Actors, costs and values: the implementation of the Work-Life Balance Directive

Abstract

The European Union (EU) launched the European Pillar of Social Rights to improve social rights for EU citizens. However, little is known about the domestic dynamics of implementing these new rights. This article examines the implementation of the Work-Life Balance Directive in three member states with different policy traditions: Denmark, Germany and Poland. Based on an actor-centred approach, the article demonstrates that two main factors were crucial in motivating national actors to shape the implementation of these rights. First, the economic costs, including new permanent costs emerging from the provisions and the administrative costs of implementing an EU directive, shaped actors’ positions in all three countries. Second, the values of gender equality – whether promoting gender-equal leave or traditional family values – constituted an underlying reason for actors to customise the EU rules differently. Both factors, used as a defence or driver for change, increase differentiated policy implementation.

Keywords:

• Europeanisation
• European pillar of social rights
• work-life balance
• earmarked parental leave
• customisation

Although the European Union (EU) launched the European Pillar of Social Rights (EPSR) in 2017 to improve the social rights of EU citizens, little is known about whether or how this bold aim is achieved in practice. To begin to investigate this issue, we examine one core initiative under the EPSR, namely the Work-Life Balance Directive (WLBD). The directive promotes a more equal sharing of parental leave and aims to achieve a more gender-equal participation in the labour market through flexible employment practices (Waddington and Bell 2021). However, the WLBD was contested at the EU level by national actors and ultimately watered down, leaving increased flexibility for its implementation (de la Porte et al. 2020). While previous studies have focussed on the role of national actors in ensuring compliance (Falkner et al. 2005; Mastenbroek and Kaeding 2006; Treib 2003), they fall short in explaining precisely why and how actors shape the decision making in the implementation process.

This study contributes to filling this gap in the research and poses the following research question: Why and how do actors shape the implementation of EU rules? And to what extent do actors customise this implementation? To answer this question, we apply an actor-centred approach focussing on the role of ideas as a factor when it comes to mapping the actors’ positions and their institutional setting. We examine the implementation of the WLBD in three different countries (Denmark, Germany and Poland) based on EU and national data on the implementation as well as elite and expert interviews. We focus particularly on father-specific leave (paternity leave as well as parental leave), which has proved to be the most contested part of the WLBD (de la Porte et al. 2020). Since we are concerned with the implementation of the provisions on father-specific leave, which is at the intersection of labour market and social policy, we focus on social partners (employer organisations, trade unions), interest organisations, public administration actors, and – not least – political parties.

Our findings are threefold. First, we demonstrate that two main factors – economic costs and values of gender equality – which also dominated the EU debates, were crucial in motivating national actors to shape the implementation of these provisions, but in dissimilar ways in the three cases under examination. Second, the two factors mentioned above also constituted underlying reasons for actors to customise the EU rules – that is, the tailoring of a directive to domestic institutions and policies (Thomann 2015). Thus, while previous research demonstrates that customisation is dependent on the institutional settings (Zhelyazkova and Thomann 2022), we demonstrate how gender values and costs shape the positions of the various actors. Third, our findings suggest that the varying national customisation processes in the case of the WLBD may lead to differentiated policy implementation (Zhelyazkova et al. 2023), thereby potentially undermining the EPSR’s aim of upward social convergence.

Literature review

Since the EU is characterised by a highly decentralised implementation structure and lacks its own capacity to implement EU law, it relies on the member states to fulfil this task (Treib 2014). In the process of implementing EU directives, various actors are involved at all levels of governance, which potentially leads to increased politicisation (Falkner et al. 2005). Therefore, previous implementation research revealed that the (political) preferences of national actors and their institutional positions are crucial for implementation (Mastenbroek and Kaeding 2006; Mastenbroek and van Keulen 2006; Treib 2003). For example, some scholars have argued that a negative stance towards a directive at the EU level – expressed via oppositional voting in the Council of the EU (the Council) or by government statements – leads to an ‘opposition through the backdoor’, where member states evade compliance (Falkner et al. 2004; Pircher 2017). Focussing on actors, previous studies have analysed their role – embedded in a specific policy context – in achieving compliance (Börzel and Risse 2003; Cowles et al. 2001). For example, Börzel (2003: 36) concludes that national actors may pressure the public administration to fulfil the EU requirements and thus ensure compliance (see also Mastenbroek and Kaeding 2006). In a study which highlights the role of social partners, Falkner et al. (2005: 305–306) elucidate that the national social partners push their interests in the implementation process, but without undermining compliance. Moreover, recent studies have established that actors in member states increasingly adapt EU directives to ‘better fit’ their domestic constituencies, a phenomenon known as ‘customisation’ (Thomann 2015; Zhelyazkova and Thomann 2022).

Therefore, even though previous studies have focussed on the role of governments, partisan positions (Toshkov 2007), political parties (Treib 2003), public administration (Hobolth and Martinsen 2013), and the role of social partners (Leiber and Falkner 2004), they generally link the behaviour of national actors to compliance with EU provisions. Research on the involvement of national actors in the decision-making process during the implementation phase – after the adoption of a directive at the EU level – is more scant. Moreover, these studies concentrate less on how actors influence the implementation, given institutional contexts that may vary, and who these actors actually are (see also Duina 2007).

We aim to fill this research gap, arguing that the focus on actors has become more important in recent years based on two considerations. First, individual actors impact policy processes to a greater extent in times of increased EU politicisation (de Wilde and Lord 2016) and tendencies of differentiated integration (Schimmelfennig and Winzen 2020). Second, the focus on actors in shaping the implementation of social policies is crucial in the context of ‘constraining dissensus’, where domestic contestation over social issues is more prevalent (Martinsen and Vollaard 2014). Against this background, we argue that a more differentiated implementation of social rights, may in turn, lead to unequal outcomes in terms of take-up of rights.

An actor-centred approach to EU implementation: the Work-Life Balance Directive

Institutional factors tend to become somewhat deterministic in explaining the implementation of a certain policy (Mastenbroek and Kaeding 2006). Therefore, it is crucial that the role of agency and actors be placed at the forefront of research on policy processes (Mahoney and Thelen 2010). In contrast to perceiving EU policy making as an institutional process ‘locked into’ its own developments and routines with fixed preferences by member states (Pierson 1996), we follow the notion that ideas (in this context, values on work-life balance policy) may serve as points of contestation, pointing to the importance of agency-based mechanisms (Mahoney and Thelen 2010). We argue that the role of agency is especially important in analysing family or gender-related policies, since actors manage to transmit their ideas into influential arenas, aiming to shape the content of the policy (Béland 2016; Mätzke and Ostner 2010). These actors might come from the official political arena or from other groups that are crucial in the decision-making process, such as social partners (Falkner and Leiber 2004).

However, actors are provided with certain constraints or opportunities to shape the implementation process, which is defined by an institutional structure (Mahoney and Thelen 2010; North 1990). More specifically, actors in the implementation of EU social policies can make use of different channels of influence, namely decision making, formal consultation and lobbying (see also Falkner et al. 2005: 235). Decision making refers to social actors’ autonomy, but they can be also involved through consultation (on a directive), which can be more or less formalised in different political systems (for instance, through written opinions, statements, or hearings). Yet another type of influence is lobbying, which falls outside the formal consultation process. In addition, public administration actors are central because they are responsible for the technical details of the implementation (Mastenbroek and Martinsen 2018).

In this article, we acknowledge the various institutional settings, but adopt an actor-centred approach to analyse the implementation of the WLBD in three member states. The WLBD aims to improve the balance between paid work and private life, including caring responsibilities. It was adopted in June 2019, and member states were supposed to transpose the directive by 2 August 2022. Opposition by several member states already emerged during the EU negotiations since the WLBD introduces provisions specifically targeted fathers. In specific terms, the directive introduces 10 days of paternity leave, to be paid at the level of sick pay, and a two-month period of earmarked parental leave, with an adequate level of payment (de la Porte et al. 2020). Based on the debates at the EU level, we identify factors that are likely to mobilise national actors to engage in the implementation process and ultimately customise the EU rules. Moreover, we build on previous research (Falkner et al. 2005: 306), arguing that since the WLBD provides member states with greater flexibility in implementing the directive, it provides openings for increased agency on the part of national actors with regard to customisation (Mahoney and Thelen 2010).

The engagement of actors in customising the Work-Life Balance Directive

Previous studies established that the degree of compatibility (fit or misfit) between European and national rules (Börzel and Risse 2012; Steunenberg and Toshkov 2009; Thomson et al. 2007) is crucial for the implementation of social policies (Ferrera 2014; Vollaard and Martinsen 2014). Therefore, we present the varying degrees of policy misfit in relation to the WLBD for the countries presented in Table A1 (Appendix). All the member states, with the exception of Sweden, have had to adapt their national law to comply with the WLBD. However, we argue that the concept of policy misfit does not adequately explain how policy-makers strive for change (Knill and Lenschow 1998). In this study, we refrain from linking national actors’ involvement in the implementation process to (non-)compliance outcomes, as there is an extensive and detailed literature on this subject, which explains how member states adapt to EU rules and ultimately whether they comply with them or not (e.g. Börzel 2003; Falkner et al. 2005; Falkner and Leiber 2004; Mastenbroek and Kaeding 2006; see also Treib 2014). Instead, we analyse factors that mobilise actors to engage in national decision making.

We argue that in the case of the WLBD, there are two factors that have dominated the EU debates and are likely to engage actors in the implementation process. These factors are costs, resulting from the level of parental leave compensation, and values, resulting from the introduction of an earmarked part of leave. The two factors are connected because a high remuneration for parental leave also provides incentives for fathers to take up these rights and thus contributes to enhancing gender equality. Indeed, the level of compensation for parental leave is an important proxy for likely take-up (Duvander et al. 2022). In countries with high levels of compensation for parental leave (80% to 100% of income), the take-up of leave is high, and gender equality in relation to leave and the labour market is also higher, while in countries with lower levels of compensation (under 60% of income), the take-up is low (Karu and Tremblay 2018). Still, we argue that it is important to distinguish these two factors analytically as each generates insights into why actors shape the implementation process. For example, values can be used as a justification by actors to either support or hinder change, which we aim to examine in our analysis. Thus, by making this distinction we are also able to examine the direction of change.

Actors’ engagement based on economic costs

The first factor that motivates actors to engage in the implementation process and customise the EU rules in the case of the WLBD is the economic costs that result from the EU acting as a regulatory welfare state (Levi-Faur 2014). Since the directive includes paternity leave, paid at least at the level of sick pay, and a paid non-transferable parental leave of two months for each parent, the costs to be incurred by member states could be considerable. Indeed, the economic costs in the case of the WLBD include not only the adaptational costs associated with the implementation of the provisions, but also new permanent costs incurred as a result of the extension of social rights (Benish et al. 2017; Falkner et al. 2005; Levi-Faur 2014). These costs may be borne by the states, by employers and/or employees (e.g. in collective agreements) or by social insurance schemes (de la Porte et al. 2020; Falkner and Leiber 2004; Obinger et al. 2005). Therefore, countries with large industrial sectors, such as Germany and France, opposed additional allowances, arguing that they could imply a financial burden for social services, employers, or insurers (Agence Europe 2018c; de la Porte 2021). As a result, the original proposal by the European Commission (the Commission) to compensate parental leave at the level of sick pay was contested and watered down to an ‘adequate’ level, without further specification (Recitals 29 and 31, Directive 2019/1158). Therefore, member states have a greater discretion in this regard, and we therefore anticipate that the perceived costs resulting from the compensation for parental leave will trigger the engagement of domestic actors in the implementation process.

Actors’ engagement based on values

The second factor that motivates actors to engage in the implementation of the father-specific provisions of the WLBD is values. The father-specific provisions of the WLBD enhance gender equality, thus promoting a dual-earner model, rather than traditional gender roles based on the male breadwinner and female carer model (Esping-Andersen 2009). This is even reflected in the preamble of the WLBD, which states that ‘…the use of work-life balance arrangements by fathers […] has been shown to have a positive impact in reducing the relative amount of unpaid family work undertaken by women and leaving them more time for paid employment’ (Recital 11, Directive 2019/1158). This prompted criticism at the EU level, mainly from centre-right parties, which claimed that the directive derogates from family autonomy (de la Porte et al. 2020). While political parties on the left typically support policies that promote gender equality, right-wing political parties tend to oppose them and regard them as illegitimate attacks on traditional gender roles (Oesch and Rennwald 2018). It follows that values of gender equality drive actors in their strategies for implementing the WLBD, especially when such issues are politicised.

Methodology: case selection, data and methods

We focus on three member states – Denmark, Germany and Poland – based on several considerations (see Table 1). First, all three member states raised concerns during the negotiations at the EU level. While Denmark generally opposed the directive based on the principle of subsidiarity, Germany refused the additional costs that would potentially result from the new provisions, and Poland opposed gender equality policies and further criticised the EU with regard to the question of subsidiarity (de la Porte 2021). Second, these three countries represent various different gender equality regimes, namely the promotion of gender equality (Denmark), a mixed model (Germany) and a conservative model (Poland). Third, each of these countries had different parental leave policies in place prior to the directive, thus creating different adaptational pressure in relation to the WLBD. While Denmark and Poland had fully paid paternity leave and highly paid parental leave in place, their schemes lacked the earmarked part, indicating a medium degree of policy misfit. By contrast, Germany already had an earmarked part of two months but no legal exclusive right to paternity leave, indicating a low policy misfit. Finally, the three countries represent different types of social actor involvement, namely decision making (Denmark), consultation (Germany) and lobbying (Poland).

Table 1. Features of cases.

	Countries
Features of institutional setting	Denmark	Germany	Poland
Institutional feature of parental leave	32 weeks of parental leave, no earmarking, 100% replacement rate of wages for those covered by collective bargains, and flat-rate unemployment benefit level for others. 2 weeks of paternity leave, full wages for those on labour market	12 months of parental leave with 2 ‘partner months’ earmarked, remuneration at least at the level of sick pay (ceiling). No paternity leave	32 weeks, no earmarking (high replacement rate). 2 weeks of paternity leave (full wages)
Gender equality regimes	Promotion of gender equality	Mixed model	Conservative model
Degree of policy misfit	Medium	Low	Medium
Involvement of societal actors	Decision making	Consultation	Consultation and lobbying

With regard to methodology, we undertake structured, focussed analyses (Bennett and George 2004), identifying the role of agency in relation to the factors of values and costs, to be able to make a systematic comparison of the importance of these factors in the implementation process. For each case, we identify the actors’ positions and ideas around the WLBD, as well as the characteristics of the institutional settings allowing actors to shape the implementation of the WLBD. When it comes to the actors involved in the process, we focus on the role of relevant actors, namely political parties, social partners, and other relevant interest organisations with a stake in the WLBD.

We draw on different data sources in the national settings. First, we analyse reports, opinions and statements by different national actors and official governmental documents. Second, we include national newspaper articles, where relevant. Moreover, we conducted a total of 14 elite and expert interviews (five for Denmark, five for Germany and four for Poland) with relevant policy-makers, including ministry officials, interest groups and social partners. The interviews were conducted between February 2020 and October 2022. Our interviews were recorded and transcribed in full, as well as anonymised. When required – mainly for the purposes of data triangulation – additional informal interviews or follow-up questions were documented with notes. All data points were analysed via qualitative content analysis, where our focus was on the role and positions of national actors in the implementation process.

How actors shape the implementation of the WLBD

Denmark

In Denmark, many actors are involved in the decision-making and implementation process in relation to parental leave, which is considered a labour market policy. The government decides on the formal legal transposition of the WLBD, including paid parental leave, but social partners are involved in the political process and they decide on additional parental leave provisions for their members (e.g. compensation level to top-up statutory flat-rate, which is at the level of unemployment benefit).

Mobilisation of actors around costs and values

When the WLBD was proposed, Denmark voted against it in the Council, criticising the Commission’s initial draft for denying member states a discretion in implementing the directive. This is because in Denmark, decision making on labour market issues is carried out by autonomous social partners (de la Porte 2021; Interview Denmark 5 2022). At the time, all political parties, except the Social Liberals and the Socialist Party, opposed the directive, due primarily to concerns relating to subsidiarity. However, the opposition, consisting of parties to the right of the political spectrum – the Danish People’s Party, the Conservative Party, and Liberal Alliance – opposed the earmarking of leave for ideological reasons, arguing that it undermined parents’ autonomy in deciding on parental leave (de la Porte et al. 2020; Interviews Denmark 2, 3 and 5 2022). This reflects the fact that earmarked leave is a politicised and contested issue. Prior to the WLBD, Denmark was the only Nordic country that did not have earmarked parental leave (Rostgaard and Ejrnæs 2021). As discussed below, in Denmark, the discussion after the adoption of the directive has predominantly been driven by values, although concerns relating to increased costs hampered the inclusion of some self-employed.

In the implementation process of the WLBD, the implementation committee, headed by the Employment Ministry and including the main umbrella organisations of social partners, started by examining how the directive could be implemented. In line with the common practice in Denmark, the initial plan was to ensure a minimal implementation of the directive. Concerning the earmarked part of leave, an agreement was reached that no additional weeks should be added to the parental leave scheme. Instead, the reserved weeks for fathers were to be taken from the pre-existing leave. Subsequently, the social partner representatives – DA (Dansk Arbejdsgiverforening) and FH (Fagbevægelsens Hovedorganisation) – developed a formal political proposal and presented it to the government, suggesting nine weeks of earmarked leave per parent (Beskæftigelsesministeriet, 2021). In addition, fathers should have two weeks of paternity leave when the child is born, while mothers should have four weeks’ leave before giving birth (these provisions existed before the directive). This was an unsurprising and cost-neutral solution, implementing the directive at a minimal level. In addition, as a base-line, 24 weeks leave per parent were proposed as a starting point, including the right to a flat-rate benefit (unemployment benefit level). The motivation was value-driven, with a view to achieving greater gender equality in the parental leave system.

After the political proposal was presented by the social partners and approved by the government, the detailed technical transposition began in the implementation committee (headed by the Employment Ministry, with the involvement of social partners). Normally, this process, although a formality to ensure that political aims are respected, is closed, which means that no new political demands are brought to the table. Following the draft of the legislative transposition tabled in December 2021, a formal hearing with societal actors was organised (Beskæftigelselsministeriet 2022). The Employment Ministry took account of the positions of the societal actors, but since the WLBD was highly politicised in Denmark, new demands from political parties also had to be integrated, which complicated the process (Interview Denmark 3 2022; Interview Denmark 4 2022). The motivation of the political actors was value-based, namely to make gender neutrality in the system even more pronounced. This included an adaptation of the social partner proposition of the 24/24 model, which was not required by the EU. This was strongly advocated by the Social Liberals, the Socialist Party, the Social Democrats, and the Red-Green Alliance; it built on what had been proposed by the social partners, but went even further, as it was to include more individuals as eligible parents. The left-of-centre parties wanted non-biological parents, who have a role in parenting, to be included – that is, LGBTQ + persons as well as foster parents. It was agreed that these groups should have the same right to financial compensation, at the level of unemployment benefit. The discussion also included grandparents, proposed by the Red-Green Alliance; however, grandparents were not regarded as eligible for the right to paid parental leave (Interview Denmark 5 2022). The inclusion of LGBTQ + and foster parents reflects values about gender – that is, the role of parent that can be performed by different individuals. As the political demands pertaining to foster parents were not required by the directive, the Employment Ministry asked the Commission whether it was legally compatible with the maternity leave directive, before it was technically transposed (Interview Denmark 5 2022).

All parties on the left supported the revised 24/24 proposal, and the Liberal Party (Venstre) also expressed support for this position, thus changing its traditional standpoint in an attempt to appear as a modern liberal party that supports gender-equalising policies (Svane 2021). This change in position could also be explained on the basis that Danish Industry (DI) – the most powerful employer organisation – was the first social partner organisation to change position, arguing that earmarking leave was good for the labour market and productivity (de la Porte et al. 2020). This position was seconded by other strong private sector actors in the Danish public debate, particularly the director of the Danish National Bank (Finans 2019). In fact, the director of DI even encouraged social partners to go further than the two-month period of paid earmarked leave stipulated in the directive with a view to supporting increased gender equality in the labour market.

The statutory costs were not a major issue during the negotiations since the flat rate of the unemployment benefit level is compliant with the directive, and no extra leave weeks were added to the current system. From the outset, however, the social partners discussed possibilities for their members to top up compensation to the full wage level (FH 2021). Even if not centre-stage, costs were a hidden concern with regard to the self-employed; only certain self-employed individuals – those who have their own company and are considered as employees of their company – will have the right to earmarked parental leave. The centre-left parties wanted coverage for a larger proportion of the self-employed, but this was excluded, partly for reasons of costs.

The implementation outcome of the WLBD in Denmark

After the WLBD was adopted, the left-of-centre parties identified an opportunity to change the parental leave system in Denmark, rendering it more equal in terms of gender. While the 24/24 model may not have much effect on take-up by fathers, as it is not earmarked, the inclusion of foster parents and LGBTQ + persons who perform the role of parent displays an openness towards the role of parent among the influential left-oriented political parties. Yet many self-employed individuals are not included, which could possibly lead to unequal social rights between those covered by collective bargaining and in regular jobs versus those who are not covered. The legal transposition culminated in the adoption of the law on changed parental leave on 3 March 2022. In the course of 2022, parallel to the work of the implementation committee, social partners, building on the legally transposed framework, actively negotiated additional aspects for their constituents in the labour market – in particular, a top-up to the full wage level for the legal minimum of the earmarked parental leave benefit, which is at the level of the unemployment benefit. The social partners agreed that eight of the nine weeks of earmarked leave should be with full wages (Interview Denmark 1 2022; Interview Denmark 2 2022). Thus, the WLBD has made it compulsory for full-wage compensation for earmarked leave to be included for all workers, which previously applied only on an ad hoc basis in certain sectors. Overall, the main change motivator for complying with and going beyond the provisions of the directive was values on gender, carried forward by the left-of-centre parties. In addition, private sector actors supporting gender equality prompted a change in position by the political party Venstre.

Germany

In Germany, different actors play a role in implementing EU social policies. In addition to the government, the responsible ministry and public administration, the institutional setting provides for the participation of social partners via consultation. Therefore, social partners expressed their positions in the EU decision-making process and in implementing the WLBD at the national level. While Germany lacks the autonomy granted to social partners in Denmark, these partners nevertheless play a crucial role in the decision-making process.

Mobilisation of actors around costs and values

Germany already complies with the WLBD in terms of an earmarked paid and non-transferable part of parental leave. However, implementing paid paternity leave was contested since such an explicit legal right to paternity leave does not exist in Germany at present (Agence Europe 2018a). The 2018 government – consisting of the Christian Democratic Union of Germany (CDU) and the Social Democratic Party of Germany (SPD) – even slowed down the decision-making process in the Council due to the additional costs that were perceived as possibly emerging from the introduction of paid paternity leave (Agence Europe 2018b; Interview Germany 1 2021). Therefore, even though the policy misfit was low, the question of costs was regarded as an obstacle. Despite these concerns, however, Germany voted in favour of the directive in the Council since there was overall agreement on the need to improve parental leave conditions across Europe (Interview Germany 1 2021).

With regard to the issue of costs, the government perceived the introduction of paternity leave as a major change to the current national regulations and thus feared high administrative costs (Agence Europe 2018a). Moreover, this view was supported by German industry and employers, who feared high additional costs resulting from a separate paid paternity leave and consequently lobbied extensively against the measure (Interview Germany 1 2021). Consequently, it was not only the government, but also employers’ associations that opposed the directive. The additional costs stem from the fact that the duration of parental leave (Elternzeit), which is up to three years after the birth of a child for each parent, is decoupled from the parental leave benefit (Elterngeld). The latter is at the level of sick pay in relation to the individual parent’s income (65% of former net earnings) to a maximum of €1,800 per month. However, while the second parent is entitled to two exclusive parental leave benefit months in practice – which are often taken up by fathers directly after birth (Blum et al., 2022) by granting an additional bonus – this regulation does not constitute an exclusive right to paternity leave. Introducing such a right thus creates additional costs. However, the argumentation was about the perceived costs since the concrete financial costs – specifically in the case of Germany – were unclear; only general estimates from the Commission’s impact assessment existed (European Commission 2017: 82–83).

As a result, the question of costs mobilised actors to become engaged in the domestic implementation, where the conservative side adopted a more sceptical stance towards the directive. Even the SPD supported this sceptical stance at the outset. Moreover, the ministry in charge (the Federal Ministry for Family Affairs, Senior Citizens, Women and Youth – BMFSFJ) further argued that Germany was already compliant with the directive and that no transposition measures were therefore necessary (Die Welt 2021b; Interview Germany 4 2021). As a result, Franziska Giffey (SPD), the minister in charge at the time, argued that current legislation already supported fathers’ taking leave in excess of the 10 days. Her reasoning was based on the fact that in terms of the shared 14 months’ leave, one parent can take a maximum of 12 months, and thus two months are already earmarked for the second parent (Die Welt 2021c).

Nevertheless, trade unions and lawyers in the field argued that Germany should implement the regulations on paternity leave (DGB Frauen 2021; Die Welt 2021c; Treichel 2021) and stressed the importance of adding new rules on paid paternity leave on top of the existing parental leave (Interview Germany 1 2021). The pressure to follow this suggestion increased as the left camp of the CDU supported the trade unions’ position (Die Welt 2021a). Before any implementation plans were drafted, the general election took place in September 2021. The political parties to the left made family policy a topic in their election programmes, with the most far-reaching suggestions stemming from the Alliance 90/The Greens, who proposed extending parental leave to 24 months (Die Grünen 2021).

The newly constituted government – consisting of SPD, the Greens and the Liberals – resulted in changes in the positions of the various actors and in government support for an exclusive paternity leave right. This could be due to the exclusion of the conservative party from government or to the fact that the ministry in charge (the BMFSFJ) passed from the control of the SPD to that of the Greens. Interestingly, the previous finance ministry (SPD) – backed by industry – initially adopted a sceptical stance on the issue when discussing the directive at the EU level (Agence Europe 2018a), which changed with the new government. In the new coalition agreement, the reconciliation of family and work became an important area (Die Bundesregierung 2021: 79). The previous minister, Anne Spiegel from the Greens, therefore proposed that two weeks of specific leave should be granted to the second parent after the birth of a child, with full wage (Die Zeit 2021). However, these debates were decoupled from the matter of the WLBD. In April 2022, Spiegel resigned and was succeed by Lisa Paus.

The implementation outcome of the WLBD in Germany

When the transposition deadline passed, Germany adhered to the status quo and did not undertake any transposition measures. The controversy over whether or not Germany needs to introduce new legislation on paternity leave to comply with the directive is also reflected at the EU level. While the Commission did not initiate any infringement procedure against Germany for failing to provide notification of the transposition measures (European Commission 2022), a report by COFACE – Families Europe found that Germany is not compliant with the WLBD at present, due to the lack of paternity leave (COFACE Families Europe 2022).

As at October 2022, the ministry had plans to introduce paid paternity leave. Other major plans included (i) increasing the so-called partner months (extra months if both parents take their reserved parental leave) by an additional month; (ii) the ‘dynamisation’ of the minimum and maximum allowance (without further specification), and the introduction of an entitlement to parental leave allowance for foster parents. In addition, the ministry envisaged that the entitlements for self-employed individuals would be improved and that parents with a prematurely born child would receive up to three additional months’ parental leave. Moreover, LGBTQ + persons would have the same rights and easier legal conditions to claim parenthood (Die Bundesregierung 2021; Interview Germany 4 2021).

Although the proportion of fathers who took parental leave was only 43.5%, and the average duration of the parental leave taken by men was 3.7 months in 2021 (Destatis 2021), the approach that favoured increased gender equality was subordinated to the issue of costs in the implementation process. Since the relatively low number of fathers who take parental leave is thought to be due to potential negative career consequences and losses in family income, a higher parental leave allowance could enhance gender equality (Interview Germany 2 2021; Interview Germany 3 2021; Interview Germany 5 2021). In summary, the primary motivation of the actors has been to maintain the status quo as far as possible, and here costs have been the primary driver.

Poland

Decision making in Poland is centralised as the government takes the main decisions after a consultation process. Unlike in Denmark, the involvement of social partners in designing and implementing social policies is limited. The Council of Social Dialogue (an official tripartite body) needs to be consulted, but does not play a critical role in the legislation process.

Mobilisation of actors around costs and values

Unlike Germany, Poland complies with the directive when it comes to the provision of paternity leave (two weeks, fully paid, since 2012). However, when it comes to parental leave, it still has a family-base entitlement. To comply with the previous 2010 Parental Leave Directive, one month per each parent was reserved after 2010 (Kurowska, 2019). Overall, although approximately 60% of fathers use their individual right to parental leave (FRD/Share the Care, 2020), it is estimated that fathers constitute only 1% of everyone who makes use of parental leave (Chądzyński, 2021). Therefore, the system favours maternal care and does not support gender equality.

The mobilisation of various actors around the WLBD in Poland should be viewed from the perspective of the political changes that took place in 2015, after the right-wing party Law and Justice (Prawo i Sprawiedliwosc, PiS) took office. Since they assumed office, Law and Justice has presented an agenda for reforming social policies in line with its ideology of conservative gender roles, thereby explicitly rejecting gender equality. Therefore, it came as no surprise that the non-transferability of four months of parental leave became the most controversial issue when the Commission’s proposal was announced in 2017. The issue of costs was regarded as a serious one even before the WLBD was adopted, when the former deputy minister stressed that adding four months in addition to the existing solutions would generate ‘unbearable costs’ (Senat RP 2017). Employers’ organisations accepted the general goals set by the EU; although, like most employer organisations across the EU, they did not want regulation to hamper business activity (Interview Poland 4 2022). At the same time, major trade union organisations expressed their support for the directive, including the pro-government NSZZ Solidarność, which pointed to the EU’s role in enhancing social policy standards. The All-Poland Alliance of Trade Unions (OPZZ) tacitly supported the EU’s initiative (Interview Poland 1 2020; Interview Poland 2 2022), stressing the EU's role in enhancing social policy standards and positioning themselves against the government and the parliament. In June 2019, the Polish government chose to abstain during the final vote in the Council. Following the adoption of the directive, the government then switched its position. As one of the deputy ministers of family commented at the time, the adopted directive allowed member states to decide on the level of benefit for earmarked leave, which he perceived as a positive development that did not necessarily entail a costly transposition (Żebrowski 2019).

There was no public discussion about the directive until late 2021, when the current state secretary responsible for its implementation, Barbara Socha, started to organise unscheduled meetings with experts, social partners and conservative non-governmental organisations (NGOs) (Interview 3 Poland 2022). At the stage, when the draft was still being prepared, the implementation plans included the addition of nine weeks to the existing scheme, accompanied by remuneration at the level of 70% of the previous wage throughout the whole leave period (as against the current 67.5%) for both the mother and the father. Socha is well-known for her promotion of conservative views of family and marriage and the traditional role of fathers (wPolityce.pl 2017). Obliged to implement the directive, the government started to frame the reforms by removing the link to gender equality, as stipulated by the directive, and instead pushed for a conservative approach towards gender roles. With reference to the age limit of the child mentioned in the directive (eight years), the ministry reformulated and repurposed the original argumentation contained in the directive (which emphasised gender equality) to stress the role of engaged (but conservative) fathers (Interview Poland 3 2021), thereby excluding LGBTQ + persons. The ministry suggested that the father could consider using his nine weeks ‘later, when the need arises’, emphasising that the reform is aimed at ‘strengthening the family’ (TVN24, 2022, no pagination). Finally, the transposition plans were discussed by the Council for Social Dialogue, where all partners supported the proposed changes with minor comments on the details of changes with regard to the issues other than parental leave (Rada Dialogu Społecznego, 2022).

The implementation outcome of the WLBD in Poland

The first draft was officially registered to follow the legislative process on 15 February 2022, with social consultations open only until 17 March 2022. Importantly, there is no separate law implementing the WLBD. Instead, there is one bill that is planned to introduce both new EU directives: the WLBD and the Directive (EU) 2019/1152 on Transparency and Predictability of Working Conditions (TPWCD). The draft incorporated the solution proposed earlier in the semi-formal consultations – that is, the addition of nine weeks’ leave at the level of 70% wage replacement. Despite the short period of time allowed for consultations, the ministry waited until July 2022 to publish the list of policy amendments, together with its response. The proposals relating to the earmarked leave received very little attention in the document that consolidated the various social partner opinions (RCL 2022).

The lack of political conflict in relation to the directive was confirmed by three interviewees, which suggests that there is no major discussion about its implementation – even at the level of consultation (Interview Poland 2, 3 and 4 2022). This lack of discussion gave the governing party even more leeway to customise the implementation according to its political needs and ideological affiliation. There were two reasons for this: first, such wide space for manoeuvre resulted from the institutional features of the policy process system in Poland – that is, the centralisation of decision making, with only a small role being played by social partners and social consultations in general; second, it stems from the fact that there is no separate bill introducing the WLBD. Therefore, the implementation of the TPWCD gave rise to much more discussion and controversy – for example in relation to compensation levels for the employees, thus effectively crowding out any discussions about the WLBD (Lesniak, 2022, Interview Poland 2 and 4 2022).

The legislative change has not yet been enacted. As at October 2022, the draft was being discussed by the Standing Committee of the Council of Ministers, with the progressive NGOs trying to argue for a higher replacement rate, stating that 70% may be too low to incentivise fathers to utilise their part of the leave. (Interview Poland 3 2022). This proposal may not be adopted; however, at the beginning of October, Donald Tusk, the leader of the main opposition party, Civic Platform (which introduced the current scheme of parental leave in 2013) proposed to guarantee a 100% replacement rate for maternity benefit (Polcyn 2022). Moreover, Tusk is likely to support proposals increasing the level of parental leave benefits when the draft reaches the final stage of the legislative process in the parliament. In summary, facilitated by the centralised mode of public policy decision making, the right-wing populist government has reoriented the goal of the WLBD from supporting gender equality to preserving conservative gender roles, in line with their values and despite the high costs involved.

Comparative discussion and conclusion

This study has analysed how national actors shape the implementation of the WLBD in three member states (Denmark, Germany and Poland). Based on an actor-centred approach focussing on actors’ mobilisation in implementing the WLBD, we outline below our principal contributions to the literature and reflect upon the significance of the findings for the study of differentiated implementation and for further EU implementation research.

In summary, our actor-centred analysis demonstrates that costs and values have been significant in different ways, which it would not be possible to fully capture through a structure-oriented framework. Nevertheless, while both factors represent drivers for actors’ mobilisation in the implementation process, the value-driven mobilisation is more likely to push for reforms to increase gender equality, whereas the actors’ mobilisation based on costs – to avoid any additional costs – may be used as a justification against reforms. On this basis, we propose the theorisation presented in Table 2, which is derived from our empirical findings, associating the motivations for actor-mobilisation with likely outcome.

Table 2. Theorisation of empirical findings.

Reasons for actor mobilisation	Values as change	Values as defence
Costs as change	A. High likelihood of change [Denmark and the Netherlands for employees and different types of parents]	B. Medium likelihood of change [Poland]
Costs as defence	C. Medium likelihood of change [Denmark and the Netherlands for self-employed]	D. Low likelihood of change [Germany and France]

For example, if actors are mobilising on the basis of values and costs and thus argue or allow for changes in both areas, a change of the policy is highly likely (Table 2, cell A). This happened in Denmark, where change was triggered by actors that were driven by values and allowed changes in the costs for the employees. This prompted the left-of-centre political parties to customise the directive during the technical implementation process, which is normally closed. The 24/24 model was adapted to include non-biological social parents as eligible recipients of parental leave rights, with compensation. Moreover, social partners advocated for full wages for those in the labour market, thus incurring new costs. Yet the Danish case has also revealed that if actors use costs as an argument against policy change, it hinders change; therefore, self-employed individuals, who are poorly unionised, have been excluded from the policy (Table 2, cell C). We see a similar tendency in the Netherlands with the implementation of the WLBD. In this case, too, there is a value-driven motivation for change where most political parties, in conjunction with societal actors, significantly altered the pre-existing parental leave system to render it more gender equalising – despite new additional costs. Like in Denmark, the question of costs has also been a defence against change by excluding self-employed individuals (de la Porte et al. 2022). The Dutch example corroborates the theoretical novelty found also in Denmark, demonstrating that actors are driven by value-based motivation to change policies in a fundamental manner, rather than merely following their own positions.

By contrast, if both factors (costs and values) are used by actors as a defence to maintain the status quo, change is very unlikely to occur, which we have seen in the case of Germany (Table 2, cell D). In contrast to Denmark or the Netherlands, actors in Germany were mobilised primarily on the question of costs, whereas the question of values was subordinated. This allowed for a customisation of the rules by the political parties, which became possible due to the election of 2021 and the rather turbulent political situation, with four different ministers responsible for implementing the WLBD. However, the WLBD was not politicised, and we could not identify any value-driven mobilisation of actors since Germany already had a more gender-neutral parental leave system than Denmark. While focussing on costs, national actors aimed primarily to maintain the status quo. Similarly, preliminary findings on the current negotiations on the implementation of the directive in France imply the same strategy of maintaining the status quo, where the political and public administration actors use costs as a reason to avoid change (de la Porte et al. 2022). Thus, actors focus on ‘costs as status quo’ in both cases. Moreover, the WLBD has not been politicised in either France or Germany, and values have not been identified by actors as an issue on the basis of which to propose change.

When actors use values as a defence to maintain the status quo while at the same time allowing for changes in the costs, there is a medium likelihood of change, as outlined in the case of Poland (Table 2, cell B). In Poland, the mobilisation of actors was value-driven. Yet values were taken as a defence of the prevalence of conservative values relating to gender roles, and the ministry stressed the need to keep mothers’ leave intact, even though this meant additional costs in practice. Adding nine weeks of paid leave, the government further highlighted the role of the conservative father, stressing the secondary caring role in the case of men and suggesting that the leave could be used at a later stage in the child’s life. In Poland, values were a defence used by the government actors to preserve the role of parents in Poland, whereas in Denmark, values created the impetus for a wider view of the role of parents. However, if the envisaged system were to be adopted in Poland, it could nevertheless lead to some change, provided that fathers who are motivated to change the traditional gender roles are knowledgeable about the system.

Overall, our findings suggest that differentiated implementation – defined as diversity in the existence and use of discretion during legal and practical implementation – is strongly determined by national actors, their motivations for mobilisation, and, of course, their degree of mobilisation. Moreover, the increased flexibility and openness of the WLBD’s provisions to find a compromise at the EU level, led to a higher density of EU rules in national implementation, as well as to more variation in practical implementation. The question of practical compliance remains to be seen, but there are currently reasons for concern since the Commission initiated infringement procedures against 24 member states in September 2022 for failure to provide notification of their transposition measures (European Commission 2022).

In each of our three case studies, we demonstrate that values and costs play a decisive role in the process of implementing EU policies. While we focussed on EU social policy, these factors could also play a decisive role in implementing environmental policies or internal market provisions (e.g. when implementing the European Green Deal or the Recovery and Resilience Facility).

Acknowledgements

We would like to thank the editors of the special issue ‘Differentiated Policy Implementation in the European Union’, Sebastiaan Princen, Eva Ruffing, Eva Thomann, and Asya Zhelyazkova for valuable comments and suggestions. Moreover, we thank Dorte Sindbjerg Martinsen and two anonymous reviewers for their constructive comments.

Disclosure statement

No potential conflict of interest was reported by the authors

Additional information

Funding

This work was supported by the European Union’s Horizon 2020 Research and Innovation programme under grant agreement number 870978.

Notes on contributors

Brigitte Pircher

Brigitte Pircher is an Associate Professor and works at Linnaeus University and at Copenhagen Business School. Her research focuses on EU institutions, EU policy making, and EU law and its implementation and has appeared in the Journal of Common Market Studies, European Union Politics, Politics & Governance, and the Journal of European Integration. [[email protected]]

Caroline de la Porte

Caroline de la Porte is Professor at Copenhagen Business School. Her research focuses on EU social policies and comparative welfare state reform. Recent publications involve special issues on Next Generation EU in Comparative European Politics and on the Nordic model in Regulation & Governance. [[email protected]]

Dorota Szelewa

Dorota Szelewa is an Assistant Professor and works at the University College Dublin and at Copenhagen Business School. She specialises in comparative social policy research from a gender perspective and her articles have appeared in International Journal of Social Welfare, Comparative European Politics and Social Inclusion. [[email protected]]

Appendix

Table A1. Policy misfits on principle issues (earmarked part and remuneration) between the WLBD and the national law in the different EU countries.

	Level of regulation
Parental leave policies	State	Social partner	Policy misfit
Earmarked part; remuneration min. at level of sick pay	Germany*, Luxembourg	Sweden**	Low
No earmarked part; remuneration min. at level of sick pay	Austria, Croatia, Czech Republic, Estonia, Finland, Hungary, Poland, Romania, Slovenia	Denmark**	Medium
Earmarked part; remuneration below level of sick pay	Belgium, Cyprus, France, Germany, Greece, Ireland, Portugal, Spain	Italy, Netherlands	Medium
No earmarked part; remuneration below the level of sick pay	Bulgaria, Cyprus, Latvia, Lithuania, Slovakia	Malta	High

* Germany has an earmarked part, but no explicit legal right to paternity leave after the birth of the child.

** Basic parental leave on state level, combined with regulation through social partners depending on collective agreement.

List of Interviews:

Interview Denmark 1, 3 F, 21.01.2022.

Interview Denmark 2, Fagbevægelsens Hovedorganisation, FH, 30.04.2022.

Interview Denmark 3, Dansk Arbejdsgiver organisation (DA), 12.04.2022.

Interview Denmark 4, HK, 15.08.2022.

Interview Denmark 5, Ministry of Employment, 16.08.2022.

Interview Germany 1, DGB Bundesvorstand official, 21.5.2021.

Interview Germany 2, Family policy expert 1, 15.5.2021.

Interview Germany 3, Family policy expert 2, 10.6.2021.

Interview Germany 4, BMFSFJ, Federal Ministry for Family Affairs, Senior Citizens, Women and Youth official, 03.09.2021 (provided additional, more informal information on 06.09.2022 and 17.10.2022).

Interview Germany 5, pme Familienservice official, 10.5.2021.

Interview Poland 1, Trade Union expert 1, 10.02.2020.

Interview Poland 2, Trade Union expert 2, 17.05.2022.

Interview Poland 3, NGO representative 1, 2.12.2021 and 10.05.2022 (provided additional, more informal information on 26.09.2022.)

Interview Poland 4, Representative of Employers’ Organisation 1, 16.05.2022