DGIII/DCS (2005) 03

Reconciling labour flexibility with social cohesion values

2^nd meeting of the working group for the 2005 FORUM

on Social Cohesion

January 21st and 22nd 2005, Council of Europe

European Youth Centre, Meeting Room 2^nd floor

Contribution of Dr. Jean-Claude Barbier

Jean-Claude Barbier

Directeur de recherche CNRS

Centre d’études de l’emploi

[email protected]

Citoyenneté, flexibilité et formes d’activation de la protection sociale en Europe

[Citizenship, flexibility and the diversity of approaches to activating social protection in Europe]

(titre provisoire/ provisional title)

Contribution pour le Conseil de l’Europe

Project:

Reconciling Labour Flexibility with Security and Social Cohesion

First draft: do not quote

December 2004

1.         Introduction/Summary

No assessment of the consequences of the trend towards greater flexibility in the world, at all events in developed countries, can ignore the key role played by monetary considerations.  The latter's influence extends beyond the labour market to the reform of social protection systems. One sign of this is the well-attested trend towards the activation of these systems.

Setting aside ideological considerations and the notion of a universal policy recipe advocated by certain international bodies, examination of the facts shows that these systems are being transformed by an "activation" process under the common stimulus of a search for greater flexibility, or flexibilisation.  However, this activation process is far from being a fatal condition leading unavoidably to the disintegration of institutional arrangements going back to the late nineteenth century and takes a variety of forms, including two ideal types that are easily identified and a third that is still much less clearly discernable.  Contrary to insistent claims of convergence, which turns out to be highly superficial, there has been no diminution in diversity in the last twenty years.

Criteria and indicators (such as those developed by the Council of Europe to measure social cohesion) are available to aid our understanding of the effects of activating social protection on civic life.  These criteria are similar to those applicable to any normative assessment of the problem of so-called transitional labour markets, as we understand it.

2.         Economic imperatives of flexibility and activation

As we have shown elsewhere (Barbier and Nadel, 2000; 2002)[1], the general trend to greater flexibility is a consequence of the economic crisis of the 1980s.  This economic context traces the boundaries of what reforms of social protection, and in particular its activation, are possible.

            Economies governed by monetary rules

The key variable in the current transition process is wages and salaries, which are now strongly influenced by international competition.  Broadly speaking, there is an increasingly rigid relationship between rising pay and productivity gains and prices.  In the new economic situation, controlling labour costs has become a strategic tool and justifies the requirement for flexibility.  The link between pay policies and the monetary system is crucial.

In broad terms again, under an international monetary system with fixed but adjustable exchange rates, countries can maintain their internal employment structure and arrangements by devaluing their currency (competitive devaluation), to the short-term detriment of their position in the international economy.  This was the situation in the period of the Bretton Woods Agreements.  The final abandonment of this system in 1973 led to a flexible exchange rate regime, under which it must be possible for wages to fall.  So whereas in a fixed exchange rate system national wage levels determine countries' international economic position, the situation is now reversed: wage levels are governed by the flexible monetary system.  Flexibility has become the key factor.

The crisis of the European monetary system and the growing interdependence of their economies persuaded certain European countries to adopt convergence criteria and a single currency.  Within the Euro zone exchange rates are now fixed and cannot be adjusted.  The setting up of the zone in the late 1990s was accompanied by efforts by all the candidate countries to limit their public deficits and restore their competitiveness.  A strategic choice was therefore made to implement wage restraint and employment and labour flexibility.  At the same time, massive under-employment in Europe was matched by more varied and unequal working conditions and diversification of employment status.  In the early 1970s, working careers tended to follow an upward trend, often with the same employer and/or in a limited number of posts.  This is no longer the case for a significant section of the population, starting with young people entering the labour market and employees suffering interruptions to their working careers.

Over the last fifteen years, in an effort to make wages more flexible and reduce the total wage bill, including taxes and social contributions, European countries have, to varying degrees, sought to reduce indirect charges, particularly on low wages.  This reduction in wage costs has been accompanied by a certain decline in social protection.  However, the extent of differences between countries and between social protection sectors makes it impossible to generalise.  In most countries – Scandinavia is an exception – the disadvantages of more flexible employment and working conditions have been concentrated on certain categories of the population.  These include less secure employment, working hours that are more fragmented (or insufficient for an adequate income), more on-call work and uncertain social coverage.  In some countries, the losers have been young and unqualified people – elsewhere, the elderly and migrants.  In most of Europe women are the first to suffer the effects of flexibility.  At the same time, other categories, such as shareholders, managers and the liberal professions reap the benefits of higher incomes, more interesting work and working conditions and greater freedom and autonomy.  The result is that employment and labour flexibility increases inequality without posing any threat to significant privileges.  In some countries the social protection system limits these growing divisions but in others individuals affected by employment and labour flexibility are forced back onto their family or community networks.

            Employment flexibility and labour flexibility: a crucial distinction

However, we have also shown that in certain cases, the effects of greater flexibility can be controlled.  To achieve the right conditions for a collective form of control which limits the negative social consequences while allowing new and potentially creative forms of work opportunities to arise, a clear distinction must be drawn between two concepts that are often confused: employment flexibility and labour flexibility.  The negative consequences are largely connected to the employment issue and concern wage rates and employment rights, whereas the positive aspects relate to work as a human activity.

French sociologists[2] were slow to draw the distinction between work and employment (Maruani and Raynaud, 1993).  For them work is the individual or collective productive activity within organisations.   Employment, on the other hand, represents a position occupied within a productive organisation, with associated duties and responsibilities, relationships of varying degrees of stability with other operators and individual and collective technical obligations.  In France probably more than elsewhere, sociologists associate employment with a status, involving rights and obligations, which in modern society now largely takes the form of paid employment.  Social protection lies at the core of these associated rights.  Moreover having a job signifies not simply temporary or permanent responsibilities or duties in a productive organisation but also a social identity.  In the 1980s, sociologists considered employment and labour flexibility from the standpoint of atypical forms of employment and employment status linked to particular markets[3].

In practice, applying flexibility to work and to employment has very differing consequences.  Making a job or employment more flexible means making its features more variable, in terms of working hours, where it is carried out and under what conditions and its legal aspects.  More flexible work in contrast means that the specific human activity (or "factor of production") becomes more malleable and adaptable to changing productive circumstances.

Table 1 – Flexibility in management, economics and sociology

	Flexibility of work	Flexibility of employment
Notions common to management, economics and sociology	Flexible organisation of production, and adaptable learning capacity and skills, hierarchical structures and working conditions and hours
Economic approach	Variable wages and labour costs	Variable level of employment in the firm or the economy and labour mobility
Sociological approach	Adaptability of the individual and collective productive activity	Adaptability of qualifications, contracts, employment status and associated rights

©Barbier and Nadel, 2000.

A common body of employment and labour law is essential to ensure that diversification of forms of work and employment does not lead automatically to growing inequalities (Barbier and Nadel, 2000; 2002).  This should go beyond the establishment of minimum rights by offering a range of substantial rights applicable to all workers, whether or not in paid employment.  By taking account of local arrangements, it would recognise flexibility of work and a certain flexibility of employment.  It should also include a definite European element since flexibility and globalisation have changed the previous national framework for enacting law, even though it has not disappeared.  Discussion on this subject is under way and exploratory proposals exist, even though they are slow to be implemented.  The proposal to institutionalise transitional labour markets (Schmid and Gazier, 2002) deserves further attention and we will show how, from the standard setting standpoint, it shows numerous similarities to the social cohesion criteria developed by the Council of Europe.

            Activating social protection in the context of greater flexibility

The scope for reforming social protection today is determined to some extent by changes in the global economy since the mid-1980s.  There is a close link between such reforms and more flexible labour markets, though this is not automatic and the results are not always unfavourable, as Pierson (2000) has very clearly shown, with reference to restructuring.  We have adopted an analogous approach.  We do not accept the thesis of a transition from welfare to workfare state[4].  Restructuring takes on a number of dimensions[5] according to welfare regimes and countries so it is hardly surprising that the activation process, as a specific element of this wide-ranging restructuring process, takes different forms and has widely varying consequences for rights and the components of citizenship, and for the quality of and access to social protection.  This is shown empirically by many works, of which only a few may be cited (Barbier and Gautié, 1998; Torfing, 1999; Morel, 2000; Wood, 2001; Goul Andersen et al., 2002; Jørgensen, 2002; Schmid and Gazier, 2002; Van Berkel and Möller, 2002; Serrano Pascual, 2004; Barbier and Ludwig-Mayerhofer, 2004).  To remove any ambiguity from the very definition of activation, the current trend has to be placed in a historical perspective.

3.         Activating social protection: a wide-ranging process

Before identifying the common characteristics of the trend towards activation, the relevant policies must first be considered briefly in a sufficiently long historical context.  Such a backward glance is essential for defining an authentic concept of activation, as distinct from the normative and political use of the terms, which has become widespread in recent years.  This political use is often particularly limited in that it focuses on certain specific policies and programmes that reflect political fashion while ignoring their potential diversity, something this section considers essential.

            A brief overview of relevant policies and programmes

The background to policies and programmes that are either expressly concerned with activation or can reasonably be placed in this category have been reviewed elsewhere (Barbier, 2004a).  Here we will focus solely on the main stages.  Recently, the political notion of activation has been associated with policies requiring persons in receipt of unemployment or social assistance benefits to behave in a particular way, often accompanied by increased penalties.  The debate has been strongly influenced by ideas emanating from international organisations such as OECD and the European Union.  Most of the material has come from English-speaking economists, with a particular emphasis on programmes operating in the United States and Britain.  Hence the success, in the literature, of the term workfare, which originated in the United States.  Even certain Scandinavian academics have felt forced to use it in international presentations of the reforms in their countries, while emphasising that the latter differ fundamentally from the American approach (Torfing, 1999). A superficial reading of international organisations' publications might suggest that activation has its roots in English-speaking countries, and that it is solely concerned with measures to influence the labour supply.  This is doubly incorrect: such programmes have diverse origins and reforms under this heading go well beyond programmes confined to "activating" the unemployed.  The older term "active employment and labour market policies" is also very unclearly defined.

The first international assessments used the term workfare, taken from American usage of the 1970s[6], without offering any precise definition.  One example is Lødemel and Trickey (2000), who lumped together a series of quite heterogeneous programmes, based on the minimalist criterion that they were “programmes and schemes that require people to work in return for social assistance” (2000, p.6). Torfing (1999, p.17) uses the term workfare but shows how substantially the Danish reform process differed from American experience[7].  Later work has stressed the variety of forms taken by activation policies (Barbier, 1996; Kosonen, 1999; Geldof, 1999; Abrahamson, 2001; Morel, 2000).  Britain has increasingly been presented as a laggard in activation while the American workfare system introduced under President Clinton's 1996 welfare reform has turned out to be much more marginal than another important form of activation, the Earned Income Tax Credit (EITC) scheme, initially introduced as far back as 1975.

In essence, activation in a broad sense is not a recent phenomenon.  It first emerged in Sweden and is closely related to the Scandinavian notion of a fully active society (Jensen and Halvorsen, 2004).  This is what might be called traditional activation. Active employment policy is Swedish in origin. The theoretical basis for the so-called Swedish model was laid down In 1951 by two economists with the LO trade union, Gøsta Rehn and Rudolf Meidner, who advocated a new policy mix that could efficiently reconcile economic and social justice.  In addition to an active labour market and a policy of employee solidarity, the third main ingredient of this policy mix was a restrictive budgetary policy.  This economic policy has to be seen in the context of the full employment requirement that formed a key element of the social-democratic credo and that persists in various forms to the present day, but is too often forgotten when current reforms of activation in Scandinavian countries are discussed (Barbier, 2002b; 2004a; 2004d).  In its original version, subsequently popularised by Rehn, activation was an overall economic and social policy objective[8].  Rehn himself, as head of the OECD Directorate for Manpower and Social Affairs from 1962 to 1973, introduced this notion into the Paris-based organisation's publications, though it was to take on a very different meaning from the 1980s.  In those years, European countries regularly drew inspiration from Sweden.  For example, France's main temporary employment support scheme, the so-called "employment-solidarity contract", introduced in 1984 and reformed in 1990, was directly based on Swedish practice (Barbier and Théret, 2001).

Much later, Denmark became the precursor of contemporary activation policies, which borrowed the term from that country's 1992-1994 reforms.  However the word activation had a different meaning, because it initially alluded, in connection first with unemployment insurance (a-kasserne) and then with assistance (kontanthjaelp), to the principle that benefit recipients must be "activated" at least after a certain time and for a certain duration.  This led to the so-called aktivering programmes, well before the Clinton reform and the British New Deals and Welfare-to-Work initiatives.  Although the Danish policy increased unemployed and assisted persons' obligations it also improved the supply of services and assisted contracts (jobtraening).  This is one of the main concerns of the current reforms, which seek to universalise the provision of services in Danish municipalities.  Sanctions have always been minimal (Barbier and Ludwig-Mayerhofer, 2004).  Employment rates in Denmark are high, and this extends to young people and the elderly, and jobs are of a higher quality than in other countries, as European Commission comparisons have shown.  Activation in its new Danish form has always, at least until now, involved more than just the application of an orthodox labour supply policy recipe (Jørgensen, 2002).

In France, the introduction of the minimum integration income (revenu minimum d’insertion - RMI)[9] in 1988 was the opportunity to innovate (Barbier and Théret, 2004).  The logic is that of the state as employer of last resort, which was re-emphasised in the reforms under way in 2004 (Barbier, 2004a), under the banner of integration (insertion).  Far from being punitive in intent, the original purpose of this approach was to integrate citizens into society through their participation in work and employment (Barbier, 2002c), in accordance with a tradition that can be traced back to the origins of the French revolution.  Unlike the Scandinavian model, however, the French system has always avoided a universalist approach (Barbier and Fargion, 2004). Even in Germany, a country often presented as being particularly "passive (Ludwig-Mayerhofer and Wroblewski, 2004), the Hilfe zur Arbeit principle was introduced into social assistance legislation well before the "new activation" reforms.

These few historical references, however simplified, suffice to show the variety of sources of inspiration for programmes that seek, not only to combine entitlement to benefits with assistance with entering or re-entering employment, but also, and more broadly, to "activate" social protection and, in the case of the Scandinavian countries, society as a whole.

The policies introduced following the arrival of the new Labour government in 1997, based on the welfare-to-work policy, which is directly derived from the American model (Deacon, 1999; King, 1995; Peck, 2001)[10], makes the United Kingdom something of a latecomer to the activation process.  Here again, the focus on programmes designed to change the behaviour of unemployed and other assisted persons was insufficient and the country, with strong impetus from the Chancellor of the Exchequer, Gordon Brown, entered into an in-depth reform of the social protection system, through the introduction, again based on the American model, of tax-credits and in-work benefits, with the aim of "making work pay", a slogan later to become common throughout Europe (Wright et al., 2004).

            A concept of activation

In its current form therefore activation cannot be reduced to a system of punishments and incentives for unemployed and assisted persons.  From an analytical standpoint, activation equates rather with the introduction (or strengthening or reactivation) of an explicit link between social protection (including employment and labour market policies) and individuals' occupational status and activities.  This link leads to a critical redefinition of social programmes, in the sense of a systematic bias towards beneficiaries' participation in the labour market, i.e. economic activity, or else a form of activity specially introduced to establish eligibility for benefits.  The main aspects of social protection concerned are unemployment benefits, employment policy and social assistance and solidarity benefits.  However, retirement, and in particular early retirement, provisions may also be affected (Guillemard, 2002).  Finally, the process has to be seen in the wider context of the financing of social protection and reform of the tax and benefit system.  At national level, the link between labour markets and social protection has been a structural feature of social protection systems since their inception and a key element of different types of welfare state or regime (Titmuss, 1974; Flora, 1986, Esping-Andersen, 1990; Merrien, 1997; Barbier and Gautié, 1998).  It is hardly surprising therefore that as the latter evolve, this central link - part of what economists of the French school of regulation call the "wage relation" – is being increasingly questioned.

The process described reflects a common phenomenon throughout Europe, and more generally in all the rich countries of the world.  As such, at this general level it is reasonable to speak of a convergent trend.  However, according to Pierson it is clear that at the more detailed level of the restructuring of the various types of social protection there is little convergence.  The common general trend is in fact of much less significance than the country to country variations in approach to and content of reforms.  Later we will show how these persistent variations are the key to understanding the options available for recasting social protection in Europe[11].

The difficulty of broadening our definition of the scope of activation should not be underestimated.  In empirical terms, we can identify three main categories of intervention by the various levels of government or by social protection agencies directed by the social partners.  The first concerns benefits, including unemployment insurance and social assistance to persons of active age who have exhausted their insurance entitlements and recipients of social benefits who can be deemed potentially active, such as lone parents and those receiving invalidity of family benefits.  In some countries the net may include early retirement, or even retirement, benefits or benefits in general[12].  The second category of intervention concerns employment policies, and includes measures to assist the young unemployed, temporary non-market employment, employment subsidies for the victims of stigmatisation, vocational training for the unemployed and so on, on all of which OECD has produced comparative statistical data for the last twenty years.  These are generally referred to as active labour market policies.  Finally, the third category comprises an intermediate range of measures between fiscal and social policies.  These are readily identifiable in liberal, or free-market, systems where there is a continuous tax and benefit system.  In Britain for example a family benefit formerly known as family credit was transformed in 1998 into the working families tax credit.  This category does pose a problem however in continental type systems as in France, Italy or Germany.  In France it also includes so-called incentive supplements for benefit recipients who find work, as in the case of the minimum integration income (RMI) or the "reduced activity" unemployment insurance allowances, introduced by the 2001 tax credit.  This type of measure must also be seen in conjunction with efforts in many countries, which are becoming increasingly marked, to reduce the burden of social charges for certain groups of employees and to reform the financing of social protection, as in France, where tax income has been used to compensate for reduced income from employees' employment contributions (Barbier and Théret, 2004).

These composite elements define the scope of activation, whose boundaries in practice are constantly changing in response to current reforms.  Although it focuses on social protection[13], the scope for applying activation policies is clearly circumscribed by other legal aspects of the organisation of the labour market, the most important of which is labour/employment law.  However the example of Italy, which has experienced difficulty in "activating" its social protection system even though it is making its labour market more flexible shows that the two processes are quite distinct (Barbier and Fargion, 2004).

            Two types of activation and an emerging third version

We have isolated two ideal types to offer a stylised picture of current activation reforms (Barbier, various references).  We will consider their main features and then analyse these "models", with a view to drawing certain normative conclusions.

The free market or liberal regime is based on the relationship between individuals and the labour market, which is supposed to produce the best employment outcomes in terms of effectiveness and equity.  Individuals who are not integrated into the market are in a temporary situation or are no longer employable and are dependant on assistance.  To increase the incentive element of the system benefits and taxes are integrated, through the development of in-work social and fiscal benefits.  Active labour market policies only play a limited role in this approach, with a relatively small number of participants being encouraged to make the transition to the ordinary market, the emphasis being on "work first", which does not provide for long-term training.  Employment on the ordinary labour market is intended to replace assistance.  Workfare is just one – American - element of the liberal type of activation.

In contrast, the universalist social-democratic regime offers the market a somewhat reduced role.  Social/employment policy retains its traditional aspect, oriented towards individuals' well-being, which it combines with economic objectives.  Replacement benefits for employment income are high and long term.  To make the system more coherent and sustainable, activation is applicable to all citizens, whatever their employment situation – be they employed, unemployed or on social assistance.  It is based on a diversified and extensive range of social services, including long-term training,  access to which is based on a contract negotiated between the individual and the social and employment services and including incentive, educational and penalty elements.  Fully active societies organise employment to take account of individual possibilities.

Obviously, no country adheres strictly to one or other model.  For example, Sweden and Denmark, which are close to the universalist schema, differ significantly with regard to employment relations and labour law (flexibility of contracts).  Neither has fully universalised the activation principle (Jørgensen et al, 1998; Barbier, 2001a).  The liberal activation approach in the United States, where only a marginal proportion of the population is covered by assistance (see the notion of underclass), takes a very different form to that in Britain where, despite the activation reforms, a very high percentage of today's active population is still considered to be welfare dependant.

Francecombines aspects of both activation approaches, and is hesitating between a disciplinary and a negotiated strategy, which underlay the conflict over the 2000 unemployment insurance reform.  It uses tools taken from each stylised policy type in a hybrid mixture that is clearly illustrated by the coexistence of several minimum incomes with no obligation to search for work, a range of penalties specifically applicable to unemployment insurance recipients, a large number of temporary jobs in the non-market sector and so on.  Germany has similar features, with the implementation of the so-called Hartz reforms and encouragement of mini-jobs (geringfügige Arbeit).  As we have seen, Italy has so far been incapable of carrying out a major reform in this area (Barbier and Fargion, 2004).  This is why, despite the fact that this "continental mixture" shows serious signs of the emergence of a third approach, we consider it premature at this stage to develop such a third ideal type (Barbier, 2004c; 2004b; Serrano, 2004).  It is now time to compare our existing ideal types and the empirical data available, with reference to a number of analytical and normative criteria that relate to Council of Europe principles.

4. Assessing the forms of activation in the light of the European Council indicators [This part is directly written in English]

The ‘transitional labour market’ (TLM) perspective is certainly among the most comprehensive frameworks existing for considering the implementation of policies aimed at combining flexibility and security on the labour market. We have confronted it to the analysis of activation types and articulated it with an ‘analytical’ notion of citizenship[14]. It is essential to note that the basic TLM framework considers four normative criteria that should be implemented so that various types of ‘transitions’ could be considered as ‘good transitions’. It is interesting to also note that the four normative criteria devised by Günther Schmid have obvious similarities with the grid of criteria established by the Council of Europe.

Activation and transitional labour markets

The basic approach to ‘activation’ present in the TLM framework is via the activating of five types of transitions of the individuals (from education/training to work; between various forms of employment; between employment and household or ‘civil’ activities; between employment and retirement/disability; between employment and unemployment). G. Schmid especially stressed (2002b, p. 426) that this goal requires “a shift from an active to an activating labour market policy”. ‘Activating measures[15]’ also feature in the picture of the four ideal-typical forms of risk-management according to whether risks arise from individual choices or circumstances and according to who is accountable (Schmid and Schöman, 2003, p. 20).

Normatively, the direction of reform suggested by the TLM approach has to abide by the general requirement which has informed G. Schmid’s proposals as to how to accommodate and adapt to the flexibility requirement of contemporary economic systems: “Wie kann die Flexibilität des Arbeitsmarktes unter den Bedingungen der Globalisierung und Individualisierung erhöht werden, ohne das Bedürfnis nach sozialer Sicherheit und Gerechtigkeit zu beeinträchtigen?” (Schmid, 2002a, p. 175). This normative perspective also fits the objective of the Council of Europe.

The stress put on ‘soziale Gerechtikigkeit’ on the same level as ‘soziale Sicherheit’ is probably a distinctive feature of G. Schmid’s normative approach, which has to be taken into account, and which implies a substantially different perspective from the traditional ‘flexicurity’ approach. But ‘Gerechtigkeit’ is difficult to translate exactly into other languages. It is certainly very different from the English notion of ‘fairness’ and it is closely related to a particular understanding of one of the four criteria (solidarity) brought forward by G. Schmid, along which ‘good transitions’ could be promoted from a ‘social choice’ perspective (2002b, p. 398; 2002a, p. 239).

His four criteria are: (i) freedom (or autonomy) of individuals; (ii) solidarity in risk sharing, which comprises the inclusion of the high-income groups via redistribution; (iii) effectiveness of coordination and cooperation including a public-private mix in implementation; and (iv) finally, efficiency, including decentralization and management by objectives. For all their normative content, these criteria may be also used as analytical criteria to assess certain existing activation strategies against others.

Indeed activation (in its broad and encompassing sense) can be seen as one of the key elements in a reform of social protection able to fit the normative goals of providing ‘good’ transitions in a changed framework, although it is certainly not the only one. This entails the necessity of not only identifying ideal-types but also systematically assessing them against criteria so as to possibly rate one type as better fitting the TLM approach than the other. But before this eventual stage of normative evaluation, analysis has to confront activation with the TLM perspective.

Table 2 – TLM and Activating Social Protection

	TLM	Activating social protection
Main Focus	LM transitions	Social protection systems and their linkage to employment/ labour market participation
Main mechanisms and instruments involved	Insuring risks by Providing rights to services, incomes Devising Incentive structures Opportunity structures	Restructuring the linkage of employment to Services Funding mechanisms Benefit eligibility Making systems ‘employment friendly’
Main social protection areas concerned	Emp/U insurance Training/Education	Assistance, family benefits, pensions, labour market programmes, unemployment insurance, training activities + social protection funding, and relations between taxing systems and social protection
Questions for the future/for the social norms Is the labour market link to pervade every area in the future? (centrality or universality of ‘paid work’?)	Are all social protection programmes potentially affected by the TLMs? Social services /welfare? Family benefits? only disability and pensions? what about assistance? Basic income?	Are all social protection areas to be affected by activating reforms? Social welfare? Health?

From table 2 it appears that there are many ways in which both the TLM approach and our analysis of activation ideal-types strongly connect with one another.The main programmes and policies (benefits and services) envisaged from the point of view of the TLMs are active labour market policies and a possible future ‘employment insurance’, which leaves rather open the question as to what other fields of social protection should attract attention for the implementation of TLMs.

On the other hand, focusing on the activation of social protection has generally entailed scarce attention to initial education and training and no explicit stress for a specific and sustainable support of transitions between ‘employment situations’.

If one assumes that current observed tendencies mean that, gradually, an ‘active society’ will encompass more and more people in the future, including working age individuals who in the previous systems would have remained outside the labour market/employment (women, early retirees, students, etc..), then the question arises, both in the TLM approach and in the domain of activating social protection, as to what are the limits where an active society ends? Are there areas of society where work/employment does not (or should not) reach? In this respect, the recent notion of ‘benefit dependency’ which is rooted into the normative and derogatory meaning attached to ‘dependence’ – at least in an Anglo-Saxon perspective (Sennett, 2003) should be challenged. In the normative TLM perspective, access to benefits seems to be a key point in the future architecture of social protection, which can no more be seen as ‘dependency’ and the Council of Europe’s approach clearly concurs on that point. The connection between activating social protection and the TLMs can be further explored, confronting both ideal-types to the criteria which Schmid has defined as essential for his general framework. From a normative perspective G. Schmid (2002b, p. 398; 2002a, p. 239-240) has suggested “four criteria that institutional arrangements have to fulfil in order to support ‘good transitions’ and to prevent ‘bad transitions’ ”, thus justifying a particular mix of soziale Gerechtigkeit and soziale Sicherheit. However, these criteria can be confronted with both our activation ideal-types using them as analytical instruments.

Table 3 - Activating social protection and the TLM criteria

G. Schmid’s criteria	Activating social protection (ideal-types)
	Liberal	Universalistic
Freedom/ autonomy [Entscheidungsautonomie (Freiheit)] [empowering people and establishing entitlements not only to transfers but also to participation in employment decisions – in exchange, individuals accept greater risks, duties and obligations]	Individualism, self reliance Disciplinary approach Incentives primary Sanctions essential	Demands of society/individuals balanced Reciprocal engagement strategy Incentives, sanctions, services and benefits Sanctions marginal
Solidarity [Förderung der Solidarität] [generality and inclusiveness in risk sharing; some ex-ante redistribution from high-income groups]	Market provision primary Safety net assistance and welfare Welfare dependency [welfare to work] Unequal exposure to risks of ‘bad jobs’/ working poor/worklessness/gender and age	State provision /high socialisation of risks High commodification/high de-commodification Universal provision of services and benefits [universal activation] High degree of equality (income/access to employment) (including state as employer of last resort)
Effectiveness [Effektivität] [improving effectiveness: specialization, coordination and cooperation; public-private mix]	Underemployment when quality is not achieved (collective and individual) Social partners not involved Public-private mix	Quality full employment Reciprocal engagement means individual success of contracts Social partners involved Public-private mix
Efficiency [Effizienz] [increasing efficiency through risk management techniques; evaluation and self-regulation, decentralization and management by objectives]	NPM (re-commodification) Dominant: central state	NPM (cost containment, updating, re-commodification) decentralisation

From table 3, it would appear that G. Schmid’s four criteria are significantly more compatible with the universalistic type of activation, although they fit with the liberal type to a lesser degree. Concerning the freedom/autonomy criterion, things are rather clear. In the TLM approach, existing entitlements to rights (Bürgerrechten) are either maintained, reformed or extended in many new areas, which is not the case in the liberal type of activation, where entitlements tend to be only residual or available in case of last resort. Hence, with respect to rights, benefits, and most importantly, services, the liberal type of activation tends to contradict the choice and freedom substance within a ‘contract’ where sanctions are prominent. The meaning of “accepting risks, duties and obligations” in a context where generous entitlements are available is bound to be different from a context – such as the liberal, where only residual entitlements exist.

However, it is probably more relevant to use the first two criteria (freedom/solidarity), as a couple: the comparison between the liberal and the universalistic types of activation demonstrates that, because of poor collective (aggregate) outcomes (inequalities of access to income replacement and employment) and because of the great pressure put on individuals, inclusiveness in risk sharing is not achieved in the former type, while at the same time redistribution from the high income groups is problematic (Korpi and Palme, 1998) and the targeted system constantly lacks sufficient support in society.

The confrontation of both types of activation tends to highlight the key question of what exactly are the entitlements and their conditionality, and of the forms and degree to which they imply risk sharing. Here we find the difficult question of the distinction between ‘active’ and ‘passive’ benefits. As a tendency, in the liberal type of activation, all benefits are seen as second-best to an absence of ‘dependency’, and consequently disqualified as ‘passive’, which is not the case in the TLM perspective, but the difficult question is where the line should be drawn from ‘active’ to ‘passive’, or, in the context of the more precise term, ‘activating’ as against ‘passive’. For instance, Esping Andersen and his colleagues (2002) have noted that, even in its present form, unemployment insurance can be a very positively activating instrument, because it allows choice for good quality employment.

It must be stressed here that the question is not only of any form of ‘income compensation’ but of a special type of compensation, organised through social insurance, which implies trust and solidarity across society, or even some form of social nexus, a very different indeed way of addressing what the Anglo-Saxon approach calls ‘welfare’ or ‘assistance’. G. Schmid has insisted on this aspect, which can also be linked to the French regulation school premises (the ‘wage/labour nexus’, Boyer, 1986): speaking about the Solidarität condition, he explicitly writes : “Eine zuverlässig solidarische und gerecht empfundene soziale Sicherung ist daher eine zweite Anforderung an eine moderne Arbeitsmarktpolitik” (2002a, p. 239) and he comments in a foot-note: “ Diese Maxime ist zentral etwa für die Beurteilung, ob die soziale Sicherung nur (“gezielt“) den Bedürftigen oder berechenbar allen Bürgerin/innen (oder zumindest einer qualifizierten Merheit) gewährleistet wird. Davon hangen beispielsweise die Legitimation und Unterstützungsbereitschaft für eine solidarische Arbeitslosenversicherung ab“.

When confronted with our two ideal-types, we are led to interpret G Schmid’s criterion of ‘effectiveness’ along with an additional dimension: effectiveness, we suggest, should also encompass the overall effectiveness of social and employment policy – actually linked to effective solidarity, including the capacity of policy to effectively deliver public goods.

Only in the universalistic type are citizens able to fully rely upon the contract with society, because when they abide by collective rules, their reciprocal engagement will be met by provision of some sort in a context of quality full employment[16].

Activation and  the Council of Europe’s perspective

The Council of Europe’s framework can be explored in the light of the previous discussion. Its four ‘dimensions of welfare’ can be related to the discussion about both citizenship and the substance of activation policies (revisited here in the light of the TLM perspective). In the present section we try to wrap up a comparative presentation of all these approaches, that, to our view, constitute similar endeavours to build a normative solidaristic framework. Operationalizing the notion of citizenship is useful here as a sort of bridge from the activation/TLM perspective to the Council of Europe project. We discussed the relationship between activation ideal-types and citizenship (Barbier, 2004e) in the light of Goul Andersen’s (2004) grid displayed in table 4.

Table 4: Criteria for analysing the impact of activation reforms upon citizenship

Goul Andersen’s analytical definition of citizenship	Our citizenship criteria in the perspective of activation
Rights (civil, political, social) actually implemented social rights : eligibility, level, adequacy, duration, conditionality, sanctions	1 - Freedom of choice and autonomy 2 - Quality of services 3 - Generosity and duration (benefits) 4 - Sanctions and conditionality
Political and social participation	5 -Participation to the formulation and the implementation of policies and programmes (role of actors, social partners, democratic debate, demonstrations..)
Identities
	6 - Access to quality full employment
	7- Inequality across genders and ages; differences insured/assisted

©Barbier (2004e) adapted from Goul Andersen (2004). Five dimensions are explored: (i) individual freedom of choice and autonomy (to what degree individual choices are taken into account by the administration); (ii) the availability of associated various quality services (counselling, support, social services, etc.); (iii) generosity (and duration) of benefits; (iv) conditionality and sanctions for non compliance; (v) participation of the social partners and other actors in policy formulation and implementation. We add the impact of full employment and the question of equality: is equality of opportunity actually existing across genders, ages, and categories of welfare recipients?

Goul Andersen’s analytical concept helps comparing transformations which have effectively occurred across the European nations as consequences of the ‘activation’ of social protection systems. Although the main aspect considered here pertains to social rights (benefits and services), their linkage to political and social participation cannot be eschewed. However, to this point, we have no corresponding criterion for his category of ‘identities’.

Moreover, the question of ‘rights and obligations’ ought to be contextualised, in order to avoid a simple extension of what is empirically observable in the Anglo-Saxon world (Barbier, 2002a). In other words, whilst developments in various countries show that entitlements to social benefits and services have been redefined across the board, it is impossible to capture the meaning of this in terms of a universally downgraded ‘social citizenship’. Indeed, de-contextualised concepts of ‘rights’ as against ‘obligations’ fail to help us grasping how the actual substance of ‘social citizenship’ and, more broadly, how citizenship, has been affected. This, we contend, is the reason why it is important to bring ample empirical material to the discussion in the future. It is obviously much easier to identify the procedural altering of rights and entitlements of persons eligible to ‘welfare-to-work’-type programmes than to take into account their actual implementation (Barbier and Ludwig-Mayerhofer, 2004). And it is even more difficult to assess the far-reaching consequences of the reforms of tax and benefit systems (including tax-credits or the subsidising of labour costs).

Moreover, as a determining factor for citizenship, full employment certainly does not belong to all contexts: empirically this is well documented. Yet, the impact of activation policies like the UK ‘welfare to work’, the Danish ‘aktivering’ or the French ‘insertion’ programmes upon ‘rights’ is bound to be profoundly shaped by the overall economic circumstances of the country where the policies are implemented. When ‘quality jobs’ are available and numerous on the market, the quality of citizenship cannot but be very differently affected by programmes which activate people, when compared to cases where jobs are missing, or where only ‘dead-end’ or ‘precarious’ jobs are available.  The substantive nature of the obligation to take a job is different in both cases. Here we again encounter the necessity of contrasting ideal-types of activation and of privileging the universalistic one, in the present normative discussion.

All in all, the successive steps of our assessment leads to positioning the elements reviewed in table 5, which, at this stage, can be seen as a contribution to the insertion of a debate on activation strategies within the Council of Europe project.

Table 5 - Comparing normative frameworks

TLM good transitions	The universalistic ideal-type of activation	↔	Citizenship (as affected by activation strategies for social protection)	↔	Council of Europe Social cohesion criteria
Freedom, autonomy ↕ Solidarity	Balance (society-individual) (rights/obligations) and reciprocal engagement		Freedom of choice + participation to the formulation and implementation + in the context of sanctions and conditions		Autonomy Personal development/dignity Participation Dignity and recognition Equity of access Equity of access Equity of access
	High socialisation of risks/redistribution		Degree of equality/solidarity across categories
Effectiveness ↕ Efficiency	Macro-quality full employment Micro – quality services/benefits		Effective access to employment Quality of services, benefits; generosity, duration
	Costs are contained Efficient public management		Access to sustainable full employment and social protection

5. Conclusion

            The universalistic type of activation is consistent with criteria of good transitions on the labour market because it fits quite well the two couples of criteria ‘freedom and solidarity’ and ‘effectiveness and efficiency’. Secondly, citizenship is bound to be affected by the current reforms under the banner of activation. If the activation of social protection conforms to the universalistic features identified here, it is logic that citizenship dimensions should be protected from substantive degradation (by comparison to traditional social protection before activation reforms).

Thirdly, the analytical dimensions of citizenship presented here may be related to the normative criteria proposed in the Council of Europe’s social cohesion normative framework. A social cohesion perspective then appears as globally compatible with a universalistic activation strategy, on the basis of freedom/autonomy (autonomy and personal integrity) and solidarity (equity of access), but also on the basis of an actual implementation (actual equity).

The universalistic activation ideal-type consequently appears as the best template for normatively orienting solidaristic and social cohesion strategies. True, it has not been exactly implemented in any country but the Scandinavian systems of social protection are very close to it, despite real differences existing between Norway, Denmark, Sweden and Finland (Jensen and Halvorsen, 2004). Thus, the key challenge for the future is to learn from their systems, bearing in mind that ‘policy transfer’ or ‘imitation’ has always been extremely narrow and bounded by the constraints inherited from the belonging to a welfare regime. But the challenge is great because these countries are the only ones which have been able so far to reconcile flexibility and security in the context of the ‘flexibilisation drive’ described in section two: they actually have constantly featured among the best pupils in the class within the European Employment Strategy, combining orthodox monetary and fiscal policies with solidaristic and generous social policies.

References

Abrahamson P., 2001, “L’activation des politiques sociales scandinaves : le cas du Danemark”, in Daniel C. et Palier B. ed., La protection sociale en Europe, le temps des réformes, MIRE, Documentation Française, Paris, p. 123-140.

Barbier J.-C., 1996, “Comparer workfare et insertion?”, Revue française des affaires sociales, n°4, octobre-décembre, p. 7 - 27.

Barbier J.-C., 2001, “Temporary subsidized employment in the public and non profit sector: a European comparative study”, report for the Evaluation Committee, March, 75p. Noisy le Grand  (http://www.cee-recherche.fr/fr/fiches_chercheurs/texte_pdf/Mesanom%20English%20Final.pdf).

Barbier J.-C., 2002a, « Marché du travail et systèmes de protection sociale, pour une comparaison internationale approfondie », Sociétés contemporaines, n°45-46, p. 191-214.

Barbier J.-C., 2002b, « Peut-on parler d’activation de la protection sociale en Europe ? » Revue française de sociologie, n° 43-2, avril-juin, p. 307-332.

Barbier J.-C., 2002c, « Des modèles d’insertion en Europe ? », in Ruptures, segmentations et mutations du marché du travail, Tremblay D.-G., Dagenais L.F. dir ., Presses de l’Université du Québec, Ste Foy, p. 27-46.

Barbier, J.-C. (2004a) ‘The European Employment Strategy, a channel for activating social protection?’, in Zeitlin J., Pochet P., with Magnusson L. (eds) The Open Method of Coordination in Action: The European Employment and Social Inclusion Strategies, Brussels: PIE-Peter Lang (forthcoming).

Barbier, J.-C. (2004b) ‘Systems of social protection in Europe: two contrasted paths to activation and maybe a third’, in J. Lind, H. Knudsen, H. Jørgensen (eds) Labour and Employment Regulation in Europe, Brussels: PIE- Peter Lang, pp 233-254.

Barbier, J.-C. (2004c) ‘Activation Policies: A Comparative Perspective’, in Serrano Pascual A. (ed) Are Activation Policies Converging in Europe, The European Employment Strategy for Young people, Brussels: ETUI, pp 47-84.

Barbier J.-C., 2004d, (with a contribution by Sylla N. S.), La stratégie européenne pour l'emploi: genèse, coordination communautaire et diversité nationale, rapport pour la DARES (Ministère du travail), janvier, http://www.cee-recherche.fr/fr/rapports/stategie_europeenne_emploi.pdf.

Barbier J.-C., 2004e, “Citizenship and the Activation of Social Protection: a Comparative approach”, in Andersen, Jørgen Goul; Guillemard, Anne Marie; Jensen, Per H.; Pfau-Effinger, Birgit (eds): The New Face Of Welfare. Social Policy, Marginalization and Citizenship, COST A13 Book Series, Bristol: Policy Press, forthcoming.

Barbier J.-C., 2004f, “Research on Open Methods of Coordination and national social policies: what sociological theories and methods?”, paper for the RC19 International conference, Paris, 2-4 September.

Barbier J.-C. et Fargion V., 2004, “Continental inconsistencies on the path to activation : consequences for social citizenship in Italy and France”, European Societies, (6) 4 : 437-460.

Barbier J.-C. and Gautié J. (1998) Les politiques de l’emploi en Europe et aux Etats Unis, Cahiers du CEE, Paris: PUF.

Barbier J.-C. et Ludwig-Mayerhofer W., 2004, « The many worlds of activation », European Societies, Introduction to the special issue (eds) (6) 4: 423-436.

Barbier J.-C. and Nadel H., 2000, La flexibilité du travail et de l’emploi, Domino, Flammarion, Paris [2002, La flessibilità del lavoro et dell’occupazione, introduzione di L. Castelluci e E. Pugliese, Donzelli, Roma.]

Barbier J.-C. and Théret B., 2001, “ Welfare to Work or Work to Welfare, the French Case ? ” in Gilbert N. and Van Voorhis R., Activating the Unemployed : A Comparative Appraisal of Work-Oriented Policies, Transaction Publishers, Rutgers, N.J., p. 135-183.

Barbier J.-C. and Théret B. (2004) Le nouveau système français de protection sociale, Repères, Paris: La Découverte.

Boyer R., 1986, La flexibilité du travail en Europe, Paris :La Découverte.

Deacon A., 1999, “The Influence of European and American Ideas upon ‘New Labour’ Thinking on Welfare Reform”, paper presented to Conference on ‘Global Trajectories: Ideas, Transitional Transfer and Models of Welfare Reform’, Florence, 25-26 March.

Esping-Andersen G., 1990, The three worlds of welfare capitalism, Cambridge: Polity Press.

Esping-Andersen, G., Gallie, D., Hemerijck, A., Myles, J. (2002) Why we need a New Welfare State, Oxford:  Oxford University Press.

Flora P., 1986, ed, Growth to Limits, the Western European Welfare State since World War II, Berlin: de Gruyter.

Geldof D., 1999, “New Activation Policies: Promises and Risks”, in Linking Welfare and Work, European Foundation, Dublin, p. 13-26.

Goul Andersen J., Clasen J., van Oorschot W., Halvorsen K., 2002, Europe’s New State of Welfare, unemployment, employment policies and citizenship, The Policy Press, Bristol.

Goul Andersen J., 2004, “Citizenship, unemployment and welfare policy” in Goul Andersen, Jørgen; Guillemard, Anne Marie; Jensen, Per H.; Pfau-Effinger, Birgit (eds): The New Face Of Welfare. Social Policy, Marginalization and Citizenship, COST A13 Book Series, Bristol: Policy Press, forthcoming.

Guillemard A.-M., 2002, « L’Europe continentale face à la retraite anticipée, barrières institutionnelles et innovations en matière de réforme », Revue française de sociologie, n° 43-2, avril-juin, p. 333-368.

Hort S., 2001, “Sweden – Still a Civilised Version of Workfare?”, in Gilbert N. and Van Voorhis R., 2001, Activating the Unemployed: A Comparative Appraisal of Work-Oriented Policies, Transaction Publishers, Rutgers, N.J., p. 243-266.

Jensen P. and Halvorsen K., 2004, “Activation in Scandinavian welfare policy”, European Societies, (6) 4 : 461-484.

Jessop B. , 1993, “Towards a Schumpeterian Workfare State? Preliminary remarks on post-Fordist political economy”, Studies in Political Economy, n°40, p. 7-39.

Jessop B., 1996, “Post-Fordism and the State”, in Greve E., Comparative Welfare Systems, the Scandinavian Model in a Period of Change, Mc Millan, Basingstoke and London.

Jørgensen H., Larsen F., Lassen M. and Stamhus J., 1998, “La politique active du marché du travail au Danemark : réforme et décentralisation” in Barbier J.C. et Gautié J., dir., 1998, Les politiques de l’emploi en Europe et aux Etats Unis, PUF, Paris, p. 155-178.

Jørgensen H., 2002, Consensus, Cooperation and Conflict, the Policy Making Process in Denmark, Edward Elgar, Cheltenham.

King D., 1995, Actively Seeking Work, The Politics of Unemployment and Welfare Policy in the United States and Great Britain, University of Chicago Press, Chicago.

Korpi W., Palme J., 1998, « The paradox of redistribution and strategies of equality : welfare state institutions, inequality, and poverty in the western countries », American Sociological Review, vol.63, n° 5, October, pp. 661-687.Kvist J., 2002

Kosonen P., 1999, “Activation, incitations et workfare dans quatre pays nordiques”, in Comparer les systèmes de protection sociale en Europe du Nord et en France, rencontres de Copenhague, Mire-Drees, vol 4, tome 1, Paris, p. 453-480.

Lallement M., 2001, “ L’emploi des sociologues ”, in Pouchet A., coord., Sociologies du travail : 40 ans après, Elsevier :Paris, p. 173-190.

Lødemel I. et Trickey H., 2000, An Offer You Can’t Refuse, Workfare in International Perspective, Policy Press, Bristol.

Ludwig-Mayerhofer W. and Wroblewski A., 2004, “Eppur si muove?”, European Societies, (6) 4 : 485-510.

Maruani M. et Reynaud E., 1993, Sociologie de l'emploi, La Découverte, Paris.

Merrien F.-X., 1997, L’Etat-providence, Paris :PUF.

Morel S. (2000) Les logiques de la réciprocité, les transformations de la relation d’assistance aux Etats Unis et en France, Paris: PUF.

Nicole Drancourt C., 1990, “  Organisation du travail des femmes et flexibilité de l’emploi ”, Sociologie du travail, n°2.

Peck J., 1996, Workplace, the social regulation of labour markets, New York :Guilford Press.

Peck J., 1998,  “Workfare, a Geopolitical Etymology”, Environment and Planning, Society and Space, vol 16, p. 133-161.

Peck J.,  2001, Workfare States, Guilford Press, New York.

Pierson P., 2001, ed., The New Politics of the Welfare State, Oxford University Press, Oxford.

Schmid G., 2002a, Wege in Eine Neue Vollbeschäftigung, Übergangsarbeitmärkte und aktivierende Arbeitsmarktpolitik, Campus Verlag, Frankfurt.

Schmid G., 2002b, “Towards a New Employment Compact”, in Schmid G. and Gazier B., eds, The Dynamics of Full Employment, Edward Elgar, Cheltenham, p. 392-435.

Schmid G. and Schöman K., 2003, “Managing Social Risks Through Transitional Labour Markets: Towards a European Social Model”, TLM paper, Berlin conference.

Schmid G. and Gazier B. (ed) (2002) The Dynamics of Full Employment, Cheltenham: Edward Elgar.

Segrestin D., 2001, “Entreprises et organisations, un état des lieux”, in Pouchet A., coord., Sociologies du travail : 40 ans après, Elsevier :Paris, p. 257-274.

Sennett R., 2003, Respect, the formation of character in an age of inequality, Allan Lane The Penguin Press, London.

Serrano Pascual A. (2004) ‘Conclusion: Towards Convergence of European Activation Policies?’, in Serrano Pascual A. (ed) Are Activation Policies Converging in Europe, The European Employment Strategy for Young people, Brussels: ETUI, pp 497-518.

Titmuss R. M., 1974, Social Policy, an introduction, London: Allen and Unwin.

Torfing J., 1999, “Workfare with welfare: recent reforms of the Danish welfare state”, Journal of European Social Policy, 9 (1), p. 5-28.

Van Berkel R. and Möller I. H., 2002, Active Social Policies in the EU, Inclusion through participation?, The Policy Press, Bristol.

Wood S. (2001) ‘Labour Market Regimes under Threat? Source of Continuity in Germany, Britain and Sweden’, in Pierson P. (ed) The New Politics of the Welfare State, Oxford: OUP, pp 368-409.

Wright S., Kopac A. and Slater G., 2004, “Continuities with paradigm change”, European Societies, (6) 4 : 511-534.