Europeanisation: the shadowy corners of a concept

5At the beginning of the 2000s, James Caporaso, Maria Green Cowles, Thomas Risse and their co-authors proposed that Europeanisation be studied using an approach called “goodness of fit”. [13] According to this idea, the European process produces an “adaptive pressure” on member states which is more intense the more national structures are distant from EU orientations. For these authors, it is medium-intensity adaptive pressures that have the most effect, because they imply a significant but acceptable transformation of national structures. If the state is too far removed from what the EU requires, national structures may well resist pressure (inertia), or even reinforce their particular arrangements (distancing). On the other hand, if the distances between domestic structures and European prescriptions is too small, then the national system won’t have far to go and change will be marginal (absorption).

6The analytic framework proposed by Caporaso, Cowles, Risse et al. enables us to theorise the various effects of Europeanisation. However, it tends to assume the existence of clear European pressure that has a basis in law, and that pre-exists domestic change. Within this framework, how can we account for the fact that certain states profoundly modified their system of public policy, even though the pressure for change was weak and did not have a solid legal base? [14] The second generation of studies on Europeanisation showed that the process cannot be reduced to either downward pressure from Europe to states, nor to regional integration and the transfer of sovereignty to the EU. [15] European decision-making can also be an opportunity for national elites to push through recommendations “from above” that reinforce their own projects (blame-shifting strategies [16]). In sectors where the EU has no legal jurisdiction, learning and sharing “good practice” can encourage the socialisation of European elites; national political elites are Europeanised to the extent that they absorb the rules of the game, the recommendations at European level, and the practices of other member states [17]. The classic definition proposed by Claudio Radaelli incorporates these different parameters of Europeanisation:

7

“Processes of (a) construction, (b) diffusion and (c) institutionalisation of formal and informal rules, procedures, paradigms, styles, ‘ways of doing’ and shared beliefs and norms which are first defined and consolidated in the making of EU public policy and politics and then incorporated in the logic of domestic discourse, identities, political structures and public policies.” [18]

8This definition enables us to link two levels of analysis (European decision-making and its domestic effects) and to take into account formal and informal rules, beliefs and cognitive frames. It allows for the possibility of studying interests and the conflict between them through analysis of domestic “logics”. However – and even though Radaelli emphasises the value of a bottom-up [19] research design – this definition tends to provide an image of Europeanisation that is clearly sequenced and vertical: process is initiated at the European level to then be applied at the domestic level. [20] More interactive, Europeanisation appears today to be a phenomenon involving reciprocal influences between politico-administrative national actors conscious of European issues [21] and institutions in Brussels which are in contact with the national level. [22] Radaelli notes that “it is important to consider this as an interactive process, not as a simple process of unidirectional reactions towards ‘Europe’”. [23] How can we express this continual process of exchange, of (re)definition of problems, norms and solutions between the European and national levels? How can we emphasise interaction, interdependence and cross-influences? The definition proposed by Bruno Palier and Yves Surel, although less exhaustive than Radaelli’s, is more flexible and less vertical:

9

“Within Europeanisation we propose to see: all the processes of institutional, strategic and normative adjustments resulting from European construction.” [24]

10This definition incorporates all the adjustments resulting from the process of European construction; local public policies, institutions, and political actors are all potentially affected by Europeanisation. It also leaves room for studies of Europeanisation beyond, or even “without Europe” – of actors or states which are remote from the perspective of integration and yet constrained to modify their policies or their governance to develop commercial or diplomatic relations with the EU, [25] or to draw on the EU as a model to follow. [26] The approach in terms of “construction” and “adjustments” shows a Europeanisation in progress and the modifications that this process leads to for all the actors concerned. Whether these changes are voluntary or constrained, correspond to the recommendations or stray from them, is not pre-empted as such. [27] In taking adjustments and thus “effects” as a starting point, this framework enables a “bottom-up” sociological approach which first studies actors and discourses on the national level, and then moves up to the European level in instances where the European variable seems relevant. [28]

11Since the work of Caporaso et al. the development of studies on Europeanisation might suggest that the EU is increasingly understood as a simple intermediary between the states. The EU is not a neutral filter however, and Palier and Surel’s definition also focuses precisely on what results from European construction. That the process is more complex than initially envisaged makes the search for causality more difficult but all the more necessary. [29] What exactly do European institutions rely on? Even without legal pressure, has “European construction” had an effect on national policy, and, if so, what effect? It is precisely this question that is raised in the employment sector where, as we will see, it is not the coercive mechanisms that have encouraged changes in national policy.

Europe and employment: a history of non-intervention?

12Noting a persistent divergence in social regimes within Europe, commentators have expressed their doubts as to the effects of the EU on domestic employment policies. [30] We might also question the precise nature of the EU’s role, given that the social sector remains the jurisdiction of member states.

13The Treaty of Rome (1957) already posed the question of the coordination of social security systems between the states; Article 117 indicated that “Member States agree upon the need to promote improved working conditions and an improved standard of living for workers, so as to make possible their harmonisation”. In order to achieve this, Article 118 allowed for “close co-operation” in terms of employment, training, law and working conditions – but without formally indicating how this might be achieved. The Maastricht Treaty specified that the Union set itself the goal to “promote economic and social progress as well as high levels of employment” (Article 2), but the notion of “high employment” had still not been defined at the beginning of the 1990s. It is unclear whether it refers to the employment rate (the proportion of individuals employed compared to the working-age population), the activity rate (the ratio between the number of active persons, including jobseekers, and the corresponding total population), or a low unemployment rate. The words “work” and “unemployment” are not mentioned explicitly [31] until in 1997 the Treaty of Amsterdam made the promotion of employment “a matter of common concern”. [32] The EES was thus launched but even within this framework the tools of the EU remain weak: Article 128 of the Amsterdam Treaty provides the EU with the power to study the employment situation in member states and make public the results in an annual report by the Commission and the Council. The “guidelines which the member states shall take into account in their employment policies” are established in view of this study. However this evaluation and the guidelines adopted are not formally binding, each member state only has to produce an “annual report on the principal measures taken to implement its employment policy”. If member states do not respect these guidelines, the Council makes recommendations to the states following the propositions of the Commission, [33] but these recommendations have no legal standing either.

14The Nice and Lisbon treaties confirmed the national prerogative in the social sector [34] and the lack of legal authority led the European institutions to concentrate on flexible instruments of governance within the framework of the Open Method of Coordination (OMC). Articles 126 and 127 of the Amsterdam Treaty leave the EU the possibility of “encouraging cooperation between member states, by supporting and, if necessary, complementing their actions”. [35] This is an intervention by default however and the Commission itself deplored the weakness of the instruments available to it. A director in the DG for Employment and Social Affairs thus declared himself “personally sceptical as to the impact of what you call the exchange of good practice, the instruments of soft law, and even the methods of open coordination” (E47). The frankness in this interview is not surprising given that the Commission officially complained about the weakness of its prerogatives: “the catch-all nature [of the guidelines] and the lack of internal prioritisation limited the impact of this instrument on national policy-making”. [36] This summary by the Commission uses the criticisms that were formulated as early as 2004 in the Kok report – this latter condemning both the lack of political will by member states and the weakness of European instruments in the employment sector. [37] In order to “ensure that member states assume their responsibilities”, [38] the report recommended “ratchet[ing] up the political consequences of non-delivery”. [39] The Commission took these criticisms (E47 and E49) on board to demand the reinforcement of EES instruments:

15

“There is significant scope for mutual learning and sharing good practice.” [40]

16The distinction between “hard law” and “soft law” is typical: [41] the concept of “hard law” refers to the area of law applied and controlled by legal and police instruments of verification and sanction. [42] “Soft law”, as defined by Snyder, refers to “rules of conduct which in principle have no legally binding force but which nevertheless may have practical effects”. [43] Although we accept today that the OMC encouraged the circulation of elites and projects within Europe, [44] and that there was an exchange of ideas in certain Europeanised sectors, [45] it still often remains difficult to determine the effects of these exchanges on specific mechanisms. [46]

17Some researchers doubt that “soft” instruments of governance have had any effect at all. [47] The question posed here is thus an old one in European governance: are non-legal instruments an efficient way of reforming systems of public policy? [48] Situating my research within debates around “non-legislative modes of policy-making and modes of governance” [49] this article will compare the most recent developments in European governance with the reforms adopted in the member states, thus interrogating whether it is possible to provide proof that certain instruments have indeed played a role in directing, encouraging and catalysing [50] social policies in France, Spain and the Netherlands.

The harmonisation of national welfare regimes

19My research adopts a “most different systems design” [51] approach and aims to understand convergences between historically different systems. I chose France, Spain and the Netherlands because they represent the most different cases in terms of the situation of the labour market, [52] welfare regimes, [53] and traditional administrative approaches to unemployment and poverty. [54] Access to basic unemployment benefits is not the responsibility of the same institutions in these three countries. In France, the state has been responsible for RMI (Revenu minimum d’insertion) [55] since its creation in 1988, with the administration and payment of benefits being decentralised to the conseils généraux [departmental authorities] in 2004. In Spain, the central state is not responsible for minimum welfare payments: over the 1990s the autonomous communities decided to finance social welfare systems. In the Netherlands minimum revenue has been financed by taxes since 1963, but the running of the scheme and its payments are the responsibility of local authorities. The amount of such payments is determined by the state in France and the Netherlands, and by the autonomous communities in Spain; funding is guaranteed by the autonomous communities in Spain and by the state in France and the Netherlands. [56]

20In the 2000s, however, the three countries fundamentally revised their policies for revenue support for the unemployed, and those out of the labour market. In the Netherlands the Work and Social Assistance Act (Wet Werk en Bijstand, WWB) has provided financial incentives to local authorities since 2004 to accelerate the return to work of individuals receiving benefit payments. The government has demanded that Dutch municipal councils take the necessary steps to reduce the number of people dependent on minimum revenue payments by 25% over four years. [57] The councils – such as that in Utrecht where I conducted my research – proposed “substitute jobs” to those currently out of the labour market. These jobs were low-paid (roughly half the Dutch minimum wage) but it was possible to combine them with social welfare payments. In the interviews, the fact that individuals did not lose their former benefits if they took up employment was seen as a “trigger” for the return to work (E26). The Ministry confirmed that making work more “attractive” (E21) meant that it had to “be financially worthwhile for those receiving benefits”. [58]

21In France, the RSA (Revenue de solidarité active) replaced the RMI in 2009. The main difference between these two forms of social welfare payments is the possibility of combining payments with income from work, with no limitations as to duration. If a beneficiary returns to work, he or she loses part of the benefit, but it decreases less rapidly than the salary increases. [59] This gradual decrease was implemented to make work “pay” more:

22

“The RSA will encourage people to return to work and provide unemployed people with a clearer idea of how their income will change if they return to work.” [60]

23A member of Martin Hirsch’s cabinet confirms this: “with the RSA, when you go back to work, you make money. That’s really it […] That’s what we really based things on – the financial benefit of returning to work” (E3). In ensuring that returning to work is financially advantageous for those receiving welfare payments, the High Commission hopes to stimulate “employment rate growth” in the French economy. [61]

24In Spain, welfare policies are the responsibility of the autonomous communities. Catalonia – one of the first autonomous communities to have created a framework for minimum revenue at the beginning of the 1990s – introduced its “Renta Mínima de Inserción” in 2006:

25

“It was considered necessary to introduce new complementary forms of economic assistance linked to getting a job in the hope of stimulating the efforts of those receiving the Renta Mínima de Inserción to find work.” [62]

26In Catalonia the theme of “encouragement to work” also led to the creation of a new scheme, “ayuda complementaria de inserción laboral”. Unlike the RSA in France, this benefit does not decrease gradually but is paid at a flat-rate over twelve months. The beneficiary of minimum revenue payments who returns to work can combine his or her salary with a fraction of the former benefit, with that benefit payment representing roughly one third of the Catalan Renta Mínima amount. [63]

27What do these reforms have in common? First, we should note the persistence of distinct public policy paths: numerical targets for those coming off benefits were set in the Netherlands, but not in France or Spain. The return to work bonus is paid at a flat-rate in Spain, but in France it gradually decreases, and in the Netherlands it is proportional to the number of hours worked. The decision to reform the scheme was made centrally in France – the Haut Commissariat aux solidarités actives [High Commission for Active Solidarity against Poverty] was especially created and appointed by the President of the Republic and the Prime Minister; [64] but was decentralised in Spain – the communities remain autonomous in the management of welfare. In the Netherlands the decision was shared – the reform of the WWB was decided on and piloted by the government, but choices about practical initiatives were made at the local level. Each system of public policy was thus reformed according to specific histories and institutional logics. [65] Yet, beyond these specificities, instruments of public policy and representations within social administrations have indeed tended to converge in Europe.

Instrumental harmonisation

28In the three reforms studied, the guiding instrument was neither legal obligation nor administrative orientation towards employment, but a bonus/penalty system. Although recipients of basic social benefits who return to work see their incomes improve (“such that work provides all with the guarantee of being released from and protected from poverty” [66]), those without work experience no such improvement and remain below the poverty line. [67] In all three cases, poverty is not addressed by an increase in social payments for the unemployed, nor by preferential access to housing or transport. On the contrary, the advantages associated with minimum revenue welfare payments must be less than those associated with working. The increasingly technical nature of the frameworks (micro-econometric calculations of the income-replacement rate, [68] of the “financial gains of returning to work” [69] and “marginal rates of pressure of work” [70]) does not hide the fact that these new instruments are “vehicles for values, which are fuelled by a particular interpretation of the social, and by specific conceptions of the proposed mode of regulation”. [71] The incentive instruments thus translate the same shift towards making the care of the poor an economic matter [72] – guiding them towards the market and integrating them into society through their participation in productive activity. [73] Even though systems of basic social welfare have inherited an approach that is more concerned with living conditions and the social objectives of being in work, [74] the reforms adopted tend to break with these traditions and move the three systems closer to a more “active” model of social protection. [75]

Cognitive harmonisation

29If the choice of an instrument is the technical expression of a vision of the social world and ways of acting upon it, then instrumental change can tell us something about cognitive change [76] in a sector of public policy. Arguments are increasingly similar: in 2007, in France, the Haut Commissariat aux solidarités actives condemned the fact that returning to work did not enable people “to increase their revenue”. [77] During an interview the manager of the “inserción laboral” programme in Catalonia confirmed that the gap between the revenue from benefit payments and the revenue from work is not wide enough:

30

“Because wages have not increased much and the minimum wage is what it is, well sometimes just the fact of working was not… motivating enough for someone benefiting from the Renta Mínima… to go to work.”

31For the head of the “Work” division at the Dutch Ministry of Social Affairs and Employment (Ministerie van Sociale Zaken en Werkgelegenheid, SZW), the problem with social welfare in the Netherlands is that “the benefit is too comfortable, too comfortable, comparing to the salary you can earn with the new job” (E18). For these politico-administrative managers, the small gap between revenue from social welfare and revenue from the labour market is a “disincentive to looking for work” (E44). For the person responsible for employment at the Social Services department in the town of Utrecht, the financial bonus is a “trigger, it’s a trigger for them” (E26). For the person in charge of “inserción laboral” at the Generalitat de Catalunya the solution would be to “encourage the beneficiaries so that they don’t get used to receiving the payment and so that they are more…inclined to get work” (E45). For the High Commissioner for Active Solidarity in France, “when a person has limited means, you cannot ask them to work for resources that do not increase”. [78] This common definition of “the problem” of basic social benefits is based on a utilitarian understanding of beneficiaries’ behaviour. According to this perspective, poor people make a rational calculation as to the costs and benefits of work compared to welfare. For the Catalonian manager “the relationship between the wage… between working and not working, depending on the money they were receiving, well, that’s it: it wasn’t encouraging that person to work” (E45). According to the person responsible for the “Work” division at the Dutch SZW ministry, “it’s a comparison people make. A calculation” (E18).

32This type of argument is based on the economic concept of “traps”. [79] Here is how the Dutch National Reform Programme (NRP) presents it:

33

“To make working – and working longer hours – more appealing, it has to pay off financially. However, the rewards are not always sufficient. Accepting work from a situation without any personal income (i.e. the re-entry trap) is sometimes less than appealing due to high marginal pressure. Precisely because this group has the financial leeway not to have to work – often involving partners of a breadwinner – it is important that the increase in income represent a sufficiently high incentive. High marginal pressure may also present obstacles to increasing the number of hours worked (i.e. the part-time trap). Among the unemployed, accepting work may appear to offer relatively few advantages in the shorter term compared to receiving benefits (i.e. the poverty trap).” [80]

34The Dutch NRP does not limit itself to oblique references to the theory of “traps”; it develops a neoclassical argument to explain how beneficiaries are kept in part-time work (“the part-time trap”), in long-lasting inactivity (“the re-entry trap”) and in poverty (“the poverty trap”). The neoclassical model of the labour market is explicitly appropriated via the “literature on incentives” (E2), and is continually evoked through:

• a critique of the negative side effects of social welfare, which are accused of creating the phenomenon they are designed to combat: “passive” benefit payments leading to poverty;
• a utilitarian conception of social behaviour: in a market system making the “choice” to work is seen to depend on the cost and benefits of the job, with each individual comparing the opportunities for work and leisure after having established their own subjective preference curve. [81]

35Even though the regimes protecting against the risks of unemployment or poverty were based on the socialisation of social risks and the exclusion of the guilt-inducing notion of “voluntary idleness”, [82] “active” social policies personalise the causes of unemployment and poverty. [83]

36This “activation” frame of reference may provoke reservations among academics and intellectuals, [84] but it does not seem to provoke political opposition or any structural ideological cleavages. In Catalonia, it was the socialist Minister of Employment Jordi Valls i Riera who drafted the decree on the active reform of 2006; [85] in the Netherlands it was a Christian-Democrat minister for social affairs, Aart Jan de Geus (CDA), who defended the 2004 welfare reform; in France, the government that developed and voted in the RSA was conservative, following on from a previous conservative government (UMP majority since 2002, renewed in 2007). The parliamentary debates did not show an increased cleavage on this issue. During the vote on the 2004 law on the WWB in Holland, the Dutch labour party (PvDA) was concerned about the funding of the system but did not oppose the “work above benefit” principle. [86] In Catalonia, the welfare reform of 2006 was not the object of a parliamentary vote because it was passed as a decree, but according to my interviews, this “consensual” reform did not provoke any major political opposition. [87] In France the socialist parliamentary group abstained from voting on the RSA, but did so for reasons of funding rather than because of the mechanism or principle, which they otherwise “approved” [88] of. One socialist MP declared: “The RSA could have had support from all sides of parliament – but on one condition, on one condition only: they would have had to make sure the funding was equitable” [89]. The RSA bill was finally passed with a majority of 306 votes for and only 20 votes against. [90]

The content of the EES: a “European doctrine”?

37If we define “cognitive convergence” as a “progression towards shared values, norms and conceptions of causality”, [91] then we might consider that this is indeed what is at work in terms of “active” proposals in Europe. However, convergence is not sufficient proof of Europeanisation. Perhaps political actors influence each other without the EU ever being either a resource or a constraint; perhaps a “downloading” process is at work here, but through the intermediary of international institutions distinct from the EU. [92] My bottom-up research approach has been as follows: as a result of observing a cognitive and instrumental convergence between three member states of the EU, I began by asking whether the direction adopted by national structures corresponded to a hypothetical “European doctrine” in this area. The final part of this article goes on to explore the practical mechanisms of Europeanisation.

38The homologous nature of European recommendations and national implementation is clear when we study the EES texts. From the first guidelines on employment in 1997, priority was given to “active” social measures rather than so-called “passive” policies. For example, to combat unemployment European states should adopt “tailor-made responses systematically giving precedence to active employability measures rather than passive support measures”. [93] The delimitation of these “active” measures remained vague in the first guidelines however, which made the original direction proposed by the EES ambiguous. [94] Gradually the economic tone of the EES became more influenced by the neoclassical approach. Few changes were made in the texts of 1999 and 2000, after which the notion of the “active welfare state” was enshrined in the guidelines for 2001. The first of the new “horizontal objectives” even explicitly mentioned “providing adequate incentives”. From this point on, each member state was encouraged to “reform its benefit and tax system to reduce poverty traps, and provide incentives for unemployed or inactive people to seek and take up work”. [95]

39Encouragement, promotion and incentives to work: over the course of the 2000s, these watchwords were repeated in every new European document. Unemployment thus appeared to be a problem linked to the insufficient “value” of work for those receiving benefits. The 2003 guidelines explain: “[T]he balance between income from work against income in unemployment or inactivity should be such as to encourage people to enter, re-enter and to remain in the labour market”. [96]

40The guidelines, which aim to use “financial incentives with a view to making work attractive”, are specifically focused on the argument of “poverty and inactivity traps”:

41

“Member States will reform financial incentives with a view to making work attractive […] Member States will in particular review replacement rates and benefit duration; ensure effective benefit management, notably with respect to the link with effective job search, including access to activation measures to support employability, taking into account individual situations; consider the provision of in-work benefits, where appropriate; and work with a view to eliminating inactivity traps. In particular, policies will aim at achieving by 2010 a significant reduction in high marginal effective tax rates.” [97]

42The concepts of “replacement rates”, “inactivity traps” and “marginal effective tax rates” are taken directly from the neoclassical theory of the labour market. [98] Whereas they were simply alluded to in the first guidelines, all stages in the OECD’s reasoning on “making work pay” are now formally inscribed in the European texts. This includes financial incentives, [99] a decrease in the “utility” of unemployment via reducing the level of benefits and restricting conditions for receiving them, [100] increasing the differential between income from work and welfare income, [101] and decreasing tax levels on low incomes. [102] From 2003 the guidelines were only revised every two or three years; they reproduced the incentive arguments and developed them still further. In 2003, numerical indicators supported the recommendations concerning the return to work of those receiving unemployment benefits. [103] In 2005 the concept of “traps” was enlarged to include “unemployment, poverty and inactivity traps”. [104] In 2008, precedence was given to the mobilisation of those “those most excluded from the labour market” as part of an “active inclusion” program. [105]

43The relative openness of the early versions of the EES [106] gave way to a more stringent political approach – “passive” policies were increasingly criticised in the European texts in favour of activation and of incentive measures. This conceptual re-organisation was linked to the establishment and progressive reinforcement of a doctrine specific to the DG for Employment at the beginning of the 2000s. [107] A communication from 2003 presents the model supported by the Commission: “"Making work pay" refers to policies [which aim] to reform tax and benefit systems with a view to providing appropriate financial incentives to take up jobs, to remain in work, to increase work effort […].” [108] The recommendations of the Commission are clear:

44

“There is therefore a need for tax/benefit policies to be re-designed in such a way as to reduce the risk of inactivity, unemployment and poverty traps. In other words, a high level of social protection is desirable as long as incentives to work are maintained.” [109]

45Pointing to the side effects of overly “generous” social welfare systems, the Commission also condemned the alleged behaviour of welfare recipients who apparently preferred to “choose to live on benefits rather than be employed”. [110] Challenging the so-called “passive” benefits is at the heart of this “comprehensive approach contributing to making work pay”. In its English version, the title of the 2003 Communication is highly evocative: “Modernising social protection for more and better jobs – a comprehensive approach contributing to making work pay”. The expression “making work pay” refers to an OECD report of the same name published in 1996 in which that organisation explained that “cutting benefits is the simplest way of increasing the incentive to work”. [111] In referring to the OECD [112] publications, and repeating the expression “making work pay” twenty-two times in the 2003 Communication, the European Commission makes the theoretical inspiration for its economic approach to the labour market very clear. Subsequent communications by the Commission on the subject of employment and social affairs confirm this doctrine. In 2007, the Communication on “active inclusion” declared:

46

“In order to make work pay for job seekers, it is also necessary to continue reviewing the incentives and disincentives resulting from tax and benefit systems, including the management and conditionality of benefits.” [113]

47The other Communication of 2007, on “flexicurity” developed these recommendations still further:

48

“Social security systems would offer incentives to low-skilled benefit recipients and monitor the conditionality of such benefits in order to ensure that taking up work pays, if necessary by providing supplementary benefits or gradual phasing out of benefits.” [114]

49The increasing specificity of the Commission’s guidelines and communications is the result of this process of trial and error, over the course of which the Commission had to invent and reinforce its doctrine in the employment sector. This analysis confirms qualitative studies on the reception of the EES in France. Interviews conducted at the beginning of the 2000s already suggested that a “European doctrine” [115] was gradually coalescing: “a ‘shared frame of reference is developing’, based on a few key themes – active policy against unemployment, decreasing contributions, investment in human resources, the ‘make work pay’ concept, and making use of all the human potential of the labour force”. [116] Another interview confirmed the Commission’s focus on a few themes in particular (“incentives to work, making work pay, decreasing contributions”). [117]

50The original tone of the EES reflected the historical context of the time in which it was written. The OECD had an active role in writing the EES, as a result of the circulation of its experts in Brussels technocratic circles in the 1990s. [118] The ambiguity of the first EES texts was also the result of the political equilibrium of the time, with “activation” initially being seen by national leaders as a “compromise” (E46, E47) between the Nordic model of massive investment in employment policy and the Anglo-Saxon model which makes the individual responsible for their unemployment. [119] The DG for Employment seized upon the concept of “activation” to construct its “modernised” model of social protection, supporting a specific synthesis of ideas and progressively reinforcing the content. [120] Such as it was subsequently understood by the Commission, the “active welfare state” [121] combines reinforced spending on incentives with stricter control over those receiving payments. [122] Over the course of the 2000s, the Commission drew more and more explicitly on algorithms, [123] promoted instruments of reform [124] and set objectives [125] which stemmed from the neoclassical theory of the labour market. How has this orientation had an impact on member states since then?

Europeanising the actors

52Elite administrators within social policy sectors are socialised at the European level, [131] and consequently absorb and re-use shared arguments when they return to their national contexts. [132] On 24 September 2008, three Franco-Spanish leaders who played a significant role in these “active” reforms met in an OMC seminar in Brussels. The first, Alfonso Prieto, was the deputy director for studies on employment in the Spanish ministry of employment. He was an advisor to the ministers Celestino Corbacho and Valeriano Gomez on the active reforms of the Spanish labour market in autumn 2010 and spring 2011. The second, Jean-Denis Combrexelle, was the director general for employment at the French ministry of employment and advisor to the employment minister Xavier Bertrand during the creation of the RSA in 2009. [133] The third, Christian Charpy, was director of the French employment service and was a frequent participant in discussions with the office in charge of the RSA, in his capacity both as the head of Pôle Emploi [134] and as someone close to Martin Hirsch. [135] On 24 September 2008 the European Commission invited these three national leaders to discuss “national policies of flexicurity”. However, these meetings in Brussels are not as informal and open as the principle of the OMC and “peer review” might suggest. Organised and managed by the DG for Employment, the meeting on the 24 September opened with an introduction by the DG’s director of employment, Xavier Prats Monné, followed by the European Commissioner in person, Vladimír Špidla, to spell out the Commission’s line in terms of active labour market policy. There were then presentations of national examples, concluded by Xavier Prats Monné once again for the DG for Employment. Even within these seminars “amongst peers”, eminent members of the Commission were present, and met with the administrative elite piloting these social reforms at the national level. There were regular exchanges and meetings between the participants. The minister Aart Jan de Geus, asked Aart Bette, as the chief adviser to the Dutch department of work, to set out the parameters of the “WWB” reform, and Bette was also invited by the DG for Employment in 2005 to a seminar in Brussels entitled “Extending working life”. These ministerial leaders frequented European seminars, met at events organised by the Commission, and exchanged points of view with directors general and with the commissioner. As a result of being socialised in the European context, they adapted the principles of Brussels to their national contexts.

53Martin Hirsch, then High Commissioner for Active Solidarity against Poverty, was a “pivotal” actor who succeeded in translating the micro-economic models and the theory of “traps” discussed in Brussels into a language that was acceptable in the French national context. A senior figure in the Office of European Affairs at the Ministry of Social Affairs described Hirsch’s relations with the European level thus:

54

“He was in demand, actually, at the European level. Firstly there were all the events that came through under the French presidency and the round table on ‘poverty and exclusion’ which under the French presidency was dedicated to active inclusion, the meeting of ministers on the fight against poverty and exclusion which was held for the first time in Marseille…. He went to inaugurate the opening ceremony for the European year of the fight against poverty and exclusion, and in fact there he talked about the RSA… […] And what’s more he was also regularly invited because, between the RSA and his personality, which was very… well, it meant that he managed to get the objective of reducing poverty adopted at the national level, and he was very supportive of that approach at the European level too, well, he was one of the Commission’s ‘favourite’ speakers and presenters.”

55What is interesting in this long extract is that a national actor made strategic use of European resources in pushing through domestic reform: [136] Hirsch used the European forums and arenas to promote and justify the RSA reform, both at the European and at the national level. During France’s Council presidency, he was the “pilot” of international events that he organised himself, but during which “he in fact talked about the RSA…”. The EU was used as a political argument (legitimising usage [137]) to make political and social actors accept the reform:

56

“When we prepare notes for the minister, or for our director, we always stress that it is a strategy that we used during the French presidency within an EU framework.”

57But Martin Hirsch did not only use the EU as a source of external pressure; he drew on it directly when devising the RSA. As early as the Green Paper on the RSA in 2007, Hirsch used the Commission’s arguments on the “trilemma of the welfare state” – in other words the need to “conceive of frameworks that reconcile incentives to return to work, reduction of poverty, and financial sustainability”. [138] This was an idea that had been formulated four years earlier by the European Commission in its communication on “making work pay”. [139] The links between Hirsch and the European level encompass both “downloading” (integrating European resources and requirements concerning “activation” in order to construct the “Revenu de solidarité active”), and “uploading”. Hirsch defended the “French approach” in Brussels and was involved in European decision-making. He was indeed “in demand” at the European Commission; a “favourite” speaker (E4), he set a European example, and was mobilised by the DG for Employment as an international spokesperson for a reform validated by the Commission. These links between the Haut Commissariat and the EU remained strong – the assistant director in Hirsch’s cabinet (E1) joined France’s permanent delegation to the EU in 2010. These are some of the reasons why Europeanisation cannot be considered as a one-way mechanism, but rather as a process of mutual interdependence and influence. National and European actors use and influence each other reciprocally to promote, justify and progress their own interests. [140] In the case of the French Haut Commissariat, their interest was in promoting this reform and improving its social and political acceptability in France, given that the revolving presidency of the EU Council coincided with the parliamentary debate on the RSA. For the Commission it was the opportunity to draw attention to an example of “good practice” and show that the national application of European recommendations (here making social policy “active”) did indeed work.

Changing constraints and resources in internal political affairs

58Although they are not binding, European recommendations can have an effect on domestic policies by modifying the resources and constraints that weigh on actors. One officer for international affairs at the Dutch ministry of social affairs regularly collaborated with the director general of the DG for Employment in Brussels. Although in the course of their discussions with me they agreed that the weakness of the OMC instruments was a shame, they also indicated, however, that the EES had been very effective during internal political debates in “unblocking” crises and getting the reforms passed:

59

“What is so interesting [about] the Employment Strategy is that, that locks in the national parliaments, because you can reach an agreement on reforms in your labour market policies. That locks… can be solved, because in the European arena, you get recommendations from the Commission on which path to follow in your labour market policy, and they are not in the forms of directives or strong… hum… powerful instruments, but no politicians at home, in the national arena, like to be told by Brussels what it should do […]. It’s always difficult to, to… exactly pinpoint what role was played by European recommendations. But we noticed that it was mentioned in the Parliament a number of times. We noticed that it increased the sense of urgency with policy makers, with the cabinet, to, to introduce their reforms.”

60In the Netherlands, the parliamentarians whose projects for reform and amendment followed the direction outlined by the Commission benefited from extra resources during political debates. The dramatisation of European warnings and recommendations gave an impression of urgency and had an accelerating effect on domestic reform. [141] The Catalan case is not comparable in this respect because the reform of the Renta Mínima de Inserción was passed by decree. In France, however, the RSA was the object of a parliamentary discussion and the European stakes were indeed mobilised during these exchanges. During the vote on 8 October 2008 in the National Assembly, the MP Bruno Le Maire suggested that the creation of the RSA was a “necessity” that was part of a framework of “tough and determined action coordinated with all our European partners”. [142] Hirsch adopted this argument to repel an unfavourable amendment to his proposal, telling the MPs that they ran “the risk of contradicting our European engagements”. [143] Before voting on the bill, the French parliamentarians all received a letter from the minister outlining the “reasons to vote for the RSA” and which drew on European arguments:

61

“The RSA is part of a programmme for inclusive action, which is at the heart of EU thinking, and which will be debated under the French presidency.” [144]

62Hirsch thus situated the RSA within the European strategy for inclusive action, which was a project supported by all of France’s European partners, and by the French presidency of the EU. One parliamentarian suggests that this implicit argument (“a vote against the RSA is a vote against Europe”) exerted its authority over MPs:

63

“The RSA, I felt that it was, both at the Parliament and… it was awkward for everyone, it was awkward for everyone. Both on the left and the right.”

64Indeed, when it came to the final vote on the RSA, only members from the communist and republican groups dared to vote against it. [145] “We can’t say no to the RSA” said a member from another opposition party (E9). European stakes were thus a lever for social reform – in France and in other member states [146] – by lending extra resources in national debates to actors responsible for getting these changes accepted.

Organising horizontal transfers

65How are transfers of public policies actually organised in the European context? The theoretical framework of policy transfer studies [147] has been criticised for apparently not sufficiently taking into account the asymmetries and power relations between actors. [148] However, David Dolowitz and David Marsh stress that they envisage the possibility of compulsory transfers: [149] receiving actors can be encouraged to transfer by the institutions behind the models or by external bodies. [150] In this article, the Spanish case suggests just how much the European Commission promoted particular models and encouraged transfers between regions. Here is how an official in the Catalan welfare system described a study trip to France:

66

“– We had a meeting. Because in France what they have done is to decentralise minimum welfare benefits too. So when they decentralised, they called us from Eastern Catalonia [Pyrénées-Atlantiques] so that we could have a day together, and talk about this. […] And from there, we had several meetings to, let’s say, exchange on these questions. And that’s when we talked about these complémentos. […] This idea, comes from there.
– This idea?
– Yes, this idea of the supplement for returning to work.”

67Of course the ideal transfer occurred region to region (“this idea came from there”), but the EU was an unavoidable intermediary in transfers because it encouraged, supported and financed the exchanges within the framework of European cooperation programmes. The series of meetings that resulted were part of the EU Interreg III program, the agenda for which (“implement common instruments on employment issues”) was established by the European Commission. [151]

68In France the idea of the RSA was also influenced by European transfers – as a member of Martin Hirsch’s cabinet explained:

69

“We had a study trip to England. So we all went to see how it worked there, both their ANPE, which at that time anyway was called ‘Job Centre’ […] and their tax mechanisms, the Working Tax Credit and the Child Tax Credit. So well, it’s always the same idea, it’s the idea of decreasing benefits with a form of cap like the RSA Activité. When you don’t work you have nothing. When you start working you earn and earn until you reach a maximum, and then, if you keep working it starts decreasing. But it decreases progressively. And so yes, yes, we were very inspired by the British model.”

70It is rare that a minister acknowledges that a reform wasn’t invented from scratch but that it drew on external examples, which is what makes Martin Hirsch’s public statement particularly valuable:

71

“We were, I am not ashamed to admit it, inspired by the ‘working family tax credit’.” [152]

72The other members of the Haut Commissariat confirmed this when I met them: “the RSA was a type of intervention that existed in other European countries, and which aimed to have an impact on labour supply” (E1); “we looked at programs that were ‘similar’ in inverted commas, such as the Working Families Tax Credit” (E2). In the Netherlands, the WWB was inspired by these same British experiments in terms of work bonuses, as the officer for European Affairs at the Dutch Ministry for Employment indicated:

73

“It’s true to say that our incentive policies were inspired mainly by English experiences.”

74Are these demonstrated borrowings between countries sufficient for us to talk of “Europeanisation”? In other social service sectors, such as health cover, exchanges between states were organised at the end of the 1990s, but without EU involvement in the process. [153] Here, the two levels of transfer are intertwined – the Commission encourages transfers both of “good practice” and of the national examples it targets. It is also worth noting that the two models cited by the French and Dutch officials – the Working Tax Credit and the Child Tax Credit – are precisely the two reforms that the European Commission has used in its publications since 2000 as examples of effective “incentive” reforms. In 2003 the Commission’s communication on “making work pay” explicitly focused on these “good practices”:

75

“The UK introduced two means tested schemes, the ‘Child Tax Credit’ and the ‘Working Tax Credit’ (WTC) to replace previous tax credits. They have the twin stated aims of helping to eliminate child poverty and to make work pay.” [154]

76As early as 2000, the Commission had invited European partners to come to London to observe the effectiveness of these systems in terms of the incentive to return to work. [155] In bringing about cooperation and discussion on these targeted themes, the EU acted as a vector for the transfer of public policy. The exchanges between member states on the “activation” of social protection were encouraged and financed by European institutions.

Providing financial incentives

77Although the EU has no legal jurisdiction in the social sector, it does have a tool for directing budgets:

78

“The European Social Fund is the main financial instrument in the European Employment Strategy.” [156]

79In France the ESF supported the RSA trials by financing “many supporting programmes set up at local level. At the time the RSA was developed [local authorities] were asking for money in addition to the ESF” (E2). Once the law was definitively passed, the RSA continued to be supported by the EU. In 2009, 39% of ESF spending on France was monopolised by initiatives for inclusion in the labour market, [157] mainly measures for “supporting the most vulnerable persons, implemented via the RSA” (E2). In the Netherlands too, programs aiming to accelerate the return to work were financed by the ESF when the WWB was launched in 2004. [158] What does this support represent in terms of local budgets? The Netherlands is one of the countries in Europe where ESF investment is the lowest; the ESF encouraged the “activation” of the Dutch welfare regime [159] but it only acted as an additional lever and not a primary source of funding. [160] In France, welfare cost 6.3 billion euros in 2007 [161] and the funding of the ESF was then 700 million per year – 11% of the total budget. It was in Spain above all that the ESF funding was the most decisive. Between 2007 and 2013, Spain was to receive 8 billion euros from the ESF – of which 285 million euros were allocated to Catalonia. In terms of the total budget devoted to active labour market policies in Catalonia, the ESF funding appears fundamental for local authorities:

80

“The European budget provides 25% of spending on active labour market policies in Catalonia”.

81When one quarter of “active” social expenditure is financially dependent on the ESF, it is logical that local politico-administrative officers are very attentive to the requirements of the Fund and apply its guidelines carefully. One interviewee in the Catalan government declared that, “it is clear that the active labour market policies in Catalonia stem from the European Strategy” (E41). When I asked if it would be possible for the autonomous community to decide not to apply the principles of the EES, reactions ranged from outraged denial to simple incomprehension of the question. [163] For the officers in the department of labour at the Generalitat, it seemed incongruous to want to distance oneself from European recommendations. [164] The ESF, after all, does not finance any type of social reform: all the Fund’s interventions must be linked to the labour market. [165] It is the European Commission which administers the ESF, determines its “strategic direction” and establishes its “axes of intervention”. [166] In keeping with Article 4 of the ESF regulations, member states should ensure that “actions set out in their operational programmes are consistent with, and respond to, the challenges identified in their national reform programmes, as well as, where relevant, in their other national strategies that aim to fight unemployment, poverty and social exclusion, and also in the relevant Council recommendations adopted in accordance with Article 148(4) TFEU, in order to contribute to achieving the headline targets of the Europe 2020 strategy on employment, education and poverty reduction”. [167] On the five axes of intervention of the ESF in Catalonia [168] the axis dedicated to “employability” alone attracts 50% of the European subsidies. [169] On the other hand, initiatives that do not conform to the strategic direction of the Commission are not fundable. Those responsible for the ESF programs in the Catalan community have stressed that an “unconditional measure of financial support for unemployed people” could not be supported by the ESF. [170] In only promoting “active” labour market measures, the ESF encourages local officials to accelerate the activation of their social protection regimes, to encourage programmes that are conditional and directed towards the labour market (improvement of “employability” and “adaptability”, incentives to return to work and to self-employment), rather than unconditional revenue support programs. As the head of the ESF unit at the DG for Employment and Social Affairs at the European Commission confirmed:

82

“Let’s say that we don’t do social policy. In other words, I don’t know… the point is not to put people in social housing, or give them food rations. But by training initiatives, leading to a job. […] In all that the Social Fund does, the objective is to find, to lead to work. We act within the framework of the labour market.”

83* * *

84Evidently, member states retain their jurisdiction in social matters and the distribution of roles remains strictly limited by successive treaties. Evidently, member states continue to experiment with differentiated social and employment policies, a legacy of their specific welfare state regimes, administrative traditions, and socio-political negotiations. Nonetheless, it is noticeable that the policy instruments for revenue support, as well as the causal conceptions and preferred solutions for combating unemployment and poverty have tended to converge in France, Spain and the Netherlands since the beginning of the twenty-first century. In the three reforms studied here, people in poverty are increasingly considered as utilitarian actors whose choice to return to work, or not, depends on the comparative benefit of each option. The aim of the RSA in France, the Ayuda Complementaria de Inserción Laboral in Catalonia, and the Work First programs in the Netherlands is increasingly to incentivise those receiving welfare payments, in the hope that by increasing the financial gap between benefits without work and revenues from work, they will hasten claimants’ return to work. These reforms are hybrid entities borne of specific socio-political systems and logics on the one hand, and comparable “activation” programs which aim to to “make work pay” on the other. The concept of “convergence” takes into account the diversity of points of departure and situations at a point in time (t), whilst observing a direction, or shift, towards instruments, objectives, procedures and cognitive conceptions that are merging within Europe.

85Once we accept this twofold observation of a lack of EU jurisdiction and the instrumental and cognitive convergence between European countries, the question remains of what is responsible for this change. If these three states have made comparable and synchronised reforms to their public policy systems, perhaps the reason for this can be found in horizontal transfers, without the EU being involved in the process at all? Without rejecting an explanation of this change as a result of infra-European (domestic logic) and supra-European (supranational pressure, shared economic challenges) variables, my research reveals clear elements that indicate the specific impact of the European variable on these social reforms. We have seen that the guidelines and recommendations of the Commission proposed a coherent and structured strategy around objectives, instruments and algorithms inspired by the neoclassical theory of the labour market. These cognitive schema and public policy solutions have spread throughout national Europeanised elites, who use them to argue for and justify these “active” reforms at the domestic level. Europeanisation has been an interactive process that has assumed four main dimensions in the cases we have studied: the European socialisation of national actors, the changing constraints and resources in internal political affairs, the organisation of horizontal transfers by the Commission, and financial incentives. Although they are not formally binding, these flexible instruments of European governance have cumulative effects on social reforms. For example in Catalonia we have seen that it was the conjunction of European interregional exchange programs, the assimilation of European vocabulary and theory on “active policies”, the conditionality of ESF funding, and local budgetary issues that led to the decision by the Department of Employment of the Generalitat to “activate” the Renta Mínima de Inserción.

86European initiatives have an impact on local public policy – through the diffusion of norms, cognitive schema, and public policy solutions as well as through providing direction and financial incentives. When studied and theorised as an interactive process, Europeanisation informs us as to the nature of political power today and the ways it attempts to influence behaviour, through conditionality and incentives as well as via legal or police constraint. [171] For this reason we believe that the concept of “Europeanisation” remains useful in working towards the normalisation [172] of Europe as an object of study, both top-down and bottom-up.

87Top-down, would mean analysing what “Europeanisation” teaches us about contemporary forms of governmentality, beyond the European context. Bottom-up would mean analysing public policy, actors and the meaning of reforms at the national level, before analysing causalities and re-introducing – among other potential explanatory factors – the European hypothesis.