New forms of governance in welfare-to-work policies Rik van Berkel and Willibrord de Graaf Work in progress Paper prepared for the 2nd RECWOWE conference, Warsaw, June 2007, strand 4 Introduction All over Europe, activation or active labour market policies and programmes are one of the core instruments to promote the transition from welfare to work, i.e. to (re)integrate people dependent on social insurance or social assistance in the labour market. These policies and programmes are expected to play a key role in making welfare states more ‘activating’, and in reducing welfare state dependency. The European Employment Strategy, that contains specific guidelines with respect to the introduction of activation programmes, clearly endorses this new orientation of social policies. Of course, between European countries significant differences exist regarding activation: in public funds available for programmes, in the nature and content of programmes, in the coverage of programmes and the treatment of various target groups, etcetera. But this does not take away that throughout Europe, the social problems of unemployment and welfare state dependency are dealt with in quite different ways nowadays compared to, say, 2 decades ago. Social policy analysts have written extensively on this new ‘trend’ in welfare states, on the contents of activation programmes, on the consequences of this trend for unemployed people, on the effects of programmes on labour-market (re)entry, etcetera (see, for example, Lo/demel and Trickey, 2001; Van Berkel and Hornemann Mo/ller, 2002; Gilbert, 2002; Serrano Pascual, 2004; Handler, 2004; Serrano Pascual and Magnusson, 2007). In this paper, our main focus is not on the substantial aspects of activation programmes, but on the ways in which the provision and delivery of these programmes are organised or, put differently, on the governance of social security and activation policies. In the study of social policy reforms and welfare state transformation processes, social policy analysts traditionally paid relatively little attention to issues of governance. These were the research domain of scholars of public administration and public management who, on their turn, were less concerned with issues of policy substance. More recently, social policy researchers started to cross traditional (sub)disciplinary boundaries, and to show an increasing interest in integrating, as it has been called, the study of ‘formal’ policy (that is, the substance of social policies) and ‘operational’ policy (that is, the organisation and management of policy making and policy delivery processes) (for example, Considine, 2001; Carmel and Papadopulos, 2003; Struyven and Steurs, 2005; Van Berkel and Van der Aa, 2005; Bredgaard and Larsen, 2005; Sol and Westerveld, 2005; Newman, 2005; Henman and Fenger, 2006; Van Berkel and Valkenburg, 2007). There are several reasons why we believe that this development in the focus of social policy research should be welcomed. First of all, the introduction of new modes of governance of social policy programmes in general, and of activation programmes specifically, has become an ever more important element of welfare state transformations. Here again, international institutions such as the EU (see Visser and Kouw, 2006) and the OEDC (for example, OECD, 2003) play a significant role. Although the new governance ‘hype’ started much earlier, the last five years has seen the launch of major reform programmes in various European countries to reform the delivery and provision of social security and employment services. To mention a few examples: the introduction of the Jobcentre Plus concept in the UK, the SUWI reforms in the Netherlands, the Hartz reforms in Germany, and the NAV reform programme in Norway (see below). Of course, the growing appeal of new governance affects the entire public sector and in that sense takes place independent of specific social policy reforms, as a result of increasing dissatisfaction with the functioning of state bureaucracies, a growing trust and belief in the virtues of the market, a neo-liberal outlook on the role of the state and of public policies, etcetera. Partly, however, new ways of governing social policies are directly related to substantial policy reforms. In the process of social policy reform, many governments have experienced that transforming welfare states requires institutions and agencies that actually implement and ‘produce’ these reforms, for example, by developing new social services, integrating benefit and employment services, introducing new criteria for testing benefit eligibility, practicing new attitudes towards claimants, etcetera – and it is one of the objectives of new modes of governance to promote this process of institutional change, either by promoting or enforcing transformations of existing welfare agencies, or by establishing completely new ones. In other words, the social policy and governance reform agendas are closely linked, at least to a certain extent. Secondly, and maybe more importantly, the increasing emphasis in social policy research on issues of governance should be welcomed because governance reforms are not without consequences for the content of social policy programmes[1] (for marketisation, see Bredgaard and Larsen, forthcoming; for decentralisation, see Finn, 2000; Van Berkel, 2006). Several of these consequences will be mentioned elsewhere in this paper. Generally speaking it is important to recognise that the nature and content of social policies and social services in practice are not simply a product of ‘official’ policy programmes, but also of their governance and implementation – an insight that is often neglected in policy evaluation research. Thirdly, the feasibility of social policy reforms may be influenced by specific governance configurations. As Clasen and Clegg argue, “different institutional configurations may themselves create quite varied openings for political action and leaderships” (2006: p. 547). They substantiate this through an analysis of reforms of unemployment protection schemes in four countries, concluding that “devolved governance structures coupled with a high degree of self-financing (rather than tax funding), or tight institutional links between unemployment insurance and political economy structures, serve as obstacles for a smoother reform path towards adapting unemployment support to post industrial labour markets” (ibid.: pp. 545-546). We see the ways in which the provision of social services is organised and managed as one of the elements of these ‘institutional configurations’. The rise of new governance is often interpreted as a move away from the traditional bureaucratic, centralised and hierarchical ways of providing public services towards a service provision model guided by ‘the three E’s’ of economy, efficiency and effectiveness, inspired by market mechanisms and based on separating the roles of purchasing and providing services (see Osborne and Gaebler, 1993). In public administration terms, the market model replaces the hierarchical model of organising policy. In this interpretation of new governance, it is practically synonymous with New Public Management and its doctrine of ‘running government like a business’. Nevertheless, ‘defining’ new governance as a shift from a hierarchy to a market or NPM model of service provision would be a simplification. If we want to use the concept of new governance to capture the changes taking place in the provision of activation services and the administration of income protection programmes, we think it is more adequate to define it as a hybrid model of service provision; that is, as a mix of hierarchical, market and partnership/network based modes of coordination, which may vary from country to country, from social service to social service area, or even within social service areas (see, for example, Newman, 2001). As Martin (2001: p. 209) wrote: “This suggests not the arrival of a new, hegemonic ‘outcome-focused paradigm’ but a more gradual transition characterized by the co-existence and interaction of hierarchical, market based and collaborative frameworks for co-ordinating service delivery”. The question of what mode of coordination of the provision of social services is used, is not simply a ‘technical’ issue about what actors to involve in service provision, or how to structure the relationships between them. It is also based on, or has consequences for, a range of other issues, such as the values that will guide the service provision process, the role and position of service users in service provision, the accountability of service providers, the discretion of frontline workers involved in service provision, opinions on how the behaviour of service providers and service users should be steered, etcetera. This is clearly illustrated by the following table developed by Denhardt and Denhardt (2000: p. 554). The table distinguishes three models of service provision, which roughly coincide with the three modes of coordination (hierarchy, market, network) mentioned before. Table 1. Three models for the provision of social services +--------------------------------------------------------------------------------------------+ | |Old Public |New Public Management |New Public Service | | |Administration | | | |-----------------------+----------------------+----------------------+----------------------| |Primary theoretical and|Political theory, |Economic theory, more |Democratic theory, | |epistemological |social and political |sophisticated dialogue|varied approaches to | |foundations |commentary augmented |based on positivist |knowledge including | | |by naive social |social science |positive, | | |science | |interpretive, | | | | |critical, and | | | | |postmodern | |-----------------------+----------------------+----------------------+----------------------| |Prevailing rationality |Synoptic rationality, |Technical and economic|Strategic rationality,| |and associated models |‘synoptic man’ |rationality, ‘economic|multiple tests of | |of human behaviour | |man’ or the |rationality | | | |self-interested |(political, economic, | | | |decision maker |organisational) | |-----------------------+----------------------+----------------------+----------------------| |Conception of the |Politically defined |Represents the |Result of a dialogue | |public interest |and expressed in law |aggregation of |of shared values | | | |individual interests | | |-----------------------+----------------------+----------------------+----------------------| |To whom are public |Clients and |Customers |Citizens | |servants responsive? |constituents | | | |-----------------------+----------------------+----------------------+----------------------| |Role of government |Rowing (designing and |Steering (acting as a |Serving (negotiating | | |implementing policies |catalyst to unleash |and brokering | | |focusing on a single, |market forces) |interests among | | |politically defined | |citizens and community| | |objective) | |groups, creating | | | | |shared values) | |-----------------------+----------------------+----------------------+----------------------| |Mechanisms for |Administering |Creating mechanisms |Building coalitions of| |achieving policy |programmes through |and incentive |public, non-profit, | |objectives |existing government |structures to achieve |and private agencies | | |agencies |policy objectives |to meet mutually | | | |through private and |agreed upon needs | | | |non-profit agencies | | |-----------------------+----------------------+----------------------+----------------------| |Approach to |Hierarchical – |Market-driven – the |Multifaceted – public | |accountability |administrators are |accumulation of |servants must attend | | |responsible to |self-interests will |to law, community | | |democratically elected|results in outcomes |values, political | | |political leaders |desired by broad |norms, professional | | | |groups of citizens (or|standards, and citizen| | | |customers) |interests | |-----------------------+----------------------+----------------------+----------------------| |Administrative |Limited discretion |Wide latitude to meet |Discretion needed but | |discretion |allowed administrative|entrepreneurial goals |constrained and | | |officials | |accountable | |-----------------------+----------------------+----------------------+----------------------| |Assumed organisational |Bureaucratic |Decentralised public |Collaborative | |structure |organisation market by|organisations with |structures with | | |top-down authority |primary control |leadership shared | | |within agencies and |remaining within the |internally and | | |control or regulation |agency |externally | | |of clients | | | |-----------------------+----------------------+----------------------+----------------------| |Assumed motivational |Pay and benefits, |Entrepreneurial |Public service, desire| |basis of public |civil-service |spirit, ideological |to contribute to | |servants and |protections |desire to reduce size |society | |administrators | |of government | | +--------------------------------------------------------------------------------------------+ Source: Denhardt and Denhardt, 2000: p. 554 It is not difficult to find all these models simultaneously present in the debates on, and practices of, the provision of activation services. The administration of social benefits, even though these have been increasingly linked to activation, is still taking place according to an ‘old public administration’ logic in most countries. The provision of activation services itself also often reveals traits of the old public administration logic, even in countries that have introduced market mechanisms in the provision of services (see below). For example, countries develop regulation that should promote the participation of the most vulnerable people in activation programmes; people that in a ‘free activation market’ may be excluded from participation through processes of creaming and parking. Several countries that have started to introduce the New Public Management logic in the provision of activation, fell back on elements of the old public administration logic, partly to promote typical ‘bureaucratic’ values such as equal access and inclusion, partly to ensure that ‘typical’ market values, such as efficiency, are actually realised. The bureaucratic mode of service provision has also lost strength as a consequence of the increasing importance attached to networking and partnerships in activation. At the same time that competition is promoted and introduced to improve service quality and reduce service costs, cooperation is advocated in order to integrate social security agencies and Public Employment Services, or to deal with complex problem situations and so-called ‘wicked issues’ (cf. Newman, 2001) in a way that fits the needs and situations of individuals, especially the most vulnerable, and of deprived areas and communities. Evidently, mixed models of service provision produce tensions and contradictions. Nevertheless, we believe these mixed models to be the rule rather than the exception of the new modes of governance currently being practiced in the provision of social services. Because of this, we think it is more useful to study countries in terms of the governance mixes they have adopted, the tensions and problems these produce, the effects these have on the nature of activation and the ‘publicness’ of activation services, the measures taken to cope with ‘perverse’ effects of modes of governance, etcetera, than in terms of their transition from one ideal-typical model of governance to another. In the remaining parts of this paper, we will discuss some elements of the new forms of governing activation programmes and policies. Subsequently, we will pay attention to the aspects of decentralisation, marketisation and competition, networking and partnerships, and individualisation and contractualisation. Finally, we will try to indicate some more general questions which pertain to these aspects: the implications of who counts as a stakeholder, the issue of rights and obligations which affects the status of the client as a contractual party, the relation between local approaches and globalising developments, and the consequences of the individualising strategy. We do not intend to provide a full and exhaustive overview of the developments and problems in this area. Our aims are less ambitious: to illustrate some of the elements of the new modes of service provision, and to provide some theoretical insight into the impact models of activation service provision may have on the content and nature of activation programmes. We hope that these insights may become an integral part of future research projects and will be connected to the debate about the policies in this field. Decentralisation Decentralisation can in general be described as a devolution of responsibilities for, and/or of authority in the design and delivery of social policy programmes from the national state to the regional or local level (‘territorial decentralisation’, often distinguished from ‘functional decentralisation’). The issue of decentralisation in social policies is an important one in all member states of the EU (Thuy et al., 2001; Sultana and Watts, 2004; Giguere, 2005), even though it is being debated and implemented in very different contexts, ranging, for example, from the traditionally highly centralised UK or French welfare states to the very decentralised – and often fragmented – Italian welfare state (see, for example, Borghi and Van Berkel, 2007). Focusing specifically on the decentralisation of activation (or employment) services, Greffe (2003; also see OECD, 2001)) has listed several reasons why national governments decide to opt for decentralisation. Most of them relate to societal changes and policy responses to these changes that require policy making and/or policy delivery processes in proximity to the contexts and collective and individual actors involved: the multi-faceted nature of unemployment, the increasing volatility of the labour market in a globalising and knowledge-based economy, the aim to develop innovative solutions through local partnerships, the transformation of passive into active welfare states and – for programmes aimed at promoting social inclusion in a broader sense than paid employment – the identification and development of activities that can promote social participation. Other reasons that are mentioned are: firstly, a reduction of welfare state expenditures by separating (national) policy making from (local) policy implementation and by introducing forms of new public management, including management by objectives, local policy autonomy in the context of national guidelines, etcetera; and secondly, a revival of forms of social dialogue that, at least in some countries, lost meaning at national levels. In literature on decentralisation in the context of activation or labour-market policies, two forms of decentralisation (which may occur simultaneously) are distinguished (e.g., Giguere, 2005). On the one hand, nations with country-wide PES organisations may decide to give regional/local agencies more room in making decisions regarding programme design and programme delivery, in order to increase their possibilities to adapt national policies to regional/local circumstances (e.g., Germany and Spain). On the other hand, national governments may devolve policy competences to regional or local governments (for example, the Netherlands and Denmark). In order to explore relationships between decentralisation and substantial characteristics of activation programmes, other characteristics of the decentralisation process are important as well. One of them concerns the ways in which national governments try to ensure that regional/local actors act in accordance with national policy objectives. This ‘principal-agent’ problem is a recurrent issue in the context of new forms of governance (also see below). Rather than solving this problem by strict rules and regulations, several national governments nowadays use (financial) incentives to control agents, for example by introducing performance indicators (proportion of benefit recipients enrolled in activation programmes, proportion of participants in activation (re)entering the labour market, etcetera). These indicators aim at steering the priorities local actors set, as well as at indirectly influencing decision making processes at local level regarding the nature of programmes, the target groups of activation etcetera (for an example, see the Swiss case analysed by Bonvin and Farvaque, 2007). Funding regimes are another example of how national governments try to steer local decision making. In the Netherlands, the budgeting of social assistance (since 2004, Dutch municipalities receive a fixed budget for social assistance expenditures) has made the reduction of the numbers of social assistance recipients priority number one, which has had an impact on the nature of activation programmes (more ‘Work First’ like programmes), the target groups of activation (emphasis on ‘easy to reintegrate’ clients), etcetera (Van Berkel, 2006)[2]. Thus, even though decentralisation does increase the room for local decision making and for tailoring activation programmes to local needs, the conditions under which decentralisation takes place may put considerable constraints on the decisions local actors make. Generally speaking, there is support for the aim to create opportunities for local actors to adapt policies to local circumstances. However, decentralisation also raises some issues that are more controversial. First, it may increase differences between municipalities and the treatment of unemployed citizens living in those municipalities. Of course, one could argue that this is an inherent and intended consequence of decentralisation. But this does not mean, that all intermunicipal (or interregional) differences with respect to activation and the treatment of unemployed people can be legitimated in terms of different local labour markets, different target groups, or differences in the situations and needs of people. They may also be the effect of, for example, local/regional political preferences or financial situations. More generally, decentralisation raises issues about the meaning of rights, equality for the law and citizenship, which increasingly are shaped by local/regional policy decisions. In order to counteract differences that are regarded as undesirable, national governments may decide to (re)centralise certain policy elements – Finland, for example, has introduced regulation in order to reduce differences between municipalities (Keskitalo, 2007). Secondly, decentralisation requires a lot of the capacities of local actors whose role and responsibilities in policy making and service provision increases significantly. In case of capacity deficiencies, local actors may resort to examples set by other municipalities or to the ‘good practices’ that have been identified or developed by, for example, expert agencies. However, whether these examples and practices are adequate approaches to the social problems experienced in specific local contexts and, thus, contribute to tailor-made solutions for these problems, remains to be seen. Marketisation and competition Another example of new forms of governance in welfare-to-work programmes is the introduction of sub-contracting, marketisation and competition in service provision, accompanied by dividing up the roles of service purchaser and service provider[3]. The idea behind this is that promoting competition will have a positive impact on the efficiency and effectiveness of services, on their quality and flexibility or responsiveness to local and individual needs, and on their price. Who acts as the purchaser, may differ. In some countries, the PES acts as coordinator of publicly funded employment services, without being any longer the (monopolist) direct provider of these services (Barbier et al., 2003). In others, benefit agencies are responsible for the activation of their clients and for purchasing services on the market. The fact that these institutions rather than the service users (i.e., unemployed people) act as purchasers is one of the reasons why activation markets are qualified as quasi-markets. Only in a very limited number of countries (for example in Germany and the Netherlands) and in specific cases do service users more or less act as purchasers, among others through vouchers, personal reintegration budgets or individual reintegration agreements (also see below; Sol and Westerveld, 2005). As far as the providers are concerned, they may involve (semi-)public institutions (for example the PES, that has to compete with other providers), private non-profit institutions and NGO’s (Southern Europe, the trade unions in Denmark), or private for-profit companies (the Netherlands). Activation markets are not a typical ‘liberal welfare state’ phenomenon, but occur in all types of welfare states, even though the US and Australia were the first to introduce this new form of governance. At the moment, it is the Netherlands (not a typical liberal welfare state) that has marketised activation services most radically – thereby underlining its ‘hybrid’ character. Of the various new forms of governance, marketisation has probably been discussed and investigated most. Research has made clear, that marketisation does not automatically result in the positive outcomes expected of it in terms of the quality, effectiveness etcetera of services. Some of the problems that the literature mentions are (Considine, 2001; Bredgaard and Larsen, 2005; Struyven and Steurs, 2005; Van Berkel and Van der Aa. 2005): - transaction costs may be high, depending, among others, on the number of providers, the knowledge providers have of the market, the degree to which purchasers try to steer providers by detailed calls for tender. Purchasers may try to reduce these costs, but that may have a negative impact on the quality of services; - in highly competitive markets, purchasers may be hesitant to take risks and to innovate, leading to standardised rather than tailor-made activation services. The same effect may occur, when contracts are granted for a relatively short period. This may discourage investments in service development, building networks with relevant actors, etcetera; - purchasers often do not have a clear idea of what kind of services are adequate for what types of clients groups. Evidence-based decision making is not particularly easy to realise in this area. That makes it, of course, difficult to decide on substantial grounds what providers tender the best service package; - mechanisms for dealing with the ‘principal-agent’ issue may have undesirable consequences. Bruttel (2005) distinguishes three mechanisms. First, incentive mechanisms such as no cure, no pay or no cure, less pay contracts. The definition of ‘cure’ is a crucial issue here, of course. For example, when contracts contain incentives to place participants in activation programmes in a job as soon as possible, the most vulnerable people are likely to receive no adequate support or no support at all. The second mechanism is information through benchmarking or monitoring. As Bruttel argues, the availability and nature of information is crucial if information is to avoid processes of parking and creaming. The final mechanism involves control through regulation. Evidently, increasing regulation may have consequences for the flexibility of service provision (and increase transaction costs, as was mentioned before). Thus, up until now the market of activation services has not lived up to its promises (see Bredgaard and Larsen, 2007). For some, this is the inevitable consequence of the fact that the market of activation services is an imperfect, quasi-market; for others, it is evidence that the market is an inadequate mode of coordinating the production of social services, that not only should be efficient and effective, but also contribute to social justice and reducing inequalities. The responses to these ‘market failures’ so far point towards less, rather than more market. For example, regulations are introduced to correct for failures of the market (e.g. Australia), a ‘preferred supplier’ system is introduced (which potentially makes purchaser-provider relationships less competitive and more cooperative), and purchasers are no longer obliged to purchase activation services from private companies (e.g., municipalities in the Netherlands). Networking and partnerships A third new form of governance refers to forms of cooperation through networking and partnerships. In a more narrow sense, this involves the cooperation between the PES and agencies responsible for the administration of benefits and social assistance (in countries where employment services and benefit administration are delivered by separate agencies). Cooperation between benefit and employment services agencies seems a logical consequence of the activation of welfare, as it fits with the aim to develop benefits into springboards towards labour-market participation, to prioritise work before income, and to make benefit entitlements conditional upon participation in activation efforts. European countries have introduced various strategies to promote cooperation between the agencies involved in benefit administration and activation. The most far-reaching strategies have been used in the UK, where the introduction of the Jobcentre Plus involved a merger between the agencies providing employment services and administering benefits; and in Norway, where the NAV – the Norwegian Labour and Welfare Organisation – is a merger of the National Insurance organisation, the National Employment Service, and the municipal Social Welfare System (http://www.nav.no/page?id=1073743655). Less far-reaching is the introduction of one-stop-shops, where benefit and employment service agencies are brought together to provide integrated services. The Dutch Centres for Work and Income are an example of this strategy. In Germany, where the Federal Employment Agency has traditionally been responsible for benefit administration and the provision of labour market programmes for the insured, the debate concentrated on how to provide integrated services to the long-term unemployed dependent on social assistance, nowadays Unemployment Benefit II. Most municipalities have introduced a one-stop-shop system (the so-called ARGE), where PES and local social agencies cooperate (Mosley, 2005; Tergeist and Grub, 2006). Evidently, these structural reforms tell little about what actually happens in the merged agencies or one-stop shops. The Dutch CWI experience shows that promoting cooperation and service integration is a long-term process during which many problems need to be solved (different cultures and histories, different management styles, different ways of servicing clients etcetera), which does not leave the service level and service quality unaffected. An evaluation of the early experiences of the UK Jobcentre Plus showed that although job entry outcomes have improved, the quality of benefit related services decreased (Karagiannaki, 2007). Cooperation and partnerships are not limited to PES and benefit agencies. They are also promoted to deal with complicated social problems, for example in cases where unemployed people are not only confronted with problems regarding labour-market participation, but also with problems concerning housing, drug addiction, mental health, etcetera; or to deal with problems of deprived urban or rural areas. Geddes and Benington argued that partnership is a concept “to promote integration both vertically (between different tiers of government – European, national, regional, local and grass roots) and horizontally (between different spheres of society – public, private, voluntary and civil)” (2001, p. 2). Partnerships do deal with problems of unemployment, but often use a wider approach to social problems, also focusing on poverty and social exclusion. More or less by nature, these partnerships are local, which contributes to their variety with respect to pursued objectives, actors involved, effects, etcetera. A study of partnerships carried out in the 1990s by the European Foundation for the Improvement of Living and Working Conditions (on which Geddes and Benington’s book cited above is based) mentioned several ways in which partnerships may contribute to alleviating problems of unemployment: through job creation, training, promoting local enterprise; through promoting improved and better targeted services; and through involving and empowering communities and excluded groups. The study identified several factors that contribute to building partnerships: an identification of the benefits gained by cooperation; strong leadership; skilled management and staff; strong shared local identity; active participation of all partners in strategic decisions and implementation; seeing new solutions to problems; obtaining new resources as well as maintaining a solid resource base; appropriate skills training and development (Geddes, 1998). Individualisation and contractualisation One of the ‘buzz words’ frequently used in debates on the provision of activation services is that these services should be individualised (or personalised), which generally means that individual circumstances, capacities and wishes should be taken into account when designing activation plans. In practically all policy documents on activation the need to individualise activation offers is emphasised. For example, the OECD Employment Outlook 2005 states in its editorial that “[p]roviding the right individualised services for displaced workers is part of the general challenge of designing effective employment services (…)” (OECD, 2005: 13), and identifies individual case management as one of the activation strategies with the largest impact. The importance attached to individualised activation services is echoed in EES documents as well: for example, the Employment guidelines for the period 2005-2008 recommend in guideline 19 “active and preventive labour market measures including early identification of needs, job search assistance, guiding and training as part of personalised action plans (…)” (EU, 2005: 25). At the same time, the meaning of ‘individualised service provision’ is all but clear. Several ‘individualisation discourses’ can be distinguished, each of which affects the provision of services to a greater or lesser extent and in different ways (see Van Berkel and Valkenburg, 2007). In the first individualisation discourse, individualising activation services means that social services should be de-standardised, differentiated, flexibilised and adapted to individual circumstances. As activation programmes deal with heterogeneous target groups, standard programmes aimed at broadly defined categories hamper effective activation, as they do not allow to take into account characteristics of the individual. Against this background, many countries introduced profiling procedures to map the individual circumstances and characteristics of clients, in order to fine-tune the process of decision making regarding activation offers. The second discourse puts individualisation in the context of the introduction of markets for the provision of public social services, and advocates that individual service users should enter (quasi-)markets of competing service providers as individual customers or consumers. As was mentioned before, markets of activation services are not only characterised by a split between purchasers and providers, but also between purchasers and users. Benefit or employment agencies buy activation services ‘on behalf of’ the actual users of activation services. In these circumstances, service users do not enter the market of activation services as customers who select their own services and service provider: they have little or no choice, control, and purchasing power (Wright, 2006). But there are some exceptions (see Sol and Westerveld, 2005). In Germany, clients can make use of placement vouchers to buy services on the market. In the Netherlands, the so-called Individual Reintegration Agreement (IRO) gives clients (mainly the insured) opportunities to develop their own reintegration trajectories and to select the service provider they prefer. Several municipalities experiment with so-called personal reintegration budgets for social assistance recipients. Outside the EU, we can point at the Ticket to Work in the US for people with disabilities. The third discourse on individualisation that we can distinguish emphasises that social entitlements should be granted on an individual basis, contingent upon the individual’s conduct, responsibilities and compliance with obligations. This discourse has had a very clear impact on the nature of the individualised provision of activation services. It is often accompanied by a process of ‘contractualisation’ of relationships between the unemployed and the state or, more specifically, benefit or social assistance agencies (Sol and Westerveld, 2005). Contractualisation has taken place in many countries, formalised in the form of individual action plans. The contractualisation tendency is related to attempts to make activation more tailor-made, and is simultaneously embedded frequently – although not necessarily – in a process of increasing the obligations of unemployed people as well as the conditionality of income support: the contract contains a ‘tailor-made’ definition of the responsibilities and obligations of the individual and facilitates the surveillance of his or her attitudes and behaviour and, subsequently, the evaluation of social security claims. Nowadays, in many countries (for example, the UK, Finland, Germany) clients are obliged to enter a ‘negotiation process’ with a personal adviser unless they are willing to accept sanctions or even benefit withdrawal – which for many is not a realistic option, of course. Clients have few institutionalised resources at their disposal to ensure that an activation offer is made to them that fits their needs and circumstances, or to force agencies to provide the services agreed upon in the contract. During the negotiation process itself, the main ‘weapon’ that the individual client disposes of is that personal advisers depend on their information and cooperation in order for activation to be effective. Because of this lack of ‘checks and balances’ in the activation process, Freedland and King (2005) – in a critical analysis of the British Jobseekers’ Agreement – argue that client contracts may tend towards ‘illiberal contractualism’. In their opinion, the justification of the decisions made and the sanctions imposed by personal advisers, is a critical issue here: when non-compliant behaviour of the client is automatically interpreted and dealt with as an offence against contractual agreements rather than a response to an inadequate intervention of the advisor, contractualism runs the risk of resulting in authoritarianism. Evidently, the question “who is in charge?” is a crucial one in the activation process in general and in negotiations on activation contracts specifically (Van Berkel and Valkenburg, 2007). This issue is also explicitly at stake in the final individualisation discourse we mention here. In this discourse, service users are seen as reflexive, competent citizens. Services should support them in realising their individual life projects, and individuals should be put in charge of the service provision process: they should be involved in defining the problems that activation should solve and the objectives it should realise, as well as in selecting the means through which this should be accomplished. Even though traces of this discourse are present in policy rhetoric (‘empowering the user vis-`a-vis service providers’, ‘taking the individual as starting point in service provision’, ‘putting customers in the driver’s seat’, etcetera), in most cases the client is not in charge in this sense. Reflections and perspectives The overview of the forms of new governance in activation policies has already indicated that the promises and expectations of this approach have not been met unequivocally, and that serious problems have come to the fore. In this concluding section we want to elucidate some of these problems and to offer possible directions for future research. The first point which in our opinion has been underexposed in the literature is the role of the social partners. The welfare state can, amongst others, be characterised by its involvement of employers and employees organisation in both the formulation of policies and the execution thereof, although this involvement has taken quite different forms in specific national contexts. One of the most striking reforms of the welfare states pertained to changes in this involvement. The power of these partners has been challenged as being an obstacle for the necessary liberation of market forces, while at the same time it is recognised that without these partners and their engagement social peace could be endangered. But the main course in the process of activating the welfare state has been to find means that could reduce or limit their influence in national policy making and execution of social policy. This has been for instance clearly the case in England, where the role of the unions has been changed massively, or in the Netherlands where the social partners have been pushed to marginal positions in the administration of social security because they where judged to use these in their own advantage in dealing with social problems of unemployment or illness. At the same time, the shift towards territorial decentralisation is motivated by the expectation that on the local level social dialogue can be restored. It is not quite clear why social partners can be more effective on the local level, when their role on the national level is contested or reduced. There are arguments for the thesis that influence on the local level is especially detrimental for the unions because without general agreements, local disputes are hard to win. This refers to the more general problem of the ways in which national representative organisations of employers and employees can reach new forms of agreement, an issue that cannot be discounted in the analysis of the administration of social services. In general, this issue of the social partners points to the question of which stakeholders are to be recognised as relevant and in what extent in the forms of new governance. On the local level, municipal authorities are definitively important players, but also the Chambers of Commerce or locally dominant businesses are having their say, and it may be very interesting to look into the dynamics between these actors and the established social partners in analysing the effects of decentralised activation. In this regard the interesting question is in what ways new governance, defined as a local/regional affair, is corresponding to democratic criteria of publicity, accountability and correction (see Flyvbjerg, ). Research into the process of which stakeholders are seen as legitimate partners and why in decentralised forms of governance is extremely useful when talking about the principal-agent problem: the outcome of money flows may be determined by the degree of the perceived adaptability of the different partners to the dominant policy paradigm. The issue of who are the stakeholders on the local level also poses the question in what way they can be effective. One of the hallmarks of the new forms of governance is territorial decentralisation. This is based on the assumption that the local or regional level is the best place to find solutions for imbalances in the labour market, and for reintegration. This assumption is, however, inconsistent with globalising developments that have indeed effect on local levels, but can not be overcome at these levels. Changes in the employment structure, the emergence of new qualifications and the rise of more flexibility and mobility in labour create territorial differences (for instance disadvantaged regions) that are transcending the influence of local agencies. More general still, it seems hardly credible that conditions of full employment can be created at the local level when these conditions are seen as difficult to realise at national levels. Problems that manifest themselves at the local level are thus not simply and automatically solved at this level. In this respect the question arises to what degree this move to the local level is a way of not only reducing welfare state expenditures, but also of evading the national task to provide satisfying conditions of living and working. A third problem refers to the status of the individual in relation to the forms of new governance. The most immediate observation is the ambivalent, not to say contradictory position of the user/client. At the one hand the user is the target of interventions aiming at participation in work, at the other hand (s)he is seen as the choosing consumer of these interventions, even as the director of these interventions. The interventions should be tailor-made which seems to imply steering by the user. But at the same time the user is supposed to fulfil obligations which make sure that (s)he is meeting the requirements of the activating programme. This ambivalence is crucial in the analysis of the rights and obligations which are ascribed to the user of activation programmes. Our overview has made clear that the implementation of activation is to be situated between two poles: the readiness of the individual to be integrated (to be enhanced by tailor-made solutions) and the obligations of the individual to be responsible for integration (to be enhanced by sanctions). This implies a redefinition of rights in terms of a conditionality regarding responsibilities and obligations for users. This shift in rights seems to be countered by the process of contractualisation, where users are treated as a ‘party’. But users are hardly an equal party vis-`a-vis the agencies, as they are the target of interventions which are based on the instrument of conditionality. In this sense the tendency to contractualism is reaffirming a shift away from rights to obligations. A fourth problem, which is connected to this new conditionality of rights, regards the social identity of the clients. The question of social identity comes up when income rights are replaced by or at least made conditional on activation programmes. This shift redefines the normative category of persons in risk and the way they are recognised as such. By stressing the importance of individual responsibility to grasp opportunities, the outcome of failing in this respect is almost directly blamed on the individual. Social security and labour market policies in terms of collective contributions and/or taxes in order to protect the less favoured imply the recognition of the ‘deserving poor’. Social security in terms of individual responsibility to reinsert oneself implies the recognition of the ‘successful poor’ and the rejection of the ‘new undeserving poor’. This different type of recognition is expressing a new norm for citizenship. The tendency is to grant social rights a provisory status: they have to be earned by showing satisfactory social motivation and behaviour. This would mean that identity as a worker is not only dependent on having work but also on taking up the individual responsibility to invest in qualifications which facilitate transitions in work and workplaces during the life course. This individualising outcome finds its continuation in the described new forms of governance, where the agencies have to evaluate the individual willingness to reintegrate, and to make decisions on the basis of this evaluation. The new forms of governance are thus not conceived as sustaining social or collective regulations for continuous qualification and mobility during the life course. Such a perspective would deal with the new dynamics of the labour market and processes of individualisation without placing the task for adaptation on individual workers or unemployed people. Defining social rights regarding work and qualification and connected risks in a collective perspective requires the production of a different social and political discourse, in which linkages between rights and resources would imply the formulation of flexible arrangements without loosing social security out of sight. In this paper, we have discussed the introduction of various new forms of governance in the provision of welfare-to-work programmes. We have argued that the study of service provision models should be an integral part of research into the practical nature, content and outcomes of activation services and their contribution to reconcile work and welfare. From a broader perspective – and following the argument developed by Denhardt and Denhardt in their analysis of service provision models – we tried to show that rather than being mere technical ways of ‘doing things’, new modes of governance may have far-reaching implications for citizenship and democracy. References Barbier, J.-P., Hansen, E. and Samorodov, A. (2003), Public-private partnerships in employment services, ILO, Geneva. Bonvin, J.-M. and Farvaque, N. (2007), A capability approach to individualised and tailor-made activation, in Van Berkel, R. and Valkenburg, B. (Eds), Making it personal. Individualising activation services in the EU, Policy Press, Bristol. Borghi, V. and Van Berkel, R. (2007), New modes of governance in Italy and the Netherlands. The case of activation policies, Public Administration, Vol 85 No 1, 83-101. Bredgaard, T. and Larsen, F. (forthcoming), Implementing public employment policy: What happens when non-public agencies take over?, International journal of sociology and social policy, Vol 27 No 7/8. Bredgaard, T. and Larsen, F. (2005), Employment policy from different angles, DJOeF, Copenhagen. Bruttel, , O. (2005), Contracting-out and governance mechanisms in the Public Employment Service, Berlin, WZB. Carmel, E. and Papadopoulos T. (2003), The new governance of social security in Britain, in Millar, J. (Ed.), Understanding social security: Issues for social policy and practice, Policy Press, Bristol. Clasen, J. and Clegg, D. (2006), Beyond activation: reforming European unemployment protection systems in post-industrial labour markets, European societies, Vol. 8 No. 4, 527-553. Considine, M. (2005), Enterprising states. The public management of Welfare-to-Work, Cambridge university press, Cambridge. Denhardt, R.B. and Denhardt, J.V. (2000), The new public service: serving rather than steering, Public Administration Review, Vol 60 No 6, 549-559. EU (2005) Council decision of 12 July 2005 on guidelines for the employ-ment policies of the Member States, Official journal of the European Union 6.8.2005, Brussels: EC. Finn, D. (2000), Welfare to work: the local dimension, Journal of European social policy, Vol 10 No 1, 43-57. Freedland, M. and King, D. (2005), Client contractualism between the Employment Service and jobseekers in the United Kingdom, in Sol, E. and Westerveld, M. (Eds), Contractualism in employment services, Kluwer, The Hague, 119-139. Geddes, M. (1998), Local partnership: a successful strategy for social cohesion? Office for official publications of the European communities, Luxembourg. Giguere, S. (2003), Local employment development, decentralisation, governance and the role of government, in Giguere, S., Higuchi, Y. and The Japan Institute for Labour Policy and Training (Eds), Local governance for promoting employment. Comparing the performance of Japan and seven countries, The Japan Institute for Labour Policy and Training, Tokyo. Gilbert, N. (2002), Transformation of the welfare state. The silent surrender of public responsibility, Oxford University Press, Oxford. Handler, J. (2004), Social citizenship and workfare in the United States and Western Europe. The paradox of inclusion, Cambridge University Press, Cambridge. Henman, P. and Fenger, M. (Eds) (2006), Administering welfare reform. International transformation in welfare governance, Policy Press, Bristol. Karagiannaki, E. (2007), Exploring the effects of integrated benefit systems and active labour market policies: evidence from Jobcentre Plus in the UK, paper presented at 5^th ISSA conference: Social security and the labour market: a mismatch? Warsaw, March 5-7. Keskitalo, E. (2007), Between universal policy and individualised practice: analysing activation policy in Finland, in Van Berkel, R. and Valkenburg, B. (Eds), Making it personal. Individualising activation services in the EU, Policy Press, Bristol. Lo/demel, I. and Trickey, H. (Eds) (2001), An offer you can't refuse: Workfare in international perspective, The Policy Press, Bristol. Martin, S. (2001), Implementing ‘Best Value’: local public services in transition, Public administration, Vol 78 No 1, 209-227. Mosley, H. (2005), Job Centers for local employment promotion in Germany, in Giguere, S., Higuchi, Y. and The Japan Institute for Labour Policy and Training (Eds), Local governance for promoting employment. Comparing the performance of Japan and seven countries, The Japan Institute for Labour Policy and Training, Tokyo. Newman, J. (2001), Modernising governance. New Labour, policy and society, Sage, London. Newman, J. (Ed.) (2005), Remaking governance. Peoples, politics and the public sphere, Policy Press, Bristol. OECD (2005) Employment outlook 2005, OECD, Paris. OECD (2003), Managing decentralisation. A new role for labour market policy, OECD, Paris. Osborne, D. and Gaebler, T. (1993), Reinventing government. How the entrepreneurial spirit is transforming the public sector, Penguin, London/New York. Serrano Pascual, A. (Ed.) (2004), Are activation policies converging in Europe? The European employment strategy for young people, ETUI, Brussels. Serrano Pascual, A. and Magnusson, L. (Eds) (2007), Reshaping welfare states and activation regimes in Europe, Peter Lang, Brussels. Sol, E. and Westerveld, M. (Eds) (2005), Contractualism in employment services. A new form of welfare state governance, Kluwer, The Hague. Struyven, L. and Steurs, G. (2005), Design and redesign of a quasi-market for the reintegration of jobseekers: empirical evidence from Australia and the Netherlands, Journal of European social policy, Vol 15 No 3, 211-231. Sultana, R. and Watts, A. (2006), Career guidance in Public Employment Services across Europe, International journal for educational and vocational guidance, No 6, 29-46. Tergeist, P. and Grubb, D. (2006), Activation strategies and the performance of Employment Services in Germany, the Netherlands and the United Kingdom, OECD, Paris. Thuy, P., Hansen, E. and Price, D. (2001), The public employment service in a changing labour market, ILO, Geneva. Van Berkel, R. (2006), The decentralization of Social Assistance in the Netherlands, International journal of sociology and social policy, Vol 26 No 1-2, 20-32. Van Berkel, R. and Valkenburg, B. (2007), Making it personal. Individualising activation services in the EU, Policy Press, Bristol. Van Berkel, R. and Van der Aa, P. (2005), The marketisation of activation services: a modern panacea? Some lessons from the Dutch experience, Journal of European social policy, Vol 15 No 4, 329-345. Van Berkel, R. and Hornemann Mo/ller, I. (2002), Active social policies in the EU. Inclusion through participation? Policy Press, Bristol. Visser, J. and Kouw, M. (2006), Ensuring effective implementation through better governance. Synthesis report, Mutual learning programme EES, Brussels. Wright, S. (2006) The administration of transformation: a case study of imple-menting welfare reform in the UK, in Henman, P. and Fenger, M. (Eds), Administering welfare reform. International transformations in welfare governance, Policy Press, Brussels, 161-183. ------------------------------- [1] This is the core topic of a recently established international researchers’ network, Reform of Employment Services Quorum (RESQ). [2] Although our focus here has been on territorial decentralisation, similar problems occur (and similar solutions are sought) where functional decentralisation is concerned. [3] As Sultana and Watts (2004) argue, sub-contracting does not always have to involve competition between providers.