Abstract
The European Employment Strategy (EES) has opened up new dynamics of Europeanization in the area of social policy. This article proposes to pay more attention to national core executives and their strategic use of the EES when considering its impact. Through core executives, the EES may not only affect domestic policy agendas, but also decision-making on reform. A comparative case study of activating employment policy reforms in the Netherlands and Germany evaluates expectations about how these agents upload and download ideas to and from the EU level. The findings indicate that uploading is facilitated by holding the EU presidency and a good fit with EES prescriptions, while downloading does not seem to depend on prior uploading and degree of fit but on other domestic factors. True strategic use of the EES by core executives following a sequence of uploading and downloading appears to be contingent on several contextual factors.
Introduction
Since the late 1990s, the EU has started to play a greater role in the area of social policy. Although the bulk of legislation is still made by member states, social policy-making no longer takes place in splendid isolation. In contrast to ‘traditional’ ways of EU policy-making by directives and regulations, EU social policy has been associated with new modes of governance, such as the Open Method of Coordination (OMC) (Daly, 2006). The aim of this article is to shed more light on the nexus between EU-level and domestic social policy. We argue that a focus on national core executives and their strategic use of EU-level policy enhances our understanding of Europeanization processes in the context of the OMC.
The OMC’s prototype is the European Employment Strategy (EES), formalized in the employment title of the Amsterdam Treaty. The EES requires member states to draw up annual National Reform Programmes (NRPs, formerly National Action Programmes, NAPs) for employment policy in line with guidelines proposed by the EU Commission and adopted by the Council of Ministers. These guidelines have been the basis for an iterative and legally non-binding process involving monitoring, benchmarking and peer pressure. 1 Taking this particular route of governance was designed to help transform Europe’s social protection systems from safety nets defending acquired skills and social status into springboards for new skills and new jobs. To reach this goal, the EES included rather specific targets whereby member states should gradually reach the average of their three most advanced peers.
As a new mode of governance, the EES has opened up new dynamics of Europeanization in the area of social policy. We argue that to understand these dynamics, one needs to take the role of national core executives into account: the EES guidelines may be influenced by their ideas on policy content and these executives may subsequently refer to the EES strategically to legitimize domestic reforms. This kind of perspective is innovative in two respects. First, it suggests that the OMC not only has an impact on the early phases of national policy cycles (Lopez-Santana, 2006), but also, via prior influence on the EES (the ‘European route’ of input), on national decision-making processes. Second, it directs attention to how agency may relate to the OMC’s cognitive dimension, to which many scholars ascribe cognitive and discursive shifts in member states (Heidenreich and Zeitlin, 2009).
The following section situates our argument in the literature on Europeanization and the OMC/EES and develops the case for considering core executives as a nexus between EU-level policy and domestic policy-making. We then formulate some hypotheses on how these executives influence and/or make strategic use of EU-level policy and present some methodological considerations. The third and fourth sections examine whether Dutch and German core executives influenced and/or made use of EU-level social policy in major activation reforms after the launch of the EES. The conclusion highlights that in both countries, EES-related arguments added an extra justification to activating changes of unemployment policy – which differed in their intensity and timing – while uploading depended on specific enabling factors.
Europeanization and the EES
Since the debates on the nature of European integration between intergovernmentalists (Moravcsik, 1998; Taylor, 1991) and supranationalists (Weiler, 1981), consensus has grown for the idea that the relationship between the EU and its member states is not only ‘bottom-up’ or ‘top-down’ in character and that it varies across policy fields and over time. Member states are not merely passive takers of EU-level policy but affect the goals and instruments of policy at an earlier stage. Thus, Europeanization could be seen as a ‘two-level game’ (Putnam, 1988), where a systematic relationship emerges between domestic and EU politics with governments (and other actors) in between. Such a conceptualization attempts to connect the ‘shaping’ (or decision-making) and the ‘taking’ (or implementation) phases of EU policy, which allows a better grasp of the relationship between member states and the EU (Börzel, 2002).
Joining EES (the first OMC) to the EU’s repertoire of policy instruments, by now ‘an established form of socio-economic governance at EU level’ (Tholoniat, 2010), also added a new dimension to research on Europeanization. Much of the work undertaken on the OMC to date reflects a divide between optimists and pessimists concerning its effectiveness and legitimacy (Büchs, 2009; Kröger, 2009), which we cannot discuss here further. In contrast to the community method, where the European Court of Justice can use coercion in case of non- compliance (such as infringement processes), compliance with the EES is voluntary as it relies on ‘soft’ ways of monitoring and enforcing commonly agreed objectives (Trubek and Trubek, 2005). Therefore, the ‘implementation’ of non-binding regulation can hardly be compared with compliance with traditional EU law with its institutionalized, rigid prescriptions and binding legislation (Lopez-Santana, 2007: 5). Rather, according to a typology of governance modes with emphasis on the policy dimension (Treib et al., 2007: 14), the EES might be classified as an example of ‘voluntarism’, combining a non-binding instrument and flexible implementation.
Importantly for our purposes, the EES offers opportunities for member states to carry out reforms in line with their preferences and strategic objectives as long as these are congruent with the EES. Recent empirical work on the influence of the OMC more generally has found evidence of creative appropriation by domestic actors for the purposes of governments (Heidenreich and Zeitlin, 2009). Because of its open, deliberative and negotiated character, the OMC can serve as a ‘creative resource’ for member states by providing them with formal and informal platforms, networks and (limited) financial support through the European Social Fund (Van Gerven and Beckers, 2009: 78). Moreover, as the OMC process involves periodic revision of guidelines, governments have the opportunity to influence them in line with their preferences. But how and when does the OMC matter to domestic policy-making and where is political agency located?
Core executives as nexus between EU and domestic social policy-making
With regard to the ‘how’ question, it is widely argued that ‘soft’ forms of governance have an impact on policy through processes of socialization and learning (for example, Sabel and Zeitlin, 2008; Visser, 2009). However, the lack of convincing empirical proof of policy learning and problems of definition leaves critics to stress that claims about learning remain tentative (Kröger, 2009). Yet, the exchange of ideas and the dissemination of certain discourses about reform are deemed important for the impact of the OMC. Some scholars claim that through discourse and framing, policy ideas influence policy adoption in their own right (Schmidt, 2000; Radaelli, 2000).
To better understand how the cognitive content of the EES affects domestic policy-making, we propose to focus on those actors who typically initiate reforms and subsequently legitimize and gather support for them, that is, cabinet ministers. They possess the authority to launch policy proposals and are likely to convey policy ideas both in the process of re-formulating OMC objectives and when publicly defending reforms in domestic arenas. Ministers have been shown to successfully adopt welfare state reforms by stressing cognitive and normative aspects of policy ideas (Stiller, 2007, 2010). Moreover, the OMC has engendered an increase of executive federalism, that is, a channelling of activity toward the executive branch of government (Duina and Raunio, 2007). While cabinet members are surely not the only actors who try to influence the OMC process (the same goes for national and EU civil servants and other members of the labour market epistemic community), we argue that their role stands out because of (1) their final responsibility for member state input to the EES, and, (2) their capacity to influence domestic reform agendas.
This focus on core executives also indicates when the OMC matters to domestic policy-making. The periodic revision of the EES offers opportunities for cabinet ministers to push their policy agendas. Moreover, they are the ones who can use the EES strategically, that is, by legitimizing domestic reforms with policy ideas inspired by the EES guidelines. For instance, core executives have made explicit reference to EU policies to overcome domestic opposition to pension reform in Italy (Stiller, 2006) and in other social policy reforms in Southern and Eastern European countries (Kvist and Saari, 2007). Our argument adds to earlier evidence on the OMC’s impact on the first phases of the national policy cycle, where it facilitates the inclusion of a policy item on the domestic agenda and/or reformulation of existing policies and institutions (Lopez-Santana, 2006: 482). We specify how the EES, through the role of core executives, could have an impact both on the phase preceding the national policy cycle and on the national decision-making phase. In so doing, the argument adapts the meaning of Europeanization to the context of the OMC/EES by stressing the role of time and sequence in policy-making in the following way. The cyclical EES process provides cabinet ministers with periodic opportunities to influence the guidelines during the recurring review of objectives. Accordingly, we hypothesize that cabinet ministers at time T1 transfer policy ideas held at the domestic level to the EU level. Building upon the literature on the role of actors and ideas in policy-making, we also expect that ministers subsequently use the EES guidelines to legitimize controversial domestic reforms. The corresponding hypothesis states that cabinet ministers at time T2 make use of EU-level ideas to facilitate the adoption of domestic reforms. In sum, the EES process may serve as a cognitive and strategic resource to core executives: to push their policy agendas at the EU level by influencing the cyclical OMC process, and, at a later stage, to legitimize reforms once they have been put on the domestic policy agenda.
Methodological considerations
Our criteria for concluding when uploading or downloading has (not) occurred are as follows. First, following Börzel (2002: 195), we see uploading as a process of change initiated by core executives followed by an actual change at the EU level. In our analysis, uploading has occurred when core executives have transferred ideas about activation of labour market policy to the EES guidelines. Second, drawing on Radaelli (2004: 10) we argue that downloading has occurred if a process of policy change at the domestic level is preceded by usage of the EES. Usage means that core executives either legitimize reforms of policies with direct (EES) or indirect (circumscribing the EES) reference to the EES, 2 or point to the availability of financial resources of the European Social Fund (ESF) in accordance with the argument about the OMC being a ‘creative resource’ stated above. We acknowledge that empirical analysis of our hypotheses is not straightforward. Evidence of uploaded ideas can be at best partial or indirect, as many other factors may influence the cognitive content of EES guidelines and objectives. Furthermore, research confronts ‘an administrative merger of EU and national levels’ (Van Keulen, 2006), which makes it unclear which actors and institutions, at different levels and in different phases of policy- making, are responsible for which policy outcomes. Similarly, it is difficult to find clear-cut evidence for the downloading of ideas from the EU level for justifying domestic reforms. Actors may use ideas underlying the EES guidelines without explicit acknowledgment or hold them even before the EES was adopted. In our analysis, we obviously cannot resolve all these problems but we seek to present as much evidence as possible, resorting to indirect evidence if needed.
Our cases are the adoption and implementation of the first and second EES guidelines in the Netherlands and Germany. These guidelines contained the activation approach towards unemployment and encouraged member states to offer job seekers a new start after a certain period of unemployment. 3 The time period considered lasts from 1997, when the guidelines were formulated, until 2003, when both countries had passed legislation to implement them. We focus on the issue of activation since it is considered to be the first paradigm change visible in EU social policy. The two countries selected to illustrate our argument traditionally shared broadly similar corporatist structures and Bismarckian welfare state institutions (Haverland, 2001), yet they differed on their approach to activation when the EES was launched. The Netherlands had embraced activation for some groups even before the EES, ensuring a relatively good fit with the corresponding part of the guidelines. In contrast, Germany re-orientated its labour market policy much later towards activation, which made for a relatively large policy gap with the guidelines. We assume that these initial differences are likely to make for quite different experiences with core executives using the EES strategically. Specifically, in the Netherlands we should see more uploading of ideas concerning activation because of pre-existing experience with activation measures, while Germany can be expected to upload other types of ideas concerning unemployment policy. In this vein, the difference between these countries may be rather qualitative than quantitative. Furthermore, we expect that downloading should be more likely in Germany than in the Netherlands because of the larger policy gap when the EES was introduced. Here, the differences between countries may well be of a quantitative rather than a qualitative nature. We admit that these two cases only allow a preliminary exploration of our hypotheses but may reveal contextual factors that influence up- or downloading. The data comprise 19 interviews with Dutch policy-makers, trade unionists and civil servants (both at a national and EU level) and were conducted in the framework of studies evaluating the effect of EES on Dutch employment policies (Van Gerven, 2008a; see also Zijl et al., 2002 for an earlier study). For the German case, 15 respondents from political parties, ministries, trade unions and labour market policy experts were interviewed between 2005 and 2006, primarily for a study of the role of cabinet ministers in overcoming reform obstacles (Stiller, 2007). Most interviewees were dealing both with the EES and domestic policy in their work, which is why we do not expect a bias in responses judging the strategy’s impact. Furthermore, secondary literature, speeches and interviews of core executives, and official policy documents also served as sources.
Fighting against the windmills: Minister Melkert and Dutch activation policies
The Dutch system of social security has traditionally been characterized by a strong Bismarckian orientation including earnings replacement and protection of male workers (Van Gerven, 2008b). To date, the Dutch unemployment benefit programme consists of two pillars, unemployment insurance (Werkloosheidsverzekering, WW) and unemployment (social) assistance (Bijstand, or WWB). Within the unemployment insurance scheme, workers are covered by generous insurance benefits that are administered publicly. Apart from requirements to register at the employment office and to accept a job or training offer, the WW contained very few provisions related to activation up to the mid-2000s. Those unemployed ineligible for WW benefits can make a claim to means-tested assistance under the WWB scheme. In the mid-1980s, the overall number of benefit recipients was increasing rapidly, which led to a gradual increase in activation measures. 4 However, the initial activation measures were directed at recipients of the unemployment (and social) assistance benefits, not at claimants of social insurance benefits. In particular, cures for the ‘Dutch disease’ of low economic growth and high levels of social security beneficiaries were sought in activating the young and long-term unemployed. In 1984, subsidized temporary work programmes for the youth (Jeugdontplooingsbanen) had been already introduced and the idea of creating temporary jobs for unemployed youth was included in the coalition agreement of the Lubbers cabinet of Christian Democrats and the Liberal Party (1986–9). Subsequently, a plan for youth employment (Jeugdwerkgarantieplan) was drawn up with the Association of Dutch Municipalities and representatives of employer and employee organizations. In 1991, existing piecemeal measures were formalized under the Guaranteed Work for Youth Act (JWG), which codified the nationwide approach for the activation of those up to 23 years of age, providing them with temporary part-time jobs or training during the first 6 months of unemployment.
After electoral defeat of Lubbers III, Prime Minister Wim Kok’s ‘purple’ coalition of Social Democrats and two liberal parties took office in 1994. 5 The core of its socioeconomic policies and the main goal of Minister of Social Affairs and Employment, Ad Melkert (1994–8), was to increase labour market participation. Although previous governments had been rather keen on the idea of activation since the mid-1980s, social insurance remained relatively transfer-oriented, except for benefits for the young and long-term unemployed. Next to youth programmes, subsidized jobs (Melkertbanen) had also been created for the long-term unemployed in 1996. The dominant principle underlying Dutch unemployment protection prior to the EES was status maintenance through generous insurance benefits with little concern for activation of adult workers. Since 1994, the Dutch Public Employment Services (PES) applied a diagnosis instrument for jobseekers, measuring their distance from the labour market (Van der Meer and Visser, 2004: 49), yet the majority of activation measures (training schemes, job mediation, subsidized jobs and work experience places) were aimed at the long-term unemployed. In contrast, the PES offered little guidance to newly unemployed people except for registration as jobseekers and recording vacancies.
Indeed, the Dutch political climate in the mid- 1990s was hostile towards extending activation to all adult workers. The coalition parties, the Netherlands Bureau of Economic Policy Research (CPB) and the Ministries of Economic Affairs and Social Affairs and Employment were opposed because of cost considerations (interviews EU2, NL8 and NL9). Despite high unemployment figures in the early 1990s, it was difficult to find support for measures that increased social expenditure. Rather, the Ministry of Finance argued that resources should be directed at unemployed youth and the most disadvantaged long-term unemployed (Visser, 2005). Notwithstanding this unfavourable context, Minister Melkert had a vision to extend activation to all adult workers and he kept on pushing his ideas despite the opposition within the coalition, and by his own civil servants. According to one EU-level interviewee (interview EU2), Melkert was ‘a man with a clear vision in the sense that he had a goal, the motivation, and he was able to articulate his ideas’.
Uploading ideas: strategic use of the EU during the agenda-setting phase
Melkert wanted to change Dutch social policy from ‘being a hammock to a springboard’ (interview NL4). Together with his fellow Dutchman Hans Borstlap, chairman of the European Labour Market Committee (the predecessor of the Employment Committee), Melkert sought to introduce the ‘comprehensive approach’ to jobseekers through the European route (interviews NL1, EU2). Other interviewees indicated that he did not pursue his ideas in order to merely improve the European social model. Rather, his intentions were more pragmatic and strategic: after encountering opposition to his plans at home, Melkert turned to Brussels (interviews EU2 and NL7). Most opposition came from his liberal coalition partners, who held the Ministries of Finance and Economic Affairs. Their hesitation was rooted in fear of pressure to increase European funds and a view of the EES as being overly bureaucratic. Instead, they favoured the Broad Economic Policy Guidelines and the Economic and Financial Affairs Council (ECOFIN)’s guardianship of the Stability and Growth Pact, and they preferred the views of the Organisation for Economic Co-operation and Development (OECD) on labour market and employment policy over those of the EU (interviews NL4, NL6, NL7). 6 As the fragmented nature of the Dutch political system (Van Keulen, 2006) gives ministries rather strong departmental autonomy, this facilitates the promotion of novel ideas and innovations at the European level. Using this autonomy made it possible for Melkert to push his goals forcefully and successfully when faced with strong opposition by the other ministries to implement his plans through domestic policy-making. 7
Notably, Melkert was offered an excellent window of opportunity to promote his goals during the Dutch EU presidency in 1997. The Intergovernmental Conference on Social Policy and Economic Performance was an important moment for the Dutch to push their vision of social policy as a productive factor. That vision saw social policy as contributing to rather than inhibiting economic performance, as is commonly argued in the economic sphere. Inherent to this view was that, in addition to cutting costs, modernizing the welfare state would necessitate social policies that increase labour market participation. At the conference, top EU social policy-makers met with representatives of trade unions and employers’ organizations and welfare state scholars. The outcome was a strengthening of convictions about the economic value of social policy (Hemerijck, 1998), as confirmed by our interviewees (interviews EU2, NL1, NL2 and NL3). According to one EU-level respondent (interview EU5), ‘for the Employment Chapter in the Amsterdam Treaty it was crucial that Melkert came up with the argument “labour market policy is economic policy”’ (see also Van Riel and Van der Meer, 2002, for the important role of Melkert in the advocacy coalition behind the emergence of the EES). Another vital factor in this change was the political climate in Europe, as many countries, including the UK and France (later in 1998 to be joined by Germany), had centre-left governments in place. In addition, the EU accession of some Nordic countries had path-breaking effects: to reassure Finland and Sweden that EU enlargement would not end the social democratic model, active labour market policy and the goal of full employment would have to be ‘exported’ to the EU. Arguably, these favourable circumstances made it much easier to achieve member state support for the proposed employment chapter. Based on the foregoing, we conclude that Melkert, armed with a clear vision, good examples of youth activation, a favourable venue during the Dutch Presidency, strategic allies such as the chairman of the European Labour Market Committee, and important pro-welfare state governments in power, managed to upload his idea on the comprehensive approach to the EES guidelines.
Downloading ideas: strategic use of the EES to introduce the comprehensive approach
Melkert’s term as minister ended in August 1998 and he quickly withdrew from social policy matters thereafter, although remaining parliamentary party leader of the Social Democrats. Still, he was given credit for engineering a turning point in labour market policy by consistently arguing for more activation, as our interviews indicated. In the 1998 coalition agreement, the second Kok Government embraced the idea of extending activation to adult workers at risk of long-term unemployment and of modernizing unemployment insurance and employment services. It did so by explicitly referring to the agreement at the 1997 Luxembourg European Council (the beginning of the EES) about employment growth and fighting long-term unemployment. In his speeches prior to this paradigm change, Melkert rarely referred to the EU, and if doing so, he emphasized that the Netherlands had played an important role in (European) decision- making. ‘We have made an important decision on employment at the European Top in Luxembourg last year … . Europe has entrusted us to extend in the next five years the comprehensive approach to all jobseekers’. 8 Although politicians and civil servants saw the importance of complying with the EES, interviews (and policy documents, see next paragraph) show that extending activation to a wider group was considered as a ‘national’ innovation. 9 The EES process, and the preparation of NAPs in particular, were seen to have speeded up the actual implementation of EES guidelines 1 and 2, and consequently, the extension of activation measures to all newly unemployed (interviews NL1, NL2 and NL3). As two respondents put it (interviews NL8 and NL11): ‘If the EES had not been in place, the Ministry would never have had the money … . Before the EES, politicians were not ready to spend money on the newly unemployed.’
With an additional financing of 250 million guilders by the government, the comprehensive approach for adult workers was introduced by an administrative reform and fully implemented in 2002, one year earlier than proposed by the coalition agreement. Next to government funding, the costs of the reform were borne by the social partners and the EU, through ESF funding. As to the usage of the EES during the decision-making process, few direct references can be found in government documents. For instance, a letter from Minister Melkert to Parliament, which set the scene for the extension of the comprehensive approach, made no explicit reference to the EES, but it does refer to ESF funding (Tweede Kamer, 1998a). Similarly, the 1998 coalition agreement had stated that government expenditure could be limited not only because of expected savings on benefits but also because of resorting to the ESF (Tweede Kamer, 1998b). Regarding the status of EU-related arguments, both the coalition agreement and the Ministry’s policy programme for 1999 (Ministerie SZW, 1998) stress that the necessity for extending activation was rooted in domestic problems, that is, the failure of existing policy and the socio-economic situation. Interviewees from the civil service respondents echo this view, arguing, for instance, that ‘national policies remain national affairs. Civil servants are influenced by matters coming from Brussels, knowledge spreads around Europe, but Dutch policy is formulated by the Dutch government’ (interview NL4). Where references to the EES were made, they served as an additional argument to legitimize policies that would have been introduced anyway. As some interviewees put it (interview NL4, NL8), ‘they [EU social policies] strengthen the case to be presented before Parliament … as a civil servant, you seek support for the things you want to do, and if such support is lent by the OECD or EU, so be it … the EES offers justification for national activities … But it is solely an extra argument. National policy-making dominates’.
To conclude, Dutch Minister Ad Melkert used the European route to get the issue of activating adult workers out of a blind alley. Although Dutch governments since the 1980s had been rather keen on activation in general – irrespective of changes in government composition – a translation into concrete policies faced major political opposition. It took vigorous action by a politician with a clear vision to overcome this resistance. Subsequently, unemployment protection was underpinned not only by the income compensation principle, but also by improving people’s chances to find employment through the comprehensive approach. Our expectation about the uploading of ideas about active labour market policy from the domestic level holds in the Dutch case. Melkert succeeded in placing ideas about the comprehensive approach to unemployment on the EES agenda. With respect to our expectation on downloading ideas from the EES, we found that its extent was limited compared with other arguments for reform. Minister Melkert made few direct references to the EU in the domestic policy process. However, indirectly, the EES helped the Kok Government to speed up activation reforms. It provided core executives with an extra justification before parliament and against opposition to extend activation and helped the Ministry to finance such extension.
Belated activation in Germany: the Hartz IV reform and Minister Clement
Until well into the 1990s, German unemployment insurance was predominantly transfer-oriented, seeking to maintain the former living standard of the unemployed. 10 It was only after 1996 that the German government enacted reforms that started to change the character of this policy towards a more active stance (Büchs and Hinrichs, 2007). Subsequently, the Red–Green Coalition (1998–2002) introduced the 2001 Job-Aqtiv Act, which thoroughly reformed active labour market measures and introduced some degree of activation, such as profiling unemployment risks, ‘inclusion agreements’ and extending possibilities for job-creating measures. After its re-election in 2002, the Hartz reforms added a truly activating dimension to the unemployment protection system. Here, the ‘Hartz IV’ reform, which fundamentally changed existing arrangements, was most notable. Until then, German unemployment protection had consisted of two pillars: unemployment insurance (Arbeitslosengeld) and unemployment assistance (Arbeitslosenhilfe). The social assistance scheme (Sozialhilfe), the de facto third pillar, provided a guaranteed minimum income available to anyone in need if registered as unemployed or not. The tendency among these pillars to shift responsibilities for recipients back and forth slowly led to the consensus that further co-existence of the three programmes was untenable. Moreover, German reunification and massive shifts on the East German labour market had contributed to successive incremental reforms of the unemployment benefit schemes during the 1990s. Since the mid-1990s, the obligations of job-seekers to seek work and accept job offers and placements in activation programmes as well as sanctions for non- compliance had been extended across pillars without solving the underlying problem.
Hartz IV was the first attempt at a comprehensive reform. It merged unemployment and social assistance benefits to create a single scheme called ‘secondary unemployment benefit’ (Arbeitslosengeld II or ALGII). Since 2005, the scheme has provided means-tested benefits for households of registered unemployed who do not qualify for unemployment insurance benefits. 11 Regarding activation, requirements for jobseekers with respect to accepting suitable job offers were tightened, as those eligible for the new benefit must in principle accept any job offer (with some exceptions). In addition, harsher sanctions for those refusing job or training offers, especially for young people, were created. Together with the Job-Aqtiv Law, Hartz IV changed passive unemployment protection into a far more activating system of unemployment provision, implementing the comprehensive approach specified in the first and second EES guidelines. For the passing of Hartz IV, the involvement of Minister Wolfgang Clement was very important. Following the re-election of the Red–Green Coalition, he was asked to head the newly created Ministry of Economics and Labour Affairs. Chancellor Schröder’s strategy was to reinforce this organizational change with a strong ‘modernizer’ to tackle the huge reform agenda in labour market policy. Clement, known within the Social Democrats as a business-friendly reformist, took on the task with great commitment. His biggest challenge was to implement the recommendations of the Hartz Commission in a difficult political context. This included high unemployment, bad performance of the Social Democrats in the polls and internal resistance from the left party faction to far-reaching reforms, and a powerful conservative–liberal opposition (holding the majority in the Bundesrat) keen to sharpen the Ministry’s reform proposal. Had previous German governments tried to influence the EES, and did Clement (and Schröder) use EU-related arguments in order to pass Hartz IV?
Uploading ideas for the EES?
Before the EES was created, activation was hardly on the agenda of the conservative-liberal Kohl Government (1982–98). Therefore, it is highly unlikely that ideas about activation – beyond restricting and cutting back benefits – were brought to the EU-level by then Minister of Labour and Social Affairs Norbert Blüm. At the same time, the position of German governments has been generally positive towards EU social policy, including the EES (Büchs and Hinrichs, 2007). At the earliest, evidence for the uploading of ideas might have been found during the 1999 German EU presidency under the new Red–Green Coalition. In this period, the EU’s agenda featured the introduction of the Euro, the conclusion of Agenda 2000 on agriculture and financing reforms, the strengthening of the Common Foreign and Security Policy and the resignation of the Santer Commission. In addition, one of the declared goals of the German presidency was to fight mass unemployment. By proposing a new ‘Employment Pact’, Germany sought to intensify member state co-ordination and to complement measures taken on a national basis to raise overall employment levels (Verheugen, 1999). The pact was supported by a new macroeconomic dialogue between the social partners and the European Central Bank. However noteworthy, these initiatives resembled practices of the German social market economy rather than initiatives furthering activation, as the latter was not yet on the German policy agenda at the time. Therefore, there is neither evidence for the uploading of ideas about activation to the EU-level during the late 1990s, nor actors comparable to Minister Melkert in the Dutch case. As for more general policy ideas about promoting employment, it can be argued that Germany was keen on keeping EES objectives in line with the principle of its social market economy, by stressing the value of subsidiarity (Büchs and Hinrichs, 2007: 23).
Downloading ideas for activation: from the Job-Aqtiv Act to the Hartz IV reform
Until the early 2000s, the Red–Green Coalition made labour market policy with a distinct Social Democratic flavour, combining social security transfers with active labour market measures and tripartism at the EU level. Yet growing unemployment and a rising urgency to act led the government to rethink policy, including the emergence of themes as activation, employability and ‘flexicurity’. Since then, German governments have referred to the EES to support reform initiatives stressing those themes (Büchs and Hinrichs, 2007). For instance, the 2001 Job-Aqtiv Act largely corresponded to measures in the EES guidelines, unlike previous labour market reforms (Zohlnhöfer and Ostheim, 2005: 156). As some of its measures, for example, accentuating the preventive approach, were in line with the general direction of policy but had been absent in coalition parties’ election manifestos, the EES arguably had had ‘intensifier effects’.
Two years later, Hartz IV, the core of the Hartz reforms, became the greatest single step towards activation so far. The Ministry’s promotion slogan ‘Fördern und Fordern’ (‘carrot and stick’) denoted a cognitive and normative reorientation of labour market policy, amounting to a paradigm change in thinking about jobseekers’ rights and obligations. Accepting work below previous levels of training was seen to be ‘more just’ than granting unemployment benefits over an extended period of time. Minister Clement played a key role in getting Hartz IV adopted, supported by Chancellor Schröder. The Minister both pushed for ideational change in labour market policy and engineered political majorities in Parliament for the particular way of implementation championed by the Ministry. He provoked resistance by left-wing Social Democrats and trade unions, which criticized – but could not prevent – the tightening of acceptability criteria and growing demands on jobseekers. He also fairly successfully negotiated with the conservative-liberal opposition, which had criticized (and delayed) the PES-centred implementation plan and achieved even tighter activation criteria.
Against this background, to what extent did German core executives use the EES to justify this reform? 12 As for direct and indirect references, Chancellor Schröder presented his reform programme, Agenda 2010 – including Hartz IV – as a comprehensive answer to the EU Lisbon Strategy in documents presented in a European context (Deutscher Bundesrat, 2003a; Deutscher Bundestag, 2003b, cited in Büchs and Hinrichs, 2007). In addition, the Ministry’s draft bill referred to the EES, albeit as an additional justification (Deutscher Bundestag, 2003c: 44): ‘this labour market reform contributes to the long-term goal of achieving full employment … according to the objectives of the EU employment guidelines’. Notwithstanding these references to the EES in documents, Minister Clement consistently cited the failing existing scheme as the main reason for Hartz IV. His argumentation vis-à-vis Parliament and the media focused on domestic necessities instead of acknowledging the role of the EES: ‘We will merge the two systems because their co-existence is inefficient and debilitates employment … . The most important goal of our reform – one cannot stress it often enough – is a faster and more precisely tailored placement of people fit to work’ (Clement, 8 April 2003 13 ). The focus on domestic reasons is echoed by interviewees who closely observed the reform process. As decisive factors in the passing of the reform, they most often mentioned ‘high and growing unemployment’, ‘inefficiencies from the coexistence of two schemes’ and ‘urgency to act politically’, but hardly the EU/EES, reflecting the clear dominance of domestic factors over the EES in the reform debate.
At the same time, Clement repeatedly called for a normative redefinition of the goal of social justice: ‘“Fördern and Fordern” at a one-stop agency is more efficient and more equitable than anything we have achieved in labour market policy so far … who does not make an effort, should not be sustained by the grace of the public. Social justice is no one-way street’ (Clement, 1 July 2004 14 ). Such arguments tell us that core executives were trying to change the existing frame of thinking about unemployment, in line with EES prescriptions. Rethinking the meaning of social justice reflects a paradigm shift towards employability (Preunkert and Zirra, 2009), through which the EES, stressing such a shift, contributed to substantial policy change. In this view, both the Job-Aqtiv Law and the Hartz reforms (including Hartz IV) strongly resembled (and referred to) EES guidelines, due to a learning effect on the part of the Ministry, owing to a high degree of readiness to change the current policy paradigm and a favourable organizational coupling of competences within the ministerial structure (Preunkert and Zirra, 2009). Several other studies contend, yet without referring to core executives, that the core of Hartz IV, merging several benefits, corresponds to the EES guidelines, even though this was not publicly acknowledged (see Behning (2006) and Heidenreich and Bischoff (2008); Zohlnhöfer and Ostheim (2007) stress EES references in the Hartz Commission report). Heidenreich and Bischoff (2008: 518) detect ‘a subtle, but nevertheless quite effective influence on German … employment strategies’, including Hartz IV, amounting to ‘learning by irritation’, that is, a mutual influence of European and national patterns of perception and behaviour (2008: 520–1). As for the use of the ESF in reform debates, this aspect of downloading could not be observed with regard to core executives of the federal government. If supportive references to the EES and the ESF were made, it was in documents of Länder ministries (in particular the less wealthy ones, for example, Berlin), whose labour market policies, separate from and complementary to federal policies, are supported by ESF funding (Büchs, 2005: 241).
To conclude, our expectation about the uploading of policy ideas about activation cannot be confirmed. Yet, Germany has clearly sought to keep the EES in line with its particular ideas about a social market economy and to preserve national jurisdiction as far as possible. In addition, we found that core executives (Schröder more so than Clement) downloaded from the EES in order to speed up reforms that they justified primarily by domestic necessities, albeit without references to ESF funding. Interestingly, while references to the EES were found in legislative texts and related documents, Minister Clement’s argumentation in favour of Hartz IV did not rely on the EES. Therefore, there was downloading by core executives, although its extent was limited compared with the wealth of domestic factors used to legitimize the reform. We also note that – in accordance with the existing literature – downloading ‘activities’ were likely preceded by processes of policy diffusion of EES content during problem-definition and agenda-setting phases to prepare the national policy arena for activation reforms.
Conclusion
This article aimed to conceptualize and empirically illustrate the ways in which national core executives may be agents of change in the context of multi-level social policy-making the EU knows today. The analysis has signalled, despite its exploratory character and its methodological limitations, that national core executives may act as a nexus between the EES and domestic policy-making on activation, advancing our understanding of Europeanization in the context of the OMC.
The uploading hypothesis held in the Dutch case, where Minister Melkert successfully positioned ideas about comprehensive activation on the EES agenda. Regarding the downloading hypothesis, we found evidence for both criteria. It gave Dutch policy-makers an extra justification to extend activation and provided them with access to extra funding to do so, making for arguments that underpinned and speeded up domestic activation reforms. Melkert indeed used the EES strategically to prepare subsequent changes in domestic policy. In contrast, the German policy agenda was still largely geared towards passive unemployment protection, and ideas about activation available for uploading were not available. German core executives used EES-related arguments as additional justification on top of systemic problems (after the EES had possibly served as a source of inspiration for core executives for how to define and solve labour market problems), while references to ESF funding were absent. Regarding the sort of arguments used during decision-making processes, we would like to note that Dutch and German governments were intent on keeping the EES from encroaching on national autonomy, observing the principle of subsidiarity. The latter possibly explains why the EES appeared as additional justification although for various reasons. Germany’s federal structure (and Länder sensitivity to subsidiarity) and strong adherence to the social market economy (Büchs and Hinrichs, 2007) goes some way towards explaining the relative reluctance to point to the EES. Somewhat differently, in the unitary state of the Netherlands, growing dissatisfaction with European integration and communication deficits of the political elite vis-à-vis citizens about EU policy (Hemerijck and Sleegers, 2007) might explain why core executives sought to keep social policy-making a national affair.
How do these findings fit in with our expectations on up- and downloading strategies given the initial differences between the two countries? As regards the policy gap between the EES and national policies, the Dutch case indicated that the positive policy experience with activation helped Minister Melkert and his allies to upload national ideas to the EU-level. At the same time, holding the EU presidency by itself was important. As a consequence of the Dutch ‘agenda-setting advantage’, the comprehensive approach towards the unemployed was more easily taken up in the EES guidelines, adjusting EU prescriptions towards Dutch wishes and being implemented some years later in the Netherlands. One could argue that the initially smaller gap between the domestic policy structure and EES prescriptions made uploading more likely to occur but would also reduce the need for downloading. Yet downloading did occur through (in-)direct references (and subordinate to domestic necessity arguments) and the argument of ESF funding. In contrast, in Germany uploading was not forthcoming during the 1999 presidency because the German debate on activation was still in its infancy, and ideas on activation did not start to permeate the policy discourse until the early 2000s. Therefore, our expectation that uploading would be less likely there than in the Netherlands proved to be correct. More interestingly, the role of downloading in the efforts of German core executives to implement activation was less prominent (and the funding argument was not used), contrary to expectations based on the large initial gap between existing policy and EES prescriptions. As indicated above, this possibly had to do with the saliency of the EU and national attitudes toward EU social policy, but the question certainly deserves further investigation.
The conditions identified above that promote downloading are both related and complementary to factors explaining variation in EES impact discussed in the recent literature. A low level of compliance (that is, the goodness of fit argument), related to the mechanism of peer pressure, and a low level of consensus between key actors on reforms and a high level of Europeanization (as evidenced by a high EES profile and a low extent of Euro-scepticism), both related to the mechanism of strategic use of the EES, are shown to facilitate its impact in four countries with widely varying employment policies different from the ones considered here (Mailand, 2008). Moreover, openness of the government towards EES and a high degree of reform necessity (Zohlnhöfer and Ostheim, 2007) are proposed to facilitate EES policy transfer. Revisiting our findings for our country cases in the light of these factors, the compliance argument did not hold in any of them, while the Europeanization argument does help to interpret the evidence in both cases (if taking specific state structures and attitudes towards subsidiarity into account). The nature of consensus on reform required in both countries varied due to different veto point structures, but in both cases consensus was reached with substantial support from domestic context-specific arguments. The relative openness of governments towards the EES was evident with Germany’s Red–Green Coalition (where reform necessity was especially visible) as well as the Dutch Purple Coalition (where reform necessity was less evident due to the better goodness of fit with EES objectives), although it did not translate into overtly strategic use of the EES, as we have shown.
To conclude, the findings show that favourable conditions for uploading include holding the EU presidency and a relatively good policy fit with EU prescriptions, while some degree of downloading seems to occur irrespective of prior uploading of ideas and degree of fit, although conditioned by a number of factors. More precisely, policy fit appears to play a greater role in uploading ideas than in downloading them, and holding the presidency encourages uploading if a country has already had positive experiences with the policy ideas it wishes to promote. Therefore, true strategic use of the EES, that is, core executives who follow a sequence of uploading and downloading is contingent on several contextual factors. Our theoretical expectation about a two-step sequence of strategies by core executives could not be confirmed by our investigation of activation reforms in both countries but only for the Netherlands. This raises the need for further and more systematic examination of our propositions on up- and downloading, including the conditions under which they apply to see whether the Dutch experience was a singular event. In doing so, it would be desirable to include a larger sample of countries, to investigate a longer time period and to use a more fine-grained operationalization of core executives’ up- and downloading strategies.
Footnotes
Acknowledgements
We wish to thank the two anonymous reviewers for their critical and constructive comments.