Cash for childcare schemes in the Nordic welfare states: diverse paths, diverse outcomes

Family policy reform has been on the agenda everywhere in recent years (Mätzke and Ostner 2010). Policy developments in mature welfare states display increasing complexity and diversity that may blur conventional family policy typologies (Daly and Lewis 2000). ‘Scandinavization’ or ‘farewell to maternalism’ characterise certain aspects of the ongoing change (Mätzke and Ostner 2010). Some European welfare states (e.g. Germany, the Netherlands and the UK) are ‘path shifters’, moving away from a strong male breadwinner regime by reforms facilitating mothers’ employment (Morgan 2013). On the other hand, all the Nordic welfare states, typified as ‘earner-carer’ models advocating gender equality in paid and unpaid work, have added cash for childcare (CFC) schemes supporting parental childcare at home or private care arrangements, motivated by ‘parental choice’ (Sipilä et al. 2010). This ‘de-Scandinavization’ has led some scholars to talk about Nordic childcare politics ‘edging’ towards a neo-liberal free choice model; others to consider the CFC schemes a challenge to the ‘traditional goals’ of the Nordic welfare states, while yet some see such schemes as a reflection of increasing regime ‘hybridisation’ (Knijn and Smit 2009; Eydal and Rostgaard 2011a; Mahon et al. 2012; Ellingsæter 2012b).

In comprehending the causes and consequences of growing family policy complexity, it is crucial to address both the underlying policy dynamisms and parents’ responses. Countries engage in reforms in different ways, depending on their model currently in place (Mätzke and Ostner 2010). Timing of reform may vary, and policy processes may influence policy outcome. Moreover, the meaning of seemingly similar national policies can be quite different. Often, the motivation behind them varies, and the context in which they are inserted is often different (Daly and Lewis 2000). Policy outcomes need to be addressed separately from policy output, or content of reforms, as the relationship between intentions behind family policy initiatives and parents’ responses is not straightforward. In exposing the dynamics shaping policy content and policy outcome, national case studies attending to complexity, context and chronology within the national setting are indispensable (Daly and Lewis 1998). Our approach to untangling the substance of policy hybridisation is to compare processes and outcomes of similar family policy reforms in similar welfare state models: CFC schemes for children under the age of three who do not use publicly subsidised childcare services in the Nordic welfare states. What does the insertion of a familistic policy measure in earner-carer models actually mean? The present comparison of the CFC schemes in Denmark, Finland, Iceland, Norway and Sweden aims at explicating the policy dynamics behind the institution of reforms and parents’ responses to this change in the institutional opportunity structure.

Change in family norms and mode of state involvement in family affairs is an ongoing process, not quite settled in scope and character (Mätzke and Ostner 2010). Mätzke and Ostner find that the timing of policy innovation, the role of ideas in processes of change and the agents of change appear to be significant factors in recent shifts in family policy, and these influences will be highlighted in the present analysis.

Policy analysis should be ‘placed in time’ (Pierson 2004). Cross-sectional correlations between policy arrangements and outcomes do not take into account the ‘temporal horizon’ of the resulting patterns. Policy timing and sequences are crucial for comprehending the introduction of reforms like CFC schemes and for policy outcomes. The temporal structures of explanatory accounts involving causes and outcomes may reveal substantial diversity: Some causal processes are slow moving and can be termed incremental or cumulative and some contain threshold effects (Pierson 2004). This means that outcomes of similar policy reforms may vary significantly across countries and over time, also because policies interact with other political and societal development. Thus, policy outcomes may vary depending on quantity, coherence and timing as Neyer and Andersson (2008) point out. This may be seen as path dependencies up to the actual situation. Many existing social arrangements are likely to exhibit inertial qualities; pressures may build up for some time without generating immediate effects. Social processes may take a long time to unfold and may be mediated through indirect effects on other social institutions (Pierson 2004; Neyer and Andersson 2008).

The main idea supporting CFC schemes is ‘choice’. This usually refers to more choice for families in arranging childcare, that is, a choice between childcare services and cash transfers for taking care of your own children (e.g. Rantalaiho 2010). The political rationale of ‘choice’ is associated with a liberalist anti-state paradigm setting limits for the state's interference in the private affairs of families; families should decide for themselves. Popular support of ‘parental choice’ may, however, also have other origins, for example, a desire for autonomy, or as a symbol for recognising a variety of life styles. Protagonists and antagonists of CFC schemes tend to follow traditional political party cleavages: A common denominator of the countries that have introduced CFC schemes is that centre-right governments stand behind (Morgan and Zippel 2003). CFC schemes have been criticised from parties on the left for being detrimental to gender equality as well as being a hindrance for groups that are already weak on the labour market, such as immigrant and lowly educated mothers.

The idea of ‘choice’ in the area of childcare is of complicated nature (Lewis 2008). In explaining parents’ responses, conceptualising preference formation, that is, circumstances shaping and changing preferences, is vital. Individual preferences are emerging in a gradual but continually adaptive process (Fagan 2001). Childcare choices can best be understood as contextualised patterns of action, that is, accommodations to family and employment demands, financial and other resources, filtered through norms about acceptable parenting and gender behaviours (Meyers and Jordan 2006). Choices depend on childcare ideals, access, costs and experience of various childcare options, as well as labour market opportunities. Individual choice is thus constrained on both the individual and structural levels (McRae 2003). Policies have a role in addressing constraints on choice when legitimising certain practices and making particular choices easier to realise (Lewis 2008). Thus, CFC is commonly expected to reinforce traditional gender roles by recognising and institutionalising the role of mothers’ care (Morgan and Zippel 2003). However, if policies are far from norms on behaviour, or even contradicts norms, they may have insignificant effects (Neyer and Andersson 2008).

CFC schemes can also be associated with scholarly cleavages. They can be classified as a familistic measure, supporting the family's caring function (Leitner 2003). For scholars who advocate gender equality and a ‘symmetrical’ earner-carer model where women and men engage in both paid work and unpaid caregiving, CFC benefits are considered a counterproductive measure. Inequalities in the domestic division of labour, especially caregiving, are seen as main obstacles to achieving gender equality, and CFC presumably reproduce gendered patterns (Gornick and Meyers 2008; Crompton 2009). By contrast, other scholars give preference to expanding parental ‘choice’ to accommodate various gender arrangements among diverse citizens (Orloff 2009). It is argued that women's emancipation does not necessarily demand dissolution of the gendered division of labour. Moreover, some scholars contend that CFC schemes should not be portrayed as a conservative reaction only; although it reflects traditional values, it can also be seen as an expression of neo-liberal and neo-familist values (Sipilä et al. 2010).

Gender equality is usually acknowledged as a hallmark of family policy development in the Nordic welfare states, supporting earner-carer models that encourage mothers’ labour force participation and redistribution of care; from mother to father in the family, and from family to society (Korpi 2000). It has been pointed out that the recognition of the value of care has been crucial in this development (Lewis 2001). The main childcare policy elements; legislation on paid parental leave and publicly subsidised childcare services have evolved gradually since the 1970s. The pace of development has varied across the countries, however, and differences between the Nordic family policy models have been noted (e.g. Lewis 1992; Leira 2002; Ellingsæter and Leira 2006). Historically, Norway's ambivalence and slower progress towards an earner-carer model have been characterised as a ‘puzzle’ (Sainsbury 2001), while Finland in recent years has been described as moving towards a ‘neo-familist’ path (Mahon 2002). Differences between Norway and Sweden have been interpreted as the two countries being on different paths, Norway as a modernised male breadwinner model and Sweden on the path of a dual breadwinner and public childcare (Pfau-Effinger 2005). By contrast, Danish family policy dynamics are considered subjugated to a ‘productivistic’ paradigm (Abrahamson 2010). Family policy development also in the Nordic countries is thus far more complicated than a straight line towards an adult worker model (Lewis 2002; Daly 2011). CFC has been seen as part of the path towards a 1 ½ worker model where women's work is still inferior to men's (Daly 2011).

Family policies have been an important field of partisan struggle in the Nordic countries, also in recent years, in a context of minority or coalition governments (Ellingsæter 2014a). Parental ‘choice’ versus ‘equal parenthood’ has been a main axis of conflict, in which CFC schemes sometimes feature prominently. The Nordic CFC schemes, where cash is given instead of day care, can be considered social policy newcomers, and the reform processes have been slow and far from easy (Sipilä et al. 2009). The schemes have been promoted by centre-right parties/coalitions and met with opposition from parties on the political left (Ellingsæter 2003, 2012a; Hiilamo and Kangas 2009; Duvander and Cedstrand 2012).

The following elucidates the political processes surrounding CFC reforms – timing, ideas, and role of political parties – followed by an account of the policy outputs of these processes.

The first country to introduce CFC schemes was Finland. The schemes date back to the early 1970s, when some municipalities started to pay cash for care benefits in order to limit the growing demand for childcare services (Rantalaiho 2009). Parents’ ‘choice’ with regard to childcare was then rapidly absorbed into the national family policy discourse; when the CFC (hemvårdsstöd/kotihoidontuki) was introduced in 1985, it was presented first of all as an alternative to childcare services (Rantalaiho 2009). Costs have been an overarching concern, as CFC is much cheaper than the alternative, childcare services (Hiilamo and Kangas 2009). Centre-right advocates framed the CFC with the value of work at home and parents’ opportunities to choose the form of care they need (Hiilamo and Kangas 2009). Also important to the dominant Centre Party (agrarian base) was that childcare services did not satisfy the needs of rural families, the benefit was to compensate lack of services (Lammi-Taskula and Takala 2009; Rantalaiho 2010). The CFC was opposed especially by the Social Democrats, but eventually, the reform was introduced as a compromise in combination with expansion in municipal childcare services. The CFC has become institutionalised as a main element of the national childcare policy model; there have until recently been no strong political initiatives to abolish the benefit (Hiilamo and Kangas 2009; Rantalaiho 2010). Yet, some scholars argue that the consensus is ‘relative’ (Salmi 2006). Recently, negative effects of CFC have been discussed, and in 2012, a proposal from the left-centre-right majority coalition to cut the allowance for the two-year-olds caused heated debate (Salmi and Lammi-Taskula 2013).

Profound political conflict characterised the CFC benefit (kontantstøtte) reform in Norway, launched in 1998 by a minority centre government, the Christian Democrats being the leading protagonist, supported by the right-wing parties. The Social Democrats and the left were fierce antagonists. The main aims of the CFC were to give families choice regarding care arrangement and more time to care for their own children, as well as a more equal distribution of public subsidies to families (Ellingsæter 2003). The gender equality concern was prominent in the left's opposition to the reform; it was expected to lead to a decrease in mothers’ employment, lower demand of childcare services, and a slowdown of service expansion. Polarised views among the political parties have persisted, but the emphasis of criticism of the reform has shifted, to paying more attention to the negative impact on integration of immigrant mothers and their children (Ellingsæter 2012a). This must be seen in light of changes in the profile of benefit recipients (next section). In 2012, the red-green government abolished the CFC for the two-year-olds. In 2014, a new right-wing minority government with support from the Christian democrats and the Liberal Party increased the benefit significantly for the one-year-olds, and future reintroduction of the benefit for the two-year-olds was announced. Some centre-right municipalities have continued to provide the benefit for two-year-olds (Ellingsæter 2012a).

Also in Sweden, CFC has a long history of political struggle. After several years of heated debate, a CFC (vårdnadsbidrag) was introduced in 1994 by a centre-right government (Nyberg 2010), but it was fast abandoned six months later by the new government headed by Social Democrats. In 2008, a new scheme was introduced by a majority centre-right coalition. The reform was a compromise among the coalition parties where the main protagonist was the small Christian Democratic Party. Part of the compromise was also a new gender equality measure: a gender equality bonus (tax break) in the parental leave system (Ferrarini and Duvander 2010). The main rationale for CFC was to increase families’ freedom of choice and it was made voluntary for municipalities to offer it. In 2012, about 40% of all municipalities had introduced a CFC benefit, mainly in centre-right local governments (Statistics Sweden 2012). The benefit is supposed to enable parents to spend more time with their children, and to support flexibility in the combination of employment and care. The political opposition considered CFC a ‘housewife trap’. Other worries voiced were that children would be losing out on early educational and pedagogic stimuli, and that CFC might take funds that secured the quality of publicly subsidised childcare. Critique was raised also from within the parties in government, for example, from the women's associations of the liberal and conservative parties. Concerns have also been raised that the overrepresentation of immigrants among users will increase segregation (Segendorf and Teljosuo 2011). The new government in 2014 led by Social Democrats has expressed critique of the CFC, mainly from a gender equality perspective. In the autumn 2015 it was decided to abolish the CFC in 2016.

The existing schemes in Denmark and Iceland are seldom mentioned in comparative studies of childcare (Rantalaiho 2010). In Denmark, the first version of CFC (Tilskud til pasning af egne barn) was introduced on a national level by a right-wing government in 1992. The main argument was to create new jobs in a time of economic crisis by creating rotation in the labour market. There was no clear opposition and the new centre-left government in 1993 kept the benefit. When a centre-conservative government reinstituted the CFC in 2002 together with a reform of the parental leave, the arguments were flexibility and choice. There has not been much debate regarding the CFC in Denmark, however (Duvander and Cedstrand 2012).

Iceland is the only Nordic country without national legislation on CFC (Heimagreidsla). It is up to the municipalities to decide upon. The majority of the municipalities introducing CFC have been run by conservatives (Eydal and Rostgaard 2011b). Thus, no national political process on CFC has taken place, but the argument put forth was equality between parents who had been offered a childcare place and those who were waiting. CFC can thus in many cases be seen as an interim arrangement. Since 2006, some municipalities have introduced variations of CFC with different setups and payments; in 2009, 13 out of 77 municipalities provided it, including Reykjavik (Eydal and Rostgaard 2011b). As a consequence of the economic crisis, most municipalities have now abolished the CFC.

To sum up policy processes then, Nordic CFC schemes have been introduced at different points in time; they have been subject to conflicts and amendments to different degrees (see overview in Table 1). While ‘choice’ is the main rationale, some schemes are also underpinned by valorisation of parental care and more equal distribution of state support between families. Another motive behind introducing CFCs is efforts to retrench public expenditure as CFC benefits are less costly than childcare services. Some schemes have developed from national claims for a mother's wage, and thus connote with a support of the stay-at-home mother (Leira 2002). However, this idea is modernised and repackaged as parental ‘choice’, in line with the influential choice discourse entering Nordic social policy debate in the 1990s. In general, the motivation of Nordic CFC benefits is gender neutral; there is no explicit aim of reinforcing complementary gender roles, though this might be the implicit consequence (see Leitner 2003).

There are some similarities regarding policy output, that is, content of the schemes, but policy processes have also resulted in some important differences that potentially may impact on parents’ responses (for details, see Table 2). Finland, Norway, and Sweden have adopted national legislation, although provision is voluntary for municipalities in Sweden. The benefit is about 10% of the income of an average wage earner in these countries (Bakken and Myklebø 2010; Eydal and Rostgaard 2011b). CFC benefits thus involve low payment, presuming a main (male) breadwinner in the family and are therefore likely to support a traditional family model and gendered division of labour. Whether CFC can be used together with other cash transfers is likely to be more important than the difference in benefit amounts between countries. In Finland and Norway, the CFC can be combined with other transfers, which is not the case in Sweden. It is for households with two earners, with a large income difference in the couple that CFC becomes most affordable (Tunberger and Sigle-Rushton 2011). In combination with other cash transfers and household income, the variation in disposable incomes will influence what groups are typical users of the CFC. This is important not just for households with unemployed or sick parents, but perhaps primarily to one-parent households. Single-parent households are about as common in all Nordic countries, but the risk of poverty may vary between countries. By contrast, usage of publicly subsidised childcare makes less distinction between one- and two-parent households.

In Finland, the current CFC pays benefits for children up to age 3, in combination with other transfers or earned income. The benefits partly depend on family income and vary between municipalities. Extra municipal supplements provide incentives, especially for low-income families to rely on the allowance as an alternative to childcare services (Sipilä and Korpinen 1998; Repo 2010). In addition, the allowance structure encourages families also to care for their older children at home. In Norway, the benefit currently is paid only for one-year-olds, but can be used part-time and be combined with earned income and other transfers from the state. Until 2012, CFC could be used for 1–2-year-olds. For single parents (read women) out of labour market work, the CFC may provide a further source of income. In Sweden, in the municipalities that offer CFC, the benefit can be used until the child turns 3. But it cannot be combined with any other social transfers in the household, and over half of the parental leave days have to be used before CFC can be claimed. These restrictions essentially limit the users to families with one other adult in work. The CFC can be used part-time and also in combination with individual earned income to pay for privately for childcare, for example, a nanny (Duvander and Cedstrand 2012). In Denmark, the current CFC can be used for maximum one year at any time before the child starts school. Municipalities set the benefit at maximum 75% of the cheapest cost for childcare services and the benefit varies substantially between municipalities.

The national outcomes of reform in terms of parents’ responses to the new incentives have been quite diverse. On the one hand, perhaps not unexpectedly, the more restricted Danish and Icelandic schemes have had little practical influence on parents’ childcare choices (Table 2). On the other hand, there is wide variation in parents take up of the CFC benefit in Finland, Norway and Sweden.

In Finland, the use of CFC is highest among the Nordic countries and it has been quite stable since its introduction (Repo 2010). Around 50% of all eligible children receive CFC per year; about 9 out of 10 children receive CFC at one point in time (The Social Insurance Institution of Finland 2012). About 95% is paid to women. Municipalities where the benefit is highest have the largest share of users (Kosonen 2011). After the introduction of CFC in Norway, a high proportion of parents received the benefit, but it has since plunged: At the end of 1999, 75% of all parents of 1- or 2-year-olds received the benefit, compared to only 25% of parents at the end of 2011 (Ellingsæter 2012a). The significant increase in the amount for the 12–18-month-olds in 2012 resulted only in a slight increase in the take up rate for this age group. About 80% of those receiving the benefit are women. The share of recipients has declined in all groups, but those receiving the benefit have become more homogeneous: Parents with low income and education, immigrants from Asia and Africa and mothers with weak labour market attachment are overrepresented (Bakken and Myklebø 2010). In Sweden, few parents use the benefit. In the municipalities that have introduced the scheme, the proportion of parents with children in eligible ages who received it sometime during 2012, was 3.9% (7082 children) (Statistics Sweden, data on demand). This represents 2.1% of 1- to 2-year-olds in the whole country. The use varies among the municipalities who have CFC; from under 1 to over 10%. Even if numbers are low, they represent an increase since the CFC was introduced in 2008. More than 90% of the benefit recipients were women. Young parents, parents with low education, parents of more than two children, and foreign-born parents are overrepresented among the users (Statistics Sweden 2012).

Variation in the contents of the schemes described in the previous section can to some extent explain differences in take up rates. However, there are also other potential explanations. One could be the changed demographic profile with an inflow of non-Western immigrants; the use of CFC is higher among mothers with immigrant background. But the use of the benefit is much lower in Sweden, the country with the largest immigrant population, compared to the high take up in Finland, with the smallest immigrant population. Another alternative explanation could be variation in national childcare norms. However, in general, Nordic differences in gender equality norms are small, and when they do vary, as in the case of care for the youngest children, they seem to interact closely with differences in structural settings. This indicates that the CFC reforms may interact with different institutional contexts shaping parents’ opportunity situation. In the following, this point is elaborated by considering the interaction between CFC schemes and other elements of the national childcare regimes, and with national labour market structures and demand.

Two aspects of national childcare regimes may be particularly relevant: One is the length of the paid parental leave period and the degree of flexibility in the transition to work. The other is access to childcare services after the parental leave.

The potential duration of use of CFC for children under 3 is influenced by the length of the leave, as the CFC benefit usually follows after the expiry of parental leave. Sweden stands out has having the longest leave of 13 months with 80% wage replacement, plus 3 months at a flat rate (Duvander and Lammi-Taskula 2011). In addition, this leave can be extended further if a lower replacement level is accepted, which is often the case. This is a major reason that childcare enrolment is low before age 1 in Sweden. While Denmark, Finland and Iceland have a history of shorter parental leave, the Norwegian leave has been about a year since early 1990s. In both Denmark and Iceland, and to some extent in Norway, the CFC seems to be used while waiting for public childcare. This waiting time did not exist when the CFC was introduced in Sweden as childcare services were (more or less) fully expanded (see below).

The flexibility of the parental leave system has been less developed in Norway and Denmark than in Sweden, but even less so in Finland. In Sweden, it has been possible to use the leave until the child is 8, and it may be used in many different forms of part-time leave, combined with employment or unpaid leave. This flexibility is used by Swedish parents to a large degree (Duvander and Viklund 2014). In Denmark, one may use the leave until the child is 9 years old. In Norway and Iceland, the leave may be used within three years from birth, while parents in Finland have to use the leave directly after birth. The degree of flexibility in the parental leave system makes the CFC play different roles in the various countries, especially in its capacity of bridging the gap between parental leave and childcare services. The gap may occur because childcare is not available or the norm of ideal age at childcare start does not coincide with the length of parental leave.

Publicly subsidised childcare is the alternative to CFC in most cases, thus access to affordable, high-quality childcare is crucial to how mothers perceive the choice of childcare arrangements. In Finland and Norway, the CFC was introduced in a period when public childcare was not a readily available alternative, while in Sweden, the CFC came long after the main expansion of childcare. Today, in all Nordic countries, children have a universal right to a place in childcare services; Iceland is the only exception. The right to childcare is from age 1 in Norway and Sweden and from 6 months in Denmark. Maximum fees that parents pay for childcare are relatively low (Eydal and Rostgaard 2011b). However, the CFC may influence the supply and demand of childcare services significantly. Only 29% of 1-year-olds in Finland are enrolled in childcare services, compared to 51% in Sweden and 69% in Norway (Table 3). The general decline in use of CFC in Norway can be understood in the light of changes in childcare services since the early mid-2000s: strong expansion of places for children under 3, lower fees and introduction of the right to a place (Ellingsæter 2014b). Denmark stands out as having the highest enrolment rates also at early ages. The Icelandic numbers are low for the youngest children, mainly because of lack of available childcare services (Ingolfsdottir 2013).

Labour market structures are likely to contribute to the high CFC uptake in Finland. By contrast to the other Nordic countries, Finland is among the OECD countries where employment rates for mothers with children under age 3 are markedly lower than for mothers with older children (OECD 2010). In Norway, the introduction of CFC benefit was associated with a standstill in employment rates among mothers with children under 3, and estimation studies indicated a decline (Rønsen and Kitterød 2010). But employment has been increasing in parallel with childcare expansion in the second part of the 2000s (studies referred in Ellingsæter 2012a). According to National labour force surveys, three out of four mothers with children under 3 years old are employed in Norway and Sweden, compared to one in two in Finland (available numbers at the homepages of the National statistics bureaus; www.scb.no; www.scb.se; www.stat.fi). In Finland, long care leaves have gained strong legitimacy and are considered in the best interest of the child (Rantalaiho 2009; Repo 2010). It is suggested, however, that almost half of all mothers receiving the CFC have no real choice between paid and unpaid work, as they have no job to return to (Haataja and Nyberg 2006). Temporary employment is common in Finland and often work contracts end at the occurrence of parental leave (Salmi et al. 2012). The highest long-term unemployment rates have been found among mothers with children aged 3–6, that is, after the CFC cannot be used anymore (Haataja and Nyberg 2006). Moreover, it is less common to reduce work hours in Finland. In 2014, only 19% of women in the labour force worked part-time, while the same figure was around 40% (30% in Iceland) in the other Nordic countries where parents with young children have the right to reduce hours temporarily (see ec.euoropa.eu/eurostat/).

To sum up, childcare alternatives and labour market structures develop over time and it is important to recognise the context at the introduction of CFC as well as the potentially dynamic interplay over time. In Finland, the CFC became an important alternative to unemployment during the economic crisis of the 1990s. In Norway, the introduction of CFC met a shortage of public childcare, but CFC got less important when childcare services expanded for this age group in a context of very low unemployment levels. Sweden introduced CFC in a context of (more or less) full coverage of public childcare and strong incentives to labour market work. Also in Denmark, public childcare has been available and CFC has not been seen as a viable alternative. In Iceland, with a much shorter history of generous family policy, it was possible to abolish CFC at time of economic hardship.

Labour market structures leave very different choices to women in the Nordic countries and CFC is given different meanings in different contexts.

Regime hybridisation is likely to build up through different national policy dynamics, which might result in differences in both policy output and policy outcome. This study of the CFC schemes in the Nordic countries adds to our understanding of hybrid family policy models, studying the causes and consequences of inserting a familistic policy incentive into developing earner-carer models. Policy dynamics generating Nordic model hybridity display both similarities and differences, while policy outcomes vary widely.

CFC evidently is a contested policy in the Nordic welfare states, enmeshed in political conflicts and negotiated compromises. Sometimes this has delayed the introduction of reform and compromises and adjustments have affected the contents of the schemes. It has been emphasised in recent social policy studies that neither parties nor their ideologies are unitary but the result of continued struggles over interpretation (Hiilamo and Kangas 2009; Mätzke and Ostner 2010). Nevertheless, Nordic CFC reforms mainly follow a partisan logic; the left versus centre-right distinction continues to be important in family policy, ‘equal parenthood’ is competing with ‘parental choice’ (Ellingsæter 2014a). The particular driving force of small confessional parties in both Norway and Sweden is noteworthy though.

To summarise national trajectories: Finland has the longest history of CFC, starting in the 1970s and becoming national in 1985. The use has been stable and high as almost all mothers of young children use CFC at any point in time. Only recently has the policy been challenged. Thus, Finland has continued on a ‘familist’ path identified a decade ago. Norway has provided CFC since 1998. At the beginning, parental take up was high, but the share of recipients has decreased dramatically and the time period CFC can be used had been restricted. Norwegian parents have definitely abandoned the male breadwinner model. In Sweden, the first CFC reform in 1994 had a very brief existence, but in 2008, municipalities got the right to reintroduce it. Parents’ use has since been very low, although slightly increasing. Denmark's CFC has a long history but is limited and with many changes and minimal use, indicating that it never got a foothold. In Iceland, CFC can be seen as an interim solution, also volatile in its construction as it was broadly taken away by the event of the economic crisis. Neither policy content nor take up rates are fixed once and for all: Norwegian policy is being limited and Finland's at least challenged, while Sweden's policy was introduced in a similar form as the other two countries. The take up of the benefit still diverges, where Finland continues at a high level and even if the use is reduced in Norway, it is still much higher than the use in Sweden.

Variation in the content of the schemes represents different choices for parents and influences take up rates. The limited reach of the Danish and Icelandic schemes are likely in part to explain the limited influence on parents’ use of childcare. The Swedish scheme also has been more restricted than those in Finland and Norway, both in terms of eligibility, that is, the possibility to combine it with other state cash transfers, and the take up is very low also in municipalities where CFC is offered. Not just content, but also timing of policy reform is likely to influence outcomes. In Sweden, high coverage rates of public childcare were established before the CFC was introduced, and the Swedish parental leave is both long and flexible. Different timing of reforms means that they have interacted with different institutional opportunity structures, especially access to childcare services and labour market opportunities.

Yet, we conclude that CFCs are unlikely to fundamentally undermine the Nordic earner-carer model, at least short term. Combining work and childbearing is still the dominant pattern.

We also conclude that conceptualisation of CFC schemes needs to be contextualised. In developing earner-carer models like the Nordic ones, CFC schemes partly will result in policy compromises that may differ from other types of family policy models. Moreover, the degree of maturity of the earner-carer opportunity structure is likely to affect mothers’ choices. Still, the question about ‘choice’ and equity needs be addressed. Parental ‘choice’ in childcare policy has an implicit gender dimension with regard to who should take care of the children. Alternative care and employment opportunities mainly affect women's choices, although the presence of a main provider is sometimes a prerequisite for the use of CFC. Choice of a long period of childcare may limit women's future labour market integration. Thus, choices of childcare may augment socio-economic polarisation between women as well as between households. User profiles across the Nordic countries suggest that the choice of CFC often is a result of constraints among women with the least labour market resources: young mothers with low education, foreign background and many children. If CFC is chosen because of lack of alternatives, this seriously questions the promotion of CFC as a real ‘choice’.

We wish to thank Sofie Cedstrand and the anonymous reviewers for helpful comments.

No potential conflict of interest was reported by the authors.

Ann-Zofie Duvander is Associate Professor of Sociology and Demography at the Department of Sociology, Stockholm University. Her research interests include family policy as well as the family and work-connection. She is an expert on parental leave policies in Sweden, and she has conducted several studies of variation and change in parental leave usage and its effects. She has published in journals such as Journal of Marriage and the Family, Demographic Research, European Sociological Review and European Journal of Social Policy.

Anne Lise Ellingsæter is Professor of Sociology at the University of Oslo. Her research interests include family policy, work-family relations, gender and work, and fertility. She has published in journals such as Journal of Social Policy, Social Politics, and Journal of European Social Policy. Her most recent book is The Social Meaning of Children and Fertility Change in Europe (with A.-M. Jensen & M. Lie, Routledge 2013).