Welfare-through-work and the re-regulation of labour markets in Denmark.

Introduction

This paper is positioned within theoretical perspectives that focus on welfare states as systems of power and negotiation between key social forces acting in and through the state apparatus. In this context, we suggest that Labour's welfare-to-work programme is beginning to generate considerable debate on the re-regulation of labour markets. Although there is an emerging consensus that this strategy is deeply problematic, there appears to be reluctance, within the UK debate, to discuss viable alternatives to neoliberalism. This paper is intended to stimulate discussion on this issue and to contribute, in doing so, to theories of welfare state restructuring by focusing on the social regulation of labour markets in Denmark. In stark contrast to UK and North American strategies, Denmark has adopted a 'welfare-through-work' model, which is built around a more inclusive system of welfare reform. The paper discusses the emergence of that model, focuses on the importance of 'Job Rotation' as its leading-edge socio-economic strategy, highlights recent conflicts and tensions within Job Rotation, and suggests lessons for Britain.

Recent attempts to reform the British welfare state through Labour's welfare-to-work initiative are generating considerable debate on the re-regulation of labour markets (Finn, 2000; Lister, 2001; National Audit Office, 2002; Nativel et al., 2002; Peck, 1999; Sunley et al., 2001; Turok & Edge, 1999). Welfare-to-work represents a significant strategy to develop behaviourist and supply-side models of labour market regulation, and this policy is becoming commonplace in developed capitalist societies (Lodemel & Trickey, 2000; OECD, 1999; Peck, 2001).

Debates here have highlighted the fact that supply-side initiatives represent a new mode of social control that leads to widening income inequalities and a downward spiral of low skills and low pay, which ultimately impacts on social cohesion and economic competitiveness (Carlson & Theodore, 1995; Grover, 2003; Pascual, 2002).

In this paper, we suggest that Denmark's 'activation' reforms have embodied elements of workfare, but that they also incorporated a more 'social inclusive' model, which holds lessons for those seeking to address the contradictions of neoliberalism (also see Jorgensen. 2002). This model involves three elements in what we have termed, elsewhere, a 'welfare-through-work' political strategy (Bewick et al., 1997; Etherington, 1998; Etherington & Jones, 2004; cf. Totting, 1999; Ploug, 2002).

First, social partnerships have been strengthened in policy formulation and implementation at all levels of governance. Second, financial planning and decision-making has been decentralised to regionally-based institutions. Third, the unemployed have been given rights to counselling, an individual action plan and, more importantly, access to a comprehensive package of job training, Job Rotation, education and childcare leave schemes.

This strategy is underpinned by the central role of the public sector and local government in the implementation of work- and education-based programmes.

Following a broad discussion on theoretical frameworks for comparing the Danish and UK welfare state, this paper addresses the key features of Denmark's welfare-through-work model. It then focuses on an initiative, Job Rotation, which has been integral to the labour market reforms and was conceived by the Danish labour movement. This is followed by an assessment of some of the current tensions within the Danish model, and specifically of Job Rotation within this, after which potential lessons for the OK are drawn out. Before starting this discussion, however, we need to make it clear that we are not suggesting that the Danish model can be uncritically exported through policy-transfer, and cloned in Britain. Instead, Denmark offers a number of guiding principles that can shed light on the problems identified above. Welfare-through-work provides a discourse and political strategy for actively taking this agenda forward and, in doing so, for formulating a credible challenge to neoliberalism.

Welfare regimes and labour market regulation: A comparative perspective

When thinking about shifts within the form and function of the welfare state, we argue that public policies are contingent on political and social struggle. The underlying differences between nation states, therefore, needs to be located in theoretical perspectives that privilege the changing 'balance of social and class forces'. The modern representative state is the culmination and condensation of bourgeois political power, which brings social classes together in both harmony and conflict and, as such, each capitalist state defines a particular relationship of classes within a given territory (cf. Jessop, 1990, 2002; Moran, 1997; Poulantzas, 1978). Our comparative perspective, therefore, takes on board an analytical framework that views a welfare regime as embodying historically-formed class (struggle) relations, and policies as contingent upon the balance of social forces and specific forms of political struggle (cf. Huber & Stephens, 2001; Lavalette & Mooney, 2000). Engaging in such theoretical debates around this issue is important, because the key reason for comparing policies' trajectories is to understand the role and strategic particularities of political mobilisation in the context of welfare-to-work policy formation. As we explore further, below, in Denmark the retention (until now) of its particular model lies with the fact that the balance of political and social forces have been such that capital has been unwilling, or unable, to impose the particular strategies that have characterised more than twenty years of a relatively transparent class offensive against labour in the UK.

Within the framework of welfare regime analysis pioneered by Esping-Andersen (1992, 1999), the Danish welfare state can be characterised as 'social democratic', because of its strong orientation towards income redistribution and the role of the public sector in the provision of welfare and social services. Esping-Andersen's concept of welfare regimes as systems of power and negotiation between key interests and actors is useful in terms of understanding the social and political dynamics of labour regulation. In this respect, and in contrast to the 'liberal' regimes found in the us and Britain, the social democratic welfare regime consists of strong labour movements and trade unions, reflected in their relatively high employment to union membership ratios. The power configuration of this welfare regime is, therefore, frequently constructed around corporatist networks and institutional arrangements in which trade unions are key 'bargaining partners' in the formulation and implementation of economic and social policy-making. Through this approach, welfare regimes can also be explored as 'labour market regimes', whereby institutions and policies in a social-democratic context are geared towards labour market integration. Crucially, labour regulation comprises employment rights and protection, wage regulation and the minimum wage, and active labour market policies. Furthermore, other aspects of welfare policies that enhance integration, such as childcare provision and regulations on maternity rights, are also of importance within a social democratic model.

There are, however, problems with exaggerating the positive aspect of certain 'models' and underplaying some of their internal contradictions and instabilities, which are endemic features of the Keynesian and social democratic strategies favoured by Esping-Andersen (1999; cf. Cochrane, 1993; Hamnett, 1996). These themes are explored by Coates (2000), who takes a different perspective and suggests that capitalism comprises different types of 'models' of economic growth, each involving different types of welfare state, which are conditioned by the roles played by labour in relation to economic competitiveness. This approach is novel in that the examination of different 'models', as such, is framed within a perspective that emphasises the inherently contradictory and unstable nature of capitalist accumulation.

In particular, 'left' visions of a more democratic, high skill, high social investment model frequently ignore the realities of global competition, and especially strategies for exporting capital to exploit uneven development (Coates, 2000: 9-10, 244). Furthermore, Coates also argues that skills development and training needs to be analysed within a wider understanding of class relations, within which the power of capital is seen as a motivating force for shaping training-based institutional structures and social relations.

In this respect, strategies for training and labour market adjustment by social democratic parties have largely ideological, rather than practical, components as mechanisms for regenerating economies. For example, active labour market policies which Esping-Andersen (1999) seems to accept a priori as inevitable, and even positive, can reflect, or indeed embody, a retreat by the centre-left parties from influencing capital and confronting dominant neoliberal explanations of unemployment, skill shortages and employment problems (Coates, 2000: 120).

The usefulness of Coates's approach, then, is in its assessment of capitalism as an economic and social system, which also acknowledges the limitations and barriers for achieving 'socially progressive' policies within the general framework of liberal or social democratic economies.

Set against Coates's concerns, comparative frameworks must also consider the role played by globalisation as a neoliberal political strategy for establishing greater control by capital over labour power through restructuring the state apparatus (see Edwards & Elger, 1999). But in this context, we also would maintain that the influence of class, social and gender interests on processes of state restructuring has, in turn, actively produced different welfare settlements, such that a capital-logic approach becomes untenable. Following Jessop (1990, 2002), state power has to be viewed relationally and with respect to the different forces acting in and through the variegated state apparatus--especially in terms of struggles over spatial scale and the resulting politics of territory.

In this respect, according to Jessop (1993, 1999), key transitions and changes in capitalist welfare regimes have being taking place in recent years, from Keynesian Welfare National States to Schumpeterian Workfare Post-National Regimes (SWPR). Such shifts are related to the crisis in the traditional post-war welfare settlement of redistribution and regulation: a more contradictory and unstable regime, based on the subordination of policies to the needs of competitiveness, and involving privatisation and re-commodification, is deemed to be occurring. Within the SWPR, the national territory is no longer the sole 'power container'.

Instead, policy-making functions are being shifted (or 'hollowed-out') upwards, sideways and downwards, and in this respect state restructuring holds implications for the geographies of policy formulation and implementation (Jessop, 1999). Crucially, sub-national politics are increasingly influential in shaping regulation, and the different trajectories involved in state and policy restructuring are contingent on the balance of political forces, institutional legacies, and changing economic and political conjunctures.

Following the logic of Jessop's argument, emphasis, within the SWPR, is placed on sub-national governance, and regional and local scales appear to be playing more of a role in policy delivery. Critically though, this scalar shift does not necessarily mean a whole-scale devolution of power. It is often the case that a complex reordering of the relations between different levels of governance is occurring.

Denationalisation, therefore, does not imply the end of national state power: it signifies its scalar re-articulation, and the form of this is dependent on the particular policy-sector being analysed and the various social forces acting through the state therein (MacLeod & Jones, 1999).

In some respects, this could mean an increase in the control functions of the state through centralisation, a strategy whose use has been geographically uneven throughout North America and Western Europe, in the case of labour market policy.

For Jessop, then, although welfare regimes are shifting in a general direction towards 'workfare' and disciplinary forms of social policy, there are marked differences between different national states such as the OK and Denmark, due to the specific balances of class and social forces acting through the state apparatus.

In the UK, the configuration of class relations and the balance of social forces that has emerged since the 1930s needs to be understood in relation to an inherently reformist and 'conservative' labour movement, and a relatively weak industrial relations system relying heavily upon voluntarism.

In essence, the post-war consensus and compromise was based on a highly unequal form of corporatism, whereby the trade union and labour movement's influence upon social policy was weak (Elger & Edwards, 1999).

Furthermore, patriarchal aspects of welfare entitlements, and the notion of full-employment being related to male employment, were never fully challenged by the labour movement.

We would argue that New Labour's welfare-to-work programme builds on these contradictions and legacies, and that in many ways it is based on furthering labour market deregulation. And, when combined with the privatisation of public services, this particular neoliberal strategy is bound up with shifting the relations between labour and capital (cf. Ginsburg, 2001; Hay, 1999). The situation in Denmark is somewhat different, which we now turn to discuss.

Denmark's welfare-through-work strategy

Historical background

Danish social democracy was born out of a class struggle, at the end of the nineteenth century, which had been focused around establishing a system of employment and trade union rights. A major agreement was reached in 1901, following a bitter lock-out that established management's right to manage, whereby the trade union movement received rights to association and representation in policy decision-making, and a series of welfare reforms followed providing social insurance, health and universal benefits.

For capital, these concessions embraced a collective bargaining system, which included industrial peace agreements that placed quite severe restrictions on the organisation of strikes and industrial action. Based on this legacy, the strong links between the labour and trade union movements and labour market policy lies with the trade union management of unemployment insurance funds (UIFs), and the active involvement of the Workers' Educational Association (AOF) in both the politics and implementation of vocational training (Etherington, 1997b).

The system of both centralised and decentralised collective bargaining has, therefore, remained more or less intact, benefiting both class and labour interests. From the point of view of labour, it has retained trade union access and influence over social and labour market policies, and therefore high levels of income transfer and redistribution are possible. From capital's point of view, it has guaranteed the state underwriting of the costs of social reproduction and skill development, and the compliance of labour within the workplace (see Bender et al., 1998; cf. Lind, 2000).

Building on this legacy, Denmark's welfare-through-work programme involves three important principles: needs-orientation, with a relatively generous benefits system; decentralisation; and the active involvement of the social partners (Goul-Andersen, 1997). According to Moller (1999; see also Compston & Madsen, 2001; Slim, 1998), there have been three main strategies at work, which must be analysed in order to understand the restructuring of Denmark's welfare state over the last decade. The first of these is a liberal strategy, pursued by the conservative parties and advocating a more market-based programme, including lowering benefits and the minimum wage. The unions and the Social Democrats, however, successfully resisted this, although certain changes to collective agreements have been implemented as a compromise.

The second (dominant) strategy, and the one favoured by the Social Democrats and union leadership, relates to the compulsory inclusion of the unemployed in the labour market, but also places increasing responsibilities on the private sector to provide training and job opportunities. It stresses a central role for the public sector in managing training programmes, but in doing so also embodies a broadening of the welfare state beyond that of local government as monolithic provider.

The third key strategy is one supported by the trade unions and the labour movement, and elements of this have been strategically and selectively incorporated by the Social Democrats. The trade union movement has advocated a strategy built around social solidarity, pursuing work-sharing and changes to the balance between work and family life. The labour movement has thus supported childcare leave schemes and the expansion in higher quality vocational training.

The 1994 labour market reforms focused around 'activation'. For the first time, the Danes introduced workfare-style policies, in that access to benefits was conditioned by acceptance of various educational leave and/ or employment training offers. This principle has been extended through the Active Social Policy Act 1998 (see Danish Government, 2000; Lodemel, 2000; Ploug, 2002). Although elements of the measures introduced by the Ministry of Labour in 1994 were concerned with reducing the period during which the unemployed could receive benefits--provoking heavy criticism from the trade unions representing unskilled workers--in stark contrast to the neoliberal welfare-to-work model, the reforms also introduced an integrated training and job-placement package. The main purpose of these is not based on moving towards a work-first model of workfare; we would suggest that this human-capital strategy plays an important role in regulating the supply of labour, through a 'leave programme' and targeted training initiatives.

Paid Leave Schemes comprise educational, sabbatical and parental/childcare initiatives, although the sabbatical scheme was originally experimental and has now been abandoned (see below). Both the employed and the unemployed have rights to participate in these programmes, and to receive an income equivalent to 70 per cent of the maximum unemployment benefit. 'Job Rotation' (see below) involves the unemployed replacing those participating in the educational leave schemes, whereby both receive some form of planned vocational training, and the unemployed secure mentored and supervised work-based experience.

As part of the government's desire to improve 'family-friendly' employment policies, an extended Parental Leave initiative was introduced as a legal right for all workers (men and women), including the unemployed. This is open to parents with children under the age of nine for periods of between 13 and 52 weeks, with only the first 26-week period being a legal right, and the remainder being based on negotiation with the employer or with the Employment Service.

Parents taking leave receive 70 per cent of the maximum unemployment benefit rate for the period of entitlement (reduced to 60 per cent from April 1997). The scheme is financed by the 'labour market contribution' fund (see below), so no other financial assistance is expected from the employer.

However, the parent is not allowed to work, nor to enrol on public education or training courses for the period of leave. Furthermore, parents taking leave with very young children are prohibited from making use of publicly-funded childcare, except in special circumstances. Participation rates in all three forms of leave scheme are detailed in Table 1.

These measures have been supplemented with new initiatives to target the most vulnerable groups in the labour market, and at the same time to begin to address the problems of market failure. The Danish welfare reform strategy is genuinely concerned with issues of creating an inclusive labour market (Ploug, 2002). Again, in stark contrast to the UK's New Deal, the Job Training Scheme was introduced into both the public and private sectors, where both pay and working conditions are regulated through collective agreements (i.e. union-negotiated rates). This is buttressed by a wage subsidy, which is paid to employers. Again, unlike the British case, this operates alongside demand-side policies to regulate the rogue behaviour of employers.

As part of a 'labour market contribution' tax, employers contribute around institutional arrangements for implementing labour market and welfare policies. Policy planning is decentralised from the Central Labour Market Council and the Ministry of Labour to Regional Labour Market Councils (RLMCS), which operate on a tripartite basis, with boards comprising equal membership of the trade unions, local government and employers. The 1994 reforms had major consequences for local government, allocating responsibilities for 'activating' those receiving social security, and providing the basis for formulating labour market plans within localities in collaboration with trade unions and private employers (in addition to their existing powers, on the subject of which, see Goul Andersen, 2002).

Furthermore, in addition to local authority interests being represented in corporatist networks and forums, the role of the Local Government Association (Kommunernes Landsforegningen) is crucial in terms of its 'steering and advising' local authorities, and acting as a national pressure and lobbying organisation for local authorities (Ploug, 2002).

Set against this important institutional context, the key labour market programmes are implemented by the Employment Service (Arbejdsformedlingen), for unemployed members of the unemployment insurance system, and by local government for those claiming the social assistance administered by the local state. Fourteen Regional Labour Market Councils, whose boundaries are coterminous with the County Council system, undertake the implementation of labour market programmes.

The RLMCS are corporatist-style institutions, with planning and implementation undertaken by the 'social partners', i.e. local government, trade unions and the employers. This mirrors the composition of the Central Labour Market Council. LMC boards have executive status, are supervised by the central government Labour Market Authority, and their policies and plans are subject to approval by the Ministry of Labour. Labour market policy, therefore, reflects the geographies of local labour markets in Denmark in that a trend towards decentralisation and intervention in regional/ local economies by the state, and involving social partners, has been implemented (Ploug, 2002).

Moreover, because RLMC boundaries are coterminous with local authorities, there is an inclusive scaling of labour market politics and an inclusive politics of labour market scale. Compared with the current Regional Development Agencies, and with the former Training and Enterprise Councils in England--which provide an institutional platform for the incorporation of private-sector interests into the state apparatus--the RLMCS are more inclusive because of their statutory requirement for equal representation by trade unions, local authorities and private employers (Etherington, 1998).

Furthermore, we would argue that, because there are important connections between these structures of governance and the particular patterns of policy intervention, the public sector has remained central to processes of 'activating' the unemployed. These issues are explored further through a brief discussion on the politics of Job Rotation in the city of Aalborg.

The evolution of Job Rotation

Processes of 'hollowing out' via decentralisation, leading to new geographies of state restructuring, are associated with initiatives, innovations and mobilisations which vary according to the levels and forms of organisation and political influence acting within specific localities. Aalborg represents an interesting example, where the nature and success of labour market policy, exemplified by the operation of Job Rotation, can be partially explained by local labour movement traditions.

Aalborg, the third largest city in Denmark, is situated in the north-eastern corner of Jutland. Its economic geography is made up of Denmark's staple 'traditional' industries, such as textiles and furniture. The city is also home, in terms of employment and trade, to a declining port. Its social and class structure is closely linked to rural depopulation, following a decline in agricultural employment, which has involved population movements from the countryside into the city. Major employment changes have involved the growth of the welfare state, with the local authority and county council headquarters being situated in the city.

This has influenced the dynamics of local politics, with a broad layer of professional workers and public sector trade unions shaping the social and economic policies implemented by the city council. Set against this, the Social Democratic Party has controlled the city council for the past 30 years. The city's core manufacturing base comprises a proletarian working class who are members of the unskilled workers unions (e.g. SIP and KAD), which tend to be aligned, politically, to the left wing of the Social Democratic Party. The social and political dynamics of labour market policy in Aalborg, therefore, must be understood in relation to the relative dominance of the Social Democratic Party, and connections with trade union and labour movement organisations (Bender et al., 1998; Flyvbjerg, 1998).

The implementation of labour market policies, post-1994, is overseen by a committee or forum, which comprises the key social partners (trade unions, employers, local government, and politicians). This committee has a large degree of influence on council policy-making. This emphasises the importance of informal networks, as well as formal procedures, in the realms of power-bargaining in the Danish context. For instance, there are close local ties and relationships between an active trades unions council and Social Democratic politicians, which have been significant in shaping the particular focus of labour market initiatives, to the extent that the labour movement is heavily involved with many of the counselling services established by the local authority (Aalborg Kommune, 1997).

Thus, social institutions such as local government and the trade unions are allocated important decision-making and participatory roles within the overall framework of policy formation and implementation (OECD, 2001; Ploug, 2002). The role of labour movement organisations is particularly significant here. The Workers' Educational Association (AOF) is a major training provider at the local level, and the Unemployment Insurance Trusts (uIFs) play an important role in cooperating with local labour market institutions such as the Employment Service and, more recently, with local authorities, in terms of developing joint initiatives for labour market integration. Trade unions and labour organisations have, therefore, become incorporated into the welfare state system, mainly through the medium of the Trades Council and the activities of its individual branches.

For example, Aalborg Trades Council is represented on the various labour market programme forums, and the AOF ensures that the quality of vocational education meets the needs of the employed and the local economy. And although the planning and coordination of labour market initiatives is undertaken by the North Jutland Labour Market Council, there are many informal networks established between the different interests, where channels of representations are made in the LMC which assist in building consensus as well as in managing conflict. Crucially, the mobilisation of the unions at the grass-roots level (through the UIFs and branches) has assisted the development of employment and welfare initiatives that are both sponsored and controlled by the unions.

Following high unemployment in the 1980s, trade unions made demands on the distribution of work, and work-sharing was proposed as a solution to the unemployment crisis. Out of this strategy emerged the 'Job Rotation' model, which was piloted in a medium-sized textile factory in Aalborg (North East Jutland) during 1993, and co-ordinated by the Danish Workers Educational Association and trade unions with co-operation from the firm's management and labour market authorities. As discussed above, Job Rotation relates to a model of work-sharing where unemployed people are given direct job training experience, and unskilled workers are released to update their training and education. The unemployed receive work experience at trade union-negotiated rates, as well as additional vocational training. The employed obtain additional vocational training and the firm (or public-sector organisation) benefits through an 'upskilled' workforce, without losses in employment (EU Jobrotation, 1996; Etherington, 1997a; Job Nord, 1997).

The success of the pilot scheme was brought to the attention of the national labour market authorities, and Job Rotation has subsequently been incorporated into the 1994 and 1998 labour market reforms. Job Rotation is also being flagged as a model of good practice for employment and lifelong learning throughout Europe (see Etherington et al., 1999; European Commission, 2000).

As Figure 1 indicates, the Job Rotation model is sophisticated. It involves a detailed planning process and the active role of partners, particularly the trade unions, who represent both the unemployed and the employed. Due to this, Job Rotation is a complex and potentially time-consuming process, involving a multitude of different financial and policy programmes. This concern, however, has to be balanced by the fact that Job Rotation is a key long-term instrument for nurturing training and skills development. It represents an important framework for connecting supply-side policies with demand-side initiatives, addressing, in the process, the failure of markets to provide the necessary level of skills (cf. Parker, 2001). Accordingly, the type of educational portfolio drawn up within Job Rotation does not necessarily involve purely work-related training: emphasis is placed on broader and territorially-sensitive educational and skills development (EU Jobrotation, 1996).

[FIGURE 1 OMITTED]

Evaluation research reveals that the key motivation, for Aalborg City Council, for using Job Rotation in the area of childcare nursery provision is updating professional qualifications by releasing workers to undertake further education and training, with their replacements receiving job-related and basic vocational training, and the possibility of securing permanent employment (Nordjyllands Arbejdsmarkedraad, 1998). It is also of note that this scheme arose directly from union demands around tackling staff shortages and retraining in this sector. This exemplifies the potency of Job Rotation as a bargaining tool for trade unions with which they can negotiate change in the workplace (Aalborg LO; Job Nord, 1998).

The general impact of Job Rotation can be assessed in relation to the number of schemes implemented, and the types of projects developed therein. Using national-level data, as Table 2 shows, the number of Job Rotation schemes implemented in both public and private sectors was around 29,000 in 1995, falling to 7,500 in 2001. One way of assessing the impact of Job Rotation is to compare its participants with overall labour market participants in the same period (see Table 1). This confirms that Job Rotation has had a significant impact on the overall labour market strategy. Furthermore, the numbers of unemployed who obtain permanent jobs is also impressive--figures of between 60 per cent and 80 per cent have been reported (Grunewald & Sorensen, 2001; Kankaanpaa-Lehtinen & Lahtinen, 2001). On this measure, and compared to British labour market policy over the past 25 years (on which, see Jones, 1999), Job Rotation is a significant success story.

Conflicts and tensions within the Danish model

One of the significant trends in the Danish model in recent years is the marked decrease in Job Rotation and other 'activation' measures as instruments of labour market policy. This is partly due to reductions in the longer-term unemployed (Danish Government, 2000; Ploug, 2002). More importantly, however, the scaling down of Job Rotation can be also explained by shifts within political strategy towards a more neoliberal workfare agenda. As suggested above, there have always been tensions within the welfare-through-work model because of conflicting interests between labour and capital over the reproduction and regulation of labour power. As the former Social Democratic Government moved to the right throughout the 1990s, the ground initially gained by the labour movement from the 1994 reforms became gradually eroded. The decline in the use of Job Rotation is a good example of this. Accordingly, the sabbatical leave programme has been phased out, the education leave scheme has been closed, and there is more emphasis now on the 'duties' within activation measures. Furthermore, in 1999 legislation was passed promoting the rationalisation of vocational training programmes (Goul Andersen, 2002; Grunewald & Sorensen, 2001). Some argue that the traditional welfare model has reached its limits, with more market-based policy initiatives set to dominate future political and policy agendas (see Goul Andersen, 2002; Kosonen, 2000). Workfare thus became a dominant, though not uncontested, strategy in Danish society, especially among sections of the labour movement. Compulsory and disciplinary measures, despite the comprehensiveness of the labour market schemes, however, ideologically contradict the political strategies of the labour movement, which seeks to retain activation and benefits in relation to a more rights-based system of regulation. The Liberal-Conservative Government, elected in 2001, launched a discussion document with the slogan 'Flere I Arbejde' (More in Work), with the intention of rationalising the organisation of labour market policy, adjusting access to unemployment insurance, and making education and training initiatives more work-specific (see Regeringen, 2002).

Although welfare strategies appear to be changing in Denmark, Job Rotation remains an important instrument of struggle, for shifting the agendas within the workplace and enhancing union representation in relation to training. At the EU Jobrotation 2000 Conference, for instance, debates were centred on the links between Job Rotation and issues of social solidarity--e.g. the strengthening of links between the employed and unemployed as a vehicle for building a relevant and comprehensive adult vocational education system; the possibilities for influencing the politics of social inclusion within labour-market policy; and a mechanism for improving the delivery of public services by upgrading the skills of social and health workers (see EU Jobrotation, 2000a and 2000b). This vision, however, is somewhat at odds with an emphasis, post-1997, on a more private sector workfare-based social policy, which is partly reflected by the changing balance of forces acting in and through the Danish state apparatus. The significance of Job Rotation, and the future roles it may play, relate to the way it is situated within these ongoing political struggles in Denmark.

In summary, Lind (2000) argues that the main thrust of labour regulation post-1994 has been to create a flexible system of allocating and 'up-skilling' labour reserves in the context of a more intensely competitive (global) economy. Work-sharing can either challenge this trend, or be compatible with the search for flexibility. It can be compatible in the sense of being used by employers as part of overall restructuring strategies. Its ability to challenge relates to specific models of negotiation and bargaining, which trade unions can use to ensure that employment rights and access to training are retained as integral features of Job Rotation. Herein lies the central contradiction and problematic of Job Rotation, and the reasons why the initiative embodies a variety of strategies and discourses. Another important factor influencing its future as a way of re-regulating the reserve army of labour relates to debates on the distribution of working-time. Work-sharing is bound up with the Danish labour movement's long-standing struggle to reduce working time, which has common links with trade union demands in other European countries (see Went, 2000).

Lastly, if we use Job Rotation as a barometer of struggles and demands from the more activist and rank-and-file sections of the labour movement (and policy community), there is evidence to suggest that the offensive against labour is impacting in diverse ways, and that it is geographically constituted. Thus, for example, despite the national government's resistance to pressures from interest groups to develop further, social solidarity-style dimensions to public policy, there are examples of these initiatives being developed in places such as Aalborg as a result of the processes and practices of ongoing trade union mobilisation. This suggests that neither the previous Social Democratic nor the current Liberal Conservative regime have been able to completely jettison the Job Rotation innovation.

Conclusions

We have argued, in this paper, that labour market and welfare reforms in Denmark have been able to promote a somewhat unique strategy based on needs-orientation, decentralisation, and the active involvement of the social partners, while retaining a relatively generous level of benefits and childcare provision. Many features of the reforms have originated from trade union pressure, reflecting the importance of social mobilisation through the labour movement to the promotion and implementation of demand-side policies, and attempts to tackle market failure (cf. Huber & Stephens, 2001).

In the UK, in contrast, the machinery of the welfare state tends to be viewed as a drain on national resources, and a brake on international competitiveness. This thinking, sustained throughout the 1990S, currently dominates Labour Party thinking on welfare-state reform (compare DSS, 1998; Giddens, 2002; Labour Party, 2001). But the evidence is not convincing. Denmark, a 'high tax-and-spend country' with powerful trade unions and few natural resources, is experiencing employment growth, falling unemployment and economic growth at around 2-3 per cent (Etherington, 1998). At the same time, it has the lowest rates of poverty in the whole of the European Community (OECD, 1998) and has the joint-highest labour market participation rate in Europe (Lodemel & Trickey, 2000). Moreover, the Danish model is the only European example that 'has been able to, on the one hand, reduce unemployment significantly, and on the other hand, to increase both the labour force participation rate and the employment rate' (Ploug, 2002: 3)

Given the importance attached to 'stakeholding' and partnerships in Labour's approach to policy formulation (especially DfEE, r997), the message from Denmark is that structures of governance and coalition-building have crucial implications for the implementation and effectiveness of welfare-through-work. Strengthening social partnerships at the regional and local level could provide the basis for strengthened accountability and democracy--themes that chime loudly with the principle of consensus-building, and with the negotiated economy as 'social solidarity' (Taylor-Gooby, 1996). Relevant policies and decisions can be formulated to take account of different labour market conditions, which is why the Danish reforms involve both the trade unions and local government as key partners. The enhanced role of local government in policy formulation and implementation has implications for developing effective strategies for those particularly disadvantaged in the labour market (i.e. young people and the long-term unemployed, who also tend to receive support from social and welfare services).

Moreover, this paper has sought to highlight the fact that trade unions play a prominent role in providing counselling and welfare advice to their unemployed members, through the Unemployment Insurance Funds. In many cases, both the Employment Service and local authorities, when overwhelmed by pressure on their counselling services, have sought trade union services to undertake counselling. Thus the implementation of a key aspect of the reforms, and the maintenance of commitments to counselling and career plans, has been achieved through trade union assistance. Trade unions, through securing relevant 'rates for the job', and in their general role of maintaining employee rights for the job, provide important safeguards for people entering work-based training programmes. Trade unions also ensure the validity of returning-to-work guarantees for those participating in the educational programmes. In turn, Job Rotation relies heavily on trade union and Workers' Educational Association involvement. This is because the whole package of training involves work-based negotiation, in which trade unions play a leading role, in terms of identifying those in employment wishing to undertake further education and training. As well as providing relevant vocational training programmes, the Workers' Educational Association, which is sponsored by and closely linked to trade unions, also plays a role in seeking out companies to participate in the Job Rotation initiative (see Etherington, 1997b).

This is not to argue for an uncritical reading of corporatism and its politics. Danish corporatism, on the one hand, opened up a space in which the labour movement could formulate social clauses. But on the other hand, there is a danger that trade union involvement can be institutionalised within the welfare state apparatus. As with debates in Britain during the 1970s and the 1980s, around the Manpower Services Commission and its mode of tripartite corporatism within vocational training (see Jones, 1999), this system of institutionalised representation partly explains the 'in-and-against' strategic dilemmas currently being experienced within the Danish model. The hollowing-out of the welfare state and the realignment of the various geographies of power, then, have to be seen as complex, and sometimes contradictory, processes.

In contrast, in the UK the partnership arrangements of the New Deal do not offer the same clearly-defined opportunities as in Denmark. The Employment Service plays a pivotal role in policy formulation and delivery, but local level arrangements are left to so-called 'partnerships', which will inherit the 'failure of local programmes as well as their success' (Peck, 1999: 366; also NCVO, 1999). In other words, there is no fixed framework that can guarantee that the interests of specific groups--particularly those excluded from the labour market--are represented on local forums. The New Deal does not require, in any way, the involvement of local government and trade unions through local partnerships arrangements (see Herd et al., 1998; LGA, 2001). Instead, it underlies a neoliberal approach, set by previous Conservative governments and continued in the framework of Labour's 'third way', which provides the basis for subordinating the interests of labour to capital and its social partners, within the formulation and implementation of employment agendas.

Certain aspects of the welfare-through-work model are being debated outside Denmark. The formation of an 'Eu Jobrotation International Association' is proving important in promoting Job Rotation as a tool for business development, lifelong learning and active labour market policies fostered on supply and demand-side concerns.

Based on five years of cooperation across Europe, and building on the EU-Adapt programme, this is a 'permanent partnership for promoting know-how transfer, consulting and the implementation of joint projects among its member organisations as well as the development and mainstreaming of Job Rotation, vocational training and active labour market policy' (Jobrotation International Association, 2000). The importance of this development cannot be over-emphasised: the 'Association' has been effective in encouraging the formation of a number of Job Rotation projects in 14 European countries. The emerging lesson from the EU-Adapt programme is that Job Rotation's success, outside Denmark, is contingent on strong partnerships and close networks between social partners and the private sector (compare Bason et al., 2000; Etherington et al., 1999; Parker, 200i; European Journal of Vocational Training, 2001). In addition to this, we would argue that the important question for future critical policy and political analysis is whether initiatives such as Job Rotation will evolve as an adaptation of workfare, and a continuation of the 'new paternalism' found in North America (see Mead, 1997), or whether they will be deployed as pilots for more inclusive and radical labour market instruments, perhaps indicative of social solidarity, that could provide a space within which to challenge contemporary neoliberal orthodoxy.

Table 1: Average number of participants in active labour
market measures 1994-99

                             1994        1995        1996

Subsidised employment       59,834      51,494      48,617
Leave schemes               50,845      82,116      62,990
Education                   23,397      17,382      23,163
Other active measures        2,088       3,250       3,213

                             1997        1998        1999

Subsidised employment       48,956      46,052      41,533
Leave schemes               46,709      42,944      35,890
Education                   23,816      27,258      36,590
Other active measures        4,002       4,665       5,421

Source: Larsen & Stamhus (2000: 7)

Table 2: Number of participants on Job Rotation schemes

                    1993      1994      1995      1996      1997
Unemployed/
Substitutes        1,000     6,000     8,000     6,000     3,500
Employed           4,000    12,000    21,000    30,500    16,000

Total              5,000    18,000    29,000    36,500    19,500

                    1998      1999      2000      2001
Unemployed/
Substitutes        5,200      4500     1,800     1,000
Employed          22,000    19,900     7,000     6,500

Total             27,000    24,500     8,800     7,500

Source: Sorensen (2002: 9)

References

Aalborg Kommune (1997) Arbejdsmarkeds Indsatsen 1997: Arsrapport [Labour market programmes 1997: Annual Report] (Aalborg Kommune) Aalborg.

Aalborg LO [Aalborg TUC] (2001) Partnerskab for rummelig arbejdsmarked [Partnership for a social inclusive labour market], (Aalborg LO) Aalborg.

Bason, C.C., P. Sidelman, N. Ejler & T. Bjornkilde (2000) 'Evaluation at European level: The Adapt community initiative', Paper for the Fourth Conference on Evaluation of the Structural Funds, Edinburgh, 17-19 September.

Bender, H., K. Knudsen, E. Petersen & L. Tvede-Jensen (1998) Arbejderbevaeglsen i Aalborg 1870-1970 [Labour Movements in Aalborg 1870-1970] (AOF) Aalborg.

Bewick, T., D. Etherington & M. Jones (1997) 'Reforming welfare-through-work: Scandinavian alternatives to neoliberal policies', Paper Presented to the University of Manchester, International Centre for Labour Studies Conference Between Welfare and Work: International Exchanges, September.

Carlson, V.L. & N.C. Theodore (1995) Are There Enough Jobs? Welfare Reform and Labor Market Reality (Centre for Urban Economic Development/Chicago Urban League) Chicago.

Coates, D. (2000) Models of Capitalism (Polity) Oxford.

Cochrane, A. (1993) 'Comparative approaches and social policy', in A. Cochrane & J. Clarke (eds.) Comparing Welfare States: Britain in International Context (Sage) London.

Compston, H. & P.K. Madsen (2001) 'Conceptual innovation and public policy: Unemployment and paid leave schemes in Denmark', Journal of European Social Policy, vol. II, pp. 117-132.

Danish Government (2000) National Action Plan for Employment 2000 (Ministry of Labour/Ministry of Economic Affairs) Copenhagen.

DfEE (1997) Design of the New Deal for 18-24 Year Olds (DfEE) London.

DSS (1998) Green Paper: New Ambitions for our Country (HMSO) London.

Edwards, P. & T. Elger (1999) The Global Economy, National States and the Regulation of Labour (Mansell) London.

Elger, T. & P. Edwards (1999) 'National states and the regulation of labour in the global economy: An introduction' in P. Edwards & T. Elger, (eds.) The Global Economy, National States and the Regulation of Labour (Mansell) London.

Esping-Andersen, G. (1990) Three Worlds of Welfare Capitalism (Polity) Oxford.

Esping-Andersen, G. (1992) 'The emerging realignment between labour movements and welfare states', in M. Regini (ed.) The Future of Labour Movements (Sage) London.

Esping-Andersen, G. (1999) Social Foundations of Post Industrial Economies (Oxford University Press) Oxford.

Etherington, D. (1997a) 'Local Government and labour market policies in Denmark: Is there an alternative to free market policies?' Local Government Studies, vol. 23, pp. 48-58.

Etherington, D. (1997b) 'Trade unions and local economic development--Lessons from Denmark', Local Economy, vol. 12, pp. 267-274.

Etherington, D. (1998) 'From welfare to work in Denmark: An alternative to free market policies?' Policy and Politics, vol. 26, pp. 147-161.

Etherington, D., B. Evans & D. Esp (1999) 'Feasibility and development study in job rotation project in Sheffield' Report to Sheffield Training and Enterprise Council (University of Huddersfield) Huddersfield.

Etherington, D. & M. Jones (2004) 'Beyond contradictions of the welfare state: Denmark, welfare-through-work, and the promises of Job Rotation' Environment and Planning C: Government and Policy vol.22, pp.129-148.

European Journal of Vocational Training (2001) 'Jobrotation' Special Issue, no. 24, pp. 3-54.

EU Jobrotation (1996) EU Jobrotation Action Now! (EU Jobrotation) Silkeborg.

EU Jobrotation (2000a) Jobrotation i Danmark Ansats til status og problemfelter EU Jobrotation [Job Rotation in Denmark: Statement of status and problem areas] (EU Jobrotation) Aalborg.

EU Jobrotation (2000b) Gode ideer til viderudvikling og udbredelse af Jobrotation [Ideas for further developing and expanding Job Rotation] Paper to EU Jobrotation National Conference, Copenhagen.

European Commission (2000) 'A memorandum on lifelong learning', Commission StaffWorking Paper (Commission of the European Communities) Brussels, 30 October.

Fay, R.G. (1996) 'Enhancing the effectiveness of active labour market policies: Evidence from programme evaluations in OECD countries', Labour Market and Social Policy Occasional Papers, no. 18 (OECD) Paris.

Finn, D. (2000) 'Welfare to work: The local dimension', Journal of European Social Policy, vol. 10, pp. 43-57.

Giddens, A. (2002) 'There is no alternative--the Third Way is the only way forward', The Independent (Review), 8 January, p4.

Ginsberg, N. (2001) 'Globalisation and the liberal welfare states', in R. Sykes, B. Palier & P. Prior (eds.) Globalization and European Welfare States Challenges and Change (Palgrave) Basingstoke.

Goul Andersen, J. (1997) 'The Scandinavian model in crisis? Achievements and problems of the Danish welfare state in an age of unemployment and low growth', Scandinavian Political Studies, vol. 20, pp. 1-31.

Goul Andersen, J. (2002) 'Work and citizenship: Unemployment and unemployment policies in Denmark, 1980-2000', in J. Goul Andersen & P.H. Jensen (eds.) Changing Labour Markets, Welfare Policies and Citizenship (Policy Press) Bristol.

Grover, C. (2003) 'New Labour, welfare reform and the reserve army of labour', Capital & Class, vol. 79, pp. 17-23.

Grunewald, U. & J.H. Sorensen (2001) 'Jobrotation experience, problems and prospects in Germany and Denmark', European Journal of Vocational Training, vol. 24, pp. 47-52.

Hamnett, C. (1996) 'Social polarisation, economic restructuring, and welfare state regimes', Urban Studies, vol. 33, pp. 1407-1430.

Haughton, G., M. Jones, J. Peck, A. Tickell & A. While (2000) 'Labour market policy as flexible welfare: Prototype Employment Zones and the new workfarism', Regional Studies, vol. 34, pp. 669-680.

Hay, C. (1999) The Political Economy of New Labour: Labouring Under False Pretences? (Manchester University Press) Manchester.

Herd, D., J. Peck & N. Theodore (1998) 'Union Representation in New Deal Partnerships', A Briefing Paper Prepared for the TUC (University of Manchester) School of Geography New Deal Monitoring Project.

Huber, E. & J.D. Stephens (2001) Development and Crisis of the Welfare State: Parties and Policies in Global Markets (University of Chicago Press) Chicago.

Jessop, B. (1990) State Theory: Putting Capitalist States in their Place (Polity) Cambridge.

Jessop, B. (1993) 'Towards a Schumpeterian workfare state? Preliminary remarks on post-Fordist political economy', Studies in Political Economy, vol. 40, pp. 7-39.

Jessop, B. (1999) 'The changing governance of welfare: Recent trends in its primary functions, scale, and modes of coordination', Social Policy and Administration, vol. 33, pp. 348-359.

Jessop, B. (2002) The Future of the Capitalist State (Polity) Cambridge.

Job Nord (1997) 'Job Rotation i Nord Jylland' [Job Rotation in North Jutland] Mimeograph (Kjeld Lang) Aalborg.

JobNord (1998) 'Jobrotations projekter for Paedegogisk medhhjaelper/dagplejer' [Jobrotation projects for nursery workers] Mimeograph (JobNord) Aalborg.

Jobrotation International Association (2000) 'An introduction to the Jobrotation International Association', Press Release (EU Jobrotation Secretariat) Aalborg, 1 November.

Jones, M. (1999) New Institutional Spaces: Training and Enterprise Councils and the Remaking of Economic Governance (Jessica Kingsley/Regional Studies Association) London.

Jones, M. & A. Gray (2001) 'Social capital, or local workfarism? Reflections on Employment Zones', Local Economy, vol. 16, pp. 178-186.

Jorgensen, H. (2002) Consensus, Cooperation and Conflict: The Policy Making Process in Denmark (Edward Elgar) Cheltenham.

Kankaanpaa-Lehtinen, K. & P. Lahtinen (2001) Job Rotation--European Model (University of Tampere Institute for Extension Studies) Tampere, Finland.

Knudsen, H. & J. Lind (2000) Collective Bargaining on Employment and Competitiveness in Denmark (Department of Development and Planning, Aalborg University) Aalborg.

Kosonen, P. (2001) 'Globalisation and the Nordic welfare states', in R. Sykes, B. Palier & P. Prior (eds.) Globalisation and European Welfare States Challenges and Change (Palgrave) London.

Labour Party (2001) Ambitions for Britain: Labour's Manifesto 2001 (Labour Party) London.

Larsen, F., C. Hansen, H. Jorgensen, N. Lassen, B. Bagge & H. Hocker (1996) Implementering af Regional Arbejdsmarkedspolitik [Implementation of the Regional Labour Market Policy] (Centre for Labour Market Research, Aalborg University) Aalborg.

Larsen, F. & J. Stamhus (2000) 'Active labour market policy in Denmark: Labour market reform, crucial design features and problems of implementation', Mimeograph, (Department of Economics, Politics and Public Administration, Aalborg University) Aalborg.

Lavalette, M. & G. Mooney (eds.) (2000) Class Struggle and Social Welfare (Routledge) London.

Lind, J. (2000) 'The social pact in Denmark under a neoliberal regime', in J. Lind & D. Mortimer (eds.) Worlds of Employment (Aalborg University Press) Aalborg.

LGA (2001) Opportunity to Prosper Beyond the New Deal: A New Approach to Local Employment Policy (Local Government Association) London.

Lodemel, I. (2000) 'Discussion: Workfare in the welfare state', in I. Lodemel & H. Trickey (eds.) An offer you can't refuse: Workfare in international perspective (Policy Press) Bristol.

Lodemel, I. & Trickey (eds.) (2000) An offer you can't refuse: Workfare in international perspective (Policy Press) Bristol.

MacLeod, G. & Jones, M. (1999) 'Reregulating a regional rustbelt: Institutional fixes, entrepreneurial discourse, and the "politics of representation"', Environment and Planning D: Society and Space, vol. 17, pp. 575-605.

Mead, L. (ed.) (1997) The New Paternalism: Supervisory Approaches to Poverty (Brookings Institution Press) Washington.

Moller, I.H. (1999) 'Trends in the Danish social mode of economic regulation', in J. Lind & I.H. Moller (eds.) Inclusion and Exclusion: Unemployment and Non Standard Employment in Europe (Ashgate) Aldershot.

Moller, I. H. & J. Lind (2000) 'The labour market in process of change? Some critical comments' in J. Lind & D. Mortimer (eds.) Worlds of Employment (Aalborg University) Aalbord.

Moran, J. (1997) 'The dynamics of class politics and national economies in globalisation: The marginalisation of the unacceptable', Capital & Class, vol. 66, pp. 53-83.

National Audit Office (2002) The New Deal for Young People: Report by the Comptroller and Auditor General, HC 639 (Stationery Office) London.

Nativel, C., P. Sunley & R. Martin (2002) 'Localising welfare-to-work? Territorial flexibility and the New Deal for Young People', Environment and Planning C: Government and Policy, vol. 20, pp. 911-932.

NCVO (1999) Partnerships for the New Deal: Playing an Active Role. (National Council for Voluntary Organisations) London.

Nordjyllands Arbejdsmarkedraad [Northjutlands Labour Market Council] (1998) Evaluering af nordjyske jobrotations-projekter afsluttet I virksomhednes vurdering og maaling af effekt [Evaluation of North Jutland's Job Rotation Projects--Employer's assessment of impacts] (AMR) Aalborg.

OECD (1998) Key Employment Policy Challenges Faced by OECD Countries (OECD) Paris.

OECD (1999) The Local Dimension of Welfare-to-Work: An International Survey ((OECD) Paris.

OECD (2001) Local Partnerships for Better Governance (OECD) Paris.

Parker, E. (2001) 'The development of job rotation in the UK: Implications for London', First report for Job Rotation in London Project (Department of Geography, Queen Mary, University of London).

Pascual, A.S. (2002) 'Are European activation policies converging?' Paper presented at the IREC Conference, Aalborg, Denmark, 28 August.

Peck, J. (1999) 'New labourers? Making a new deal for the "workless class"', Environment and Planning C: Government and Policy, vol. 17, pp. 345-372.

Peck, J. (2001) Workfare States (Guilford) New York.

Peck, J. & N. Theodore (2000) '"Work first": Workfare and regulation of contingent labour markets', Cambridge Journal of Economics, vol. 24, pp.119-138.

Pedersen (2000) 'Lone mothers, poverty and employment', in D. Gallie & S. Paugam (eds.) Welfare Regimes and the Experience of Unemployment in Europe (Oxford University Press).

Ploug, N. (2002) Impact Evaluation of the European Employment Strategy--Denmark Synthesis Report (The Danish National Institute of Social Research) Copenhagen.

Poulantzas, N. (1978) State, Power, Socialism (New Left Books) London.

Regergingen [The Government] (2002) Flere I Arbejde--Et Debar Oplaeg [More in Work--A Discussion Document] (Ministry of Labour) Copenhagen.

Rosdahl, A. & H. Weise (2000) 'When all must be active--Workfare in Denmark' in I. Lodemel & H.Trickey (eds.) An offer you can't refuse: Workfare in international perspective (Policy Press) Bristol.

Siim, B. (1998) 'Vocabularies of citizenship and gender: Denmark', Critical Social Policy, vol. 56, pp. 375-396.

Sorensen, J.H. (2002) 'Job Rotation schemes: A conjunctural influences instrument of active labour market policy--and its effects on the dependence of qualification strategies, pursued by enterprises' Mimeograph (CARMA, Danish Centre for Labour Market Research, Department of Economic, Politics and Administration, University of Aalborg).

Sunley, P., R. Martin & C. Nativel (2001) 'Mapping the New Deal: The local disparities in the performance of welfare-to-work', Transactions of the Institute of British Geographers, vol. 26, pp. 484-512.

Taylor-Gooby, P. (1996) 'Eurosclerosis in European welfare states: Regime theory and the dynamics of change', Policy and Politics, vol. 24, pp. 109-123.

Torfing, J. (1999) 'Workfare with welfare: Recent reforms of the welfare state', Journal of European Social Policy, vol. 9, pp. 5-28.

Turok, I. & Edge, N. (1999) The Jobs Gap in Britain's Cities: Employment Loss and Labour Market Consequences (Policy Press) Bristol.

Turok, I. & D. Webster (1998) 'The New Deal: Jeopardised by the geography of unemployment?' Local Economy, vol. 13, pp. 309-328.

Went, R. (2000) Globalization, Neoliberal Challenge, Radical Responses (IIRE/Pluto) London.