Perceiving childcare as a labour market instrument

It is clear that childcare was primarily absent from the social and political agenda in the Netherlands until the 1990s. At that time, in a broader context of discussing work and care combinations, childcare became viewed as a possible labour market instrument to increase women's employment, both in terms of participation as well as working hours. Yet the welfare state approach to childcare remained fragmented. For example, until 2002, childcare policy fell under the domain of the Ministry of Health, Welfare and Sport (MinHWS). The exceptions to this were childcare arrangements considered to be related to policy areas maintained by the Ministry of Social Affairs and Employment (MinSAE), such as childcare for social assistance recipients. Childcare policy was divided among the ministries, and separate policies were created for distinct groups of parents and children, such as lone mothers, social assistance recipients or children in specific age groups. This shifted in the late 1990s, as a common perspective among actors developed, that market-driven childcare would improve women's employment. Two events, an initial push for marketisation and a shift in ministerial focus, are crucial for understanding how this mechanism shaped childcare policy.

First, a report from the Ministry of Economic Affairs entitled ‘Marketization, Deregulation and Legislative Qualities’ (Marktwerking, Deregulering en Wetgevingskwaliteiten; MDW), was released in the spring of 1998, just prior to the national elections. The MDW project, carried out during the ‘purple’ political coalitions of Social Democrats and Conservatives between 1994 and 2002, marked a trend towards increased marketisation in various policy fields, including childcare.

Earlier attempts at creating more market-driven childcare were not successful. One governmental policy-maker, who later became involved with childcare policy at the MinHWS,Footnote ³ explained that during the mid-1990s, a move towards market-driven childcare was discussed but not accepted.

Legislation to achieve market-driven childcare was developed during the late 1990s, although the legislation was not implemented until 2005. To develop this legislation, the state requested an advisory report from the Social and Economic Council (SER), the primary consultative body for the government, with representatives from employers’ organisations and trade union confederations, and independent political advisors. Part of a larger advisory report on work and care, the report confirmed the desire for a move towards more market-driven childcare (SER, 1998). The SER report also confirmed a unilateral agreement to manage childcare within the existing institutional structure of the Dutch welfare state, based on shared responsibility among the welfare state, employers and parents (employees). Applying this institutional approach to social risks related to reconciling work and care ‘created a breakthrough’ in childcare policy (I4, Representative in consultative organisation). Following the SER's advisory report, policy-makers developed new legislation detailing marketisation efforts intended to stimulate childcare demand and increase childcare supply. While the childcare law was outlined in 2001, parliament did not pass the legislation into law because, despite some agreement that marketisation was a sound approach, sufficient support was lacking. Support for this approach grew as insufficient childcare provision was increasingly perceived to be a barrier to women's employment. The dominance focus on women's employment and barriers to employment allowed a common perspective to develop.

This common perspective was an important factor causing a change in childcare governance. Following the elections in 2002, in which a new centre-right coalition was elected, childcare was moved from the MinHWS to the MinSAE. ‘This dossier needed an enormous impulse – to improve labour market participation – and that's why it went from MinHWS to MinSAE’ (I45, Governmental policy-maker). The change in childcare governance structures is indicative of two things: it demonstrates the focus of relating childcare to employment, rather than children's health and welfare/wellbeing and it also emphasises the central role of the social partners in creating and regulating childcare policy. The MinSAE is tasked with developing and regulating policy related to, among other things, industrial relations. In this sense, the MinSAE regulates the social partners’ ‘self-regulation’ of social policy areas covered by collective bargaining. If a social policy issue was dealt with in collective bargaining, it would be relevant to the MinSAE from an employment standpoint.

By bringing childcare under the auspices of the MinSAE, the state could maintain an open dialogue with the social partners on this topic. Childcare could now be discussed within the Regiegroep, an institutionalised informal setting in which ministerial policy-makers, employers’ organisations and trade union officials meet to discuss current and long-term policy strategies. Moreover, moving the childcare dossier to this ministry signalled the state's intent to maintain an institutionalised industrial relations approach to childcare policy.

Considered in the light of similar social policy developments in other countries, the presence of this social mechanism, a common perception that marketised childcare would improve women's employment, is not entirely unique. Linking childcare to women's employment is a common motivation for social policy developments in the area of work–family reconciliation. Lewis and colleagues (Reference Lewis, Knijn, Martin and Ostner2008) have termed this ‘employment-led’ social policy. Daguerre (Reference Daguerre, Armingeon and Bonoli2006: 225) finds similar developments in Britain and Switzerland, whereby ‘the need for female labour in the context of a booming service economy opened a policy window for British and Swiss childcare advocates’. Attempts to activate women and mothers in particular are also visible in countries like Germany, where family policies have become increasingly oriented towards employment (Fleckenstein and Seeleib-Kaiser, Reference Fleckenstein and Seeleib-Kaiser2011). Australia, one of the latest OECD countries to develop a paid parental leave scheme, did so in an attempt to improve mothers’ employment (Martin et al., Reference Martin, Hewitt, Baird, Baxter, Heron and Whitehouse2011). In contrast to other countries, however, the Dutch welfare state expanded its role in childcare policy through mandatory marketisation and placed the emphasis on parents as childcare consumers. Childcare policy was developed under two Social Democratic-led coalitions yet, rather than develop a system of publicly-funded childcare, similar to the Nordic approach, a marketised approach was taken. To understand why, it is necessary to look more closely at the role of another important mechanism, that of party politics, in the development of childcare policy.

The importance of party politics

The ideologies of political parties are argued to be driven by underlying class cleavages (Hicks and Misra, Reference Hicks and Misra1993; Huber et al., Reference Huber, Ragin and Stephens1993, Reference Huber and Stephens2000), leading political actors to pursue different policy responses to social risks. Traditionally, Social Democrats pursue policy solutions focused on public service provision and equality (Huber et al., Reference Huber, Ragin and Stephens1993; Seeleib-Kaiser et al., Reference Seeleib-Kaiser, Dyk and Roggenkamp2008). Christian Democrats favour policy solutions focused on transfer payments to male breadwinners and the involvement of social partners (van Kersbergen, Reference van Kersbergen1995). Conservative (secular) parties are likely to pursue policy solutions focused on private markets.Footnote ⁴

The presence of multiple party coalitions and a desire for consensus means distinct elements of various party ideologies combined when developing childcare policy. From 1994 to 2002, two so-called ‘purple’ coalitions held the political majority, consisting of Social Democrats, Conservatives and Liberal Democrats. Social Democrats, despite being the leading party in the coalition and preferring public provision, needed the support of the Conservatives and Liberal Democrats to achieve further developments in childcare policy. Consensus-seeking led to the combination of two dominant political objectives: greater equality in childcare policy (Social Democrats) and market-driven childcare (Conservatives). To achieve greater equality, the Social Democrats emphasised the role of individual choice as a means of challenging male breadwinner norms. Prior to 1994, male breadwinner norms dominated under Christian Democratic rule, with a clear emphasis on the family (Seeleib-Kaiser et al., Reference Seeleib-Kaiser, Dyk and Roggenkamp2008). The Social Democrats preferred greater state involvement in childcare (e.g., public provision) and were clear about this throughout the 1980s (Seeleib-Kaiser et al., Reference Seeleib-Kaiser, Dyk and Roggenkamp2008). But once in power, they needed to achieve consensus within the ruling coalition. To achieve greater equality in childcare policy, they were willing to concede to a marketised approach despite their preference for childcare being a basic public provision.

The marketised approach to childcare, with its public–private mix of collective subsidies, was thus a result of the particular mix of political ideology at the time, ruling out a Nordic style of childcare policy based on the public provision of services.

While much of what is now contemporary childcare legislation was shaped during the second ‘purple’ cabinet from 1998 to 2002, it was not implemented until a Christian Democratic-led (CDA) coalition ruled Parliament from 2002 onwards. Christian Democrats had become increasingly supportive of childcare as a means of improving labour market participation, in essence converging towards Social Democrats on the issue (Seeleib-Kaiser et al., Reference Seeleib-Kaiser, Dyk and Roggenkamp2008). Yet the party continued to emphasise the importance of the family. This ideology helped shape a more conservative, family-oriented approach to childcare policy, even as marketisation was accepted. Formal childcare was made available through market forces of demand and supply, rather than the public provision of childcare services. The latter was undesirable because it could be viewed as the state advocating full-time use of formal childcare.

Conservative views of childcare, emphasising the importance of the family stem not only from Christian parties on the right in the Netherlands, but from parties on the left as well.

These quotes reveal an underlying political tension in the development of Dutch childcare policy: between viewing formal childcare as a means of improving mothers’ employment versus viewing formal childcare as (negatively) affecting children's health and well-being. The marketised approach, by emphasising freedom of choice through demand-driven childcare, offered a solution for developing childcare policy without having to resolve this underlying tension.

The absence of a resolution to this tension means that, even today, consensus remains absent regarding the educational role of formal childcare in the Netherlands. Seeleib-Kaiser and colleagues (Seeleib-Kaiser et al., Reference Seeleib-Kaiser, Dyk and Roggenkamp2008: 148) have argued that a ‘fundamental disagreement’ remained between Social and Christian Democrats on this issue. But this political tension is more encompassing than the Social and Christian Democrats. Conflicting views on whether childcare is solely a labour market instrument or whether childcare potentially plays an important role in childhood development and education spans the entire political spectrum. The absence of a resolution makes it difficult if not impossible to emphasise education/childhood development and childcare as a labour market instrument, which can lead to a lowering of childcare quality standards.

To sum this up, the influence of a political mechanism on the development of childcare policy in the Netherlands is clear. The Social Democratic party prefers public funding of childcare, which creates greater social equality. Christian Democrats are in favour of some public funding for childcare as a labour market instrument, but prefer to emphasise the role of the family. The secular right also favours some public funding for childcare if it serves as a labour market instrument, but only in combination with market-driven demand and supply. These political ideologies combined in a context of consensus-seeking in an attempt to improve women's employment. The result was a marketised approach to childcare that emphasises individual choice (parents as childcare consumers) but remains silent on the issue of childcare as educational or influencing children's health and well-being. While the previous two mechanisms help to explain the direction in which childcare developed, they cannot explain what triggered the final change in childcare governance structures, shifting childcare away from industrial relations into the welfare state. This shift is best explained through a corporatist institutional mechanism and, more specifically, the failure of this mechanism.

Corporatism and childcare policy

The Netherlands is characterised by a strong corporatist institutional structure whereby the state shares its regulatory capacity with the social partners in many social policy areas. This structure allows policy to be developed and managed within existing corporatist institutions, such as agenda-setting through bi- and tri-partite bodies or through collective bargaining. The latter offers scope for ‘self-regulation’ by the social partners, meaning regulation by the social partners themselves in the absence of state involvement. While state regulation may be absent, there is generally a distinct interplay between ‘self-regulation’ in bargaining and welfare state policy, an interplay evident in policy areas like pensions and unemployment benefits in many corporatist countries (Ebbinghaus, Reference Ebbinghaus2011; Johnston et al., Reference Johnston, Kornelakis and d’Acri2011). While collective bargaining can be an effective means of compensating for cuts in, or lack of, welfare state policy (Trampusch, Reference Trampusch2006; Yerkes and Tijdens, Reference Yerkes and Tijdens2010), relying on industrial relations as the dominant approach to childcare policy was relatively unique to the Netherlands.

Throughout the 1990s and into the 2000s, collective bargaining was the primary means of developing Dutch childcare policy, focused on increasing the availability of employer subsidies. But self-regulation in this field eventually failed. Employers and trade unions were responsible for negotiating employer subsidies for childcare. Amidst growing concern that an insufficient proportion of collective agreements contained such subsidies, the state threatened to mandate employer subsidies if a target of 90 per cent was not met by 2006, in essence a state shadow of hierarchy (Scharpf, Reference Scharpf1997). This target was made explicit during the debates on impending childcare legislation in 2000.

Most interviewees believed this 90 per cent target was created by the social partners within the bipartite Labour Foundation. Yet while the Labour Foundation drafted childcare recommendations in 1999 and again in 2004, these texts focus on a general increase of childcare subsidies in collective agreements; no explicit target is mentioned (StAr, 1999, 2004). Independent of the origin of this target, the state confronted the social partners in 2006 with statistics that showed the target was not being met (Tweede Kamer der Staten Generaal, 2006).

Despite the strong development of childcare subsidies through industrial relations, the social partners were unable to achieve the 90 per cent target due to a free-rider problem. Even though an increasing number of employers offered childcare subsidies through collective agreements, employer subsidies were not universal, undermining the tripartite financing of childcare and signalling the failure of self-regulation.

Two solutions to the free-rider problem were discussed: a continued push for an increase in childcare subsidies in collective agreements or mandatory employer subsidies. The latter option was not favoured by employers; the three Dutch trade union confederations differed in their views. Prior to 2006, a solution to the free-rider problem was sought through improved efforts to increase the number of childcare subsidies in collective agreements, hence maintaining voluntary employer subsidies. When this proved insufficient, the Dutch government made good on its threat, thereby re-assuming its regulatory capacity in childcare policy.

State intervention after the failure of self-regulation has occurred in other policy fields in the Netherlands as well, such as smoking regulations (Weyers, Reference Weyers2010) and pensions. In fact, the failed self-regulation of childcare policy followed by state intervention is exemplary of a broader government trend towards self-regulation in combination with the use of performance norms, demanded of organisations that serve the public interest (Mackor, Reference Mackor2010). By relating childcare to employment, the social partners were indirectly serving the public interest by providing childcare subsidies to employees. The social partners’ failure to meet the performance norm, of 90 per cent of childcare subsidies in collective agreements, led to state intervention as a matter of public interest. The failure of a corporatist institutional mechanism explains why the state chose to take the route of self-regulation in childcare policy. Childcare was a relatively new policy area, with care previously taking place within the home and dominant social and political preferences favouring informal over formal care options. Choosing to develop childcare along the trusted route of industrial relations was an obvious choice – a matter of path dependent, increasing returns (Pierson, Reference Pierson2000).

Why the state then decided in 2006 to enforce the performance norm of 90 per cent in a very strict sense must be understood in the context of the various social mechanisms at play. Although the state had been concerned about childcare policy in relation to women's employment prior to 2006, in 2006 participation had become a dominant theme in political discourse. In that year, the newly appointed cabinet created a commission to investigate means of improving employment. The Commission's findings reiterated the need for improved childcare to increase women's labour market participation (Commissie Arbeidsparticipatie, 2006). The installation of this commission and the support for its findings reflects political acceptance of an adult worker family model (Lewis and Giullari, Reference Lewis and Giullari2005). Therefore, the common perception that childcare is a barrier to women's employment led to the view that the self-regulation of childcare through industrial relations had failed to serve the public interest. To realise participation goals, state involvement was necessary. Simultaneously, the shaping of political forces remained. In theory, there is support for an adult worker family model in the Netherlands. In reality, this model is absent. Most Dutch political parties support the adult worker model, yet many prefer that it take the form of a combination model, whereby paid work and care is shared between men and women (e.g., each working four days a week). But even this preferred combination model is absent. As in most industrialised countries (Cooke and Baxter, Reference Cooke and Baxter2010), Dutch women continue to be over-represented in household and care work (Merens et al., Reference Merens, Brakel, Hartgers and Hermans2010) and are much more likely to work part-time than men (Merens et al., Reference Merens, Brakel, Hartgers and Hermans2010; SCP, 2008; Yerkes, Reference Yerkes2009). In this sense, the Dutch case is exemplary for the ambiguity between aspiring to an individualised adult worker model and the reality of having a ‘dual-earner, gender-specialised, family arrangement’ (Daly, Reference Daly2011: 1–2).