Quasi-Market Regulation in Early Childhood Education and Care: Does a Nordic Welfare Dimension Prevail?

Until about 25 years ago, Norway, Denmark, Sweden, Iceland and Finland used a bureaucratic-professional governance model to reach common welfare goals in ECEC, something that arguably constituted a Nordic dimension. Since then, the countries have introduced post-bureaucratic governance models such as evaluation and quasi-markets in varying degrees. Differences in the timing of policy changes and saturation of demand have resulted in variations in the use of quasi-market instruments and divergence in the composition of public, for-profit and non-profit providers. However, common welfare goals, such as inclusion and language learning, have been strengthened. We base our analysis on existing research, statistics and policy papers


Introduction
A Nordic dimension in the politically defined goals for early childhood education and care (ECEC) can be identified in the shared history of Sweden, Denmark, Norway, Finland and Iceland. In all of these countries, the service was first established by non-profit providers, with municipalities gradually becoming the dominant providers as ECEC matured to become a general welfare service around 1970. Consequently, ECEC became instrumental for several of the main Nordic welfare goals: increased female work participation, high-quality welfare services for all and reduced social inequality through education and public subsidies (Kautto et al., 2001;Rothstein, 1998). The shared welfare goals for ECEC thus seem to reflect a Nordic dimension.
At the same time that ECEC was gradually being established as a general welfare service, the bureaucratic-professional governance tools used in other core welfare areas were also put into place for ECEC. The municipalities became responsible for funding, steering and increasing service delivery, while national authorities were responsible for inspection and regulation through subsidies, legislation and guidance documents. The Nordic dimension of politically defined welfare goals seemingly presupposed the bureaucratic-professional regulation that has dominated these welfare societies.
Over the last two decades, we have seen an erosion of bureaucratic-professional regulation in many areas of Nordic welfare . In line with the conceptual framework developed by Maroy (2012), we see that countries have replaced or supplemented a bureaucratic-professional governance system with post-bureaucratic regulation. For compulsory education, Maroy points to two forms of postbureaucratic regulation: (1) quasi-market instruments with user choice and vouchers to create competition between public and private service providers, and (2) evaluative tools with performance assessment, incentives and sanctions used to reach politically defined goals.
In this article, we compare changes in governance tools for ECEC in the five Nordic countries. Do we see a replacement of bureaucratic-professional regulation by quasimarkets and/or by assessment and incentives? If so, has this led to a redefinition of the political goals that characterised the Nordic dimension during the expansion period of the welfare state?
One hypothesis is that the inability to reach these goals may have led to goal replacement, perhaps inadvertently. In Sweden, for example, freedom of choice resulting from vouchers and competition between public and for-profit compulsory schools seems to rank higher in political discourse than the previous goal of reducing inequality (reasons for this are discussed in Svallfors & Tyllström, 2018). A contrary hypothesis is that the inability to reach defined goals has led to reforms of the governance system to better reach these goals.
First, we present our conceptual model for ECEC governance and institutional change. We then present the traditional Nordic ECEC model and the most important changes that have taken place over the last twenty years. Our inferences are based on national and Nordic statistics, public documents and a review of the existing research, all of which are more extensively presented in a previous report than this paper allows (Traetteberg et al., 2021).

Analytic approach: Governance models and institutional change
ECEC was, until twenty or thirty years ago, governed in a bureaucratic-professional way, in line with other service areas in the welfare state, through legislation, bureaucratic institutions and the directives of elected leaders and appointed officials in a Weberian fashion (Karila, 2012). To secure central control, provision was predominantly provided within the public sector organisation, with mainly non-profit providers as a supplement. When we analysed developments in Nordic ECEC, we did this considering two conceptual alternatives to the professionalbureaucratic governance model: a quasi-market model or an evaluator-state model (Maroy, 2012).
The establishment of quasi-markets for ECEC services means that the state tries to emulate market mechanisms, but within a public regime where the state sets the price, determines quality standards and funds the service (Le Grand & Bartlett, 1993). A core aspect of an ECEC quasi-market is a user choice scheme that enables parents to choose which institution to enrol their child in, which can replace bureaucratic steering in the assignment of places. A second important aspect relates to the fact that user choice directs funding to the provider, as parents, in this way, are de facto provided with vouchers. This may create a competitive dynamic, as ECEC institutions are dependent on attracting families to achieve funding. A third important factor is whether any actor who fulfils certain criteria has a right to establish new ECEC institutions. This would give market forces room to define service coverage and limit the ability of elected and appointed officials to control the composition of service providers or where they are located according to geographic and socio-economic dimensions. A fourth factor is the limits on profits that private providers can make.
Traditional, professional-bureaucratic logic may be used to counteract some of the possible unintended consequences of quasi-market developments. This can be done, for example, by regulating the user fee, the content of ECEC, standards for the educational level of staff, and children per staff ratio or group sizes. In practical terms, this constitutes an indirect limitation of profit that can be transferred to owners of nonpublic ECEC institutions. Furthermore, on the national level, there may be variation over time in the use of certain regulation measures, or even giving municipalities the right to choose which regulatory mechanisms to implement -such as allowing free establishment of private institutions or not.
In the evaluative state, authorities set goals, gather information about service provision and provide incentives. Alternatively, public agencies may implement broader policy changes or react to institutions that do not deliver according to standards. This secures some level of autonomy for ECEC institutions (Maroy, 2012). The basic idea is that this enables local units to innovate and improve their operations. It will also enable them to adapt to user preferences and adjust to local needs and conditions. At the same time, local freedom is accompanied by regulation and inspection by national or local authorities and the demands of formal internal control. This means that service providers are expected to adapt their activities to improve their scores on given measurements (Ball, 2003), which is something we see in Nordic elementary schools (Elstad & Sivesind, 2010;Wallenius et al., 2018).
The ECEC field constitutes a specific institutional set-up. Changes in this institutional set-up can be abrupt, gradual, complete or partial. According to historical institutionalism, abrupt change can often be explained by policy failure, which opens a window of opportunity for radical change (Pierson, 2000). Alternatively, gradual change can take place if different governance tools do not replace each other, but are layered upon each other (Mahoney & Thelen, 2009), with defined areas of application, or even with the option to use one tool or another according to what seems appropriate in a given situation.
Furthermore, to explain why change takes place, discursive institutionalism is centred on the role of ideas and discourse in politics. Discursive institutionalists regard institutions as something that frames the actions of agents but also as something that is constituted by the actions of agents. To explain change, we need to look at the ideas behind the actions of the people driving them. Agents create change, but they do so within the context of the institutions and the ideas conveyed in discourse, which serves as both internal coordination and external communication (Schmidt, 2008). Today, all the Nordic countries also have for-profit providers of ECEC. This is important for at least two reasons. First, the need for supervision, control and transparency may be different depending on the service-providing actors. This is especially relevant for the for-profit actors, according to principal-agent theories (Anheier, 2005). For example, ECEC institutions that have been built up with public funding of properties, buildings and training of service personnel can be sold and the proceeds kept by private owners, while the buying company plans to accumulate revenue from the investment. Second, a large non-public sector may limit the ability of the public sector to steer this service area. This can occur partly because command and control tools are not available to providers outside of the public hierarchy, but also because private actors, such as limited companies, have legitimate economic interests that merit some sort of protection from public over-reach. A large private sector can also mobilise resources and support from business federations or other interest groups to limit transparency and regulations (Svallfors & Tyllström, 2018). In this sense, it is also important whether the for-profit sector consists of many separate units or whether it is dominated by a small number of big chains that have strong incentives and the ability to influence policy.

Shared tradition -bureaucratic-professional governance to reach common goals
The goals in Nordic ECEC have historically been about children's welfare and development and enabling parents' labour market participation. As ECEC became a general welfare good in all these countries from around 1970, these goals remained central.
In addition, it was an independent goal to increase the labour supply by enabling mothers to enter the work force and, in turn, give flexibility to families. Like other Nordic welfare services, high-quality service for all users across any geographic or socio-economic divide was also important in the ECEC area (Rothstein, 1998).
Additionally, as these extensive welfare states developed services that would follow citizens 'from cradle to grave', it became important to make the services responsive to the preferences and wishes of the users (Goul Andersen & Hoff, 2001). Hence, ECEC reforms are intended to give parents opportunities to take part in governance and to choose an operator.
In order to achieve these goals, a bureaucratic-professional governance model was used, where the municipalities in all the countries were identified as the main governance actor with important responsibilities as local administrator, supervisor and, most prominently, provider of services. Yet, as the international trend of new public management (NPM) also gained traction in Nordic educational services, this approach was challenged by alternative, post-bureaucratic approaches (Gunter et al., 2016).
In later years, two further goals manifested themselves. First, the observed positive effects of ECEC on children's development of language and social skills have gained increased attention, leading to efforts to enrol children in ECEC as the first step on the educational ladder. Second, this is especially relevant as the immigrant population has grown in these countries, and ECEC has increasingly become a vehicle for social integration, language learning and inclusion of immigrant parents in the labour market (Garvis et al. 2019).

Increased use of ECEC and changes in the welfare mix
A prominent new development in the goals for ECEC that we find in all countries is that of social investment and considering ECEC as the first step on the educational ladder.
In practical terms, the national and municipal governance of ECEC was moved from social service administrations to education. This implies that more children should be enrolled in ECEC to achieve the positive benefits from the service, something that is expected not only to be beneficial for the individual person, but also for the human capital and social cohesion of the countries.
In Figure 1, we demonstrate how already high levels of ECEC attendance have increased over the last thirty years for three-to five-year-old children, who are the core users. It shows that, particularly Denmark and Iceland, but also Sweden, were early movers in expanding the service, reaching 95% coverage in 2003, and then shortly stagnated, indicating that demand for all practical purposes was covered.
Norway followed suit four years later, while Finland has only recently begun more rapid expansion, reaching 88% coverage in 2019. In Norway, difficulty reaching full demand and its later development than its Scandinavian neighbours were important as a context for a political compromise in 2003 that would allow for-profit actors to obtain resources on an equal level as public and non-profit institutions (St. meld. nr. 24, 2002(St. meld. nr. 24, -2003. In Finland, familyoriented policies are part of the explanation for later growth (e.g. Eerola et al., 2022).
In policy discourses, care at home, with all its positive characteristics, was contrasted with bureaucratic institutional care, and 'freedom to choose' was a hegemonic framing (Hiilamo & Kangas, 2009).
Although public provision, central steering and bureaucratic administration are key words used to describe traditional Nordic ECEC governance, non-public actors have always been present in these countries. Non-profit actors were instrumental in establishing ECEC as a service area, and these actors never disappeared, even if the extent to which they were able to remain important parts of the welfare mix has varied among the countries.
Since the 1980s, the role of for-profit providers has increasingly been a point of political conflict in Nordic ECEC, resulting in the five countries having different approaches to the issue. In Figure 2, we show developments in the welfare mix by the percentage of children attending public ECEC. Unfortunately, we do not have good, The original struggle over for-profit provision occurred in Sweden, where this became a highly ideologised issue that began in 1984 when the Social Democrats introduced a law -Lex Pysslingen -to ban for-profit ECEC. Over the course of time, ECEC was gradually opened to for-profits (Westberg & Larsson, 2020). Figure 1 shows that as the demand became saturated, growth in the percentage of children in ECEC gradually slowed down until it reached 98% in 2007. Figure 2 shows that from 2008 to 2020, the comparative data on the division between for-profit and non-profit provision, but the figure shows a decline in public provision from 1998 to 2020, a period when for-profit providers entered the scene in all these countries. However, for Sweden, data were only available from 2008 and onwards, and for Denmark, from 2007 to 2014. We do not know the exact percentage of private ECEC in Finland before the voucher system was introduced in 2009 ( Kela, 2021). Also, the recipients of ECEC vouchers were included in the statistics as late as 2015 even though they had already been taken into use in 2009.
The proportion of public provision between 2009 and 2015 may therefore be somewhat lower than is shown in Figure 2.  public sector share of children in ECEC was stable at around 80%. In a previous report, we showed that from 2008 to 2020, the for-profit sector in Sweden increased from 8% to 12% of all children in ECEC, while the non-profit sector declined from 10% to 8% (see Figure 4.2 in Traetteberg et al., 2021). At the same time, the public sector decreased from 81% to 79% of all children ( Figure 2).
Norway, which in many service areas is reluctant to permit for-profit provision of welfare (Sivesind, 2017), has seen radical developments in ECEC. As a means to increase ECEC coverage, the Norwegian Parliament in 2003 chose to use private providers to expand supply. Subsequently, the share of children attending ECEC increased from 85% to 97% (Figure 1). Public sector coverage decreased from 58% to 50% of all children (Figure 2). In absolute terms, employment in public, for-profit and non-  Figure 2.2 in Traetteberg et al., 2021). However, the most remarkable development has been the rapid growth of a few for-profit chains increasingly dominating ownership in the for-profit ECEC sector (Lunder, 2019).
In Denmark, there is a lack of data differentiating between non-profit and forprofit sector provision by number of children or employment. The only relevant table in Statistics Denmark's databank shows the number of institutions. This is an unreliable indicator of changes in the welfare mix, since smaller ECEC institutions may be closed down or merged with larger institutions as part of structural rationalisation processes.
The data show that from 2017 to 2021, the number of non-profit, self-owned institutions decreased from 569 to 490, public institutions decreased from 2,872 to 2,803, while other private institutions increased from 505 to 568. It is important to note that some self-owned non-profit institutions, which were included in the municipal coordination, opted for the private organisational form due to the increased autonomy allowed for this category (Børn & Unge 2014). Therefore, the number of ECECs does not necessarily reflect changes between the for-profit and non-profit sectors, but rather a small variation in how private entities are governed by the funding municipalities (Thøgersen, 2013). The most important conclusion is that municipal provision of ECEC has remained stable, around 78% from 2008 to 2014 (Figure 2).
Finland has seen a development similar to Norway's: growth in for-profit providers. In 1997, PDA became available for all families whose children were not in municipal ECEC (Kela, 2021). In 2009, the introduction of vouchers contributed to a gradual decline in the share of public provision, falling to 79% by 2019. The increasing dominance of big for-profit chains is a remarkable feature in Finland (FINEEC 2017(FINEEC , 2019. The share of private provision in Finland has bypassed that of Iceland and Sweden in recent years (Figure 2). Neither the expansion of the share of the private sector nor the increase in its coverage has been an explicit aim in national policymaking. In Finland, municipalities have strong autonomy in deciding how they subsidise private provision of ECEC, and there is a wide variety of the share of private provision in different municipalities. When municipal policymakers rationalise increased subsidising of private providers, they refer to economic reasons, greater opportunity to govern private provision with the help of service vouchers (the share, fees and quality) and increasing families' freedom to choose (Ruutiainen et al., 2020).
In Iceland, as shown in Figure 1, high ECEC coverage, with 96% of three-to five- year-olds in ECEC, was reached before new legislation was passed in 2008 that allowed for extended user choice and more deregulation and privatisation in the ECEC field.
However, there is a lack of data in Iceland differentiating between for-profit and nonprofit ECEC with regard to the number of children or employment. Nevertheless, it is evident that from the year 2000, nearly all private provision of ECEC in Iceland has been limited to for-profit companies which have crowded out the non-profits. Figure 2 shows that the public share of children in ECEC was decreasing before the 2008 legislation and continued to decrease from 88% in 2008 to 83% in 2020. The privatisation that followed was therefore not important for reaching full coverage, and policy discussion indicates that the main purpose was to increase freedom of choice. However, the ideol-

Changes in governance models
User choice is, to some extent, present in all five countries, and at a minimum in some municipalities. To the extent that the supply of ECEC places allows for it, parents can choose among different providers. In practically all places, the goal of including private providers is primarily to empower parents and increase supply, not to reduce public costs through competition. However, the dynamic is different among the countries.
In Norway, ECEC governance experienced a watershed moment in 2003, when a parliament majority decided to invest massively to reach full coverage of demand from a level where about two-thirds of children attended ECEC. The new legislation did not specify any difference among actors in the welfare mix, as the money followed the child in a de facto voucher system, and crucially, private actors had an almost free right to establish. In addition, limitations on profits that stipulated a 'reasonable level' were, in practice, of little consequence, as the public authorities did not have much oversight, few resources for auditing and few opportunities to sanction. This combination had at least two consequences. First, it led to a rapid increase in the supply of ECEC ( Figure 1). Second, it led to faster growth in for-profit providers (see Figure 2.2 in Traetteberg et al., 2021). While the right to establish became limited in 2011, strong economic incentives for expansions in supply had already made lasting changes in the market. By then, 97% of three-to five-year-old children were in ECEC (Figure 1), restricting the room for further organic growth for private chains. This resulted in a strategy of continued growth through buying up smaller independent institutions, thus concentrating ownership in the field (Lunder, 2019).
In Norway, decreased government control through reduced direct public provision and a growing for-profit presence has led to more detailed regulation of ECEC.
The background for this is partly a reaction to the criticism that the for-profit providers are able to reap an undue level of profit (BDO, 2018;Lunder, 2019;NOU, 2020: 13;Storberget et al., 2021). An indirect way to cut profits and at the same time raise quality was the adoption of a stricter regulation of staff ratios and education enacted in 2018. Apparently, some of these changes have cut profit margins in private ECEC (Bjøru et al., 2021). At the same time, a well-organised private sector has been able to block other legislation that it believes would weaken its framework conditions. In further efforts to increase transparency, a new national unit for economic supervision was established, and in 2022, the Norwegian Parliament adopted legislation that requires each ECEC institution to be a separate judicial subject that can only borrow money from regular banks (Utdanningsdirektoratet, 2022).
The Danish welfare model has traditionally had more user choices than in the other Nordic countries . The current welfare mix is rather stable, as there have been few consequential changes to the governance system. Denmark is the country with the strongest tradition for non-profit, free schools and self-owned institutions, representing an important alternative to the public service in welfare services.
In its ECEC, Denmark has a dual system where self-owned institutions are tightly integrated with the municipal service with caps on tuition and applications for places managed through municipal administration (Thøgersen, 2013). These self-owned institutions are all non-profit foundations. In addition, there is a newer category of private ECEC institution that has more autonomy from the municipalities -they can set their own tuition, and they have the right to establish as long as they fulfil certain legal criteria. They can be both for-profit and non-profit; however, we do not know how many actually are non-profits (Brogaard & Petersen, 2020). While the municipality administrates the places in municipal and self-owned institutions, private institutions have an entirely independent system of user choice.
In new legislation passed in 2020, new minimum standards for child-to-staff ratios were established in Denmark, where private providers will receive full coverage of expenses. The political compromise brokered to achieve a majority for this legislation included provisions that the entities would no longer be allowed to transfer profit to owners, which essentially excludes for-profit operations from the Danish ECEC sector. In February 2022, the legislative majority for this provision disappeared (Kirkeby Theilgaard, 2022). It is nevertheless a strong indication of the political willingness to govern the welfare mix in the Nordic state that has probably seen the fewest changes in the ECEC welfare mix, where the existing for-profit institutions are mostly small, local entities that tend to operate in a way similar to the non-profits, and the big chains are absent (Henriksen et al. 2016, p. 227).
In Iceland, a milestone was reached in 2008, when legislation affecting the entire education system was passed, enabling further private growth in ECEC. In general, the emphasis was on deregulation and autonomy, as each school was supposed to write its own curriculum based on its philosophy and methods. The main reason for allowing pri- for-profit institutions that were included in the municipalities' plans were allowed, which meant that municipalities had the option to ban for-profits. However, in 2008, the law granting free right to establishment was passed (Westberg & Larsson, 2020).
All applicants who fulfil the demands of relevant laws must be approved by the municipality (SOU, 2016, p. 78). Our data show that full implementation of vouchers in 2008 had no effect on the percentage of children attending ECEC, which had reached 98% (Figure 1), and little effect on public sector coverage at around 80% (Figure 2), since demand in practice was saturated at an earlier stage. The most important change was that for-profits increased their share from 8% to 12% of children between 2008 and profit institutions, including some that previously were non-profits, has been concentrated in five or six chains (Enfeldt, 2022, pp. 31-32).
In Finland, from the establishment of ECEC legislation in 1973 until 1993, governance of ECEC was state-led (Kinos & Alila, 2014). To obtain permission to act as an ECEC provider, the private operator needed to be accepted by the State Provincial Office. Fees were regulated and profit-making was not allowed.
Prohibition of profit-seeking was abolished in the 1980s as part of the social and health care reforms (Laiho & Pihlaja, 2022). In 1997, private day care allowances were introduced to provide wider opportunities for parents to organise childcare (Law 1128(Law , 1996. Around the same time, municipal autonomy in organising basic services increased. However, it was not until the introduction of service vouchers in 2009 that there was an increase in private provision. Prior to that, the share of private provision had been, compared with other Nordic countries, small and stable ( Figure 2). According to Laiho and Pihlaja (2022) When it comes to evaluative steering, we find less differentiation. As mentioned, even though the state regulates quality standards, the municipalities in all of the countries are the most important actors regarding supervision of the content provided (bar the public ECEC in Sweden). This contrasts with the school sector, where the Nordic countries have seen a growth in the use of national evaluation and assessment systems (Wallenius et al., 2018)

Concluding discussion: Changes in governance tools vs. the Nordic dimension
We have observed fundamental changes in the public governance of ECEC in the Nordic countries. There are significant differences between Norway, with a high share of private providers and fast growth in for-profit chains, and Denmark, which has had a large share of non-public providers for a long time, but where non-profit, selfowned institutions have dominated, and banning for-profit operation has been on the agenda. The other Nordic countries are somewhere between these two extremes. Does this mean that the Nordic dimension in ECEC is disappearing?
It is crucial that we differentiate between governance models and welfare goals. The Nordic dimension is primarily found in policy papers that define objectives for ECEC, which in important dimensions are similar to the core welfare services. Here, we see that similar trajectories seem to persist. Traditional goals of child-centred services, female participation in the workforce and flexibility for families have, in all of the countries, been reinforced and supplemented by the goals of school preparation, social investment and integration of the minority population. Despite some differences in emphasis, these welfare goals are shared across the board and constitute a Nordic dimension in their own right. Indeed, it may be that there is some sort of convergence to this dimension as, for example, Finland has taken steps to increase ECEC coverage as a reflection of increased support for ECEC and the fading prominence of familyism as a challenge to ECEC.
The quasi-markets developed quickly in Norway, but were delayed by political power struggles in Sweden; however, in both countries, there was a concentration of ownership of the private ECEC institutions. In contrast, in Denmark, the private actors mostly chose a similar organisational structure to the self-owned institutions (Henriksen et. al. 2016). In Norway, unintended consequences of weak regulations of the for-profit providers gradually became apparent, but they were difficult to manage due to concentration of ownership and lack of transparency in economic transactions.
Several reports and public investigations have pointed out that undue profits could be made, in particular during the process of selling institutions (BDO, 2018;Lunder, 2019;NOU 2020: 13;Storberget et al., 2021). A set of professional-bureaucratic measures was thus introduced to reduce the adverse effects of quasi-markets: regulations for staffchild ratios, educational level of staff and economic supervision. This means that the unintended consequences of quasi-markets with regulations that were too weak were counteracted by reverting to the governance tools of a professional-bureaucratic model. In Sweden, the political situation is more polarised regarding private welfare and freedom of choice. The minority red-green coalition that governed from 2014 to 2022 was not able pass propositions that would support the non-profit welfare providers, such as profit restrictions in publicly funded welfare services (SOU 2016: 78), as the conservative-liberal parties blocked attempts. In Finland, an attempt to restrict profits in ECEC services has met judicial resistance, and a development towards more extensive quasi-markets, as in Sweden and Norway, is possible. In Iceland, there are no restrictions on extracting profits by owners in the ECEC legislation. However, there are indirect limitations concerning, for example, regulation of fees, children-staff ratios, economic supervision and evaluations.
While the Nordic dimension in ECEC 50 years ago consisted of similar means to meet similar ends, we see today that the countries use different means to meet similar ends. This is partly the result of differences in timing. When the countries opened up for for-profit providers, a gap occurred between countries with a saturated market and those with a still expanding market. This helps explain for-profit growth in Norway and Finland. In contrast, Sweden, with the most market-emulating regulation of welfare in Scandinavia (Sivesind, 2017), has seen more modest development, as it reached full coverage before vouchers were fully implemented in 2008 (Westberg & Larsson, 2020). The pattern of seeking the same goals with different means thus adds complexity to governance schemes. The movement towards quasi-markets has been met with calls for different types of counteracting regulation. In line with institutional theories of gradual change (Thelen & Mahoney, 2009), we see today a layered governance architecture, where aspects from different policy regimes are present on both the national and local levels. The degree and form of layering are unique for each country, providing a differentiated set of governance models.
To explain the country differences, discursive institutionalism directs us towards the ideas that dominate welfare governance (Schmidt, 2008). The discursive context of ECEC developments seems to frame the course of action for policymakers, resulting in specific policy solutions. Denmark has the strongest tradition of self-owned welfare providers, making changes in the regulation of for-profits easier to countenance as long as the non-profits are not unduly hurt. In Norway, the rapid expansion of forprofit provision in ECEC was an anomaly in welfare governance. Lacking the ability to generate sufficient supply through traditional means may have constituted a policy failure that has enabled untraditional governance tools to be implemented (Pierson, 2000). As the consequences of the policy experiment have become clear, we have seen governments from both left and right seeking professional-bureaucratic measures to counteract some of the adverse effects of this development. Conversely, in Sweden, the discourse has been dominated by a struggle between user choice and political control of welfare provision. This has made it difficult to come to an agreement to rein in the frame conditions for for-profits, prompting calls for a 'Norwegian model' of regulating staff ratios and education in ECEC (Enfeldt, 2022). In Iceland, the privatisation of ECEC has not been high on the agenda in political discourse. The main emphasis has been on user choice and bridging the gap between parental leave and ECEC. Forprofit growth has been an overlooked side effect. Finland has seen some ideational change, as the historically dominant position of familyism has lost some of its importance (Hiilamo & Kangas, 2009). This has resulted in increasing ECEC coverage and a potential for growing for-profit market shares and ownership concentration, as has happened in Norway and Sweden.
An evaluator-state model is more dominant in compulsory education in Nordic countries, but there are few examples from the ECEC sector. In Denmark, there is a national test system to assess language skills. On the local level, many municipalities use management by objectives in the governance of ECEC institutions, similar to those in other service areas. Given the increased emphasis on ECEC as part of the educational ladder and the strengthening of educational ambitions for ECEC in the Nordic countries, we may see more evaluative governance measures in the future.
An open question for further studies is to what extent the Nordic dimension in ECEC welfare goals is indeed Nordic. We find that the five countries discussed here share many of the same goals, but it is unclear whether these goals are particularly Nordic or if the focus on for example social investments and integration of immigrants makes the goal structure of Nordic ECEC more on par with many other parts of the world.