challenges discussed in section 1 - the one based on unilateral initiatives of deregulation and decentralization, and the one that instead aims to achieve concerted re-regulation among central actors - yield a typology which, although it has not been explicitly proposed as such, could easily be included in the neo-institutionalist strand. Using standard terminology to emphasise the key aspects of each, one could call the two alternative responses respectively 'deregulation' (a response which seeks to create greater room for the market by eliminating the constraints imposed by other institutions) and 'concertation' (an attempt to reinforce forms of regulation which combine the roles of the interest associations and of the state to produce a mix of control and consensus), and investigate the role of the pre-existing institutions in inducing actors to choose one or the other response. By way of example, one could show that industrial relations systems with traditionally B unions, legislation on co-determination and on bargaining rights, etc., prevent employers and governments from following the deregulation road. Examples of this kind, which use existing institutional arrangements to account for trends in one or the other direction, are widespread in the political economy literature.
However, also neo-institutionalist theories, which stress the structural divergence of the responses by the advanced economies to common challenges, have their shortcomings. First, they yield a static picture, which provides little room for understanding internal tensions and attempts to change, even less for assessing their implications in terms of instability of the response adopted. The scheme of analysis that they typically use is a functionalist one, which relates a given institutional context to the policy outcomes that this context tends over time to foster or allow. Omitted from analysis, and therefore unexplained, are the actors' attempts to change such outcomes and to impose different solutions, which may not achieve immediate success, but which reveal tensions that may build up to the point of eroding the consensus necessary for the traditional solution to work.
The second and more important shortcoming of neo-institutionalist theories is that they cannot account for the variability of solutions to be found in each country, when different areas of socio-economic regulation are jointly considered. As anticipated in note 1, in the next section I will deal with three areas of particular importance for assessing the type of response made by each country to the common challenges: the labour market regime, the wage bargaining system, and the welfare state. In each country, the institutional context that influences the responses prevailing in these three areas is basically the same: whether the object is the labour market or the welfare state, the policy-making institutions are the same, and so too are the industrial relations system or the legal framework. Consequently, the responses provided by an economy to external challenges should, theoretically, move in the same direction, namely either towards deregulation or towards concertation. However, as we shall see in the next section, the processes of change that have actually occurred in the last ten years in Europe show that this almost never happens. In some countries, there have been B tendencies towards the decentralization of wage bargaining, but no serious effort to deregulate the labour market; in others, attempts have been made to unilaterally alter the welfare system, while flexibilization measures have been concerted and collective bargaining has been re-centralized; and so on. In short, analysis of recent trends shows that each national economy may adopt different solutions according to the area of socio-economic regulation concerned, so that it is not possible to identify highly consistent 'national responses'.
For reasons of space, I shall try to synthesise in a synoptic table (see Appendix) the actual responses of European countries in the three areas specified above, discussing only the most significant examples in some detail. After showing that these are indeed different responses but not sharp alternatives - even less internally consistent solutions, related to pre-existing institutions - in the final section I shall address the crucial issue of the factors that may account for these differences.
3. The responses of European economies: de-regulation versus concertation
As repeatedly mentioned, the areas of socio-economic regulation on which I shall focus are the labour market regime, the wage bargaining structure, and the welfare system. These are three of the aspects most radically affected by the new challenges, and their change is therefore at the centre of scientific debate and at the top of the political agenda. Regulation of the labour market and of the welfare system involves several institutions and rules, whose overall change is difficult to grasp. In order to enable comparison among actual trends in the European countries, along the horizontal axis of the synoptic table in the Appendix I shall focus on a few sub-areas which have been subject to major pressure for change in recent years: more specifically, as regards the labour market, I shall restrict comparison to the regulation of atypical forms of employment; and as regards the welfare system, the analysis will be confined to the reform of pensions and unemployment benefits.
Summarized along the vertical axis are the directions of change in each of the three areas - that is, the different ways in which the European countries have responded to the new challenges.
(17)
Indicated first is what have traditionally been considered to be the main alternatives for each area: rigidity or flexibility in labour market regulation, the centralization or decentralization of wage bargaining, expansion or reduction of the welfare system.
(18)
However, in recent years in Europe, one discerns no actual shift towards the former of these alternatives in any of the three areas considered. For example, although the German, French, Italian or Spanish labour markets are more rigid than the British one, recent trends in them have not been towards greater rigidity, but towards flexibilization. And since the comparison concerns not the regulatory systems of European countries but their directions of change, the traditional alternatives seem to be of little relevance. This by no means implies that the diversity of responses is disappearing - as argued by convergence theories - only that it should be reconceptualized so that it captures the real alternatives.
Thus, on closer inspection, the directions of change in the regulation of labour market entry are still divergent, even though in no country has legislation or collective bargaining sought to make it more rigid. National cases of extended or even 'generalized' flexibilization of entries, like those of Denmark and Spain to some extent (cf. Toharia 1998, Björklund 1998) - besides the British case, where various other aspects of the labour market are being deregulated (Deakin 1998) - are matched by others in which 'controlled' or selective flexibilization has occurred. In The Netherlands, for example, atypical contracts (especially part-time or fixed-term) have been greatly encouraged, but within the framework of regulations (so-called 'flexicurity accords') designed to extend to them elements of the social protection traditionally afforded to permanent employees (Gorter 1998, Visser & Hemerijck 1998). In Italy, agency temporary contracts, fixed-term contracts, etc., have been liberalized, but within a regulatory framework which specifies their recipients (young people, or workers with specific skills, or in geographical areas where 'territorial pacts' are being implemented), their amount and duration (Samek 1998, Adam & Canziani 1998).
The logic behind extended or generalized flexibilization is that of deregulating the labour market, reducing to a minimum the rules that regulate individual behaviour and the functions performed by state institutions and associations, in pursuit of a 'pure market' ideal. "The essence of the deregulatory approach is to increase the autonomy of individual firms and workers to respond to shifting markets. Within labour markets, this has usually been taken to mean that the institutional structures regulating them, be they of state or collective-bargaining origin, should be dismantled" (Marsden 1995, p. 83).
Controlled or selective flexibilization strategies instead start from the premise that not only is the sheer deregulation of labour markets not feasible; it is not advantageous in terms of greater flexibility either. Firstly, the result of attempts to dismantle certain institutions or rules may be that their place is taken by other, equally constraining regulatory mechanisms; or they may even generate hidden rigidities. Secondly, since the same institutions that impose constraints on firms may also provide them with resources or competitive advantages, the cumulative effect of deregulation may prove to be negative (Regini 1998). For these reasons, the purpose of recent changes in the regulation of the European labour markets has been less to achieve generalized flexibilization than to introduce 'controlled derogations' to the general rules, which are left substantially unchanged.
The collective bargaining structure, or the wage-determination system, has displayed more sharply diverging trends among the European countries in the last ten years. The late 1980s were marked by a general and powerful thrust towards decentralization which was especially evident in Great Britain (Edwards et al. 1998) and Sweden - a country which had long pursued wage solidarity policies through the primacy of centralized collective bargaining (on the radical changes in Sweden see Pontusson & Swenson 1996, Iversen 1996, Kjellberg 1998). Nor did countries where the central level of bargaining remained, or had become once again, predominant (Austria, Norway, Finland, Belgium, The Netherlands, on which see the respective chapters in the volume edited by Ferner & Hyman 1998) seem immune to a decentralization of the wage determination process.
In the 1990s, this trend seems to have been reversed as a result of increasing recourse to centralized social pacts aimed at promoting incomes policies able to restore competitiveness to the national economy (Dore 1994). Emblematic in this regard is the Italian experience, which has seen governments reach important tripartite agreements on incomes policies with the social partners, and successfully negotiate a number of laws reforming social security and public-sector employment (Regini & Regalia 1997). Several other European countries have engaged in tripartite concertation in these same years. Besides the countries mentioned above, in which the central level was traditionally predominant, also in Ireland (Taylor 1996, O'Donnell & O'Reardon 1997, Von Prondzynski 1998), and more recently in Portugal (Campos Lima & Naumann 1997) and in Greece, there has been a B thrust towards the centralized bargaining of incomes policy and the signing of formal tripartite agreements.
However, closer inspection shows that centralization processes are not involved in these cases either. As emerges most clearly from the contents of the tripartite agreement of July 1993 in Italy, the new social pacts do not envisage recurrent top-level bargaining as in the classical Scandinavian experience; namely, the pattern of centralized and detailed determination of wages and working conditions which, in the 1980s, broke down in Sweden because companies found it incompatible with their need for flexibility (Iversen, 1996). The new social pacts generally establish rules and procedures for the conduct of collective bargaining, but do not fundamentally constrain industry- and company-level negotiation; nor do they prevent firms from designing different incentive structures. Although the institutional mechanisms differ, the logic of incomes policy which has inspired the new social pacts is rather similar to the Austrian system. The central level of bargaining does indeed influence the overall wage dynamics, but at the same time assigns the task of determining the relative wage levels to decentralized negotiation. Franz Traxler has conceptualized this trend as 'organized decentralization', thereby contrasting it with the 'disorganized decentralization' of the collective bargaining system characteristic of countries like Great Britain, the United States and New Zealand (Traxler 1995). The way in which bargaining decentralization is brought about should be considered jointly with the existence, or otherwise, of different informal mechanisms to coordinate the wage dynamics. These mechanisms place some countries in the category of 'coordinated market economies', which according to Soskice (1990), display responses to the common challenges which differ sharply from those of the 'uncoordinated market economies'.
By combining these two concepts, we can distinguish between tendencies towards the coordinated decentralization of wage bargaining- typical both of those countries that have traditionally pursued this strategy (see the second column sub a) and of those that have more recently started along the road of social pacts - and tendencies towards uncoordinated decentralization. Among the latter, besides Great Britain and Sweden, which have made major steps in this direction, one might include (in brackets) countries like Germany and Denmark, in which a shift towards more marked wage bargaining decentralization has in reality not occurred but has for some time been a priority goal for key actors (mainly the employers' associations) which exert B pressure in this direction (see Silvia 199 , Thelen 1998). As we shall see in the final section on possible explanations for the divergences among European economies, pressures for change which have not (yet) achieved their objective are an important indicator of the actors' overall strategies - strategies on which the choice among different alternatives ultimately depends.
Finally, the welfare system has been the target of attacks, or at least of proposals for reform, in all the European countries, albeit for a variety of reasons. Together with health care systems, continental social security systems have become the largest item in public spending, which has had be cut especially in order to meet the 'Maastricht parameters'. Even when the convergence criteria have been fulfilled, the worrying demographic trends and the low labour market participation rates that characterize many European countries help to keep the objective of a substantial scaling-down of public pensions systems among the priorities on the policy-makers' agenda. As for unemployment benefits, these have been cited as one of the causes of the poor functioning of the labour market, at least in those countries, like Denmark, which in the past have made what is now deemed over-generous provision. In short, in all the European countries with social security systems that bear down heavily on public finances, some proposal for cutbacks has been made or is on the agenda.
However, since a reduction in benefits clashes with the interests of important social groups, two options are available to achieve it: either policy-makers seek to gain the consent of those groups and their representatives, or they seek to eliminate the rules and customs that enable the latter to exercise veto powers over change. In this area of socio-economic regulation, too, one therefore finds the more general alternative between deregulation and concertation. In fact, some governments have given priority to incisive and coherent action at the expense of consensus; others have preferred to pursue more limited objectives as long as they can count on being supported by interest organizations which guarantee acceptance of changes by their members.
A clear example of the two alternatives is provided by the reform of the Italian pensions system undertaken by the Dini government in 1995, on the one hand, and the attempts at reform made shortly afterwards by the Juppé government in France, on the other. The Italian solution has made the spending cuts envisaged much more gradual than is generally regarded as appropriate, but the consensus on the reform provided by trade unions able to obtain more or less convinced endorsement by workers has nevertheless yielded a positive result. In France, on the other hand, the Juppé government's unilateral action on pensions has suffered a severe setback and has spurred the longest and largest collective mobilization since 1968. In Germany, too, the government has shown much greater impatience than in Italy, imposing unilateral solutions on the problem of reforming the social security system - solutions, moreover, that do not involve radical changes. Similar alternatives could probably be identified as regards cuts in unemployment benefits, which have been produced following different approaches [Data to be provided].
This brief comparison among the alternatives pursued by the European economies in certain areas of socio-economic regulation allows some conclusions to be drawn regarding the first question asked in the introduction (namely, to what extent have responses to common challenges been uniform or instead different). Firstly, a glance at the synoptic table in the Appendix suffices to show that, in each area, countries have moved in different directions. The range of variation is narrower and more finely-shaded than the traditional dichotomies (rigidity-flexibility, centralization-decentralization, expansion-reduction) suggest, but the alternatives adopted are nevertheless real.
Secondly, the European countries are indeed moving in different directions, but not in a consistent manner across the various areas of socio-economic regulation. Apart from a few cases in which one discerns largely congruent changes (like Great Britain's move towards more extended deregulation, or Italy's in the opposite direction of concertation), most countries have shifted towards one of the two poles in one area but towards the other pole in another area. In Sweden, wage bargaining has been dramatically decentralized since the late 1980s and continues to be pushed in this direction, but the Swedish welfare state has not been subjected to analogous deregulatory changes. In Germany, attempts to cut welfare unilaterally have been made, and even Ber have been the pressures to decentralize wage bargaining, but no actor seriously considers deregulating the labour market. In Denmark, where labour market flexibility is instead rather extended, reform of the unemployment benefits system has been based on a search for trade-union consensus. And further examples could easily be provided.
Thirdly, only a few countries seem to have chosen among the available alternatives in an unequivocal, stable and widely agreed manner. In most others, choices appear to be provisional, subject to B tensions and to repeated attempts by crucial actors to redefine them. I have already cited the German case as emblematic: here, the employers have forcefully pushed for wage bargaining decentralization, while the unions have sought to compensate the welfare cuts envisaged with a Bundnis für Arbeit (Alliance for Employment). Both attempts have temporarily failed but they will keep being on the agenda, especially if the Social Democrats should win the next elections.
In short, the contrasting alternatives of deregulation and concertation are still actual poles towards which the European economies can move, as well as being ideal types useful to synthesise options often beset by contradictory choices. Yet the concrete experiences of European countries display a much more composite, confused and uncertain pattern. Above all, the different national responses appear increasingly less to be static and permanent solutions determined by pre-existing institutions, and increasingly more to be differences in the processes of change, which give rise to uncertain and precarious results. If the institutional context is not (any longer) the key independent variable, what alternative explanations can be offered?
4. How can we account for the different responses to the common challenges?
There are various partial explanations in the literature for the emergence of different alternatives in the European economies. Each of them provides some elements useful for understanding of the divergences, although they are still unsatisfactory as overall explanations. The first and simplest interpretation relies on power relationships among the actors involved to account for the different responses. Since the variability of power relationships may depend not only on structural factors but also conjunctural ones, this explanation accounts for the often precarious and unstable nature of the alternatives pursued in different countries. For example, the marked prevalence of concertation in Italy in the 1990s, unlike in the 1980s, is often explained in terms of the dramatic weakening of its actors, which therefore "needed to be able to rely on each other" (Salvati 1995). There is no doubt, in fact, that confidence in the role, and above all the public image, of Italian employers has been eroded by their close involvement in the Tangentopoli scandals. And equally weak has been the legitimacy of the political class, either because it was identified with the old regime, or because it was based on technocratic credentials which did not derive from democratic elections. However, this interpretation cannot be generalized to other countries. In Sweden, the trade unions are still very powerful (indeed, almost uniquely, they have grown even Ber in the last ten years), but this has not prevented the formation of 'inter-class coalitions' consisting of sectoral employers' associations and unions, which have led to the breakdown of centralized collective bargaining and the solidaristic wage policy (Pontusson & Swenson 1996). Conversely, in The Netherlands and Ireland the unions are rather weak, with scant presence in the workplace, but the concertation option has increasingly gained strength and become institutionalized.
A second interpretation relates, not to power relationships, but to the varying needs of the actors in the production system. What policies are most relevant for them depends on a country's economic structure. Of course, all countries combine labour-intensive, low-value added and high-skill based firms, but the relative balance of a nation's production structure is critical. If, like in Britain or in Spain, a large part of the economy has traditionally been biased towards simple mass-production, there is clearly a more pressing need for extended wage and numerical flexibility. Where the bias is towards high-value added products, selective strategies oriented to internal flexibility and an adequately educated and cooperative labour force may of greater concern (Esping-Andersen & Regini 1998).
Paradoxically, however, the needs of actors sometimes seem antithetical to those that should stem from the traditional working of their economic system. One thinks, for example, of the different positions recently taken up by the German and Italian employers' associations. It is well known that the Italian production system, in which small firms predominate, bases its competitiveness mainly on high versatility and rapid adjustment to changing markets; a versatility and rapidity made possible by what I have elsewhere called "an interweaving between weak institutional regulation and effective but unstable voluntaristic regulation" (Regini 1997, p. 107). The German productive system, by contrast, draws its main competitive advantages from the capacity of its institutional context to efficiently provide firms with such collective goods as labour force coperation, coordination of the wage dynamics, and highly- and broadly-skilled human resources. And yet the principal concern of the German and Italian employers' associations does not seem to be that of strengthening their respective competitive advantages, by enhancing the institutional structure on which these are based; indeed, exactly the opposite priorities are pursued.
Both actors are aware that, in advanced economies like the European ones, increased competitiveness and decreased costs depend on the one hand on greater flexibility of firms and labour, but on the other on the greater ability of the national system of which they are part to provide them with 'collective goods'. These two common problems, however, have been traditionally met in different ways and to different extents in each country: hence, the urgency and the priority given today to each of them varies between one country and the other. Put briefly, one can argue that Germany's current main problem is that of flexibilizing its economy, even at the risk of weakening those collective goods for firms which have traditionally been its principal competitive advantage; hence the recent obsession of German companies to redress their lack of flexibility vis-ą-vis their foreign competitors. Conversely, the crucial problem for Italian firms today is institutionalizing the production of collective goods - traditionally unstable and uncertain in that country - more than enhancing the flexibility which hitherto has been its Best suit. Consequently, in spite of frequent claims that they have to operate in a hostile political-trade union environment and in a labour market badly in need of greater flexibility, Italian employers' associations have de facto attached the greatest value - and have repeatedly stated their willingness - to maintain concertation at any cost (Regini and Regalia 1997).
This interpretation therefore suggests that responses by the European economies differ because objectively there is a trade-off, a contradiction between deregulation and concertation, between more market and a more adequate supply of collective goods. Since both these conditions are necessary for economic performance, but the mix between them is different in each economy, national actors tend to give priority to the condition that is less present in their system, even at the cost of placing their traditional competitive advantages at risk. Though attractive, however, this interpretation relies excessively on rational behaviour by actors, as well as being difficult to extend to other countries. British firms operate in a deregulated context with scant production of collective goods, but they do not seem overly concerned to redress the balance. Nor, on the other hand, do Austrian firms, which find themselves in the opposite situation, seem as determined as German firms to radically change the context in which they operate.
A third interpretation of the divergent responses to the common challenges hinges neither on power relationships among actors nor on their needs, but on the country's position in the international system. As regards the European economies, in particular, their differing propensities for deregulation or concertation may depend on the extent to which a country is peripheral or central to the process of monetary unification - that is, they may depend on the degree to which it was both necessary and difficult for the country to meet the convergence criteria in the first place and to keep abiding to them now (Garrett 199 ). Whereas countries more central and/or less beset by the problems of monetary convergence (Germany, France, Great Britain, Sweden) have displayed more marked tendencies towards deregulation, the weaker countries from this point of view (Italy, Spain, Portugal, Ireland, Belgium) have given far higher priority to concertation, as it embodies a national effort to achieve a goal which does not permit internal divisions. Although this interpretation probably also contains an element of truth, it does not fully account for a number of important national cases (for example the Dutch, where concertation has been especially pronounced even though fulfilment of the 'Maastricht parameters' was a foregone conclusion). Also, it relates the divergent responses to a process (monetary unification) whose effects became manifest after these divergences had already taken full shape.
A last interpretation is often implicitly suggested in the scientific debate, and even more in the political and journalistic ones; an interpretation which, if sound, would have the merit of reconciling convergence theories with neo-institutionalist ones. In its simplest terms it can be formulated as follows. The responses by the European economies to the common challenges differ precisely because they are intended to converge on an intermediate model. Since the pre-existing institutions have shaped national strategies giving them quite different features, a process of convergence can only follow different routes. Sweden, for example, has been forced to decentralize its wage bargaining while Italy has had to find instruments to coordinate it, simply because both economies tend to converge on an intermediate model that has proved to be more efficient, namely the German type of coordinated industry-level bargaining.
Starting from frequently opposite positions and seeking to imitate the dominant model (which may come about unconsciously through some sort of 'echo effect' or learning process that is by now Europe-wide), the other countries are obliged to move in directions that are only temporarily - and therefore apparently - divergent. There is no doubt about which is the hegemonic system in Europe according to this interpretation: the German one, of course. Martin Rhodes (1997), for example, regards recent trends in various countries to be attempts to converge on a model of 'competitive corporatism' - for which the German and Dutch institutions are best equipped - with the abandonment of both Swedish-style 'social corporatism' and English-style laissez-faire.
This interpretation, too, seems quite convincing at first sight, with the added advantages that it resolves the conflict between convergence theories and neo-institutionalist ones, accounts for processes of change rather than static outcomes, and is not tied to contingent factors. However, it is unable to explain why it is precisely the system towards which the others seem to be converging, because it has proved to be the most efficient - the German model - the one that is currently most tension-ridden, with B pressures for change applied by all its actors (primarily the employers, but also governments and unions).
Let us therefore try to recapitulate the most convincing aspects of the interpretations discussed so far and then suggest a further explanation. The new challenges require the European economies to strike some sort of balance between the opposing needs of a deregulation of labour markets, industrial relations and welfare systems, on the one hand, and of concertation able to provide a social pact for national competitiveness, on the other. Given the different points of departure, power relations, capacity for institutional learning, and country position in the monetary convergence process, the shared requirement to strike a balance between opposing needs will induce actors to adopt different strategies: decentralization or coordination, external or internal flexibility, a search for cooperation and trust or unilateral action. As they follow these different routes, it is quite possible that the European economies will achieve relatively convergent outcomes, ones able to mediate the contradictory aspects of these alternatives and give rise to something similar to an intermediate 'competitive corporatism' model.
However, a crucial element is missing in this synthesis and should be added to it. The more closely European economies approach what may be considered an intermediate model, or a reasonable point of convergence between opposing requirements, the more each relevant actor is likely to display uncertainties, ambiguities and difficulties in setting collective priorities. The German case shows that the dilemmas, internal conflict and the weakness of the basis of internal solidarity are especially manifest among employers, to the point that they bring the industrial relations system dangerously close to breakdown (Thelen 1998). Something similar now seems to be happening within the Italian employers, at just the moment when the choice of concertation - Bly supported by their association - has helped them to achieve important goals and to bring the Italian system closer to the dominant intermediate model. But not dissimilar are the tensions which, in the countries that have gone furthest along the road towards this intermediate model, are apparent in the opposing front, the trade unions.
This should not come as a surprise. As long as a national economy suffers from a clear bias towards one or the other alternative, and striking a balance still appears far off, it is relatively easy for each collective actor to reach internal agreement on priorities, on alliances, and on the strategies to pursue. The very cognitive and interpretive framework that actors adopt is Bly influenced by implicit comparison with the model deemed to be superior. This relative 'epistemic unity' which holds together a collective actor, however, begins to break up as imbalances in one or other direction are redressed and the priorities adopted lose their obviousness. Faced with these mounting uncertainties, how do actors behave when they must finally choose between the alternatives available? The answer is not easy, and would certainly require more systematic comparative research.
A preliminary analysis of the sources listed in note 2 nevertheless suggests an interesting hypothesis. To the extent that the processes described above apply to a certain economy, the choices made by the relevant actors - and ultimately the responses of that economy to the common challenges - appear less conditioned by pre-existing institutions than they are based on an interactive game among the actors themselves. The most relevant actors in the areas of socio-economic regulation considered by this article are governments, employers' associations and trade unions. In economies that approach the 'intermediate model' having found reasonable balances between opposing needs, each of these actors has less incentive to question existing institutional arrangements if also the others give convinced support to these arrangements. This has been apparent in Italy, for example, where Confindustria showed an unexpected determination to defend concertation, despite internal dissent, until the government's decision to introduce the 35-hour working week by law; a decision that could be interpreted as a breach of the cooperative game by another actor.
However, if exogenous factors induce an actor to revise its traditional approach, it is possible that the other actors will no longer be able to keep their inner ambivalences and uncertainties under control, so that internal cleavages or even fragmentation prevail. If this happens, cross-class alliances of the type analysed by Pontusson and Swenson (1996) in the Swedish case are more likely to arise. By calling the rules regulating the interactive game among actors into question, these cross-class alliances weaken existing institutional arrangements and lead to their change. Hence, in economies in which the pre-existing institutions are less able to shape outcomes, the position of each actor becomes more capable to influence those of the others. In that case, the route followed by an economy - towards greater deregulation or greater concertation - becomes the cumulative result of their interactions; a result which, more than in the past, is an unstable and precarious equilibrium.
Appendix: Synoptic Table AREAS OF CHALLENGE
Labour market regime Wage bargaining Welfare system
(e.g. atypical contracts) (e.g. pensions, unemployment benefits)
MODES OF RESPONSE (or
DIRECTIONS OF CHANGE)
a) Rigidity vs. a) b)
b) Flexibility
(Controlled flexibility vs. D GB
Generalized flexibility) NL DK
I E
F
S
a) Centralization vs. a) b)
b) Decentralization
(Coordinated decentraliz. vs. NL A GB
Uncoordinated decentraliz.) I N S
EI B (D)
P SF (DK)
a) Expansion vs. a) b)
b) Reduction
(Consensual reduction vs. I F
Unilateral reduction) DK GB
S (D)
A=Austria, B=Belgium, D=Germany, DK=Denmark, E=Spain, EI=Ireland, F=France, GB=Great Britain, I=Italy, N=Norway, NL=Netherlands, P=Portugal, S=Sweden, SF=Finland
The prospects of the Italian trade unions in a phase of concertation
by Marino Regini and Ida Regalia
Continuity and Change in Italian Trade Unions
In few other European countries has the history of trade unionism been marked by so many phases displaying such diverse features as in Italy. And probably few other countries have seen the trade-union movement so radically change its strategy of action in the course of time, albeit with B elements of continuity. This is especially striking if one considers the logic of Italian trade-union action in the 1990s, which is the theme of the analysis that follows.
However, before broaching this topic, it is advisable to outline the main features of Italian trade-unionism, highlighting in particular those that explain the relative ease with which it is able to change. As we have argued elsewhere (Regalia and Regini, 1995; 1998), this takes us back to the years when free trade unionism was reconstituted in Italy after the fascist period, and to the unexpected consequences of the contradictory and ambivalent process then set in motion, with its twofold set of dualisms.
Trade-union action between overt informality and covert institutionalisation
In the buoyant political climate of the immediate post-war years, and amid widespread expectations of political change, the Italian trade union was reconstituted in 1944 on the initiative of all the anti-fascist parties as an organization which provided general and unitary representation for the entire labour force, with no distinction drawn between members and non-members. However, this unitary trade union - which immediately attracted a large following among the country's workers - was internally divided among party factions and therefore highly vulnerable to volatile events on the political stage. In 1948 it split along ideological cleavages, giving rise to the three confederations that still today constitute the mainstay of the Italian trade-union system: CGIL, CISL and UIL, originally linked respectively to the Communists and Socialists, the Christian Democrats, the small lay parties - the Republicans, the Social Democrats and the reformist wing of the Socialists.
The organizational weakness of the unions after the split, which was exacerbated by labour's weakness in the market until the beginning of the 1960s, fostered the growth of an industrial relations system characterized by a high degree of informality and voluntarism, and by a low level of institutionalization - that is, of formalization and fixed rules regulating relations among actors. On the other hand, the initially B support enjoyed by the unions (especially CGIL, which thereafter was invariably the union with the largest membership) (see Table 1), and the union organizations' close links with the political parties, encouraged the development - covert, but with effects that worked in the opposite direction to the previous process - of the institutional involvement of the social partners - and the unions especially - in the implementation of social policies, and at the same time of widespread but indirect intervention by the public institutions in the economy and in labour relations.
The former and more overt of these features (the pronounced voluntarism and the low institutionalization of the industrial relations system) made it relatively easy for industrial relations actors - and the union in particular - to alter their strategies according to the ongoing state of the power relations among them. The other, much more hidden features (the institutional involvement by the unions and the intervention by the state in the economy and in labour relations) instead helped to consolidate the unions' influence, independently of pure power relations and market power.
We use the expression 'high voluntarism and the low institutionalization of industrial relations' firstly to denote the fact that a legal framework for the constitution and operation of unions and interest organizations in general has never been defined in Italy. The articles of the 1947 Constitution which sanctioned the rights to form trade unions and to strike, but which postponed detailed regulation on how to achieve this goal to subsequent legislation, were in fact never implemented in the new situation of competitive trade unionism based on party-ideological affiliation - dominated by the CGIL - that arose after the split-up of 1948. Therefore, the arena of representation continued to be relatively open to new-comers, and this helped the rank and file to raise challenges against the strategies of the larger organizations - as exemplified by the growth of 'autonomous' union in the services sector during the 1960s and '70s, and by the recurrent emergence of militant opposition ("cobas") against the confederal unions in the 1980s and '90s. One notes in passing that this tendency towards the dispersion and fluidity of representation has been no less marked among employers (Lanzalaco, 1995; Martinelli, 1994).
Voluntarism, or informality, and the low institutionalization of industrial relations meant, secondly, that a substantial absence of certain rules was also a long-standing characteristic (until the early 1990s, as we shall see) of workplace representation systems, whose features and names changed repeatedly over time, giving rise to numerous functional forms. In many cases, however, this absence of rules enabled the growth of bodies with sufficient flexibility to adapt reasonably successfully to the changing requirements of intervention and representation (Regalia, 1995). Voluntarism and a low level of institutionalization also meant that the lack of rules on strikes (until the 1990 law regulating strikes in essential public services) fostered recourse to conflict by both the confederal unions and the small groups which operated independently of - and in opposition to - them. Consequently, not only did Italian strike figures long occupy the top place in international statistical tables (Franzosi, 1995) but industrial action often responded more to expressive and symbolic needs than to instrumental rationality (Regini, 1981).
Thirdly, voluntarism and low institutionalization meant that, until the early 1990s, collective bargaining developed without a formalized framework which gave clear definition to procedures, rules of behaviour, levels and relevant issues. Consequently, according to circumstances and power relations, not only did bargaining repeatedly shift from the centralized to the decentralized level and back again, but issues were duplicated and overlapped at various levels according to the claimant and market power of specific groups or categories of workers.
On the other side, the institutional involvement of the unions in the regulation of the public sector and in the management of social policies had effects that were less immediately apparent. The most evident of them was a growth of trade unionism in the public sector and of a system regulating employment relationships in the civil service based on a logic very different from that prevalent in the private sector (Bordogna, 1994). Another effect, less evident but more pervasive, was the opportunities thus provided for extra resources to be introduced into the interplay of 'voluntary' relations between the parties, thereby modifying/off-setting the results of market pressures. In this way the unwritten but nonetheless binding rule arose that change and innovation promoted by the public institutions should first receive the consent of the interest organizations, and the unions in particular, before they could be implemented.
Between the central and peripheral level of trade-union action
When the unions were reconstituted in 1944, they were organizationally structured so that interests were articulated at both the central and peripheral levels along two dimensions. The first was the 'horizontal' or geographical dimension which corresponded to the general, encompassing and political logic of representation typical of the confederal unions; the second was the 'vertical' or industrial dimension, corresponding to the articulation of sectoral differences in the economy, and more sensitive to market fluctuations, typical of the industrial federations.
This complex organizational structure not only allowed trade-union initiative to switch rapidly between the centre and the periphery and between more general strategies and more sectoral ones, thereby favouring change, but it also made it possible to pursue different courses of action simultaneously. The latter property was of particular importance in a context of ideologically-based competitive unionism, since it allowed the experimental, pragmatic, non-mandatory and quasi-covert development in the periphery of strategies different from those 'officially' enjoined by the organizations at the centre, thus increasing the flexibility and adaptability of trade-union action.
If one takes account of this second dualism in trade-union action, it becomes easier to understand why the evolution of Italian trade unionism exhibits phases very different one from the other, and in which tendencies to centralize and decentralize action have alternated. In fact, a system of centralized and "predominantly political" industrial relations, characterized by the division and the market weakness of the unions (during the 1950s), gave way to a period marked by collective mobilization - which reached its historically maximum level at that time (see Table 2) - and by decentralized bargaining in which the unions were able to strengthen themselves organizationally, overhauling their relations with shop-floor representatives and making themselves indispensable as partners to both employers and governments (late 1960s - mid-1970s). This phase of predominantly conflictual and decentralized action in the market was followed by one of centralization and moderation of economic demands in exchange for benefits in the political arena, in order to tackle unemployment and inflation (the "decade of concertation" between 1976 and 1984, when a failed tripartite agreement put an end to it). And then again, with another shift in the centre of gravity of trade-union action, during the second half of the 1980s a new phase of company-level decentralized action began. Unlike the previous phase, however, this one centred, on the one hand, on the willingness of the local unions and works councils to take pragmatic account of new corporate needs to restructure in order to compete on volatile international markets, and on the other, of the willingness of the majority of managements to use the industrial relations institutions for this purpose, rather than seeking to by-pass them (the phase of micro-concertation of industrial adjustment) (Regini, 1995).
As we shall see in detail in the next section, the 1990s have seen yet another change, this time towards the revival of centralized action. First, however, it is necessary to conclude this long introduction with two brief remarks. The first concerns the advantages of the dualistic system. These advantages undoubtedly lie in the high capacity for innovation and adaptation that it imparts to the behaviour of the industrial relations actors - especially the trade unions - despite their B ideological orientations. From this point of view, and notwithstanding numerous opinions to the contrary, the unions have in fact been a factor, first of modernization and then of flexibilization, in the regulation of the economy. Emblematic in this regard is a comment made by Paul Samuelson in an interview given to La Repubblica on 3 May 1996: 'For years you've been walking along a razor's edge, and the unions could have given you the fatal push. They didn't do so. Perhaps you don't realize it, but the unions are the jewel in your crown'.
The second remark concerns the shortcomings of the system, which grew increasingly evident towards the end of the 1980s. In an economy that is much more closely dependent than in the past on trends in international markets, more narrowly centred on flexible modes of production requiring the greater involvement of labour, and closely conditioned by the criteria to be met for full membership of the European Union, there is a greater need for predictable and reliable behaviour by actors, while the advantages of informality and the play-off between declared and actual behaviour correspondingly diminish.
The re-emergence of concertation in the 1990s
In the 1990s political bargaining - which ceased after the failed tripartite agreement of 1984 - has been resumed with vigour, and a solution has at last been found for the problem of the scant institutionalization of the relationships among the actors concerned. The most significant events of this re-emergence of concertation have been the following: the laws informally negotiated with the unions in 1990 and in 1992-93 to bring the civil service into line with the Italian model of industrial relations in the private sector; negotiation over incomes policy and the collective bargaining structure, which concluded with two tripartite agreements in 1992 and 1993; and negotiation over pension reform, which gave rise first to the most severe social conflict of the last twenty years and then to an agreement between government and unions on a law which was strikingly innovative for the Italian system of policy-making
(19)
.
This revival of concertation is a crucial feature of the political system and the management of the economy in Italy during transition from the first to the second Republic, and it is also an essential factor in its stability. It is even more significant because it stands in sharp contrast to both the crisis of the centralized bargaining of incomes policies in countries with long neo-corporatist traditions like Sweden (Pontusson and Swenson, 1996; Iversen, 1996) and the failure of similar attempts to reform the pension system conducted unilaterally by the French government.
Incomes policies and collective bargaining reform: the new tripartite agreements
The absence of formal political bargaining in the second half of the 1980s did not signify that relationships among the three actors were entirely lacking. Various governments continued the practice of consulting - separately - the social partners before they introduced important economic policy measures (in particular, prior to preparation of the annual legge finanziaria, or budget law). And they occasionally reached bilateral agreements, an example being the tax agreement signed with the unions in 1989. However, any genuine revival of tripartite concertation was obstructed, besides the factors discussed above, by the major and unresolved issue of the scala mobile. Although it effectively shielded wages from the risk of inflation (until the mid-1980s, wages were automatically adjusted to cover about 80% of the inflation rate), the scala mobile was the principal problem as regards not only firms' labour costs and competitiveness but also collective bargaining, given that it left very narrow margins to negotiate wage growth and differentials. For the unions, on the other hand, the scala mobile had great symbolic significance, since it was the outcome of previous waves of collective mobilization and the main indicator of their ability to resist change in power relations.
Hence, when negotiation over the cost of labour and the collective bargaining structure was resumed in 1989, the issue of the scala mobile seemingly blocked any possible progress. However, the political situation changed radically in 1992, the year in which the Mani Pulite
(20)
investigation was launched, the April elections brought the collapse of the old political system, and the new 'technocratic government' headed by Amato found itself faced with a dramatic fiscal and monetary crisis. It was this climate of national emergency that enabled the government to mobilize the consensus necessary for the most drastic attempt to balance the state accounts since the late 1940s.
The tripartite agreement reached in July 1992, which abolished the scala mobile, was also bred by this climate. The aim was to reduce the inflation rate from the current 5.4% to 3.2% in 1993, to 2.5% in 1994, and to 2% in 1995. For this purpose, not only was the scala mobile abolished, but company-level bargaining was frozen for the entire period 1992-93. The core of the agreement was therefore the curbing of wages growth without the compensatory measures that had traditionally accompanied the political trade-offs of the early 1980s. Despite the resignation (later withdrawn) by the leader of the Cgil, who had signed the agreement notwithstanding considerable pressure from his union members, and despite the numerous wildcat strikes mounted in the most unionized factories, the agreement was generally hailed as the first true, albeit incomplete, turning-point in the relationships between the three actors. Incomplete it was because it did no more than set a temporary halt on company-level bargaining, while failing to introduce rules and procedures into the overall collective bargaining system.
The problem was tackled the following year by the new 'technocratic government' led by Ciampi. In fact, the agreement signed by this government with the social partners in July 1993 was less obviously an emergency measure with short-term solutions. It instead delineated a stable architecture of incomes policies and of collective bargaining relations. First of all, as well as confirming the abolition of the scala mobile, the agreement set out an incomes policy based on the joint but autonomous commitment by the parties involved to conform their behaviour to the expected inflation rate. For this purpose, two annual meetings were set up in order, respectively, "to define common objectives concerning the expected inflation rate, the growth of GDP and employment" and "to verify the coherence of behaviour by the parties engaged in the autonomous exercise of their respective responsibilities".
Secondly, the bipolar character of the Italian collective bargaining system - consisting of a national industry level as well as of a company or local one - was confirmed. At the same time, however, the roles of the two levels were specified, and the relations between them better defined, in order to prevent the overlappings that had occurred in the past. The national industry contract (of two-year duration as regards wages, four-year duration as regards other matters) was given the fundamental function of adjusting pay scales to the expected rate of inflation and possibly to productivity increases in the industry. The company contract (for large firms) or the local contract (for small ones) were instead expected to redistribute further productivity increases, as well as to deal with the consequences of technological and organizational innovation, although they were not to overlap with concessions obtained at the national level. In order to render this second level of collective bargaining viable, a new system of workplace representation was envisaged; a system, in fact, that was later sanctioned by an agreement between the unions and the employers' associations in the December of the same year (as we shall see below).
Thus, for the first time, "collective bargaining, and especially the relationship betwen the social partners and the government has become the core of economic and social regulation" (Patriarca and Pellegrini, 1995: 100).
Negotiation over pension reform
Since the Second World War, the Italian pension system has grown rapidly, but in an incremental and confused manner, with the progressive extension of insurance to different occupational categories, each with a diverse relationship between contributions and benefits and often with separately managed funds. Reform of this chaotic system had long been on the policy agenda for reasons of cost cutting, rationalization and equity. The latter objective was an enduring component of the concertation practices of the 1970s and 1980s. For example, it oriented the bill negotiated by the government with the unions in 1978, during the period of 'national solidarity'; a bill, however, which was blocked in Parliament by the pressure groups that it penalized (Regini, 1984).
In the 1990s, the problem of curbing expenditure on social security has grown especially dramatic. Although in Italy the percentage of overall social spending to GDP is below the European average, the proportion of that spending allocated to pensions is far exceeding the average
(21)
. When this fact is set against the background of a huge public debt on the one hand, and of constant decline in the population and labour force on the other, it is evident why reform of the pension system has become the keystone of Italy's economic recovery strategies and a central concern of the country's policy-makers.
The first to get to grips with the problem with any degree of success was the Amato government, which issued a decree in 1992 that raised the age of retirement and increased the minimum number of years of contributions necessary to qualify for a pension, made it more difficult to combine a pension with other work-related income, and introduced other changes. But a 'structural' reform which would replace earnings-related pensions with a contributions-related scheme, and abolish the 'seniority pensions' (which enabled employees to retire at any age as long as they had completed thirty-five years of contributions), was still lacking. These issues were closely bound up with the 'acquired rights' of many categories of workers, as well as the role and the power of the unions, since these hold the majority in the board of directors of Inps (the institute that manages the pensions of wage- and salary-earners in the private sector). It was therefore extremely difficult to deal with these problems without gaining some consensus from the interest organizations.
The Berlusconi government installed in May 1994 seemed initially to comply with this unwritten law when it set up a committee of experts and representatives from the social partners with the task of formulating proposals for reform. However, given the inability of this committee to go beyond agreement on generic principles and suggest shared and specific measures, the government decided to act unilaterally by including provisions in the legge finanziaria that would effectively eliminate seniority pensions. The first centre-right government of postwar years therefore tried to change the unwritten rules of the game that had hitherto regulated social security policy in Italy. The government decided, that is, to use the issue of pension reform - for which there was broad political consensus combined with B pressure by the financial markets - to test what it perceived to be altered power relations. It also wished to verify whether it was possible to cut public spending without the consensus of the unions.
However, it was precisely the nature of this initiative as constituting a general test for the new mode of policy-making envisaged by the centre-right government that gave especial impetus and incisiveness to trade-union mobilization. Although the reform proposals that the unions themselves drew up were not greatly dissimilar to those of the government, a general strike was immediately proclaimed, and it achieved notable success. The protest was vociferous and spread to every part of the country. Wildcat strikes were staged in many workplaces even prior to the general strike; nationwide demonstrations by pensioners had a major impact; and the bulk of public opinion apparently supported the mobilization. The protest culminated in a national rally in Rome organized by the unions in November. With a million and a half people taking to the streets in protest, this was the largest demonstration in postwar Italian history, and its inevitable effect was to weaken consensus for the government's manoeuvre. Even Confindustria (the main employers' association), which had explicitly supported it, and some parties in the government coalition, watched these developments with disquiet. The result was a defeat for the government, which in an agreement signed with the unions was forced to remove the provisions on pensions from the legge finanziaria and postpone them until the following year.
At the beginning of 1995, the centre-right Berlusconi government was replaced by one more 'technocratic government' led by Dini which included reform of the pension system among the four points of its programme to be fulfilled before new elections were called. Although Dini had been Minister of the Treasury in the previous government and therefore one of the architects of the failed attempt at reform, his technocratic government was based on parliamentary support from the centre-left. The new project to reform the pension system, therefore, had once again to search for consensus from the unions.
Actually, the government and the unions engaged in outright bargaining, based on the project drawn up by the union experts. Agreement was reached in May 1995. Although the employers' associations were involved at various stages of the long and difficult talks, Confindustria was generally critical of the reform, deeming it not sufficiently radical while regarding the spending cuts envisaged as excessively diluted over time. The agreement was therefore signed only by the government and the unions. The government converted the text of the agreement into a bill submitted to Parliament, while the unions put it to referendum in the workplaces, where it obtained hard-won but significant majority consensus.
The bill was finally approved by Parliament in July, and the new 'negotiated' law was hailed as one of the cornerstones, together with the tripartite agreements of 1992 and 1993, of Italy's economic recovery. There is no doubt that this was one of the most radical reforms in the history of the Italian welfare state, although its immediate effects on curbing expenditure on pensions were not particularly pronounced. The key condition for obtaining trade-union consensus, in fact, was retention of the previous pension system as far as more elderly workers were concerned, with the introduction - total or partial - of the new and more rigorous system for workers with lower seniority. This obviously meant that savings would only accrue gradually. But, as said, it was the assent given to the reform by unions able to muster the labour force's more or less convinced approval that yielded a result opposite to the setback of the previous year (and also to the result obtained a little later by the Juppé government in France). In a policy-making system like Italy's, moreover, which is based on incrementalism, the reform of social security represented a quite unusual policy innovation (Lange and Regini, 1989: 249-72).
A final important area of tripartite concertation has been employment policy. In September 1996, a 'Pact for Employment' was signed to promote employment, especially in the less developed areas of the country. The most significant points of the agreement include reform of education and training systems, promotion of temporary work and working-time reductions and - perhaps more important - the notion of 'territorial pacts' to promote new investment in areas with low rate of development and high unemployment. These agreements are to be signed by unions and employers' organizations, local authorities, banks and other private participants. They should involve greater wage and labour market flexibility; however, new divisions among the trade unions have emerged on this sensitive issue.
Actors and processes in the re-emergence of concertation
At the beginning of the 1990s, the features and strategies of Italian interest organizations appeared by and large to be a continuation of the previous model, namely a pluralist and voluntarist one combined with some aspects of institutional involvement of the industrial relations actors (Cella 1989; Regalia and Regini, 1995). Numerous actors, on both sides of industry, were engaged in complex games of cooperation/distinction, not only between the two alignments but internally to them as well. The level of formalization and institutionalization of relations between the two sides continued to be low; this restricted their degree of predictability and reliability and hampered the growth of explicit cooperation. On the other hand, the veto power of trade unions were still high enough to discourage attempts to simply ignore them, as was the case in other countries. Finally, one should recall the marked differentiation of rules and practices between the public and private sectors, which gave rise to several distortions.
In the 1980s, each of these aspects - the segmentation and dispersal of representation, the informality and scant coordination of relations between actors, the differences between the normative frameworks of the public and private sectors - prompted the various parties concerned to undertake numerous projects for rationalization. However, these attempts at reform were thwarted by the difficulty of dealing with the complex games-playing among actors and the peculiar mix of informality and institutional regulation of the industrial relations context. If innovations were to be successful, B external pressure and/or government intervention were apparently required. As we shall see, it was this that happened in the early 1990s, bringing profound changes to the initial framework.
Dealing with trade-union weakness. The reform of works councils and the debate on union unification
At the beginning of the 1990s, Italian confederal trade unionism appeared weaker than it had been ten years previously, although a proper assessment of such a weakness reveals it to be less obvious than is commonly claimed. A first signal was the decline in membership in Cgil, Cisl and Uil by active workers from 7,376,000 in 1980 to 6,149,000 in 1990, with a consequent reduction in the unionization rate from 48.6% to 38.6% (see tab. 1).
(22)0
This decline, however, seems to have stemmed more from the effects of structural changes in the economy than from explicit worker disagreement with trade-union strategies. And in any case the rate of unionization in Italy remained rather high by European standards.
(23)1
Another sign of growing weakness was the deterioration in the confederal unions' relationship with their rank and file as symbolized by the growth of Cobas (independent unions) and by expressions of discontent such as the so-called 'self-convened' councils. However, this discontent was evident in certain areas of the public sector, but it was much more sporadic in the private one, where confederal unionism continues to be predominant at the company level. Here, in fact, the unions have been actively engaged in the micro-concertation processes mentioned earlier.
Also negative were the effects of the division and competition among trade-union organizations that followed the breakdown of the Federazione Unitaria (the federative pact between the three confederal unions) in 1984 and which repeatedly blocked attempts at organizational reform. However, although this competition made the decision-making process more sluggish, it did not substantially impede united action. In March 1991, indeed, after several previous attempts, the three union organizations managed to reach an important agreement on rules of reciprocal behaviour and on a new pattern of workplace organization (see below).
More generally, until the early 1990s the role of the unions at the company level remained scattered, hardly visible and socially recognized. Distinctions and tensions among the three confederations, especially at the central level, created space for direct intervention by the state to resolve controversial issues, while also systematically impeding implementation of the various projects for organizational self-reform (including the already-mentioned one of 1991). Subsequently, however, in concomitance with the tripartite agreement of July 1993, changes in the relations between unions and workers and among the three union organizations were finally introduced. This strengthened confederal unionism and enhanced its influence and role.
A first and crucial change was the reform of the workplace representative system in 1993. Within the tripartite agreement of July 1993, the social partners decided that a single pattern of workplace representation was to be introduced all over the economy
(24)2
, under the name of Rappresentanza sindacale unitaria (Rsu), which had been coined in the 1991 agreement between Cgil, Cisl and Uil in order to stress the unions' commitment to endow themselves with a unitary body in the workplaces. Actually, like the 'old' works councils, this was a representative body elected by all workers (and not only by union members), which also represented the trade-union organizations, as they had priority in nominating candidates. The novel aspect was that, following the reorganization of the collective bargaining system established by the tripartite agreement of July 1993, the employers realized that it was also in their interest to place in-company representative bodies on a sounder footing, so that they might have a reliable partner for decentralized bargaining.
(25)3
Accordingly, in December 1993 the employers' associations and the trade unions signed a national-level agreement on the Rsu, the first to regulate such matters after thirty years of informal arrangements.
Although ambiguities in interpretation and the resistance of those who feared electoral losses still persisted, in 1994 and 1995 the workplace representative bodies were finally renewed to an extent unknown since the end of the 1970s, and thereby also the relationships with the workers were revitalized. The results were striking. According to data from the "National Observatory on Rsu", more than 70% of those entitled to do so turned out to vote. Everywhere, confederal unionism obtained large majorities of votes and of representatives (95% and 96% respectively) (Carrieri, 1995: 46-8). This led to a broad, and perhaps unexpected, turnover of workplace representatives. In Milan alone, 5,000 new delegates, i.e. without previous experience, have been elected in the last two years. As a consequence, efficient channels for voice (from below) and for consultation initiatives (from above) have been created, as evidenced on the occasion of the pension reform in May 1995. This is all the more significant if one bears in mind that, after an interval of many years, the Rsu elections have regained a wide coverage in the press, which has helped to strengthen the public image of the confederal unions.
Another new feature has been the resumption and acceleration of debate on union unification. In this case too, this is a change contemplated ever since the breakdown of the federative pact in 1984. It was partly anticipated by the joint action undertaken by the unions in the second half of the 1980s, as well as by the decision taken by the Cgil in the early 1990s to dissolve the internal party factions. The innovative aspect, however, is that the intention is now to unify confederal unionism: the idea, that is to say, is to give rise to a single union organization after half a century of either splits or ad hoc pacts (whether of non-belligerence or alliance). For this purpose, several meetings have been held among the union leaders, and the schedule formally established, although the outcomes are as yet unclear.
Overall, therefore, in the mid-1990s various changes have taken place in the relationships between representatives and represented, and among the labour organizations; changes which have consolidated confederal unionism while also strengthening its role vis-ą-vis employers and the government. In this new context, even the results of the referendum of June 1995, which abolished a number of statutory provisions particularly favourable to the confederal unions, have not - at least to date - had the damaging effects that one might have expected in other circumstances. This depends in part on the fact that these matters were also regulated by collective bargaining, which the referendum did not affect. But it is certainly significant that the employers have not used the results of the referendum to launch a campaign against the confederal unions, taking advantage of their decreasing popularity.
As said, the changes that we have discussed are to a large extent the revival or duplication of attempted reforms which failed in the past but which now achieved success. This raises the question as to the factors responsible for this change; a question that cannot be answered solely by reference to reasons internal to the actors, i.e. to shifts in their strategies and choices. The decisive factor has instead been the impact of external variables, which have functioned as catalysts of change. The first variable relates to changes that have occurred in the parties of the left since 1989, which have gradually eliminated the historical reasons for the divisions in the labour movement. A second factor has been the explosion of corruption scandals involving parties, firms and institutions, the unexpected outcome of which has been to render a break with the past socially desirable. The third factor has been the constraints imposed by the process of European unification, especially since ratification of the Maastricht Treaty.
Trying to cope with the anomaly of public employment
In the area of public employment, recent changes in regulation of the context and in definition of the actors have tended more to anticipate than to accompany the turning-point represented by the tripartite agreement of July 1993; indeed they have constituted one of its preconditions. June 1990 saw, in fact, enactment of the law regulating the right to strike in 'essential' public services. In October 1992, another law empowered the government to begin the so-called 'privatization' of the public employment relationship. Accordingly, the following year, the government issued a decree implementing the privatization process, as a consequence of which the tripartite agreement of July 1993 has become the 'first ever' agreement simultaneously applying to both private-sector workers and public employees (Barbieri, 1995; see also Garofalo, 1994). Furthermore, the law on privatization, like the one regulating the right to strike in the public services, is one of the factors to have facilitated the July agreement. These laws provide some sort of government's guarantee to the other parties (the representatives of the private sector) of the possibility and the intention to reduce the public employees' disruptive power, and the consequent difficulty of curbing public spending.
In both cases, the changes have come about following systematic consultation/negotiation with the unions - which is almost inevitable in sectors with high levels of unionization (it is estimated that confederal unions alone organize just under 50% of employees) and with a powerful presence of 'autonomous' and rank and file unions. The law regulating the right to strike in the 'essential' public services has been the outcome of prolonged negotiation based on a proposal drawn up by a group of legal experts appointed by the unions. Although this law breaks with the more than forty-year-long tradition of non intervention by the state, it builds on the trade-union practice of self-regulation and it does not affect the constitutional principle of the right to strike. It introduces a number of rules (prior announcement of the form and duration of the industrial action, compliance with measures that ensure the delivery of essential services), the implementation of which it defers to agreement between the parties, thereby strengthening their role and promoting their cooperation. Finally, a new actor super partes has been created, the Commissione di garanzia dell'attuazione della legge, which consists of experts appointed by presidential decree and accountable to parliament, with non-coercive powers but with the task of non-mandatory mediation. To be sure, this reform has many weaknesses, for which correctives have been proposed since it was first introduced (Carinci, 1992). Nonetheless, it is probably no coincidence that there has been a pronounced fall in labour disputes since its approval, not only in industry, where strike action has been in decline for more than ten years, but also and especially in public services, reversing the anomalous trend of increasing conflict in that sector during the 1980s (see tab. 3) (Bordogna, 1994).
Also the jointly-agreed legislation on the 'privatization' of the employment relationship in the public sector stems from proposals made by a group of legal experts appointed by the unions, and which was harshly critical of a previous law reforming the sector (the "framework law" of 1983), the effects of which proved extremely disappointing. This latter law, in fact, had introduced to some extent collective bargaining alongside the traditional mechanism of regulation by decree. But it had done so in a manner that simultaneously encouraged the uncontrolled growth of expectations and demands, pushed public spending up to excessive levels, created grievances among the categories of public-sector workers most able to apply pressure, and provoked these categories into rebellion against confederal unionism.
Under the new system, on the other hand, civil servants find their employment relationship entirely regulated by collective bargaining. This bargaining freedom, however, has been constrained within a legal framework that places controls on public spending and redefines the bargaining subjects, on both the workers' and the government's side. The greatest innovation from this point of view has been the creation of the Agenzia per le relazioni sindacali, a technical agency with legal status which takes the place of the traditional committees nominated by the ministries involved to represent the government as employer. Thus pursued, therefore, has been the twofold objective of centrally coordinating negotiations and of separating political from administrative responsibilities, in order to reduce clientelism and curb public spending.
Conclusions
Contrary to the assumptions of neo-corporatist theory, the re-emergence of concertation in the 1990s has not been premised on centralization and representational monopoly by the interest organizations, and trade unions especially. On the contrary, it has acquired stability precisely because it strengthens workplace representation and gives it a more prominent role. Whereas the tripartite agreements of the late 1970s and early '80s were made highly unstable by, among other factors, the excessive distance created between the unions' top leadership and the rank and file, the 1993 agreement, as well as the pension reform negotiations of 1994-95, have been accompanied by the creation of a new system of workplace representation (as discussed above) which enabled the unions to consult their rank and file and to receive legitimation in return (Locke and Baccaro, 1996; see also Carrieri, 1995).
Consequently, the potential crisis of representation has been channelled and controlled not by securing legal monopoly over representation but by creating mechanisms for the expression of 'voice' within, rather than outside, the trade unions. Moreover, the elections of workplace representatives held during this period have revealed unexpected and overwhelming consensus for the confederal unions engaged in concertation (see above), and this has also indirectly given them greater legitimation to proceed in this direction.
A widespread interpretation of the resurgence of concertation in Italy is that "the three partners involved were weak and needed to lean upon each other" (Salvati, 1995: 84; see also Bellardi, 1995: 162-3). This simultaneous weakness might go a long way towards explaining the ease with which traditionally intransigent internal oppositions have been overcome. Not all actors found themselves in the same position, however. To be sure, the public image and self-confidence of the employers have been damaged by their involvement in Tangentopoli (the corruption scandals). And the legitimacy of a political class either identified with the old regime or possessing technocratic credentials not ratified by democratic elections has perhaps been even more severely diminished. The 'technocratic governments' (those headed by Dini in 1995, by Ciampi in 1993 and by Amato in 1992) have been those most in need of the external support and social legitimation provided by trade unions as a substitute for electoral investiture. At the same time, however, they have been able to play on their image of being "neither pro- nor anti-labour" but instead entrusted with the task of coping with a national emergency. Not surprisingly, therefore, these governments have been able to find the consensus with which to achieve such goals as reforming pensions, regulating collective bargaining, and abolishing the wage indexation system, that their predecessors found impossible in spite of their wider Parliamentary majorities.
As for the unions, however, although hit by growing unemployment and by internal fragmentation, they have not been fundamentally weakened; or at least not to the extent of their counterparts in most of the other European countries, where the employers have instead grown far Ber. This difference in relative strengths helps to explain the greater prominence acquired by concertation during the early 1990s in Italy. It is the medium-strength trade unions, in fact, those which do not possess the high power resources of the Scandinavian ones but are at the same time well-rooted in the workplaces and embedded in networks of more or less institutionalized cooperation, that in the current situation can perhaps be seen as a 'precondition' for successful concertation. This type of unions, in fact, can be simultaneously a constraint and a resource for their partners: they may discourage both governments and employers from taking unilateral action which would risk confrontation, while they may convince their rank and file that existing power relations will not allow them to obtain more than the joint regulation of wage dynamics and of some economic policies.
The institutionalization of relationships between employers, unions and the state as an effect of concertation
The Italian industrial relations have traditionally been described as a system with a low level of institutionalization, in the sense of the formalization and stability of the rules regulating relations among actors. Compared with the two polar models of European labour relations - the Anglo-Saxon system based on broad voluntarism in the parties' behaviour, and the German one based instead on a high degree of legal regulation of such behaviour - the Italian model certainly more closely resembled the former, at least apparently, although it did not possess its linearity and coherence (Cella, 1989; Regalia and Regini, 1995).
The resurgence of concertation in the 1990s has radically changed this scenario. Interaction between unions, employers' associations and governments has not only yielded a large number of formal, bilateral or trilateral, agreements. Also and more importantly, unlike in the late 1970s and early 1980s, these actors have explicitly addressed the task of imposing order on the collective bargaining structure, of introducing rules into the relations among them, and of increasing their degree of formalization; objectives that were achieved with the tripartite agreement of July 1933. This drive for the institutionalization of relationships was by no means inevitable, since it conflicted with the tradition of voluntarism, with the climate of profound uncertainty that surrounded ongoing political transition, and partly with the interests of the actors themselves.
Indeed, informality may have significant advantages, as demonstrated by the good performance of the Italian industrial relations system during the economic adjustment of the 1980s. An informal system offers the possibility of flexible and adaptive industrial relations in a phase of intense change; it allows the solutions adopted in the periphery of the system to be diversified according to local conditions; it enables experimentation with innovations that would be more difficult to introduce if they entailed institutional change. Moreover, whenever general rules and therefore new standards are established, on the one hand those who fall below these standards are forced to adjust or to exit, while, on the other, those who already over-fulfil them are encouraged to lower their standards. Finally, the formal regulation of relationships may have the unintended effect of restricting space for interaction by removing issues not subject to regulation from the agenda.
Nevertheless, the choice of imposing order on relationships among actors by institutionalizing industrial relations eventually prevailed, probably because its benefits were deemed to outweigh its disadvantages. Although informality has its advantages, in fact, it also has its drawbacks, in particular the scant predictability of industrial relations and their vulnerability to shifts in power relations. Secondly, in the current phase, the unions considered it more likely that, in the absence of rules, sub-standard situations would multiply more than those in which traditions and power relations would secure higher standards. Finally the normative framework provided by the tripartite agreement of 1993 was designed so that it would favour the growth of interaction in the periphery of the system (i.e. collective bargaining at the company- or local-level) rather than suffocating it.
The early 1990s as a transition period or as the beginning of a new and stable pattern of relationships?
The revival of concertation in the 1990s stemmed from the renewed importance of public policies, not, however, as the traditional source of welfare benefits but as a crucial factor in national competitiveness. Incomes policies, like reform of social security, were regarded as an important factor in the ability of firms to compete in international markets, as well as being a compulsory choice for European governments constrained by the 'Maastricht parameters'. Moreover, although the processes of micro-social regulation of the economy that developed during the 1980s are still able to secure a certain amount of wage coordination and other collective goods for firms, the new economic context requires greater stability and the greater predictability of outcomes; objectives that can be more easily achieved by means of incomes policies and concertation.
Finally, the globalization of markets and the intensification of international competition make it more difficult for firms which compete principally on flexibility and product quality to restrict themselves solely to these. Curbing labour costs has become once again important for these firms as well. And since medium-strength trade unions may obstruct pursuit of this goal by market instruments while being a key resource for concerted policies of wage restraint, many employers have now become the most convinced proponents of an incomes policy.
These features of the external context may constitute important factors in the stabilization of concertation. But it is still not clear to what extent actors are willing to regard concertation as a now institutionalized fact, one not to be called into question whenever the economic situation or power relations change. The request formally submitted by Confindustria to the head of state prior to the 1994 elections, urging that whatever the winning coalition, it should commit itself to maintenance of the tripartite agreement signed the previous year, was certainly a signal of profound change in the employers' attitude. But this change is not yet stable, as is shown by the controversies between the social partners and with government which have re-emerged since the centre-left coalition gained office.
Although extremely important, the tripartite agreement of 1993 was an agreement on rules; it was not a social pact which committed the parties to common views on economic development and its priorities. To be sure, in this historical phase, agreement on the rules may foster the development of such views and encourage the emergence of a shared plan of action. Nevertheless, an agreement on rules only provides a framework; it is not enough on its own to change the pattern of industrial relations. And there is the risk that it will deteriorate unless filled with new contents. By 'common views on economic development and its priorities' we do not mean the old trade-off between employment and productivity; we refer instead to a conception of national competitiveness which gives rise to a joint effort for the full development of human resources. Despite the importance of recent developments, trade unions in Italy have not yet been able to force this broader view of a shared design of economic development on their partners, as well as on major components - especially in the public sector - of their membership.
What consequences for the trade unions?
One of the effects of the resumption of concertation is undoubtedly the increased role and visibility of the unions on the political stage - emblematic of which is the frequent appearance of the leaders of the three confederal unions on the mass media. Although their media presence was rather pervasive during the period of sheltered micro-concertation of the 1980s as well, during the 1990s, and especially since 1996 when the centre-left Prodi government took office, this presence has become an important element in political debate and an essential ingredient of policy-making.
From this point of view the unions have indubitably increased their influence. But this greater political influence has raised new problems, above all in relations with the rank and file.
Precisely because the unions appear Ber and more authoritative than in the past, workers' expectations of greater and more incisive protection by the unions have correspondingly increased. Indicative of these heightened expectations are, for example, the not minor difficulties encountered in the renewal of the contracts for traditionally unionized industries like transport, or the outbreaks of revolt by unemployed workers against the government as well as the unions in the high unemployment areas of the Mezzogiorno.
For the unions, however, this does not simply mean that they are confronted by the traditional dilemma between moderating their demands with a view to consolidating their cooperative relations with the government and the majority of employers, and claiming a larger share of benefits in order to augment or not to prejudice their following among workers. There is in fact a new element that the unions must consider when elaborating their strategy: the profound changes now taking place in the world of work.
For some time now, consequent on changes in the economy and in the organization of production, the traditional distinctions between regular work (on which trade-union action was shaped) and 'atypical', irregular, temporary forms of employment have been fading; and also those between dependent employment and self-employment, between work in production or services and voluntary 'activities' of greater or lesser social utility. More in general, the distinction between labour requiring collective protection and activities capable of self-protection has grown socially more blurred. Today, therefore, the Italian unions find themselves obliged to redefine the bases of their representation and to devise more appropriate strategies to pursue.
References
Barbieri, M. 1995: Le politiche contrattuali dopo il Protocollo di luglio: continuitą e discontinuitą degli assetti contrattuali nel pubblico impiego. In CESOS (ed), Le relazioni sindacali in Italia. Rapporto 1993/94. Rome: Ed. Lavoro, 295-307
Bellardi, L. 1995: La contrattazione collettiva nel 1993, Giornale di diritto del lavoro e di relazioni industriali, XVII, 65, 161-93
Bordogna, L. 1994: Pluralismo senza mercato. Milan: Franco Angeli.
Bordogna, L. 1998: La conflittualitą. In CESOS (ed), Le relazioni sindacali in Italia.
Rapporto 1994/95. Rome: Ed. Lavoro, 163-69
Carrieri, M. 1995: L'incerta rappresentanza. Sindacati e consenso negli anni '90: dal monopolio confederale alle rappresentanze sindacali unitarie. Bologna: Il Mulino.
Cella, G.P. 1989: Criteria of Regulation in Italian Industrial relations: A Case of Weak Institutions. In Lange, P. and Regini M. (eds), 167-85
Codara, L. 1988: La sindacalizzazione. In CESOS (ed), Le relazioni sindacali in Italia. Rapporto 1994/95. Rome: Ed. Lavoro, 170-81
Carinci, F.1992: L'attivitą della commissione di garanzia (L. N. 146/1990), Giornale di diritto del lavoro e di relazioni industriali, XIV, 55, 435-59
Franzosi, R. 1995: The Puzzle of Strikes. Class and State Strategies in Postwar Italy. Cambridge: Cambridge University Press.
Garofalo, M.G. 1994: Legislazione e contrattazione collettiva nel 1992. Giornale di diritto del lavoro e di relazioni industriali, XVII, 61, 163-95.
Iversen, T. 1996: Power, flexibility and the breakdown of centralized wage bargaining: Denmark and Sweden in comparative perspective. Comparative Politics, July, 399-436.
Lange P. and Regini M. (eds) 1989: State, Market and Social Regulation. New Perspectives on Italy. Cambridge: Cambridge University Press
Lanzalaco, L. 1995: L'evoluzione dell'azione imprenditoriale: il ruolo delle associazioni datoriali. In Chiesi, A:M:, Regalia, I. and Regini, M. (eds), Lavoro e relazioni industriali in Europa. Rome: La Nuova Italia Scientifica-Carocci: 111-32.
Locke, R. and Baccaro, L. 1996: Learning from Past Mistakes? Recent Reforms in Italian Industrial Relations', Industrial Relations Journal. 27(4), 289-303.
Martinelli, A. (ed.) 1994: L'azione collettiva degli imprenditori italiani. Milan: Ed. Comunitą
Patriarca, S. and Pellegrini, F. 1995: Contrattazione interconfederale,. In CESOS (ed.), Le relazioni sindacali in Italia. Rapporto 1993/94. Rome: Ed. Lavoro, 99-108
Pontusson, J. and Swenson, P. 1996 Labor markets, production strategies, and wage bargaining institutions. The Swedish employer offensive in comparative perspective, Comparative Political Studies, 29(2), 163-79.
Regalia, I. 1995: Italy: The costs and benefits of informality. In Rogers, J. and
Streeck, W. (eds), Works Councils. Consultation, Representation and Co-operation in Industrial Relations. Cicago and London: The University of Chicago Press, 217-41.
Regalia, I and Regini, M. 1995: Between Voluntarism and Institutionalization: Industrial Relations and Human Resources Practices in Italy. In Locke, R., Kochan, T. and Piore, M. (eds), Employment Relations in a Changing World Economy. Cambridge (Mass.): MIT Press, 131-63.
Regalia, I and Regini, M. 1998: Italy: The Dual Character of Industrial Relations. in Ferner, A. and Hyman, R. (eds), Changing Industrial Relations in Europe, Oxford: Blackwell, 459-503.
Regini, M. 1981: I dilemmi del sindacato. Bologna: Il Mulino
Regini, M. 1984: The Conditions for Political Exchange. How Concertation Emerged and Collapsed in Italy and Great Britain.In Goldthorpe J. (ed.), Order and Conflict in Contemporary Capitalism. Oxford: Clarendon Press, 124-42.
Regini, M. 1995: Uncertain Boundaries: The Social and Political Construction of European Economies. Cambridge: CUP
Regini, M. and Regalia, I. 1987: Employers, Unions and the State: The Resurgence of Concertation in Italy? , West European Politics, 20(1), 210-30.
Romagnoli, G. and Della Rocca. 1989: Il sindacato. In Cella, G.P. and Treu, T. (eds), Relazioni industriali. Manuale per l'analisi dell'esperienza italiana. Bologna: Il Mulino, 85-134.
Salvati, M. 1995: The Crisis of Government in Italy. New Left Review, No. 213, 84
Visser, J. 1992: The Strength of Union Movements in Advanced Capital Democracies: Social and Organizational Variations. In Regini, M. (ed.), The Future of Labour Movements. London: Sage, 17-52
Table 1 Membership of the three main confederations 1950-94 (000s)
|
Aggregate membership |
Employed membership
| Year |
CGIL |
CISL |
UIL |
Total |
CGIL |
CISL |
UIL |
Total |
Density
% |
| 1950 |
4,641 |
1,190 |
- |
5,830 |
|
|
|
|
50.8 |
| 1951 |
4,491 |
1,338 |
- |
5,829 |
|
|
|
|
50.9 |
| 1952 |
4,342 |
1,322 |
- |
5,664 |
|
|
|
|
48.8 |
| 1953 |
4,075 |
1,305 |
- |
5,380 |
|
|
|
|
45.6 |
| 1954 |
4,134 |
1,327 |
- |
5,461 |
|
|
|
|
44.6 |
| 1955 |
4,194 |
1,342 |
- |
5,536 |
|
|
|
|
43.9 |
| 1956 |
3,666 |
1,707 |
- |
5,374 |
|
| |
