MPIfG Working Paper 03/2, March 2003


From State Weakness as Strength to State Weakness as Weakness: Welfare Corporatism and the Private Use of the Public Interest [1]


Wolfgang Streeck , Max Planck Institute for the Study of Societies



Prof. Wolfgang Streeck is Director at the Max Planck Institute for the Study of Societies, Cologne.






The paper is a contribution to a book edited by Simon Green and Willie Paterson, Semi-sovereignty Revisited: Governance, Institutions and Policies in United Germany. The book will appear in 2003 or 2004. It explores to what extent Peter Katzenstein's seminal study of the "semi-sovereign" German state in the 1980s is still valid. The paper looks at one of the showpieces of Katzenstein's theory of beneficial semi-sovereignty, German industrial relations, and follows their development from the Modell Deutschland of the 1976 election campaign to Schröder's Bündnis für Arbeit. It comes to the conclusion that as far as Germany is concerned, the days are gone when it was an advantage for the governance of industrial relations to have a weak state.






Das Papier ist ein Kapitel in dem von Simon Green und Willie Paterson herausgegebenen Buch Semi-sovereignty Revisited: Governance, Institutions and Policies in United Germany, das 2003 oder 2004 erscheinen soll. Das Buch untersucht, inwieweit Peter Katzensteins epochemachende Studie über den "halbsouveränen" deutschen Staat noch Gültigkeit hat. Das Papier behandelt eines von Katzensteins wichtigsten Beispielen für die Vorteile begrenzter staatlicher Souveränität, die Arbeitsbeziehungen. Es verfolgt die Entwicklung der deutschen Arbeitsbeziehungen von "Modell Deutschland" aus dem Wahlkampf von 1976 bis hin zum Schröderschen "Bündnis für Arbeit". Dabei kommt es zu dem Schluss, dass, zumindest was Deutschland angeht, die Tage vorbei sind, als eine schwache Regierung für die Regulierung der Arbeitsbeziehungen von Vorteil war.







Context: The Rise of Welfare Corporatism
Agenda: Combating Unemployment
Process: Less Encompassing Organisation
Process: Even Less Sovereignty
From the Kohl Failure to the Schröder Deadlock
Now a Strong State?





Can being governed by a state that is not fully sovereign be anything other than a crippling disadvantage for a country? In his seminal analysis of post-war West Germany Peter Katzenstein suggests it can. Indeed he argues that precisely not having command of full state capacities afforded West Germany more effective governance than comparable countries. This was because semi-sovereignty, according to Katzenstein, protected the West German state from counterproductive illusions of omnipotence other states at the time still held, and forced it to cultivate other means of public policy than direct state control - means much better matched than traditional state intervention to the evolving problems of governance in a changing world.


Successful semi-sovereign governance, in Katzenstein's sense, rests on two pillars. First, a fragmented, decentralised state whose capacities for direct intervention are limited must learn to make deals with independent actors in civil society which command their own sort of sovereignty that cannot be ignored or circumvented. A state of this sort must therefore be a "co-operative state" (Willke 1983), one that governs more by negotiation and co-optation than by legal command (Scharpf 1993). Where this works - in a political system that has learned to build a culture and develop techniques of indirect control - independent social organisations and institutions with their guaranteed autonomy and power turn into agents of publicly licensed self-government, under the roof of a negotiated public order of which the state is just one element among others, although a pivotal one (Streeck and Schmitter 1984). Social autonomy is thus transformed into delegated public responsibility, and organised private groups become quasi-public agencies of societal governance much more competent and legitimate in dealing with the problems of their constituents than state bureaucracies could ever be.


Second, the social groups that are to become partners in governance of a co-operative state must be organised in a way that makes them suitable for the purpose. In Katzenstein's language, the deficiencies of the decentralised semi-sovereign German state were compensated by a society capable of governing itself through centralised intermediary organisations. Competition between different organisations for the allegiance of the same group does not normally give rise to responsible behaviour, nor does organisational instability of any other sort. Social groups included in semi-sovereign governance should therefore, as Olson (1982) puts it, be represented by large encompassing organisations, externally inclusive to make it impossible for them to impose the costs of their policies on outsiders, and internally heterogeneous to force them to integrate divergent special interests and learn to align them behind a broad, centrist compromise. Civil society, in other words, must be represented by a small number of organisations that can legitimately and effectively speak for their constituents, together covering the society as a whole and not excluding any significant social category. Such organisations, described as "corporatist" by the literature of the 1970s (Schmitter 1979 [1974]; Lehmbruch 1979 [1974]; Katzenstein 1984) can be expected to identify their interests, both substantive and organisational, with those of the society as a whole, or at least define them in such a way that they take the public interest sufficiently into account to enable a semi-sovereign state to avoid disruptive conflict.


One of the showpieces of Katzenstein's theory of benevolent semi-sovereignty was industrial relations. Here the German configuration of a weak and decentralised state and a centralised society connected by a variety of parapublic institutions was almost ideal-typically present. Under Tarifautonomie, the government was barred from direct intervention in wage setting. The void was filled by encompassing trade unions and employers associations safe in their positions and disciplined precisely by their power, which they were both allowed and obliged to exercise, and which could be exercised best only in co-operation with one another and with the state. Protected by the system of semi-sovereignty from radical change - for example in the form of a statutory income policy - unions and employers negotiated incremental change and took responsibility for its implementation, if only because nobody else could have done this for them. Unions in particular were held responsible by their deep involvement in a peculiar parapublic institution at the workplace level, co-determination (or Mitbestimmung) of workers, which had been extended in the 1970s and to which Katzenstein devotes much attention. The result, according to Katzenstein, was a level of "social peace" unique among large countries at the time, in turn supporting exceptional industrial competitiveness and enviable macro-economic performance.


This essay deals with what became of German industrial relations after Katzenstein. It argues that as far as Germany is concerned, the days are gone when it was an advantage for the governance of industrial relations to have a weak state. Katzenstein's successful politics of private-public coordination required economic and political conditions that it could not by itself reproduce. With hindsight it can be seen that the capacity of the German state for effective concertation of capital and labour and the social and economic benevolence of German-style free collective bargaining depended on a historical background that was taken for granted even when it was already dissolving. Among the variables that were for too long treated as constants were high and stable export-driven economic growth making full employment a normal condition, and the willingness and ability of trade unions to put their political and institutional loyalties above the articulated interests of their members. While the latter ended with the wave of worker militancy in 1969, the former finally disappeared with the first oil crisis in 1973.


In the mid-1980s when Katzenstein wrote, the specifically German configuration in industrial relations of a weak state and strong organised interests was already in crisis. The combination of institutionalised monetarism, as imposed by the Bundesbank from 1974 onwards, with an unchanged collective bargaining regime - and specially a metalworkers union sworn never to forget the "lessons of 1969" - was generating rising unemployment that presented increasingly intractable political problems to the government. For a while the German political economy managed to respond to its institutionally imposed condition of low inflation and high wages with structural change towards an internationally highly competitive production pattern that came to be referred to as "diversified quality production" (Streeck 1991). Moreover, between the first and the second oil crisis, fiscal reflation helped hide the trilemma between free collective bargaining, monetary stability and employment - a trilemma that given German institutions could be resolved only at the expense of the latter. However, when this had run its course, the very opportunity structure of semi-sovereignty and the same close intertwining of state and organised civil society that Katzenstein had celebrated, both constrained and enabled the social partners and the government, dependent as they were on social peace, to cover up the country's employment problem by diverting the welfare state to the purpose of a defensive management of the labour supply. In the process the endemic pressures of a permanently unbalanced labour market incrementally but fundamentally recast the relationship between labour, business and the state, and in the end placed the issue of state sovereignty - of the state's capacity to govern - firmly back on the economic agenda.



Context: The Rise of Welfare Corporatism


Katzenstein's semi-sovereignty thesis described a weak state successfully holding well-organised interest groups accountable to national economic objectives. But in Germany in the 1980s and 1990s, less encompassing though still firmly entrenched "social partners" learned to use their privileged political status to take advantage of the condition of semi-sovereignty and utilise core public institutions for the purpose of subsidising an increasingly untenable labour market regime. Step by step, the public use of private organised interests, as described by Katzenstein and others, turned into a private use of the public interest. By the late 1990s at the latest, in the Bündnis für Arbeit of the Schröder government, the decay of Modell Deutschland became visible in a lasting political deadlock over labour market reform, as corporatism in the sense of responsible group self-government gave way to something like corporatism in the French or Italian meaning: the pursuit of special group interests in the public realm at public expense, first in collusion with the government and later against its - reluctant - resistance.


How could this have happened? Economic semi-sovereignty - and in particular the lack of control of the federal government over the money supply - had protected both unions and the government from Keynesian temptations. As unions could not expect the state to underwrite excessive wage increases, they found it in their interest to moderate their wage demands in the light of their predictable macro-economic effects. This, however, changed to a significant degree in 1969 when in particular IG Metall attributed the unofficial strikes to its co-operation with the economic policy of the government, and with its Konzertierte Aktion in particular. In subsequent years union wage policies became more responsive to the demands of a growing and more interest-conscious membership. In 1974, in the wake of the first oil shock, the Brandt government was forced to settle for a public sector wage increase of above ten per cent; a few months later it resigned. Having lost the confidence that the government would ever be able to reign in the unions, the Bundesbank unilaterally switched to a monetarist policy that was non-negotiable for both unions and government. Immediately thereafter, mass unemployment became the critical problem of the German political economy, which it remained until the present day (Scharpf 1991).


The government of Helmut Schmidt, who became the chief propagandist of tripartite Modell Deutschland, depended on the political support of the unions and could therefore not afford politically to intervene in collective bargaining. [2] Nor was it within its powers to adjust the money supply to what would have been required for a return to full employment. Indeed faced with the unrelenting monetarism of the Bundesbank, the unions in 1978 formally walked out of Konzertierte Aktion. In the same year the government, trying desperately to find a way back to full employment, started an attempt at internationally concerted fiscal reflation. This failed when in 1979 the United States, in violation of the Bonn summit agreements, unilaterally switched to a policy of tight money while almost simultaneously the world economy was hit by the second oil shock. With unions bent on defending the high wage and low wage dispersion labour market regime that had served them and their members well, the result was rising public debt coincident with rising unemployment, which in turn caused the demise of social-liberal coalition in 1982.


Already in the late 1970s, then, semi-sovereignty had begun to turn into a liability. While it continued to prevent the government from using monetary policies to shore up the labour market, it also forced it under Tarifautonomie to leave the regulation of the labour market to unions and employers. Early attempts by Kohl to undercut union strength never got anywhere near what happened during the period in Britain, probably also for lack of trying, as Kohl understood only too well the political constraints of semi-sovereignty. Conflicts in industrial relations and between government and unions did rise. Soon, however, unions and employers learned to use the Bismarckian welfare state as a functional equivalent to Keynesianism - compensating the adverse employment effects of free collective bargaining, not by increasing aggregate demand, but by retiring excess labour and taking it out of the market (Streeck 1997, 2001b). As the government given its limited powers had nothing better to offer to contain the number of unemployed, it more or less willingly acquiesced.


In subsequent years successive Kohl governments, under electoral pressure to do something about unemployment but unable effectively to intervene in the labour market, were happy to allow the social partners to proceed with industrial restructuring under a high-wage and high-productivity labour market regime by having the social security system absorb the victims, to a point where the costs of defensive labour supply management themselves began to contribute to unemployment. This culminated in the period after unification when Kohl had to mend his fences with the unions to shore up national unity for a consensual Institutionentransfer from West to East (Lehmbruch 1998) and to prevent "radicalisation" in the East. In the process the productivistic consensus that Katzenstein had described as characteristic of German industrial relations was gradually replaced with a welfare state consensus, underpinned by the institutional entrenchment of unions and employers associations in the self-government of the social security system, until that consensus, too, began to crumble under its rising economic costs in the mid-1990s.



Agenda: Combating Unemployment


When Kohl became Chancellor in 1982, the last attempt at fiscal Keynesianism had failed, unemployment continued to rise, and public debt was at a record high. Both government and unions were aware that economically, Bundesbank monetarism had become the only game in town. More than Thatcher, however, Kohl had to fear negative electoral consequences of high unemployment. And while the unions were strong enough to resist government intervention in Tarifautonomie, they were also afraid of letting unemployment grow further. Their answer was a strategy of redistribution of work through reduction of working time.


How exactly work redistribution was to be accomplished became a matter of divisive contention among unions. For a time the metal workers union, IGM, wavered between early retirement at no loss of pension and cuts in weekly hours at no loss of pay, until in early 1984 it settled for the latter and started a campaign for a 35 hours week (Wochenarbeitszeitverkürzung). Five other unions, under the leadership of the chemical workers, instead opted for what came to be called reduction of lifetime working hours (Lebensarbeitszeitverkürzung). In part this reflected the traditional antagonism between "left" and "right" unions. But a different age structure may also have played a role: as IGM members at the time were younger on average, an equal reduction of hours for all may have been more popular with them. For their leadership, member support was central as they had concluded early on that working time reduction was not to be had without a long strike. [3]


On the employers side preferences were clear. Employers feared the cost increases and the organisational rigidities they expected from a 35 hours week, and massively urged the new government to come to their help. This it did by trying to influence the debate among the unions in favour of the so-called "Gang of Five", by legislation under which the social security system would cover part of the costs of early retirement if a replacement was hired. Still, in the summer of 1984 the metal workers went on a six week strike for the 35 hours week. The strike ended in arbitration giving the union a reduction of weekly working time at full pay to 38.5 hours, with the prospect of a later reduction to 35 hours. [4]


As it turned out, if ever there was a Pyrrhic victory, it was that of IGM in 1984 Streeck 2001b). To get a settlement the union had to accept extensive provisions for working time flexibility. Subsequently these became the major productivity tool of German employers, who became adroit at uncovering the productivity reserves in working time arrangements. As a result unions and works councils lost control of the wage-effort bargain. This made it easy for employers, especially in large firms, to compensate high wage settlements with productivity improvements that made part of the workforce redundant. Consequently the resistance of the large firms that dominate the employers associations to wage increases declined since, unlike the losses in international market share resulting from industrial conflict, rising labour costs could easily be made up internally. The indirect effect was a crisis of solidarity within the employers camp that began to undermine the organisational basis of tripartism.


Moreover, as the details of working time flexibility have to be negotiated at the workplace between employers and works councils, the 1984 agreement set in motion a substantial decentralisation of industrial relations that is still continuing. Furthermore, growing work intensity as a result of productivity-enhancing reorganisation dampened the enthusiasm of IGM members for further reductions of hours, and in the 1990s the leadership gradually switched to the position of the dissenters of the early 1980s until, under Schröder, it called for a general retirement age of 60 (Rente mit 60). Finally, employers had agreed to the arbitration award only after Kohl had promised their leaders in a telephone conversation to pass a law prohibiting unemployment benefits being paid to workers laid off due to indirect effects of an industrial dispute. The intended outcome was to make it less easy for a large union like IGM to call a strike, and indeed since 1984 there has not been any strike of national significance in the metal sector. [5]


In subsequent years the political countermeasures taken by government and employers against hourly working time reductions, together with the long-term effects of the 1984 strike settlement, issued in joint reliance of unions, employers and the government on the welfare state to reduce the labour supply through a variety of forms of early retirement. Metal industry firms used the various legal opportunities to thin out and rejuvenate their workforces as extensively as firms in the chemical and other sectors, and with no less support from their works councils. [6] In the course of the 1980s the public pension system developed into a safety valve for a labour market regime that generated employment risks for ever larger segments of the workforce. It also allowed firms to restructure in response to pressures for "lean production" and "downsizing", without having to face the resistance of works councils and local unions. Absorbing through publicly funded early retirement and other provisions the surplus labour created by high wages, low wage dispersion and strong employment protection, the German welfare state of the 1980s slowly turned into a functional equivalent of the Keynesian reflation state. [7] The price was both a growing need of the social security funds for government subsidies and, in the contribution-based German welfare state, rising statutory non-wage labour costs. Removing surplus labour from the market to fight unemployment thus made labour even more expensive, requiring the labour supply to be cut further - a downward spiral in which the method of choice to fight unemployment became another, potent contributor to it.


Roughly the same as to the pension system applied to labour market policy in a narrow sense, in particular short-term work, job creation measures and further training. Conceived as temporary devices to protect and upgrade the skills of unemployed workers until the restoration of full employment, labour market programs over the years became permanent stopgaps for a low employment labour market regime. Especially after unification, labour market policy came to support a giant "second labour market" hiding away large numbers of unemployed in make-believe jobs before they were sent back to social assistance or unemployment benefit, or went into early retirement. The Bundesanstalt für Arbeit, which administers labour market policy and is governed on a tripartite basis by the state and the social partners, today collects a hefty 6.5 per cent of payroll from all workers up to an indexed upper income limit. In 2002 it had a staff of 90,000 and a budget of Euro 50bn, around 40 per cent of which it spends on so-called "active labour market policies". [8] (The entire public university system in Germany costs about Euro 27bn per year.)


By the late 1980s the economic burden of early retirement became so heavy that the Kohl government undertook first attempts at retrenchment. Some measures, like a slow increase in the legal age of retirement, even gained the support of the Social Democrats. But then unification arrested reform and indeed turned back the wheel. The irritations of the 1980s between the Kohl government and the unions receded as the challenge of unification pulled government, employers and unions closely together. In particular, to keep East Germans from migrating to the West, the West German welfare state, with the support of all parties, was early on extended wholesale to the Neue Länder, together with the currency, the market economy, the legal, educational and health care systems and, of course, the collective bargaining regime.


When economic disaster struck, in part predictably and in part following the unexpected breakdown of East European export markets, mass unemployment caused the tools of defensive labour supply management that had been developed in the 1980s being used on an unprecedented scale. In effect this amounted to West German workers and employers shouldering the lion's share part of the costs of unification through rising social security contributions. After the end of the short post-unification boom in 1993, high non-wage labour costs began to depress employment in the West. They also were perceived by employers as impairing their competitiveness at a time of economic internationalisation and of the completion of the European Internal Market. This set the stage for the conflicts of the mid-1990s, which contributed to the defeat of Kohl in 1998, which in turn issued in the Schröder government's attempt to forge an "Alliance for Jobs" with the unions.


Funding the unexpectedly high and growing costs of unification through the social security system must have been convenient to a government that had promised not to raise taxes while being constrained by approaching Monetary Union to cut public deficits. That high social security contributions themselves increased unemployment and eroded employer confidence in the governing coalition posed a political dilemma that the government eventually tried to address by attacking other, non-statutory non-wage labour costs, especially employer-paid sickness benefits. In the early summer of 1996 this caused the breakdown of the hitherto amicable relations between Kohl and the unions, which from then on put their entire political capital on an SPD election victory in 1998. Among employers, high payroll taxes drove a wedge between small and medium-sized employers on the one hand and large, increasingly multinational companies on the other. While the former suffered from what was for them a crippling cost squeeze, the latter were able to use flexibility provisions in collective agreements to pay for high labour costs out of productivity improvements, in the same way as they had learned to respond to high wage settlements. Failing this, they could always and with growing ease relocate production abroad, buying works council support with liberal supplements to early retirement public pensions.


For the unions the matter was less straightforward. High social security contributions meant losses in employment and in real wages for union members. On the other side of the scale was, however, the interest of an aging membership in high and early pensions. Of particular importance here is that the Bismarckian German welfare state, organised on an "insurance principle", not only emphasises cash transfers over services but also tailors benefits to a worker's skill level and previous wage, to protect his or her social status. The result is long spells of benefit-supported unemployment, due among other things to numerous legal possibilities to reject job offers on the ground that they are unsuitable ("unzumutbar"), combined with lax enforcement of willingness to work rules by the employment office.


In the 1990s the Bundesanstalt für Arbeit in particular became the focal institution of welfare corporatism, a relationship between government, unions and employers different from the demand and supply-side versions of neo-corporatism that prevailed in the 1970s and part of the 1980s. Welfare corporatism is a response to lasting unemployment untreatable, for economic or political reasons, by Keynesian demand-side management as well as both neo-liberal and "left" supply-side policies. Under welfare corporatism, industrial relations and social policy become densely intertwined, with the continued operation of the industrial relations system depending on subsidisation by the social security system. Jointly using the public social security system to compensate core categories of the victims of an exclusive labour market regime - so as to make reform of that regime less politically pressing - unions, employers and the government together sustain a "low activity, high equality" employment system (Streeck 2001a) by obliging each other in complex ways that include consensual co-management of economic change at the workplace, toleration of labour market rigidities, a modicum of wage moderation, the use of social security contributions to fund core government activities (Trampusch 2003), and a strong role of unions and employers associations in the government of the social security system.



Process: Less Encompassing Organisation


The failed attempt of the unions to fight unemployment by redistribution of work and the subsequent defence of extant labour market institutions by social security undermined the encompassing organisation of business and labour and of the industrial relations system as a whole and gave rise to progressive decentralisation of the regulation of employment conditions. Over time the social partners of semi-sovereign governance corresponded increasingly less to Katzenstein's image of stable, centralised, publicly responsible private governments. Instead, forced to become more attentive to demands of their members that they found difficult both to ignore and to change, they became resistant to government efforts to make them take changing needs of public policy into account.


New divisions arose in the employers camp that were reinforced by the internationalisation of product markets and production systems. As pointed out above, small firms were less able to take advantage of the flexibility provisions of the 1984 settlement than large firms; for them working time reduction was simply an increase in costs. The same applied to early retirement and the rise in social security contributions it required; while small firms had to pay the same contributions as large firms, they often could not afford letting their more experienced workers retire. Large multinational German firms, facing unprecedented price competition in domestic and international markets, responded to rising labour costs by asking their domestic suppliers for price reductions. Sometimes this directly followed wage rises conceded by the large firms, in their capacity as leaders of the employers association, also on behalf of their suppliers. Declining resistance of large firms to wage demands, due to both foreign competition for market share and the new opportunities to compensate wage rises by productivity increases, convinced Mittelstand employers that the large companies used their dominance within the employers associations to secure labour peace for themselves, at their expense.


By the mid-1990s, after the end of the unification boom, an unprecedented revolt was under way inside the system of business associations. In the 1970s Hanns-Martin Schleyer, the President of BDA, the national association of employers associations, had been elected in personal union President of the BDI, the confederation of trade associations. During his double presidency, which coincided with the Schmidt government, the corporatist centralisation of German business associations reached its peak, and so did the relative influence of employers associations as compared to trade associations. One-and-a-half decades later the situation had almost reversed. After the short interlude of national unity in the wake of unification, the more specialised and less encompassing trade associations came to serve as representatives of the interests of Mittelstand firms, and their most forceful and militant spokesman became the President of the BDI, Hans-Olaf Henkel. Using neo-liberal rhetoric hitherto unheard from a German business leader, Henkel became highly visible by publicly confronting his counterparts at the BDA, first Klaus Murmann and then, after he had caused Murmann's resignation, Dieter Hundt.


Henkel's favourite themes, which were outside his formal jurisdiction as President of BDI, were a reform of collective bargaining, especially its decentralisation, and a substantial reduction in social benefits and non-wage labour costs. While some leaders of German industry distanced themselves from his firebrand rhetoric, Henkel was re-elected several times as his public pronouncements reflected growing disenchantment among his constituents with the Kohl government's close relations with the unions in the early 1990s. After the introduction in 1995 of compulsory old age care insurance, again to be paid for by a payroll tax, a large segment of the German business class lost confidence in Kohl's ability and willingness to accomplish the social retrenchments they considered necessary in a period of economic internationalisation.


Discontent with Kohl expressed itself in the Standort Deutschland debate started by organised business, which emphasised the alleged economic disadvantages of Germany as a production site and reached its peak after Kohl's re-election in 1994. Henkel's call for radical changes in industrial relations and for cutbacks in social costs took up the interests of smaller firms that could not easily relocate production abroad, and thus helped the BDI shed its traditional image as a representative of large industry. Later the new concerns of the Mittelstand were taken up, in addition to the leaders of various powerful trade associations, by politicians in the FDP as well as by the floor leader of the CDU/CSU, Wolfgang Schäuble, who aspired to succeed Kohl in 1998. As the cleavage between BDI and BDA deepened, many employers associations began to lose members, especially after IGM had gained a high wage settlement in 1995 when the leading large firms had, again, caved in to the union for fear of a loss of international market share (Schroeder and Ruppert 1996).


Pressures for decentralisation of collective bargaining were exacerbated by the situation in East Germany. Very few firms there had joined the employers associations in the first place. Of those that were covered by industrial agreements, a large number paid less than the official wage, often with the consent of works councils desperate to protect employment in their workplace (Schroeder 2000). Paying unter Tarif soon spread to firms in distress in the West, and so did the practice of more or less tacit concession bargaining between employers and works councils even though these are charged by law to enforce adherence to industrial agreements (Bispinck 1997). Unions often looked the other way, in the hope for better times somehow to return. Especially IGM observed helplessly the growing de facto decentralisation of collective bargaining, sometimes trying to prevent it, sometimes condoning it under pressure from the membership. With time, unions learned to insert clauses in industrial agreements that allowed individual employers, with the consent of the works council or of the parties to the industrial agreement, to suspend wage increases, extend working hours, or cut working time at reduced pay (so-called Härteklauseln or Öffnungsklauseln).


For the unions, redefining their role in collective bargaining to allow for greater local autonomy was difficult as the new challenge coincided with a rapid decline of their institutional position and organisational strength. Katzenstein had still been able to point to "an increasing membership in a shrinking workforce", which he took as showing "that in its organisational structure and political presence West German labour is a model of strength for the unions in all of the major industrial societies, except for Sweden" (1987, p. 28). But between the mid-1980s and the mid-1990s, coverage of private sector employees by works councils fell from 51.4 to 41.6 per cent (Hassel 1999). [9] In addition and partly for the same reasons, union density declined dramatically. Hardly ten years after unification, the number of union members in Germany had returned to what it had been in West Germany by the late 1980s. Low union membership reflected low employment as much as, especially in East Germany, high unemployment. It was also related to the growth of the informal economy and of non-standard forms of employment, caused not least by the high costs of labour. In addition, general demographic change and reluctance of younger workers to join unions - caused by different employment situations, work experiences and career expectations - resulted in a rapid increase in the average age of union members. Early retirement further raised the proportion of pensioners among the membership. By the late 1990s, pensioners accounted for 19 per cent of those organised in DGB unions. Disregarding the retired and the unemployed, overall union density in the private sector of the German economy fell from 27.3 per cent in 1980 to no more than 18.6 per cent in 1999 (Ebbinghaus 2002a).


While the decentralisation of collective bargaining and several moderate wage rounds in the second half of the 1990s [10] brought the membership losses of employers associations to a halt, unions watched their own organisational crisis in helpless bewilderment. The transformation of the Flächentarifvertrag proceeded against their passive resistance, driven by individual employers and works councils and beyond the control of a union movement uncertain and divided over its response (Hassel and Rehder 2001). Several union mergers were accomplished, but the bleeding was not stopped. Mergers also seem to have weakened the peak association, the DGB, which now represents a very small number of large and basically self-sufficient affiliates. Although the cleavage between the metal workers and the chemical workers in particular is not new, the DGB today seems to find it harder than ever to build a united front for tripartite national negotiations.


Welfare corporatism divided the employers and did not prevent the decentralisation of industrial relations and its encapsulation in a shrinking segment of the German economy. But by the late 1990s, it had become one of the last strongholds of a union movement no longer able to gain for its members growing wages or secure employment. Today, involvement in the running of the social security system, with its huge parafiscal budgets, assures industrial unions a national role at a time when wage setting is becoming more decentralised. Indeed welfare corporatism seems like a strange travesty of the gesamtwirtschaftliche Mitbestimmung that German unions had envisaged in the immediate post-war years. Workplace co-determination, to which Katzenstein rightly pays so much attention, was by and large accepted by employers during the 1980s (Kommission Mitbestimmung 1998), the more so as it became in practice increasingly driven by the needs and concerns of individual firms. Very much in Katzenstein's sense, Mitbestimmung continued to function as a parapublic infrastructure of co-operation between labour and capital, but in a way he could not have foreseen: by providing an institutional framework for the managements and workforces of large companies within which to negotiate the externalisation of the social costs of competitive restructuring to the public. In its 1979 ruling on co-determination, which Katzenstein quotes (1987, p. 162), the Constitutional Court still defended the presence of trade union representatives on the supervisory boards of large firms as a way of "preventing or at least diluting the 'enterprise egoism' expected in the wake of expanded co-determination". Clearly this was expecting too much.


Already Katzenstein had noted that German unions were "on the defensive both politically and economically" (p. 28). However, he failed to anticipate the looming disaster of the policy of work redistribution that unions adopted precisely to overcome that crisis. He also believed that somehow, expanded co-determination would spawn new union goals and activities, like "qualitative growth and the deliberate steering of technological change", not to mention "humanisation of work", rivalling and pushing back more traditional trade union concerns (p. 135). But as the years passed successive attempts at redefining the interests of the membership came to nothing. Unable to set new goals for themselves that would have given them a universalistic mission at a time when stable unemployment rendered the defence of high wages and social benefits increasingly particularistic, unions found themselves forced to operate as mere "wage machines" whose claim, which Katzenstein found still widely accepted, to "represent the interests of organized and unorganized workers" (p. 28) was rapidly losing credibility.


Today, having lost much of their industrial power, German unions must rely on their political power - from electoral pressure to influence inside the SPD - to secure for their members, if not employment, then an early exit from employment, and if not rising wages, then high pensions and unemployment benefits. Unlike its monetary policy, the social policy of post-war Germany's semi-sovereign state had turned out wide open to capture by safely established "social partners". During the 1990s it was above all their hold on the welfare state that enabled German unions to defend the standard employment relationship and resist labour market reform without having to worry much about the consequences. In fact given their shrinking and ageing core membership as well as the failure of all alternative programs, they also had good organisational reasons for taking the side of the recipients of benefits against those paying for them, who are mostly young and not organised. [11]



Process: Even Less Sovereignty


By the mid-1990s, less centralised and encompassing organisations of business and labour faced a state whose sovereign capacity had further declined. In Katzenstein's time, semi-sovereignty had taught the German state to make a virtue out of necessity and become adept at co-operation with what once were well-organised interest groups. Now, not only had German society become less organised, but the ability of the state to align the interests of unions and employers with its political needs had further diminished.


As already shown by the fate of Konzertierte Aktion, the German state was never particularly good at formal tripartite negotiations with business and labour. Indeed the same condition of semi-sovereignty that, according to Katzenstein, stands in the way of unilateral state action, also seems to obstruct hard bargaining with interest groups. Corporatist political exchange requires a unified position on the part of the government. But as Katzenstein has shown in impressive detail, the capacity of the Chancellery to control the ministries has always been low; coalition government further weakens the authority of the Chancellor; federalism requires the government to seek the support of a majority of Länder governments for major legislation, which limits the concessions it can offer to organised groups; the budgets of Bund, Länder and local communities are so intertwined that even minor changes in the tax system cause enormous technical and political complications; independently scheduled elections at Länder level result in an almost permanent election campaign, especially as their outcome may change the composition of the Bundesrat; and a whole panoply of courts, most importantly the Constitutional Court, can be called upon to defend established groups rights like Tarifautonomie or the social insurance principle. All of this can offer disaffected interest groups or rivalling parties and party factions ample opportunity to prevent or undo tripartite agreement.


While this had been so for some time, in the 1990s the capacity of the German federal government to impose its will on organised interests further declined. With unification, the heterogeneity of interests in the federal system as well as the number of Länder elections during a federal government's period in office have increased. Moreover, European integration, and especially budget consolidation under the Stability Pact, has ruled out rewarding union co-operation with increases in public spending of whatever sort. (Other concessions, like extended rights to workplace representation or further increases in employment protection, have become difficult to make for other reasons, in particular resistance from smaller firms that employers associations can no longer ignore.) European integration also made the government more dependent on union wage moderation, both because budget consolidation requires public sector wages to remain under control, and because wage moderation represents the main short-term protection for the government against even higher unemployment, with all the fiscal and electoral risks this involves. Aggressive wage bargaining has thus become a potential weapon in the hands of unions threatened by social policy or labour market reform.


Moreover, as Roland Czada (2003) has pointed out, party competition in the German political system has intensified, voting became more volatile, and the electoral majorities of governments have shrunk. International comparison shows that political agreement between the major parties and broad-based and stable national governments may neutralise institutional veto points and make for effective governance even where organised interests have numerous opportunities to intervene. Germany, however, according to Czada, is not only the OECD country with the largest number of veto points, but it is also least likely to produce electoral-parliamentary majorities broad enough to override group interests. As declining majorities in a more volatile electorate lead to intensified party competition, governments become more vulnerable to interest group pressures.


The entrenched centrism of the two big parties that Katzenstein praised as a pillar of political stability has not helped. Regardless of whether the Christian Democrats or the Social Democrats are in power, cuts in social entitlements are always likely to be challenged by the opposition in the next election, whether at federal or at Länder level, making it impossible for the SPD in government to corner the unions on social policy reform. Also, declining ideological coherence of the large parties seems to have reinforced the factionalism bred by a federal system that gives regional party organisations their own electoral base supporting independent leaders with a national role and, often, national ambitions. Similar reasons account for the growing independence of the parliamentary groups of governing parties from their party leaders in government, making it likely that tripartite agreements are reopened by the Bundestag and giving interest groups forced to make concessions in corporatist negotiations a chance to have them undone in the parliament.


Perhaps most importantly, there were irresistible temptations for the federal government to make its peace with welfare corporatism, regardless of the spiral of high non-wage labour costs and unemployment that it entailed (Trampusch 2002). In the 1990s a tacit deal had evolved under which the social partners tolerated the government drawing on the parafisci of the social security system to cover part of the unexpectedly high costs of unification, enabling it to avoid raising taxes or further increasing the public debt. In return for being allowed to collect what in effect was a hidden tax, the state confirmed its commitment to both the extension of the exclusionary West German labour market regime to East Germany and its public subsidisation. It also continued to acquiesce in the colonisation of its social policy by unions and employers. However, while undoubtedly public-private co-operation of this sort contributed to "social peace", it also appears to have compromised government authority and the state's residual sovereignty to an extent that made them even less effective when the skeleton of low employment in the joint welfare-corporatist closet could no longer be ignored. When tax competition in an internationalising economy required cutting corporate taxes while simultaneously Monetary Union demanded budget consolidation, the funding of the welfare state by social security contributions became more safely enshrined than ever, even disregarding the intricacies of the German Finanzverfassung and the indispensability of the welfare state for social peace and government political support in the East.



From the Kohl Failure to the Schröder Deadlock


In the mid-1990s business discontent with declining economic conditions forced the Kohl government to address the issue of high non-wage labour costs. Since incremental change in line with the logic of semi-sovereignty had caused the problem, and would in any case no longer satisfy business and key political allies, Kohl tried to get the unions in tripartite negotiations to agree to a comprehensive reform package. Although IG Metall and DGB in late 1995 came to his assistance with their offer of a national Bündnis für Arbeit, and an understanding was reached in principle to cut non-wage labour costs to less than 40 per cent, Kohl's consensual approach failed and the government in the end resorted to unilateral legislation. This, too, failed when the CDU/CSU lost the 1998 election.


Why did the Kohl Bündnis fall apart? Among business, Mittelstand pressure undermined the position of the traditional leadership. Hans-Olaf Henkel in particular, the chief of the BDI and self-appointed representative of all those dissatisfied with German corporatism, demanded that the government act on its own to reverse the trend of the past decade. With business divided, Henkel and his followers managed to win the support of the FDP, the small liberal coalition party, which was still chafing from its defeat on several social policy issues in the Kohl era. Three Länder elections in the spring of 1996, in which the FDP was unexpectedly successful and the opposition parties gained no ground, seemed to suggest that the electoral risks of unilateral reform were smaller than thought. Within the CDU, Schäuble began to champion a platform of painful social policy reforms, distancing himself from Kohl's policy of habitual compromise. In the end, Kohl was constrained to include in the government position for the Bündnis talks demands that the unions could not but reject or accept political humiliation. As was to be expected, they walked out and the Alliance never materialised.


When tripartite negotiations had come to naught, the government embarked on unilateral measures for labour market flexibility and lower non-wage labour costs. More important than its content, the proposed legislation signalled a departure from Germany's normal politics of consensus and might have opened a political space for further government unilateralism. In addition to a first step towards pension reform and several changes in social benefits, the government catered to Mittelstand sentiments by cutting sick pay, which is paid by employers directly and outside the public social insurance system, for six weeks at hundred per cent of a worker's wage. Sick pay, however, may be regulated not just by law but also by collective agreement. Legislative intervention was therefore bound to cause a conflict of principle over the right of unions to collective bargaining and the proper role of the state. Moreover, as the high level of sick pay was the result of a historical strike victory of the metal workers in the 1950s, it was of symbolic significance for the unions and considered non-negotiable by them.


Immediately after the breakdown of the Alliance, the unions in June, 1996, organised the largest protest rally in the history of post-war Germany. Unfazed, the Bundestag passed the reform package. Shortly thereafter the conflict over sick pay escalated. In October, large firms in the metal industry, like Siemens and Daimler-Chrysler, came under pressure from their works councils, broke ranks with their associations and refused to implement the cuts. The event left employers associations in disarray, intensified conflict in the business camp, and caused bitter mutual recrimination between government and business.


It is today widely accepted that Kohl's unilaterally enacted social policy and labour market reforms contributed to the outcome of the 1998 election. During the campaign, in which it received unprecedented political and financial support from the unions, the SPD promised to rescind the reform legislation. Moreover, Schröder committed himself to a revival of tripartism and to creating the Bündnis für Arbeit that Kohl had not been able to bring about. For the unions this meant that any future changes in social policy or in the labour market regime were to be conditional on their agreement, although for Schröder it may not have been much more than a convenient device to avoid questions for the details of his unemployment strategy. When the SPD had won, its left wing under the then party chairman and Finance Minister, Oskar Lafontaine, insisted that it kept its election promises and scrapped the Kohl reforms immediately. While a Bündnis für Arbeit was established in December, 1998, the legislative changes were never discussed there and were delivered to the unions without anything in exchange.


During Schröder's entire first term, the Bündnis für Arbeit, which started with tremendous publicity, remained deadlocked. Hard bargaining on employment-enhancing labour market and social insurance reforms never happened, perhaps because the government could not agree on its own strategy, perhaps because it had nothing much to offer in exchange. While the government hung on to the Alliance for the photo opportunities it offered an embattled Chancellor, both business and labour seem to have kept it alive basically to be able to threaten the government and each other with their exit. On pension reform, the CDU retaliated for the SPD election campaign of 1998, opposed the government's proposals, sided with the pensioners, and won several Länder elections in 1999. At the end of that year, Schröder and the SPD were pronounced dead and were rescued only by the Kohl party finance scandal. After the experience, unilateralism was out of the question for Schröder, and even more so for the SPD and its parliamentary party. To mend fences with the unions, which continued to refuse to co-operate on institutional reform, they were handed various improvements in co-determination, outside the Bündnis and again without explicit concessions in return.


The only major success the Schröder government achieved in the context of the Alliance was the two-year moderate wage settlement of early 2000, after pent-up demand and the Keynesian rhetoric of Lafontaine as Finance Minister had produced a high wage rise immediately after the change of government in early 1999. Indeed during its entire term, the government spent most of its political capital with the unions on wage moderation - an issue that formally never came on the Alliance agenda due to union insistence on Tarifautonomie - not least because the employers and the public expected it. Wage moderation being as important as it was to the government for domestic as well as European reasons, the unions seem to have extracted in return a promise from the Chancellor to forego unilateral labour market or social security reform. Given the logic of electoral competition and the strong position of the SPD parliamentary party, that promise may not have been difficult to make.


By the middle of Schröder's first term at the latest, the Bündnis für Arbeit was mired in deadlock. Schröder knew that Kohl's attempt to achieve non-incremental change by tripartite agreement had failed, due to unbridgeable differences between what the government felt was necessary and what the unions were willing to accept. But having himself brought upon Kohl the punishment by the voters for his subsequent turn to unilateral reform, Schröder saw no alternative but to try again for a consensual solution. Unfortunately for him, what the unions had learned from the events was that they had no rational reason to co-operate. If corporatist concertation, as has often been pointed out, depends on the government being able credibly to threaten unilateral action, then it was even less likely to happen in Germany after 1998 than before. A state unable to overcome resistance is very likely also unable to make divergent interests pull together. Consensus politics, if it is to accomplish more than the sort of change that would happen anyway, requires not a weak but a strong state. This was interestingly reflected in the frequent German debates at the time of the "Dutch model", which moved from the notion of a solidaristic "Polder society" to the observation that unlike in Germany, there is no Tarifautonomie in the Netherlands, enabling the government to take wage-setting out of the hands of unions and employers if they are unwilling to take government policy into account.


As far as the unions were concerned, neither did Schröder know what to offer them to entice them into co-operation, nor did they know what to ask for, except that everything basically remained the way it was. By the 1990s German governments could no longer count on the gentle push and pull of parapublic institutions bringing organised social interests in line with public interests, as in Katzenstein's world. Instead major parapublic institutions had turned into bridgeheads inside the state of increasingly particularistic interest groups that the government needed to take back in order to regain its freedom of manoeuvre. Growing economic and societal dualism, a tendency that already Katzenstein had observed (p. 147), was not a major concern for a union movement safely entrenched in its encapsulated domain and catering to the interests of an ageing core membership. Nor was it as such a problem for a government that had been more than willing to absorb the external costs of social peace at public expense. It was only when the costs of mutual exploitation for the state exceeded its benefits that governments began to make half-hearted attempts to extricate themselves from welfare corporatism, reasserting as public interest their own institutional interest in stopping the downward spiral of high social expenditures and low employment. But then, the congenital weakness of the semi-sovereign German state made it difficult if not impossible for any government to move unions out of their passive and active resistance that had given up any political ambition beyond the hard-headed economistic defence of their Besitzstand. Nor could the employers associations, forced to take into account the concerns of small and medium-sized firms, be relied upon on any more to help the government buy time. Never having really started, German tripartism had run its course.



Now a Strong State?


Tight interdependence between the state and organised social groups in Germany, according to Katzenstein, makes "large-scale departure from established policies an improbable occurrence" (1978, p. 35). Unable to accomplish radical change, German politics learned to cultivate consensual incremental adjustment, which for a time served the country well as it protected political trust and provided stable conditions for investment and consumption.


Nowhere was it written, however, that incrementalism would always proceed in the direction of economic and political sustainability. Social dysfunctions, especially those that are deeply rooted in stable institutional structures, may take time to accumulate before they become acute. Incrementalism continued in the 1980s in the face of persistent unemployment and growing social dualism, responding to the evolving structural problems with a series of stopgaps that only added to the drift. Imperceptibly but nevertheless fundamentally the underlying logic of interdependence changed, while the economic foundations of the social peace that incrementalism was dedicated to preserving dwindled away.


By the second half of the 1990s the radical changes that had one last time been delayed by the Institutionentransfer to East Germany had become imperative for any government interested in its capacity to govern. But turning around a long-term evolutionary trend exceeded the capacities of a semi-sovereign state whose principal achievement had been smooth co-operation with organised civil society in the very incremental adjustment that had now become obsolete. Both Kohl and Schröder had to learn that not only was there no incremental solution to the economic and political problems facing their governments, but there also was no prospect of consensual radical change. Some incremental change there was, in the right direction and even under Schröder. Labour market policy was cut back somewhat; part of the costs of early retirement was internalised in collective agreements; changes in the practice of collective bargaining allowed individual firms more space for customised workplace agreements; and a funded supplementary pension scheme was introduced to alleviate the pressure on the public pension system. But none of this did anything significant to break the spiral of high payroll taxes and low employment, or more than to prevent the economic burden of the welfare state on the labour market from growing even faster.


Incrementalism being capable no longer to align the interests of state, unions and employers - not to mention attacking the root causes of endemic under-employment - state weakness in the 1990s became exactly that: a weakness that spelt stagnation and decline. The semi-sovereign German state had fitted in surprisingly well with the managed capitalism after 1945, when the task was to embed a liberal market economy in the post-war settlement. In the high time of semi-sovereignty, an organised and disciplined society kept a weak state in check while compensating for its disabilities. But to discipline interest groups that used their political privileges to provide for their members at public expense, a weak state is not strong enough.


Today the favourite means of governments trying to discipline rent-seeking interest groups is liberalisation. But paradoxically, liberalisation seems to require a strong state, particularly in a well-organised civil society. In this respect semi-sovereignty is both too much and too little. It is too much in that it involves deep public intervention, by the state in co-operation with organised social groups, in the functioning of the market economy. But it is also too little in that a liberal economy requires a state capable of overriding organised interests that refuse to internalise the costs of their behaviour and thus have ceased to be responsive to public interests. If the taming of organised interests requires less state sovereignty over the economy, it also seems to require more state sovereignty over the society.


In early 2002, with defeat in the upcoming election almost a certainty, Schröder exploited a minor scandal in the giant patronage machine of the Bundesanstalt für Arbeit to dislodge its self-government and call for fundamental labour market policy reforms. In a striking departure from the tripartism of the by then defunct Bündnis, he appointed a commission under the leadership of VW personnel director, Peter Hartz, to devise a plan of how to cut unemployment by half within two years. Only two of the 15 members of the Hartz-Kommission were representatives of trade unions, and only one represented the established organisations of business. Under the pressure of the election campaign, with the CDU ridiculing the Kommission as Schröder's last straw, both the unions and the SPD were blackmailed by the chancellery into supporting the conclusions that Hartz, under public prodding from the Chancellor, put forward about six weeks before election date.


When against all expectations Schröder won a second term, everybody believed the brief exercise in state unilateralism to be over. For a while this was exactly what seemed to be coming, and once again, like in 1998, the unions got ready to collect the rewards for their campaign support. But then, all of a sudden Schröder effectively abolished the once so powerful Ministry of Labour - the union stronghold inside the government machinery - and merged its largest part into the Ministry of Economic Affairs. To this he appointed a political heavy-weight from the right wing of the SPD, Wolfgang Clement. Shortly thereafter, a series of incremental emergency measures necessary to balance the state and social security budgets - measures that had been put off not to disturb the collective amnesia and the good feelings that had turned around the election, but that otherwise the political class rightly considered business as usual - caused an unprecedented loss of confidence in the government. In part this was because nothing of this sort had only been hinted at during the campaign, but probably also because it somehow transpired that no end to business as usual was anywhere in sight.


Apparently terrified by the prospect of an unending series of similarly unpopular incremental changes consuming his entire second term, by the end of 2002 Schröder and Clement pushed the Hartz recommendations through Bundestag and Bundesrat, including the first significant measures ever taken to improve the employment prospects of the low-skilled. Rather than allowing the unions to veto its policy, as it had so often in his first term, the government sided with the opposition, taking its own parliamentary party by surprise. Simultaneously it appointed another commission, along similar lines as the Hartz-Kommission and once again over the head of the minister in charge, to propose fundamental reforms of the social security system. In the face of crucial regional election, Schröder chose his New Year address to announce that all groups in society would have to make sacrifices. Meanwhile reorganisation of the Bundesanstalt, under its new, government-appointed leadership, was proceeding while the unions were watching in disbelief. If Schröder continued his course - which with him could never be certain - a battle over who was running economic policy, the government or the unions, seemed finally to be approaching. [12]





Artus, Ingrid, 2001: Krise des deutschen Tarifsystems: Die Erosion des Flächentarifvertrags in Ost und West. Wiesbaden: Westdeutscher Verlag.


Bispinck, Reinhard, 1997: Deregulierung, Differenzierung und Dezentralisierung des Flächentarifvertrags. Eine Bestandsaufnahme neuerer Entwicklungstendenzen. In: WSI-Mitteilungen 8/1997, 551-561.


Czada, Roland, 2003: Der Begriff der Verhandlungsdemokratie und die vergleichende Politikforschung. In: Renate Mayntz and Wolfgang Streeck (eds.), Die Reformierbarkeit der Demokratie: Innovationen und Blockaden. Frankfurt am Main: Campus, 173-204.


Ebbinghaus, Bernhard, 2002a: Dinosaurier der Dienstleistungsgesellschaft? Der Mitgliederschwund deutscher Gewerkschaften im historischen und internationalen Vergleich. MPIfG Working Paper 02/3. Köln: Max-Planck-Institut für Gesellschaftsforschung.


Ebbinghaus, Bernhard, 2002b: Exit from Labor: Reforming Early Retirement and Social Partnership in Europe, Japan and the USA. Habilitationsschrift. Köln: Wirtschafts- und Sozialwissenschaftliche Fakultät der Universität zu Köln.


Hassel, Anke, 1999: The Erosion of the German System of Industrial Relations. In: British Journal of Industrial Relations 37(3), 483-505.


Hassel, Anke, and Britta Rehder, 2001: Institutional Change in the German Wage Bargaining System: The Role of Big Companies. MPIfG Working Paper 01/9. Köln: Max-Planck-Institut für Gesellschaftsforschung.


Katzenstein, Peter J., 1984: Corporatism and Change: Austria, Switzerland and the Politics of Industry. Ithaca: Cornell University Press.


Katzenstein, Peter J., 1987: Policy and Politics in West Germany: The Growth of a semi-sovereign State. Philadelphia: Temple University Press.


Kommission Mitbestimmung, 1998: Mitbestimmung und neue Unternehmenskulturen. Bericht der Kommission Mitbestimmung der Bertelsmann Stiftung und der Hans-Böckler-Stiftung. Gütersloh: Verlag der Bertelsmann Stiftung.


Lehmbruch, Gerhard, 1979 [1974]: Liberal Corporatism and Party Government. In: Phillippe C. Schmitter and Gerhard Lehmbruch (eds.), Trends Towards Corporatist Intermediation. London and Beverly Hills: Sage.


Lehmbruch, Gerhard, 1998: Zwischen Institutionentransfer und Eigendynamik: Sektorale Transformationspfade und ihre Bestimmungsgründe. In: Roland Czada and Gerhard Lehmbruch (eds.), Transformationspfade in Ostdeutschland. Frankfurt am Main: Campus, 17-57.


Olson, Mancur, 1982: The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities. New Haven: Yale University Press.


Scharpf, Fritz W., 1991: Crisis and Choice in European Social Democracy. Ithaca, N.Y.: Cornell University Press.


Scharpf, Fritz W., 1993: Verhandeln, wo nicht regiert werden kann: Handlungsfähigkeit und Legitimation der Politik am Ende des 20. Jahrhunderts. In: Beirat für Wirtschafts- und Sozialfragen (Hrsg.), 30 Jahre Beirat für Wirtschafts- und Sozialfragen: Daten, Fakten, Perspektiven. Studien des Beirats für Wirtschafts- und Sozialfragen, Nr. 68. Wien: Ueberreuter, 23-41.


Schmitter, Philippe C., 1979 [1974]: Still the Century of Corporatism? In: Phillippe C. Schmitter and Gerhard Lehmbruch (eds.), Trends Towards Corporatist Intermediation. London and Beverly Hills: Sage, 7-52.


Schroeder, Wolfgang, 2000: Das Modell Deutschland auf dem Prüfstand: Zur Entwicklung der industriellen Beziehungen in Ostdeutschland (1990-2000). Wiesbaden: Westdeutscher Verlag.


Schroeder, Wolfgang, and Burkard Ruppert, 1996: Austritte aus Arbeitgeberverbänden. Motive - Ursachen - Ausmaß. In: WSI-Mitteilungen 11/1996, 316-329.


Schudlich, Edwin, 1982: Kooperation statt Korporatismus. In: Ulrich Billerbeck (eds.), Korporatismus und gewerkschaftliche Interessenvertretung. Frankfurt am Main, New York: Campus, 127-175.


Streeck, Wolfgang, 1991: On the Institutional Conditions of Diversified Quality Production. In: Egon Matzner and Wolfgang Streeck (eds.), Beyond Keynesianism: The Socio-Economics of Production and Employment. London: Edward Elgar, 21-61.


Streeck, Wolfgang, 2001a: High Equality, Low Activity: The Contribution of the Social Welfare System to the Stability of the German Collective Bargaining Regime. Review Symposium on Harry C. Katz and Owen Darbishire, Converging Divergences: Worldwide Changes in Employment Systems. In: Industrial and Labour Relations Review 54(3), 698-706.


Streeck, Wolfgang, 2001b: Tarifautonomie und Politik: Von der Konzertierten Aktion zum Bündnis für Arbeit. In: Gesamtverband der metallindustriellen Arbeitgeberverbände (eds.), Die deutschen Arbeitsbeziehungen am Anfang des 20. Jahrhunderts. Wissenschaftliches Kolloquium aus Anlass des Ausscheidens von Dr. Werner Stumpfe als Präsident von Gesamtmetall. Köln: Deutscher Institutsverlag, 76-102.


Streeck, Wolfgang, and Philippe C. Schmitter, 1984: Community, Market, State and Associations? The Prospective Contribution of Interest Governance to Social Order. In: European Sociological Review 1, 119-138.


Trampusch, Christine, 2003: Ein Bündnis für die nachhaltige Finanzierung der Sozialversicherungssysteme: Interessenvermittlung in der deutschen Arbeitsmarkt- und Rentenpolitik. MPIfG Discussion Paper 03/1. Köln: Max-Planck-Institut für Gesellschaftsforschung.


Willke, Helmut, 1983: Entzauberung des Staates: Überlegungen zu einer sozietalen Steuerungstheorie. Königstein/Ts.: Athenäum.







The paper is based on a first review of information collected in the context of joint research with Christine Trampusch on the origins and the effects of the "Alliance for Jobs" of the Schröder government, and more generally on the changing relations between organised interests and the state in Germany since the 1980s. I am grateful to Christine Trampusch, Anke Hassel and Britta Rehder for constructive criticism and valuable suggestions.


Katzenstein documents a memorable article by Schmidt, published in the New York Times in May, 1976, where he points out that, "The entire process of setting wages and salaries is the exclusive responsibility of collective bargaining partners. The Government and Parliament would not even dream of changing this. In our experience, there exists no better solution" (Katzenstein 1987, p. 156).


That the chemical workers looked for alternative ways to cut the labour supply had to do with their particular experience in the late 1960s and early 1970s, which was a lost strike. It convinced their leaders that in their industry, successful industrial action was not possible (Schudlich 1982).


1984 was the same year when Thatcher crushed the miners strike, which gave rise to much comparative comment on the outstanding strength of German unions and the extraordinary stability of the German industrial relations system.


Simultaneously growing international competition and the shift to lean production made it more difficult for employers associations to call a lock-out. In spite of the decline in union bargaining power caused by the legal changes after 1984, this may have reinforced the tendency of large employers to accept high wage settlements and deal with them by driving up productivity and trimming down workforces, thus adding to unemployment.


It is important to note that apart from the one piece of legislation passed in 1984, there were several other instruments of defensive labour supply management that had been around since the 1950s and 1960s. Then, however, they had been used only to deal with individual sectoral crises, such as in coal mining and steel. The same was true for labour market policy which in the 1990s, when early retirement through the pension system had reached its limits, turned into a holding pen for the unemployed. The use of the welfare state in the 1980s and 1990s for removing excess labour from the labour market was the result of incremental change that had not been originally intended. On early exit policies in Germany and elsewhere, see Ebbinghaus (2002b).


Giving an entirely new meaning to Helmut Schmidt's statement in 1976, also in the New York Times article quoted above, that the "social peace" for which Germany was envied in the world was "primarily an achievement of our well-developed system of social security, combined with the principle of autonomy under which the labor unions and employers' associations are allowed to pursue their interests and negotiate their differences" (quoted in Katzenstein 1978, p. 156).


This compares to 3 per cent, 64,000 employees, and a budget of DM 33bn in 1985, at Katzenstein's time (Katzenstein 1987, p. 69).


More recent data on works council coverage are not available as the Deutscher Gewerkschaftsbund has not yet published aggregate information on the works council elections of 1998.


Where wage moderation was partly caused by union concern for the cohesion of employers associations, which is essential for industry-wide collective bargaining.


An important yet under-researched aspect of welfare corporatism seems to be that involvement in the management of labour market policy offers unions, and to an extent employers associations, indirect access to financial resources enabling them, among other things, to maintain their organisations even where, like in the East, they have only few members. A central role in this seems to have been played by the Bundesanstalt für Arbeit which, until its reorganisation after the 2002 election, was basically controlled by the "social partners". The Bundesanstalt spends billions every year on retraining and further training, which in part go to organisations controlled by unions and employers associations.


Semi-sovereignty, as Katzenstein often reminds us, was not created to support German prosperity but to tame German power. For the outside world, its main benefit was not its economic success but that there was no longer a strong German state. That today Germany lacks the state capacities it would require in a changed world to recover its economic strength need at first glance not be of concern to other countries, especially those that in the past were at the receiving end of Germany's extraordinary economic competitiveness. But Schadenfreude may be expensive as the size of the German economy is such that its decline inevitably drags down its neighbours as well. Ending the domestic semi-sovereignty of the German state may be not just in the German interest.



Copyright © 2003 Wolfgang Streeck

No part of this publication may be reproduced or transmitted without permission in writing from the authors.
Jegliche Vervielfältigung und Verbreitung, auch auszugsweise, bedarf der Zustimmung der Autoren.

MPI für Gesellschaftsforschung, Paulstr. 3, 50676 Köln, Germany





MPIfG:  MPIfG Working Paper 03/2

[Zuletzt geändert am 29.03.2007 11:00]