Introduction: The many worlds of activation Open Access

The authors agreed that, although it was a clear sign that interesting rhetorical change was going on everywhere, the normative notion present in the international political discourse was inadequate to analyse the current reforms systems of social protection are undergoing: a more analytical notion was all the more needed as different legacies of programmes were apparent in various countries. With its ‘Swedish model’ and the ‘Rehn-Meidner’ inspiration, Sweden certainly played an important role in the genesis of contemporary strategies, for the very notion of ‘active policy’ was invented there. In France, because of their relatively early introduction from the mid-1970s, ‘insertion’ policies were also interesting to take on board. Additionally, the usage of the notion of ‘workfare’ (Lødemel and Trickey 2000) was sidelined, as too normative, too biased upon American programmes (Barbier 2002) and too limited in its focus on assistance programmes.

Taking account of historical legacies also led to stressing that ‘activation’ certainly was not an entirely new phenomenon, and that current evolutions had indeed to be analysed in the context of national histories: in the Scandinavian countries, as well as in Germany or Austria for instance, an important part of social policy has long been devoted to vocational training, an active policy tool par excellence. Moreover, as Per Jensen and Rune Halvorsen clearly demonstrate, the policy goal of full employment was always historically deeply integrated into the social protection rationale in Denmark and Norway. It still is today – albeit in differing circumstances – in contrast to the continental countries, which abandoned their Keynesian policies since the late 1970s. Per Jensen and Rune Halvorsen put the present activation rationale in the long-term perspective of an ‘active society’. This helps to explain that ‘activation’ in the second half of the 1990s in Denmark and Norway appeared as a rather marginal adjustment of an already very ‘activated’ system of social protection with high employment rates: indeed, placed in this perspective, the current activation reform is relativised. In this respect and, paradoxically, despite huge differences in their respective substance and the nature of benefits, services and rights, the Danish ‘aktivering’ and the US ‘workfare’ or ‘welfare-to-work’ of the late 1990s display structural similarities as to their position within their respective systems.

It was possible to establish an encompassing analytical notion of ‘activation’ which, it was agreed, extended beyond the ‘activation of individuals’, in the sense of giving incentives or even compelling them to seek and take jobs. The wider rationale of the systems of social protection was at stake, in the context of a general drive to flexibilise the labour markets. Distinct from this flexibilisation process (as is particularly well seen in Italy, in contrast with the already flexible Danish labour market), current ‘activation’ concerned an important aspect of what certain authors have explored under the concept of the ‘restructuring of welfare states’ (Pierson 2001). ‘Activation’ could then be seen as a general trend indeed compatible with a variable mix of reforms⁴ in particular countries. In this sense, ‘activation’ may take on various procedural as well as substantive dimensions and outcomes; in this sense also ‘activation’ is new because contrary to the past, it entails a new and systematic transformation of the systems to make them more ‘employment friendly’.

To sum up from definitions we proposed (Barbier 2002, 2004a,b,c), the notion of ‘activation’ is here envisaged as the introduction of an increased and explicit linkage between on the one hand, social protection, and on the other hand, labour market participation and labour market programmes. For the last decade, this dynamic has implied the redesigning of previous income support, of assistance, but also the transformation of tax and social policies. This redesigning, or sometimes even overhaul of the entire social protection systems has displayed a common feature, i.e., that of enhancing the various social functions of ‘paid work’ and labour force participation, an individual behaviour which has even been seen increasingly as compulsory in certain national cases.

Hence, the domains of social protection potentially ‘activated’ are not only the programmes for the assisted or the unemployed. They comprise (i) benefit programmes (unemployment insurance and various ‘assistance’ schemes for working age groups (including disability and other family related benefits)); (ii) pension systems and, most particularly, early retirement programmes; (iii) employment (or active labour market – ALM) programmes; but also (iv) policies involving traditional social policy and tax policy, which aim at reforming the ‘tax and benefits systems’, as the European Commission discourse would have it; such reforms may be aimed at invigorating labour demand or at incentivising job taking.

From the individuals’ point of view, these programmes are deemed to provide incentives (sometimes closely associated with compulsion and sanctions), but also, in some cases, a wide array of offers of services (for instance counselling, job search, training); they also – at least in theory, indirectly or directly – extend the individuals’ choices by increasing job supply on the conventional market (especially in the case of wage subsidies). From a system perspective, social protection is ‘activated’ in the sense that the delivery of services and benefits mainly targets working age people being in some sort of ‘work’ activities (extending in some cases to debates about the range of the ‘working age’). It is also activated in the sense that funding mechanisms and the allocation of resources are designed so as to foster increased job creation.

Once a working definition was agreed, it was particularly interesting to study the countries selected from the perspective of convergence versus national diversity. In comparative social policy, a long existing tradition has stressed the similarities and commonalities of systems’ evolutions (for instance, Wilensky 1975). This tradition is still active in some authors who presume a unique trend to activation (Gilbert 2002) and take for granted the assumption of an ‘Anglo-Saxon influence’.⁵

On the contrary, the present collection of papers empirically and theoretically invalidates the essential tenets and conclusions of the convergence approach, mainly because procedural similarity does not entail substantive convergence.⁶ It is, however, difficult to organise the national diversity displayed in the articles into a limited number of types. Indeed we have earlier identified two distinct ideal-types of activation (Barbier 2002, 2004b).

The liberal type chiefly enhances the individuals’ relationships to the labour market, which, aggregated, are assumed to yield social equity and efficiency. ALMPs as well as social policies thus take on a limited role, restricted to inciting individuals to seek work, providing quick information and simple matching services, as well as investing in short-term vocational training. Individuals in and out of the labour market are also the target of the wide-ranging implementation of ‘tax credits’ and all sorts of ‘in-work’ benefits. Accordingly, activation manifests itself in the trend towards adopting measures inciting people to be as active as possible across their life-course (accepting any job on the market as it is) and implementing pension reforms dispensing with any fixed age for retirement. Very generally, for an individual, having a conventional job on the market tends to be the normal way of accessing protection (private and social) from risks, and work systematically tends to replace assistance. Activation entails both the re-commodification of the system (already highly commodified in comparative terms) and efforts to reduce social expenditure (for instance in the case of the disabled).

On the other hand, the universalistic type not only caters for the provision of complex and extended services to all citizens, but simultaneously guarantees relatively high standards of living for the assisted, and, for the lower paid sections of the labour force, benefit levels amounting to a significant proportion of minimum wages. Hence the role of the market is not unilaterally prominent (although some re-commodification may be easily implemented in systems which are at the same time highly de-commodified and highly commodified, see Kvist 2002). Not systematically submitting citizens to work requirements, social policy retains its traditional contribution to well-being. Activation applies to all citizens in a relatively egalitarian manner and the ‘negotiating’ between the individual's and society's demands appears as much more balanced. A fully active society seems to be able to yield employment opportunities tailored to a variety of needs and capacities. Activation applies to an already highly active population employed in a context of relatively good quality jobs. It entails recalibration of the previous income compensation mechanisms, in the overall context of cost containment, but the use of tax credits and subsidies play a limited role, if any.

As was expected from the start, none of the national cases presented here exactly fit either of these stylised types, but the UK appears close to the ‘liberal’ ideal-type, and Denmark and Norway, whatever their important differences, emerge as close to the ‘universalistic’ type. This is rather well exemplified in the paper by Sharon Wright, Anja Kopac and Gary Slater, who nevertheless stress the increased quality of social protection introduced by the Labour government along with mainstream activation. On the other hand Per Jensen and Rune Halvorsen extensively document the measures taken in their countries over the period to decrease the workforce, especially by allowing older employees not to be active in one way or another.

We have discussed the possibility of a third type (Barbier 2004b). True, identifying three types of activation would nicely fit in the now traditional tripartition of welfare regimes (Esping-Andersen et al. 2002). As a matter of fact, France emerged as mixing elements from both types, which stems at least partly from its hybrid legacy of Beveridgeism and Bismarckism (Barbier and Théret 2003). Italy, on the other hand seems to oscillate between both types, according to periods. Some of the measures proposed recently in Germany might seem to indicate a new ‘state–market mix’ (with the state giving money to individuals and firms to foster more choice, against the background of considerable regulation either of specific conditions for receiving or for spending these subsidies); accordingly both in Germany and in Austria the Public Employment Service is increasingly evaluated not in terms of compliance with bureaucratic regulations (as has been the case for decades), but rather in terms of efficiency. But again it would be premature to see anything resembling a systematically developed new model from this, not least because it remains to be seen how much of the proposed reforms will actually be put into practice. All in all, the papers collected here (Ludwig-Mayerhofer and Wroblewski; Barbier and Fargion) show that it is not possible to contend that a third, coherent, ‘continental’ ideal-type has clearly emerged in activation policies.

A first lesson drawn from the papers in this special issue concerns the ambiguous outcomes of activation: on the one hand, many elements of current policies are not so new, on the other hand, it remains to be seen how many of their promises will be delivered, not the least in terms of work actually ‘paying’ and of the delivery of ‘quality’ full employment, two amongst the objectives adopted by the European Union at Lisbon in 2000.

This means that appraisals (whether favourable or critical) of ‘activation’ as a recent paradigm shift in labour market and social policies must be viewed with considerable caution. As has been pointed out for some time, the emphasis on activation in the current discourse may serve ideological reasons such as legitimating the curtailment of benefits, on account of their being allegedly generous and therefore ‘passivating’ (Sinfield 2001). On the other hand, it certainly cannot be claimed that activation is nothing but ideology, as this ideology can be – and has been – very powerful in enhancing far-reaching shifts. It is obvious that in some countries – such as the UK or Denmark and Sweden – the new policies which have been introduced under the label of activation (even though with quite divergent meaning, as pointed out above) have led to a significant re-designing of systems, and that in many other countries fresh strategies are currently being discussed and implemented. However, from evaluation studies existing so far, it is not altogether clear whether and to what extent the various activation strategies have consistently and effectively achieved their formal promises: to take but two examples, controversies presently flourish in France among economists about the real impact of the very significant decreasing of employers’ social contributions in terms of the number of jobs created; secondly, to our knowledge, very little information has been published so far in the UK about the impact of the systematic introduction of tax credits upon activity and employment rates. As Per Jensen and Rune Halvorsen show, the evidence is not so conclusive in Denmark and Norway either.

Yet, as activation ideology and practice seem to invade more or less all countries, not least under the influence of the European Employment Strategy, their multi-faceted and variegated nature remains to be fully discovered. The contributions to this special issue do a very good job to enhance our knowledge in this respect, pointing out the many dimensions and social policy fields in which activation may or may not occur, with the clear message that what is termed ‘activation’ in one country may exhibit only remote similarities to what goes under this heading in another. Not least is this true for those countries which are close to neither of both ideal types outlined above – that is, the majority of those reviewed here. While the UK on the one hand, the various New Deals notwithstanding, and Denmark (and to some degree Norway) on the other, seem to correspond more or less to the two ideal-types outlined above, all other countries present various mixtures of elements of either type and in addition some specific facets. Slovenia can be mentioned as particularly interesting as a country which had developed a Bismarckian social protection system during its socialist era and now has to combine this system withM activation elements in order to cope with the problems of the post-socialist era.

In part, the wide variety of activation policies is also a consequence of considerable divergence or even contradiction between rhetoric and practice that can be observed in some cases. In France, a basic motive of many ‘activation’ policies has been to prevent the cutting of the ties between individuals and society and to foster participation in public life – thus being perhaps more oriented towards ‘citizenship’ (in the classical sense of universality) than the Scandinavian countries, where according to Rune Halvorsen and Per Jensen ‘citizenship’ is equated to ‘being in paid work’ much more than is commonly seen in the literature. But this French notion of citizenship does not necessarily translate into programmes sufficient to support all those that are in need. Similarly, the most ‘Bismarckian’ countries – including Slovenia – have inherited considerable commitment towards vocational training as a decisive pillar of activation, but their social systems are much too volatile – not only financially, but partly also politically – to provide high quality training to a sufficient number of people.

Still, it should be kept in mind that in spite of the close and deep investigation of the complex realities of activation offered here, we are still very far from being able to assess the actual working of (different kinds of) activation at street-level. More research is needed in the future, not least field research moving beyond the usual assessment of policies from official reports and statistics. And, above all, we are still lacking thorough and comprehensive knowledge as to whether or to what degree activation policies actually help increase labour market participation, how they affect the quality of jobs, whether they are indeed a solution (not to speak of being the best one) for the current problems of social protection systems, and in which ways they contribute to transformations of social inequalities and to social cohesion or inclusion. As far as implementation is concerned, studies both in Denmark (Larsen et al., 2001) and in France have already shown that there can be considerable variability according to regions and their economic circumstances, so that a ‘national model’ of activation may be compatible with differing regional or local models.

Finally, it would be careless to neglect the many counter-tendencies to activation that are prevailing or simply the many social policy fields that are not or only very slightly affected by activation policies in most countries. In this respect it could be argued that in Europe, only the UK has truly and consistently embarked on a comprehensive activation strategy including family benefits. Yet, contrary to the USA, as Sharon Wright, Anja Kopac and Gary Slater show, new programmes designed to alleviate child poverty in the UK are implemented independently of the parents’ working status. On the other hand, continental states still have to fight severe problems resulting from high unemployment levels and the concomitant burdens on public budgets; family policies (such as rather generous support during parental or educational leave and tax and social contribution regulations) in some of these countries (Germany, Italy, Austria, and to a significantly lesser extent, France) do not much support increasing female participation in paid work. In Italy, the state seems to have been quite unable to go beyond first attempts at flexibilising the labour market, a policy that has only been weakly related to activation in the sense outlined above. And even where activation policies are consistently pursued and sizeable proportions of the social budget are allocated to them, resources may still be insufficient in view of the tremendous labour market problems. Additionally, in all the countries surveyed, including the UK and the Scandinavian countries, programmes have persisted over time which cater for inactive people: consequently, for all the activation drive, an important share of the working age population has still remained inactive and eligible to some sort of disability or early retirement benefits.

Moreover, due to the persistence of widely differing patterns of labour market participation (of women, of the elderly, but also of the young in education) it is doubtful that a ‘one-size-fits-all’ type of ‘active society’ will be easily achieved in the future along models such as the one favoured by Esping-Andersen and his colleagues (2002).⁷

To what degree and in which ways have activation policies affected citizenship? For some, activation is seen as a decisive means to increase or rather to restore citizenship, as people not attached to the labour market are viewed as ‘excluded’ from social life. But the ‘workfare’ or the liberal type of activation has been accused of having opposite effects, not least because of the high degree of ‘re-commodification’ it has involved. And those who equate activation with ‘workfare’ maintain that clear limits to ‘inclusion’ or ‘enhancement of citizenship’ are a necessary consequence of activation (Handler 2003).

Only the UK seems to have undergone clear substantive changes along with its recent activation reforms, which were preceded by earlier Conservative reforms – among which the introduction of the Job Seeker's Allowance (JSE) in 1996 increased demands in return for benefits – and which introduced services of better quality (recently with the New Deals), whilst sanctions and compulsion remain privileged policy tools. At the same time, through the gradual extension of tax credits, Labour governments have generalised the rationale of ‘in-work’ benefits, which was only emerging in the previous period. On the other hand, as Sharon Wright, Anja Kopac and Gary Slater show, one important reform of British welfare policy has been the significant increase of income support for families and the consistent strategy to address child poverty, a reform which is independent of any activation rationale. In the UK case, as was again noted by the 2003 European Employment Strategy recommendations, the overarching activation welfare reform has had only limited success with regard to persisting inactivity among certain groups, among whom are the disability benefits recipients. In this particular area, the evolution of social rights has been ambiguous, because it is not clear whether the mainstream introduction of ‘work-tests’ has, in practice, entailed actual increased demands on beneficiaries.

From the collected papers of this special issue, when compared to British developments, the conclusion emerges that the demands made on citizens, in behavioural terms, have undoubtedly increased across the countries surveyed. What is now more and more considered as the norm, i.e., active engagement in labour market-related activities, can be observed in all the countries of our sample. At the same time, the conditionality of benefits has also increased in the majority of cases. Yet to interpret this general procedural dynamics as testimony for a common substantive normative turn across Europe remains highly far-fetched.

The example of the quasi-universal dissemination of a new policy technique, namely the ‘path to integration’ or ‘activation plan’ provides an interesting case in point indeed. Although the fashion has spread, and especially after the OECD in its 1994 Jobs Study hailed the Danish ‘handlingsplan’, the substantive meaning and the actual content of these plans certainly vary substantially between Norway, Denmark, Germany, Austria, France, Slovenia and the UK. They hardly exist except in some cases for the young in Italy. Moreover, in some countries little is known about how widespread they actually are (in Germany, they seem to be used much less frequently than provided by law). Nevertheless, the international literature is often content with functionally documenting the mere existence of such ‘activation plans’ without any in-depth analysis of their logic and detailed characteristics. Further research is indeed necessary to comprehensively compare the French PARE (plan d'aide au retour à l'emploi), the ‘contrat d'insertion’, the British ‘Gateway’, the Job seeker's Agreement, the German Eingliederungsvereinbarung, the Austrian individueller Betreuungsplan and the Slovenian ‘plan’.

For all the transformation and mimicry in political discourses, it must be stressed that in certain countries, the very structural characteristics of social protection de facto prevent any actual implementation of such normative orientations. It is best exemplified in the case of Italy. Developments to the contrary are certainly not precluded in the future, although they seem very unlikely in the short term in this country. In France, where the new right-wing government has embarked on a series of fresh reforms of assistance benefits, the tendency is on reducing their levels and introducing stricter demands: yet, even with the improbable introduction of another minimum income benefit (revenu minimum d'activité, RMA) for Revenu minimum d'insertion (RMI)'s beneficiaries after two years of participation, no mention has been made of dispensing with the original legal unconditional nature of RMI, nor to oblige the greater proportion of them to register as job-seekers. New stricter restrictions are also currently being introduced both in Austria, and most spectacularly in Germany, as Wolfgang Ludwig-Mayerhofer and Angela Wroblewski document in their paper. However, and despite what they term ‘bashing the unemployed’, the very notion of Zumutbarkeit still seems to act as a protective mechanism to further increases of severity (even though to a varying degree depending on the duration of unemployment). In Germany and France, moreover, social protection benefits are significantly more generous than in the UK, but at the same time, scarce job creation does not particularly encourage active administrative endeavours to find jobs for the ‘hard-to-place’.

At any rate, the dynamics of reducing benefits or tightening eligibility criteria show considerable divergence. Whereas Germany and later Austria were to reduce benefits in the light of persistent unemployment problems, from the comparison of Denmark and Norway it appears that the high unemployment level in the former country was seen as a reason not to curtail benefits. Thus, in contrast to what people are made to believe especially in Germany and Austria, the reduction of benefit levels is not a ‘natural’ and necessary step towards more employment (by allegedly diminishing ‘disincentives’ to seek work).

Except in the Scandinavian countries of our sample and Italy, the number of sanctions seems to be on the increase, thereby indicating greater severity. Here further research is certainly needed to better assess the current dynamics in more comparative terms. However, compared to the systematic implementation and planning of sanctions which was still the rule in the British public employment service recently, similar ‘workfarist’ developments have yet to occur elsewhere. When compared with Denmark and Norway, the dissimilarity of treatment of individuals according to gender, young or old age is also striking in terms of provision of benefits and services. Here the common incapacity of continental systems to universalism is confirmed, in contrast to their Nordic neighbours.

Paradoxically, another common characteristic in most countries, i.e. the high degree of discretion also in parts accounts for their diversity. This can work in a number of ways; it may be implemented in law (the degree to which regulations are mandatory or not) but may also depend on informal rules operating at street-level delivery. Whereas discretion usually is quite restricted or absent concerning benefit levels or duration, it may reach considerable levels if it comes to other ‘offers’ such as participation in training, job-creation schemes and the like. Also, the degree to which sanctions are applied (especially reducing or halting benefits) seems to be subject to tremendous discretion and consequently variation. We may note that, first, in the light of this it is often difficult to assess the actual degree of ‘getting tough’ at the unemployed (does a low number of sanctions indicate that the unemployed comply strictly with the rules set out for them or does it mean that PES (Public Employment Service) staff are very lenient towards a large number of non-compliant individuals?). Second, however, a substantial degree of discretion is endangering citizenship, as it makes the system less predictable and can lead to arbitrary decisions.

To sum up, the possible and exact causal relationship between the various substantive components of social citizenship and national values remains to be explored, but the heterogeneity of these relationships clearly emerges from the papers published here. The comparison between Norway and Denmark on one side and the UK on the other amply shows that a similar procedural (‘technical’) implementation of sanctions can be made on the basis of values which are at odds when it comes to justifying the generosity of benefits and their eligibility periods. This means that a ‘functional equivalent’ of ‘rights and responsibilities’ retains little if any meaning indeed in the present comparative research. As we have shown (Barbier 2004a,b), there is little doubt that a common language is now increasingly and pervasively used across Europe for describing social policies; yet this homogeneity of words is certainly neither based on an homogeneity of values, nor of substance of programmes, rights and entitlements. Friedberg and Ploug (2000: 348) have shown that citizens across Europe are very similar in terms of their attitudes to unemployment (and to unemployment compensation) although welfare systems differ widely and are based on very diverging values. This plurality remains a challenge for further research.

Jean-Claude Barbier is directeur de recherche (CHRS, Sociology) at the Centre d'etudes de I'emploi (CEE, Noisy le Grand, France). He works on the international comparison of social policies, social protection systems and, particularly employment policies. He also currently studies the European Employment startegy, from the point of view of dissemination of ideas. He also does research on the epistermological and methodlogical stakes of the international comparative posture.

Wolfgang Ludwig-Mayerhofer is Professor of Sociolofy – Empirical Research Method at the University of Siegen, Germany. His main areas of research are inequality, poverty and unemployment, sociology of the family and Sociology of (criminal) law. Among his recent books are Die Armut der Gesellschaft (ed., with Eva Barlösius, 2001), Soziale Ungleichheit, Kriminaliskerun (ed., 2000) and Soziologie des Sozialstaats (ed., with Jutta Allmendinger, 2000).