Abstract
Many governments have tightened the link between welfare and work by attaching conditionality to out-of-work benefits, extending these requirements to new client groups and imposing market competition and greater managerial control in service delivery – principles typically characterized as ‘workfare’. Based on field research in Seine-Saint-Denis, we examine French ‘insertion’ schemes aimed at disadvantaged but potentially job-ready clients, characterized by weak conditionality, low marketization, strong professional autonomy and local network control. We show that insertion systems have resisted policy attempts to expand workfare-derived principles, reflecting street-level actors’ belief in the key advantages of the former over the latter. In contrast with arguments stressing institutional and cultural stickiness, our explanation for this resistance thus highlights the decentralized network governance of front-line services and the limits to central government power.
Introduction
Under post-Fordist capitalism, governments worldwide have tightened the link between benefits and work. Entitlements have been reduced, conditions increased, and non-compliance punished. Quantitative performance management techniques and private contracting have been deployed to handle the resulting administrative challenges, and services for the poor joined to the benefits system in various ways. This repurposing of the welfare state as a market-oriented and market-mediated labour-market intervention has been labelled workfare (Jessop, 1999; Peck, 2002; Soss et al., 2011) or the work-first welfare state.
How widely have these principles spread? Responses to this question frequently follow Esping-Andersen (1990), arguing that workfare is specific to the ‘liberal’ regimes of Anglophone countries. Continental models are held to differ, maintaining insurance-based entitlements for labour-market ‘insiders’ while relegating ‘outsiders’ to minimal tax-funded social assistance (Emmenegger et al., 2012). Others highlight convergence, including retrenchment of German unemployment entitlements under the Hartz reforms (Dörre et al., 2013), the international spread of conditionality in jobless benefits (Venn, 2012) and their extension to groups considered ‘inactive’ (Eichhorst and Konle-Seidl, 2008).
In France, workfare-derived principles have had comparatively weak traction. For labour-market ‘outsiders’, a new entitlement has coevolved with social benefits, namely, to ‘insertion’ services, both ‘professional’ and ‘social’ (Paugam, 1993). These services also include many potentially job-ready clients outside the benefits system, such as young unemployed people. The landscape of providers, policymakers and funders is comparatively non-marketized, even where work is contracted out. Front-line workers have more labour process autonomy, and providers are often non-profit associations with an ethos of addressing needs beyond job outcomes (Divay, 2008; Fretel, 2012; Morel, 2000). Local actors control employment services which are only weakly integrated with national public employment services (PES) and the benefits system (Berthet et al., 2016). The first contribution of this article is to describe insertion in practice, with a particular focus on its implementation through ‘street-level’ actors (Brodkin, 2011; Lipsky, 1980), and the advantages they perceived in it compared to workfare.
The second contribution is to explain why France has not extended workfarist principles to insertion provision. In the Parisian suburb of Seine-Saint-Denis, this is particularly puzzling, since the area is often invoked in the ‘moral panic’ discourse of social dysfunction in the banlieues (Wacquant, 2008). Comparative social science usually explains France’s peculiarities as due to its Bismarckian welfare state (Palier and Thelen, 2010), social dialogue traditions (Clegg and Van Wijnbergen, 2011) or nationally distinct ideas and discourses (Barbier, 2002). In contrast, we emphasize the empowerment of local funder and provider networks due to decentralizing public sector reforms dating back to the 1960s, which has made it difficult for workfarist ideas to be implemented in practice (Bezes and Le Lidec, 2011; Nay, 2002). This empowerment of local employment service providers is reinforced by the weakness of central policy tools, notably the fragmentation of services and the low coverage of out-of-work benefits for under-25s (Bargain and Vicard, 2014) which takes them out of the orbit of the PES.
The next section examines social policy and public administration literatures, before a description of our research methods. The following sections examine the ethos and local network embeddedness of insertion, identifying various characteristics which are incompatible with workfare principles but which cement political support for the insertion approach. The conclusion considers implications for policymakers seeking alternatives to the work-first welfare state and for comparative research.
Workfare and insertion
The American discussion over workfare started in the 1970s and produced practices that diffused internationally. By introducing benefit conditionality, reducing entitlements, and expanding enforcement apparatuses, policymakers encouraged the unemployed to accept jobs on offer. Workers would be required to take any job offered, even unpaid ones, as in work-for-benefit schemes (Krinsky, 2008). In Germany, status-securing benefits retreated, since requirements applied to lower paid and lower skilled jobs (Dörre et al., 2013); long-term unemployed were merged with social assistance recipients into a heterogeneous group receiving a means-tested benefit (Barbier and Knuth, 2011). In Britain, work requirements have encompassed a widening population since the 1990s, including young people, lone parents, disabled people, and other long-term unemployed (Wiggan, 2015). Promising more employment and less benefits spending, workfare-inspired activation schemes have spread, despite weak evaluation evidence (Peck, 2002).
Insertion schemes, though they assist clients into work, have a different goal. The Revenu Minimum d’Insertion (RMI) was introduced in 1988 to provide a minimum income, enhance access to social services and facilitate integration among the ‘socially excluded’ (Morel, 2000). However, while insertion schemes apply to benefits recipients, they include those outside the benefits system, including most jobless under-25s. There are also differences in the design and implementation of schemes. Compulsion is weak and rarely enforced (Clegg and Palier, 2014), often meeting staff reluctance (Lavitry, 2009). Indeed, for clients outside the benefits system, the main tools of compulsion – the jobseeker’s agreement and threat of sanction – are removed. The desired outcome is a permanent job subject to social insurance contributions, preferably permanent, sometimes subsidized through ‘contrats aidés’ (Caroli et al., 2008).
France has embraced this policy orthodoxy in some areas: mandatory jobseeker agreements in 2001, contracting with for-profit firms from 2005, monthly monitoring meetings in 2006, merging benefit payments and job-search assistance into Pôle emploi (PE) in 2008 and emphasizing ‘activation’ over insertion with the shift from RMI to Revenu de Solidarité Active (RSA) in 2009. In 2010, RMI/RSA coverage was extended to certain under-25s, potentially facilitating closer articulation between insertion schemes and the benefits system. Increasing monitoring and conditionality is in step with French public opinion. While the International Social Survey Programme showed that French people oppose high levels of inequality (Larsen, 2016), 35.2 percent of French respondents agreed (or strongly agreed) that ‘The government should spend less on benefits for the poor’, more than any other European country (own calculation). As we show, however, insertion services remain resilient.
Jessop (1999) and Peck (2002) argue that workfare principles are partly defined by their implementation and governance: namely, selective marketization. Services for the unemployed attract attention from social policy and public administration writers because of governments’ use of contracts with service providers and welfare recipients (Sol and Westerveld, 2005) and competition between providers (Bredgaard and Larsen, 2008). This reflects a broader trend joining employment services more closely to the benefits system by merging benefits agencies with employment services. In extreme cases, providers are paid mainly or entirely by results, as in Britain’s Work Programme and Germany’s placement vouchers. Jantz et al. (2015) classify such arrangements as ‘market accountability’, where principles of competition, contracts and performance dominate. An effective service is one that delivers large numbers of job placements, as defined in contracts between funder and provider.
In France, governments have not imposed market relations on such a scale. While central government has experimented with marketization, it has not managed to roll it out nationwide. Berthet et al. (2016) find very limited ‘vertical’ integration of services reflecting power struggles between national bodies (such as the PE) and local bodies (who oversee RSA allocation); indeed, the latter can choose not to use PE to provide services for RSA claimants. A check on central power is thus built into the benefits system, reinforced by the European Social Fund subsidiarity requirements to share power with local and regional government (Nay, 2002). The result resembles what Jantz et al. (2015) call ‘network accountability’: where top-down accountability to policy-setting agencies is limited, and where evaluation and provider reputation are determined by local peer groups. Here, definitions of effectiveness more often imply hard-to-measure aspects of clients’ personal progress or simple participation in the programme.
These differences affect service delivery. The pressure of workfarist social policy has encouraged the speeding up and degradation of services, leading to worse jobs on the front line (Esbenshade et al., 2016). Expanded competition has disrupted local network relations (Hipp and Warner, 2008) and made non-commercial providers’ survival increasingly uncertain (Gallet, 2016). Tight management control, intensive quantitative measurement and performance management reduce the quality and quantity of staff–client interactions (Baines, 2004).
By contrast, in insertion, the assistance of unemployed people (‘accompagnement’) is anchored in an evolving set of occupations (Fretel, 2013). Their ethos emphasizes job placement as a form of professional social work, entailing counselling and support offered to voluntary participants (Divay, 2008). Standardization and central control have increased in some areas, particularly in PE and its for-profit contractors (Divay, 2009; Lavitry, 2009). But this ethos, we will show, remains intact among street-level insertion actors.
Marketization tends to encourage for-profit commercial providers; in Britain, the latter have taken over large swathes of provision and even manage the flow of central government funds and clients to non-profit and municipal agencies (Rees et al., 2014). In such systems, providers generally prioritize job-search motivation of clients, de-emphasizing more complex questions such as skills or health challenges (Friedli and Stearn, 2015). Through ‘creaming and parking’, firms reap outcome payments by focusing staff time on the job-ready and avoiding expenditure on clients deemed distant from the labour market (Rees et al., 2014).
Insertion, by contrast, remains dominated by the non-profit associations that evolved in tandem with the 20th-century welfare state (Divay, 2008; Fretel, 2007), and which operate as part of broader local networks devising and implementing social policy (Nay, 2002). Like Considine and Lewis’s (2003) ‘networkers’, they prioritize ‘negotiating with others, moving between organizations, developing new solutions to problems, or brokering deals between different parts of the public and private service-delivery systems’ (p. 138). Parking violates underlying social policy, since insertion is a citizen’s entitlement, underpinned by statute (Paugam, 1993).
Why is France different?
Insertion in practice has resisted the extension of workfarist principles, despite seemingly work-first reforms introducing jobseekers’ agreements, the PE and RSA. Coercive levers over clients are weaker (in part owing to reduced benefits coverage), as is the workfarist market structure that constrains the labour process and discourages spending on complex client needs. Why have the principles of workfare not spread to insertion services?
One explanation concerns the supposed resilience of ‘conservative’ Bismarckian welfare states (Esping-Andersen, 1990), where contributions-based entitlements manage risk within classes rather than redistributing between them. Hence, the British and American attack on welfare states attracted middle-class support which was not politically possible in France (Prasad, 2005: 390–392). Where retrenchment does take place, resources focus on people with ‘strong labour-market attachment’, neglecting ‘outsiders’ (Emmenegger et al., 2012). Indeed, in highly dualized systems, where social assistance eligibility is restricted, people outside the benefits system cannot be subject to conditionality and sanctioning.
There are two problems with this argument. First, it does not explain why some continental welfare states, like Germany, are vulnerable to retrenchment (Barbier and Knuth, 2011). Second, it does not predict the rise of an entitlement for insertion services for labour-market ‘outsiders’ as in France, especially for clients entirely outside the benefits system. Hence, the relevance of this argument to front-line insertion provision, where workers enact the citizen entitlements of outsiders, is not obvious.
Another argument is that under Bismarckian systems, insider–outsider inequalities are reproduced by the desire to avoid conflict with trade unions that represent insiders (Palier and Thelen, 2010). Unions, alongside employers, influence policy through social dialogue and institutional participation, and have generally supported RMI, job-search requirements and individual jobseeker contracts (Clegg and Van Wijnbergen, 2011: 338–340). However, unions have little role in the working of insertion; in the programmes we examine, they were present mainly as staff representatives.
A third explanation emphasizes differences in social norms and discourses. Certainly, there are challenges in translating policy concepts between national concepts (Barbier, 2002), and the republican principle of solidarité is acknowledged across the French political spectrum (Beland and Hansen, 2000). Interviewees echoed these arguments, often believing that an ‘Anglo-Saxon’ system would contradict the French concept of citizenship. But it leaves an important question unanswered. Why in local administrative practice does a discourse of republican ideals trump that of moral panic, when the latter is arguably the dominant tone of French political discourse about this clientele (Wacquant, 2008)?
Our explanation emphasizes the need for top-down pressure from central government, in order to enforce workfare principles. Indeed, street-level bureaucracy literature shows that front-line workers’ discretion cannot easily be overridden and is manifested in ways that policymakers and top managers cannot predict (Brodkin, 2011; Lipsky, 1980). Moreover, governance literature shows that there will be very different providers operating under ‘network accountability’ from those operating under ‘market accountability’ (Jantz et al., 2015). Under the former, staff pursue shared norms of what constitutes fair behaviour vis-à-vis service users (Hupe and Buffat, 2014; Maynard-Moody and Musheno, 2000), which likely diverge from quantitative job-outcome targets of work-first services. We therefore expect to find French insertion providers acting in ways that preserve insertion norms and obstruct marketization and workfarist principles.
Because such resistance is not unique to France, it is common internationally for central governments to use their authority to impose work-first principles, using tools of marketization and performance management (Greer et al., 2017). In French employment services, however, central power is fragmented and integration between levels of government is relatively weak (Berthet et al., 2016), while the importance of local and regional governance increases (Bezes and Le Lidec, 2011; Nay, 2002). As shown below, this is reinforced by the benefit system’s limited coverage and the resulting independence of employment services targeting ‘outsiders’. Hence, we can expect strict limits to the French government’s attempts to impose workfarist principles on the front line of employment services.
Methods
We chose the département of Seine-Saint-Denis partly because it is a well-known and large area of relative deprivation, with generally weak socio-economic indicators (see Table 1) and elevated social need; 600,000 of its residents live in neighbourhoods targeted for funding as ‘quartiers prioritaires de la politique de ville’ (Kanner, 2015); it currently has 66 urban regeneration projects approved and just under €1.5b in subsidies from the Agence Nationale pour la Rénovation Urbaine (ANRU), more than any other département outside of Ile-de-France (ANRU, 2015). It also plans in 2016–2017 to spend €8.7m in the European Union (EU)–subsidized Initiative pour l’emploi des jeunes targeted at regions and départements with youth unemployment above 25 percent (Anonymous, 2015).
Seine-Saint-Denis demographics, 2012.
Source: INSEE (2015) except aANRU (2015).
INSEE: Institut national de la statistique et des études économiques; ANRU: Agence Nationale pour la Rénovation Urbaine.
We utilize qualitative interviews with 34 informants conducted between May 2011 and April 2015. Interviewees came from various organizations across Seine-Saint-Denis, encompassing workers at various nodes in the insertion network, including public funders, non-profit associations and some for-profit providers (Table 2).
Organizations and interviewees.
MEDEF: Mouvement des entreprises de France.
We contacted senior figures in various organizations, who acted as key informants, granting access to further management and front-line interviews. Initial contacts were made through cold calling and unannounced office visits. Interviews were all recorded, transcribed, annotated and coded using MaxQDA. We have anonymized establishments and interviewees. Potentially recognizable actors such as the Mouvement des entreprises de France (MEDEF) and the Missions Locales all have multiple locations in Seine-Saint-Denis.
Data were gathered in two waves. The first (2011–2012) encompassed public funders, researchers, non-profit associations and for-profit service providers, exploring the overall landscape of policy and provision. Major interview subjects were the practice of insertion delivery, policy moves to encourage greater private involvement and relationships with service users. The analysis of these interviews informed the second wave’s (2014–2015) themes; here, we examined more closely the funding arrangements and relationships between funders and providers, the labour process of front-line staff and the counsellor–client relationship.
The next sections describe the ethos underpinning insertion, its embeddedness in local networks of funders and providers, and the perceived advantages which lead street-level actors to resist workfarist shifts.
Insertion services: networks rather than markets
French insertion providers have become increasingly assertive nationally, revealing policy tensions over the role of for-profit provision and service-delivery models. Insertion organizations have forged stronger collaborative links, ranging from informal information-sharing networks to, occasionally, pooling resources and liaising over pricing (Inspection Générale des Affaires Sociaux (IGAS), 2015). New advocacy institutions have also emerged, such as the Mouvement des Entrepreneurs Sociaux (MOUVES), founded in 2010. MOUVES has lobbied local and national policymakers on behalf of for-profit social providers, advocating greater plurality of provision in French social services, including insertion. One of its constituent parts is the large association Groupe SOS, which has expanded rapidly through takeovers of hundreds of smaller non-profit entities, as well as some for-profit enterprises. Its founder has been a high-profile advocate of greater plurality in social provision, and a critic of the ‘small is beautiful’ ethos in French social services (Vialle, 2013). Such arguments remain at odds with the views of many within local insertion networks, which may be structured around highly niche services often provided by small non-profits.
Accordingly, other advocacy groups have emerged taking a different line, including the Fédération Nationale des Unions Régionales des Organismes de Formation, founded in 2000, which supports greater local embeddedness among employment service providers and criticized ‘opportunistic’ aspects of for-profit provision (Remy and Thiébaud, 2015). Two interviewees worked for a national non-profit advocacy network, and criticized the more ‘managerial’ approach they perceived to be becoming more established nationally. Instead, they emphasized locally embedded ‘initiative associative’, worrying that this would be undermined by the expansion of contracting and for-profit provision (staff interview, umbrella association B). These arguments are encapsulated in local-level debates in Seine-Saint-Denis.
Our participant organizations focus on specific user groups including disadvantaged young people, lone parents or recovering drug users, counselling them on moving ‘closer to work’. Many deal primarily with young people outside of the benefits system, while others are mandated to service growing numbers of RSA recipients. It is local départements which impose these requirements; they administer RSA, and are free to refer recipients independently of PE. Conversely, most, but not all, clients are registered with PE, which may direct clients with specific needs to particular insertion providers independently of local authorities. Moreover, clients are also referred through informal networks, including other social services, word-of-mouth and street worker outreach (staff interviews, Mission Locale A and for-profit provider). Thus, users come from various separate sources rather than a single agency under central mandate.
Clients receive intensive counselling, the first step being an analysis of needs by front-line staff (staff interviews, non-profit providers C and B, and Mission Locale A) in order to develop an individual projet professionnel that defines a developmental target and pathway (such as a return to formal education, an internship or a placement application). Users may be oriented to additional social services, either to address specific needs or because they have progressed to a new stage in their projet professionnel. For instance, non-profit provider B receives young people considered ‘far from work’, and focuses on inculcating habits such as punctuality and professional appearance. By contrast, non-profit provider C supports the job search of those who are work-ready but marginalized due to social problems. A key tool is alternance, work experience in subsidized employment supported by an insertion provider (staff interviews, Mission Locale A). Interviewees saw alternance as valuable not only for experience and skills, but in changing employer perceptions of service users (MEDEF interview).
Front-line workers and managers in insertion commonly have career backgrounds in the sector, usually requiring particular internships or professional qualifications, though many had also worked previously as schoolteachers (staff interviews, Mission Locale A and non-profit provider B). While non-profit provider C was an exception in prizing private sector experience among its staff, it also sought shared social values and a long-term commitment to social work (staff and management interviews). Interviewees across our sample emphasized the importance of personal connection with service users, perceiving this as incompatible with the increased speed and sanction-oriented approach of workfare.
Unlike in many workfare systems where long-existing non-profit provider networks have been weakened (Aiken and Bode, 2009; Gallet, 2016; Hipp and Warner, 2008), insertion remains dominated by structures that are deeply embedded in local social service networks. Their specific roles had evolved alongside para-statal services in areas like housing, health and childcare. Associations have not been marginalized or usurped by public funding policies; budgetary pressures are more likely to reflect accelerating need (management interview, Mission Locale B) or expanding wage bills owing to low staff turnover and pay progression (management interview, non-profit provider C) than funding cuts.
Local insertion networks thus feature a precise division of functions and strong collaborative ties. Front-line workers themselves tend to have well-developed networks of contacts across wider social services with whom they collaborate. These ties tend to be quite informal. At non-profit provider B, most incoming clients are referred by ‘partenaires’. This could mean the Mission Locale or another association, who decides that a client would benefit from the association’s services. When staff consider that users are ready to progress, they refer them to others in turn. MEDEF is also in this network, assisting the Missions Locales with finding placements, channelling additional funds from regional actors. These relationships are termed partenariats, but they are based on meetings and discussions between staff at each organization rather than contracting:
We can go to a meeting with partners to explain our activity, to use an opinion on our pedagogical offering. So, in fact, the partnership is created through meetings … there’s nothing really formalized at written level … We use the word partenariat in the sense that we are in permanent contact with the prescripteurs for the follow-ups and advance of the method. (Staff interview, non-profit provider B)
Funding arrangements in Seine-Saint-Denis sustain these collaborative ties, encouraging stable relationships between authorities and incumbent organizations. The Missions Locales’ role as lynchpins of insertion services for disadvantaged young people is enshrined in its status within the Service Public pour l’Emploi, a prerequisite for a contrat de cotraitance with PE. Our interviewees distinguish cotraitance from sous-traitance (literally subcontracting) as implying a much closer relationship between provider and funder (interview, Public Functions A and B); this status prioritizes the Missions Locales over potential competitors. Mission Locale B could expect to receive 200 users annually, with payment per user received, rather than payment by results for job outcomes. It supplements this stream with project-based funding to organize one-off events such as careers fairs. While the latter are tendered competitively via appels à projet, organizations with public service status were prioritized. Regional funders tended to allocate funds on a relatively egalitarian basis to avoid the perception of inequalities in provision (manager, Mission Locale B).
Respondents in associations also reported non-competitive and predictable funding systems. Non-profit provider B secures funding through responding to appels à projet issued by local authorities. Unlike calls for tender (appels d’offre), which imply a detailed buyer-defined specification, these involve organizations proposing projects derived from their analysis of local needs. Because funding decisions therefore draw on providers’ own local embedding, incumbent organizations are well placed to retain funding. One manager described feeling ‘more or less sure that funding will be renewed’ (management interview).
Once again, annual funding settlements specify a certain number of users to be seen, without demanding specific outcomes. Consequently, organizations can plan throughout the year to meet minimum user thresholds, drawing on informal partenariats in case of difficulty:
[In September] … there were still 30 young people to receive before 31st December … So, we get back to the Missions Locales, saying ‘Listen, we still have a place, if you have young people to send to us, don’t hesitate’. No, it’s not a pressure … we watch and we know where we are with [the numbers]. (Staff interview, non-profit provider B) Public funders are often sympathetic to qualitative justifications for why these thresholds may have been missed. (Management interview, non-profit provider B)
At non-profit provider C, funder criteria included requirements that 50 percent of annual users must be RSA recipients. This request was an addition to a highly stable preexisting agreement with funders, rather than being disseminated through a competitive call for tenders (management interview). Thus, despite policy attempts to funnel more benefits recipients into insertion, service provision evolved through the embellishment of existing buyer–provider relationships, emphasizing initiative associative over buyer-defined calls for tender.
In sum, insertion is characterized by networks of highly specific and locally adapted non-profit providers. The relationship between nodes in this network is informal and collaborative, something encouraged by the approach of regional public funders who generally avoid payment-by-results and competitive calls for tender. Funders rely on local expertise and ongoing consultation with providers, precluding a more centralized and arm’s-length buyer–seller relationship.
Next, we identify four characteristics of this strong network accountability (Jantz et al., 2015) which are perceived as advantageous by street-level actors and which thus tend to militate against the imposition of workfarist methods. These are greater local adaptiveness on the part of providers, greater professional autonomy among staff, greater emphasis on client agency and barriers to methods such as ‘creaming and parking’ associated with profit extraction. We also highlight the recent contrat d’autonomie (CdA) as an example of centrally imposed efforts to draw workfare-influenced ideas into the insertion landscape, discussing reasons for its failure.
Organizational adaptation to local needs
Insertion providers have far more autonomy in defining their own interventions than is typical in workfare systems. Their functions are precisely defined, addressing specific and contingent needs, as with non-profit provider C, founded by a coalition of 15 local groups who had repeatedly struggled to get their users into permanent work. It seeks to act simultaneously as counsellor to its users and as human resources service provider to local employers. Counsellors match users with appropriate permanent jobs and facilitate introductions between suitable users (50% of whom are required to be RSA recipients) and employers. They then provide follow-up meetings to support users once placed. They present themselves to employers as a unique service provider that can widen recruitment, find tailored candidates and support candidates on the job. This involves ‘prospecting’, that is, searching out potential positions and building links with employers, before service users themselves are suggested. In this sense, counsellors saw themselves as possessing highly distinctive advantages: the capacity to act as a specialized consultant towards providing a bespoke service to employers, breaking the stereotype of the insertion user relying on employers’ goodwill to be given a low-end job (manager and staff interviews, non-profit provider C).
Its ‘prospecting’ function requires this provider to buck the French trend of standardizing professional social care qualifications, by recruiting workers with commercial experience (staff and management interviews). With defined quantities of annual users and no payment-by-results, this association had rejected requests from for-profit training organizations to place larger volumes of clients; our management interviewee saw this as threatening the quality of a bespoke service.
Similarly, at non-profit provider B, where clients are further from employment, funders specify broad orientations, such as enhancing clients’ professional presentation or willingness to travel (management interview). Within these guidelines, they have significant autonomy to organize workshops on various topics (including music, fashion, cooking, budgeting, among other things) aimed at engaging users. Counsellors here saw this approach as critical to generate buy-in from clients (manager and staff interviews).
The lack of standardized buyer criteria is also evidenced in clients’ progression through insertion networks. Individual counsellors, typically liaising with partenaires and the user, make judgements about when and to where a user should move, based on an assessment of their needs and development (staff interviews, non-profit provider B). These characteristics reflect the absence of payment-by-results mechanisms. We argue that ‘market-accountable’ workfare systems will likely be unable to produce organizational models and programmes that adapt and coevolve so closely with complex and contingent local needs.
Professional autonomy and initiative
Staff respondents provided a highly tailored service, typically running meetings of open-ended length with users in order to accurately ascertain their needs (staff interviews, Mission Locale B). They make qualitative evaluations regarding when and where clients should be progressed, and cultivated their own networks to this end. At non-profit provider B, staff had full autonomy over the workshops they ran, with the caveat that they were required to emphasize punctuality and professionalism among users (staff interviews). At non-profit provider C, staff employed autonomous methods of ‘prospecting’ to create networks of potential employers. Interviewees stressed the need for individual problem-solving and creativity, as with one counsellor who had arranged with another local association to hire their vehicles at low cost, helping users to overcome the inadequacy of public transportation in the banlieues. The interviewee (staff interview 2, non-profit provider C) saw this as exemplifying a key advantage of their model: discretion in problem-solving.
Administrative reporting obligations had generally increased, particularly following tighter public budgets. While this increased staff data-entry time, their effect on the front-line labour process was cushioned. Since funders tended to specify what kinds of people should be seen, rather than demanding specific outcomes, administrative obligations usually entailed maintaining descriptive records of clients’ demographic characteristics (including benefits status) rather than performance evaluation (staff interviews, non-profit providers B and C). Moreover, objectives imposed on organizations tended not to translate into individualized targets. Annual meetings with managers evidently had little connection with salaries or job security. At Mission Locale B, staff were sometimes pushed to assign users to specific events to justify funding, but there was no sense of targets that individuals could be penalized for missing (staff interviews). This evidently also reflects the relative fragmentation of welfare-to-work governance. While local authorities are responsible for imposing certain targets, such as the quotient of RSA recipients that are seen, this is established independently from PE. Thus, front-line staff are comparatively insulated from the logic of targetization and managerialism which has gained ground within PE itself.
Work intensification for staff either reflected their own professional ethics (‘the pressure we have is not so much about results. It’s more ourselves in our capacity as counsellors, we put ourselves under pressure because we want it to work … We say to ourselves, this person really has to find a solution’) (staff interview 2, non-profit provider B), or else the demands of the entire insertion delivery chain. Workshops fill up quickly, and counsellors may struggle to place users in training events run by for-profit companies, particularly in a context of high unemployment where welfare recipients are stigmatized. In insertion, staff initiative can be directed towards countering such problems, contra the routinized processes implied in workfare, where performance measurement creates incentives to neglect more complex problems (Greer et al., 2017). This was perceived as a significant advantage by managers and staff across all participating organizations.
Emphasis on the needs and agency of clients
Staff respondents greatly valued the autonomous agency of clients, and saw welfare-to-work systems as unworkable without them:
It’s really important that it comes from them … we can propose to them interesting things, interesting training, interesting projects, but if they don’t want it, it won’t work … the accompagnement won’t be convincing, neither for us nor for them. (Staff interview, Mission Locale B)
There are no sanctions for missing meetings. At non-profit provider B, the only obligation imposed on users was that they should attend each available workshop at least once. Counsellors needed to maintain the interest of users, and wanted them to progress before boredom or frustration set in.
This clearly also reflects the weaker articulation between insertion and benefit systems compared to work-first services in other countries. Many younger insertion clients do not receive benefits, removing a potential coercive lever. Staff distaste for sanctions-centric approaches is thus reinforced by fragmentation in the overall governance of insertion, since those that provide specialized employment services do not generally have the capacity to administer punitive financial measures (staff interview, Mission Locale B). However, it also reflects the wider assumption that user needs can be recognized and addressed, rather than problematized as an obstacle to payment. In this sense, the weak insertion-benefits articulation provides a context in which non-marketized local funder–provider networks can continue with comparatively weak disruption.
Insertion staff therefore spend longer on clients’ needs assessments, allowing insight into the barriers that stop them finding jobs. These barriers are similar to those affecting ‘hard to place’ clients in workfare schemes: lack of formal qualifications, weakness in ‘soft’ labour-market requirements, racial discrimination, poor housing, childcare obligations or poverty. But insertion schemes, although not significantly better resourced than providers in workfare systems, are more at liberty to address these barriers.
Participants emphasized the human relationship between staff and clients, and the sense of a shared project. The comparative labour process autonomy of front-line staff facilitates this. Many staff emphasized a high degree of intrinsic motivation in providing good-quality social services (staff interviews, Missions Locales A and B; public function B). Participants unanimously argued that for-profit work would provide incentives that would be opposed to the aims of insertion.
Non-profit provider C emphasizes counteracting the stigmatization of insertion clients, rejecting the workfarist idea that they should take any available job. Teams identify a range of opportunities, aiming to give individuals free choice. By facilitating one-to-one meetings with employers, they hope to cancel out disadvantages that might accrue to clients in panel interview scenarios. Interviewees criticized enterprises that did not demonstrate the ‘right spirit’, and who had been dismissive of their users, a growing problem given persistent unemployment (staff interviews). One interviewee described the following case:
I’m thinking of a person who … has spent his entire career for more than 15 years as an after-service technician … Following an accident at work, he could no longer do that. But in any case, he really wanted to continue in the area of activity, to remain in the same sector … And as a result we will go prospect for [that kind of work] … that’s prospection virtually dedicated to that person. (Staff interview 3, non-profit provider C)
In all cases, the emphasis was on supporting the agency of service users, including in organizations (such as non-profit provider C) where clients were often RSA recipients. The lack of sanctioning also means that insertion avoids the infliction of financial instability on welfare recipients and their families observed under workfare systems (Griggs, 2010). This more supportive approach is partly to do with insertion’s weak articulation with the benefits system, as well as the less marketized governance of local networks which thrives in the absence of stronger central instruments.
Barriers to profit extraction
These characteristics of insertion systems preclude more problematic consequences of for-profit provision, such as ‘creaming and parking’ strategies, and have also thwarted attempts to impose workfarist initiatives. France has limited experience with workfare-like programmes. Between 2008 and 2011, the CdA was piloted in zones urbaines sensibles across 35 départements, involving an appel d’offre open to for-profit providers; in Seine-Saint-Denis, six firms won the competitive tender worth a maximum of €2.25m. The appel d’offre specified the target group (unemployed 16- to 25-year-olds), a goal of placing them in employment or training for at least 6 months, and a service of ‘intensive and personalized support in the job search, in the creation of an enterprise, or in the access to vocational training, combined with the intensive job and training prospecting [by the provider staff]’ (Délégation générale à l’emploi et à la formation professionnelle (DGEFP), 2010). The CdA also specified that counsellors should have at most 40 service users on their portfolio, and provided bursaries of €300 per month over 6 months to service users, plus supplementary support. The bursary required that the recipient was between 16 and 25 years old, lived in the municipalities the providers served, had no more than a baccalauréat, came to the centre regularly and was out of work. The providers of the CdA did not principally work with RSA recipients but received a wider group of young people out of work, seeking support in their job search.
Service users could stay on the bursary programme for up to 18 months, which would include the 6 months of work or formal training and in-work support. Payment-by-results was introduced. As with other insertion programmes, service providers had to recruit on the street or through networks.
The CdA exemplifies a significant nationally driven attempt to impose a more marketized and work-first logic on local welfare-to-work systems. However, in 2011, an evaluation concluded that the service was significantly undersubscribed (25,000 contracts as opposed to the targeted 45,000), that incumbent providers performed more successfully and that the service most benefited users who were already close to the labour market (DARES[direction de l’animation de la recherche, des études et des statistiques]), 2011). While workfare in other countries has proliferated despite critical evaluation evidence (Peck, 2002), this scheme was discontinued.
Front-line staff at our for-profit provider knew that payment was contingent on job placements. But this did not lead to numerical targets. The programme demanded conditionality from service users, but due to low benefits coverage, the only sanction was exclusion from the programme. Counsellors reported that the 6-month bursary payment had ambiguous effects, as some clients would delay their job search until the money ended, or would simply leave the programme. Therefore, counsellors looked out for ‘serious’ users who they felt that they would stay on the programme and complete a formal placement. This precluded quick-sorting methods, prioritizing instead the counsellor–user relationship. The provider also began paying out the bursary only once service users were in a job or training placement, to incentivize continuing.
Why did for-profit provision not progress further? Partly because the appel d’offre limited individual caseloads to 40. But it also reflects the resilience of street-level arrangements. Private operators encountered the Seine-Saint-Denis insertion network as outsiders, whose motives were sometimes viewed with suspicion. The for-profit provider had not yet developed clear models of the kind that they had rolled out extensively in other countries and argued that the public sector itself did not have much experience defining and paying for job outcomes (manager 1). It became clear that it had weak recognition both among potential service users and within the provider network upon which insertion depends. Management interviewees at the for-profit provider reported feeling like outsiders. Our MEDEF interviewee told us that for-profit provision was politically sensitive and used in large part due to the ability to start and stop contracts, and various other interviewees (public functions A and B; staff interview, Mission Locale A) expressed mistrust of for-profit providers. This is particularly important, given that for-profit providers would need to draw on the same networks as non-profit ones, not only to win contracts but also to identify and recruit large numbers of job-ready candidates to place in jobs and claim outcome payments.
It is not surprising that for-profit operators should encounter such problems, given the emphasis on trust and stability for funders and incumbent employment service providers. This, however, is only one part of the explanation. These trust-based networks can and have been disrupted in other countries (Greer et al., 2017). In France, however, the administration of RSA by local authorities places limits on the extent to which managerialism and market-facing methods can be imposed outside of PE itself, as does the fact that many insertion recipients under the age of 25 are excluded from RSA altogether.
Discussion and conclusion
We have outlined the difference between insertion in practice and the workfare principles that have become the international orthodoxy of social policy. Insertion services reflect the strengthening of citizenship rights for disadvantaged people in France – including entitlements to support services for potentially job-ready young people outside the benefit system – and are embedded in strong local networks of associations and public bodies. Because of this local embeddedness, it may be that organization differs greatly elsewhere, but we have used the case of Seine-Saint-Denis – a large and politically significant site of investment in these services – to explore the practical dynamics of provision. Insertion services defy workfarist principles, but have several results which are greatly valued by street-level actors: strong organizational adaptability, staff initiative, a focus on client agency and limits to profit extraction strategies such as creaming and parking. These outcomes of insertion contribute to the political legitimacy of insertion systems, helping to sustain them.
In explaining the relative weakness of workfare in France, we reject the explanation that Bismarckian welfare regimes have inherent staying power. Changes to German unemployment benefits and administration illustrate the weakness of this argument: the Hartz reforms lumped jobseekers together with former claimants of social assistance in a particular means-tested benefit administered by the PES. Nor do republican ideas of solidarité immunize France from managerialism and results orientation, which do apply in the PE (Divay, 2009; Lavitry, 2009). There have also been experiments with commercial provision via the CdA, and the moral panic over the banlieues has put pressure on national policymakers to intervene locally.
Part of our explanation is the agency of local policymakers and service providers operating under ‘network accountability’ in resisting the encroachment of workfare principles. Local actors we interviewed were suspicious of commercial providers and perceived important advantages to insertion methods. A MEDEF official expressed strong support for and deep involvement in insertion programmes, and management at a for-profit provider described the difficulties of influencing key policymakers in a way that would allow it to win new contracts. The for-profit sector failed to establish itself as a powerful interest group, and its message was to treat for-profit provision as a complement (and not a replacement) to insertion.
Moreover, the ability of the French central government to overcome this resistance is limited by decentralization and fragmentation of authority. The exclusion of under-25s from RSA and the role of départements in administering benefits and coordinating services are manifestations of fragmentation and decentralization that limit the possibility of top-down reform. The restructuring of the PES along the lines of the work-first policy orthodoxy did not extend to insertion services, and we witnessed a central government initiative to introduce commercial services in this area – CdA – wither on the vine after negative evaluation. Public concern with excessive welfare spending and policymakers’ nods to social policy orthodoxy therefore did not translate into work-first services in the establishments we visited.
We do not overlook insertion’s disadvantages. The complex organizational network can be opaque, bureaucratic and difficult to navigate. And from a work-first policy perspective, the voluntaristic approach to users creates weak work incentives; indeed, the numbers of job placements are the main weakness highlighted, starting with the rollout of RMI (Paugam, 1993). This, however, is a problem with active labour-market schemes generally, including workfare; and workfare has additional costs, such as the hardship, especially for families with children, caused by sanctions (Griggs, 2010). We have not examined labour-market outcomes; a comparison between insertion and workfare in these terms is a matter for future research.
The barriers to the imposition of workfare are not, of course, insurmountable. Insertion could be vulnerable if policymakers moved further towards work-first social policy. Such a shift would require the central imposition of new contracting arrangements and measuring techniques, a more coercive approach towards clients and the marginalization of existing local networks. This would be difficult but not impossible. One recurrent theme of our interviews was the phrase ‘not yet’, which most staff interviewees appended to all remarks about staff autonomy and the counsellor–client relationship. Although we found little concrete evidence of such shifts in the organizations we visited, French policymakers may in the future impose austerity through tighter top-down control, using quantitative targets and marketization.
Any shift from ‘network’ to ‘market’ (Jantz et al., 2015) would be a disruptive process, at least in Seine-Saint-Denis, necessitating a coercive approach to providers, front-line staff and users. This disruption would be among the costs to weigh against ostensible benefits. Given these problems and the weak evidence for the success of active labour-market schemes generally, the argument for strong and centralized market-mediated control for public authorities over providers and clients may be difficult to sustain. France’s ‘failure’ to create a work-first welfare state is less a pathology of the Bismarckian welfare state and its insider–outsider dynamics, and more a benefit of strong control by workers and localities.
Footnotes
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by funding from the European Commission European Research Council (grant #313613).