The governors of school markets? Local education authorities,school choice and equity in Finland and Sweden

Abstract

As one of the key elements of the Nordic welfare model, education systems are based on the idea of providing equal educational opportunities, regardless of gender, social class and geographic origin. Since the 1990s, Nordic welfare states have undergone a gradual but wide-ranging transformation towards a more market-based mode of public service delivery. Along this trajectory, the advent of school choice policy and the growing variation in the between-school achievement results have diversified the previously homogenous Nordic education systems.

The aim of our paper is to analyse how Finnish and Swedish local education authorities comprehend and respond to the intertwinement of the market logic of school choice and the ideology of equality. The data consist of two sets of in-depth thematic interviews with staff from the local providers of education, municipal education authorities. The analysis discloses the ways in which national legislation has authorized municipal authorities to govern the provision of education.

Introduction: The Nordic model of education?

The revival of ‘traditional Nordic localism’

According to Green and his colleagues (1999), after a period of centralization from the 1930s, ‘traditional Nordic localism’ re-emerged during the 1980s. This considerable shift in the relationship between central and local government has concerned all Nordic countries alike − as a common trend, Nordic nation states have delegated control to their subordinate bodies, municipalities.

Traditionally, vesting a high degree of decision making in local authorities has been the central feature of school education in the Nordic countries (Arnesen and Lundahl, 2006; Esping-Andersen, 1990). Nevertheless, before the 1990s, state regulation was considered essential for the sake of equality (Page and Goldsmith, 1987). During recent decades, decentralization and new public management practices have increased local autonomy, which has also led to inter-municipality fragmentation. This trajectory has weakened the unifying structural principles upon which the comprehensive systems were built (Arnesen and Lundahl, 2006; Varjo et al., 2016).

As one of the key elements of the Nordic welfare model (Esping-Andersen, 1990), the comprehensive school system is based on the idea of providing equal educational opportunities, regardless of gender, social class and geographic origin. In broad terms, the Nordic welfare regime has distinctive features: centre-left coalition governments, a high level of redistribution, strong support for investment in primary and secondary education, active labour market programmes and high-quality public day care and preschool services (Iversen and Stephens, 2008). Until the late 1980s, Nordic education systems consisted of universal, non-selective basic education, generally provided through the public authorities (Arnesen and Lundahl, 2006; Erikson et al., 1987). Pupils have generally been assigned to publicly funded schools in their area of residence (Musset, 2012).

Transformation of the Nordic welfare state model

In tandem with the ‘changing central–local relations of governance’ (see Ozga et al., 2011), Nordic welfare states are undergoing a gradual but wide-ranging transformation towards a more market-based mode of public service delivery. According to Helby-Petersen and Hjelmar (2014), the gradual change from public to private production of welfare services in home care for the elderly, provision of child care, and the operation of nursing homes, among others, constitute significant characteristics of the transformation of the Nordic welfare model.

Along this trajectory, numerous education reforms across the globe have dismantled centralized bureaucracies and replaced them with devolved systems of schooling with special emphasis on parental choice and competition between different types of schools with both public and private providers (Gewirtz et al., 1995; Whitty et al., 1998). Drawing on Ball and Youdell (2008), these reforms broadly involve practices of endogenous and exogenous privatization of public education. Endogenous privatization refers to new public management techniques and the ways in which schools are evolving into more business-like entities, whereas exogenous privatization concerns commercialization, sponsorships, school–industry partnerships, and competition between private and public actors in more concrete terms.

According to Ozga et al. (2011; see also Helgøy et al., 2007), the marketization of state schooling accompanied by the devolution of responsibilities to local level is an example of deregulation, whereas re-regulation involves the reassertion of central control through such means as target setting, performance measurement and quality indicators. These re-regulatory practices are not just technical forms of control, but create interdependencies and relationships and render spaces calculable and governable.

The emergence of endogenously or exogenously formed local school markets has become intertwined with issues of social and residential segregation in Finland and Sweden. Current analyses from Sweden have demonstrated how reforms to admission policies that abolish residence-based admission rules for upper secondary schools have increased segregation. The allocation of students to secondary schools according to their grades increased, as expected. However, the reform also increased segregation in other respects. The reform, intended to ‘reverse the effects of residential segregation on school segregation, actually increased segregation along all other observable dimensions, particularly along ethnic and socio-economic lines’ (Söderström and Uusitalo, 2010: 75).

Also, in Finland, the advent of school choice policy and the growing variation in between-school achievement results have diversified the previously homogenous education system. The Helsinki metropolitan area, for instance, struggles with growing segregation in terms of socio-economic differences, neighbourhoods, and schools. In major cities, a considerable variety in learning outcomes (measured by PISA test scores) has recently emerged, and there is a trend of growing differences between schools (Kupari et al., 2013). According to Bernelius and Kauppinen (2011), the association between urban segregation and educational outcomes has appeared even in the egalitarian Finnish context.

The municipalities in general – and local education authorities in particular – operate in the intersection between the imperative of school markets and the aversion to inequality. Due to the situation in which the performance of both Finland and Sweden in the PISA survey has declined in recent years (OECD, 2016), the municipal interpretations and actions to improve quality and equality in education have gained in significance. The aim of our paper is to analyse how Finnish and Swedish local education authorities comprehend and respond to the intertwinement of market logic of school choice and ideology of equality. It is intended to contribute to the understanding of the local education authorities’ rationales and actions in two similar-yet-distinctive Nordic policy contexts.

Approaching the local school markets

Freidson’s (2001) ideal-types of logics (market, bureaucracy and professionalism) portray the ways in which work can be organized and controlled. In the Nordic countries, there has been a long tradition of organizing and controlling the education work by a combination of bureaucratic logic (such as a national curriculum and inspectorate, grades, goal achievement, legislation) and professional logic (such as an autonomous teaching profession with shared education, work culture, values and ethics) (Lundström and Parding, 2011). The introduction of market logics, characterized by business ideas and structures, such as buying, selling, competition, marketing and the idea that fully informed consumers (i.e. parents and students) choose rationally in their own best interests, has altered the balance of logics.

Following Burch (2009), we see that market logic constitutes broad cultural norms that can influence organizational behaviour. Like Powell and Colyvas, we presume that ‘organizational practices and structures are often either reflections of or responses to rules, beliefs, and conventions built into the wider environment (Powell and Colyvas, 2008: 2). At the same time, organizations exercise various degrees of agency (Scott, 2008). Municipalities are regarded as democratic political institutions (March and Olsen, 2005) and key actors in enacting education policy via an interpretation and translation process (Ball et al., 2012). Our perspective implies a critical view of policy realization as a linear instrumental process (Apple, 2004).

School choice – that is, ‘ways to increase some parents’ access to current choices or new choices that may arise as a result of the policy’ (Merrifield, 2008: 5) – now features on the political agendas of countries around the globe. Accordingly, in Finland and Sweden, school choice is understood as one of the key elements of the re-examination of the education system and as a manifestation of market logic. On the one hand, it is an example of the changing relationship between the public authorities and markets. On the other hand, the ways in which it has been implemented in various socio-historical contexts demonstrate the presence of different versions and revisions of the Nordic education model (Arnesen and Lundahl, 2006; Erikson et al., 1987).

According to Varjo et al. (2016), demands for increased opportunities for school choice and competition have commonly been underpinned by a range of arguments emphasizing social benefits, including the claim that such reforms will reduce educational bureaucracy; strengthen democracy (the right to choose instead of being assigned); improve efficiency (higher achievements and lower costs); increase accountability; and promote equality of opportunity for the poorest students in low-achieving schools in deprived areas by abandoning the strict catchment area policies (Bunar, 2010b; Chubb and Moe, 1990).

There are also shared understandings of conceivable social costs of school choice (Varjo et al., 2016). Claims for competition and choice have been challenged with arguments that view school choice as a middle-class enterprise causing social segregation and the problem of ‘failing schools’, and returning the school-specific improvements in achievement rates to their more advantageous social composition (Bunar, 2010a). Moreover, a host of research evidence indicates that school choice, the marketization of education and differentiation of schools intertwines with socio-spatial segregation (Böhlmark et al., 2015; Bunar and Ambrose 2016; Östh et al., 2013; Söderström and Uusitalo, 2010; Waslander et al., 2010; see also Ojalehto et al., 2016).

Despite the common objectives concerning social benefits and costs of school choice, different policy options have led to dissimilar institutional arrangements. Parental school choice in Finland takes place within the publicly funded education system. Conversely, in Sweden the official policy has been to promote parental choice between public and private (independent) schools.

Method, data and focus

The aim of the paper is to analyse the ways in which Finnish and Swedish local education authorities comprehend and respond to the intertwinement of market logic of school choice and equity. The paper is intended to contribute to an understanding of the local education authorities’ rationales and actions in such a policy context and, hence, lead to a better-informed political discussion of how striving for equality can be promoted by providers of education in the context of local school markets. We contrast earlier national-specific outcomes and interpretations (see Alexiadou et al., 2016; Dovemark and Holm, 2015; Kalalahti et al., 2015; Lundahl et al., 2013, 2014; Varjo and Kalalahti, 2015 for instance) and re-organize and meta-analyse our data in a comparative setting.

Our comparative approach is based on ‘the unique nature of a variety of situations in time and space, and the cultural resources available in these situations’ (Hedström and Wittrock, 2009: 8). It draws from a three-dimensional framework for analysing contingency by Simola et al. (2017; see also Kauko et al., 2012), and consists of the political situation, political possibilities and the political Spielraum (‘scope for action’).

According to Simola and his colleagues (2017), politics as a situation connotes the idea of the opportune moment at which politics can be changed and historical rupture is visible. The emphasis on changes in the socio-historical situation aims to go beyond the ‘unbearable narrowness of the national view’, and to comprehend how the national is constituted of its interconnections, meetings and crossings with the transnational.

Political possibilities are concerned with how actors find and create the ways for acting ‘otherwise’. To make change happen, policymakers must be aware of this political situation, or offer a radical re-interpretation of the status quo to seize such a moment (Palonen, 2006). Kauko (2013, 2014) claims that the occurrence of an opportune moment is dependent on how the institutional structure of the education system fits together with external developments in the political system and in society.

Framed by the political situation and political possibilities, a major element of the dynamics in politics is the Spielraum for ‘politicking’. This refers to the potential of actors to ‘play with contingency’ and to capitalize on existing situations and options in the complexities. The interplay between these three dimensions, which may vary considerably across countries and contexts, is the basis on which dynamics are analysed (Kauko et al., 2012; Simola et al., 2017).

The empirical data ¹ consist of two sets of in-depth thematic interviews with municipal experts in provision, management or evaluation of local-level lower secondary education (Finland) and upper secondary education (Sweden).

In the Finnish case, the interviewees were selected based on their institutional position. The first two groups consisted of six municipal managers/officials who were responsible for local education administration. They were either heads of department (‘Head’) or middle-rank officials (‘Official’) who specialized in issues concerning admission and selection. The third group consisted of three persons who were nominated to the local Board of Education, based on the results of municipal elections (‘Politician’).

In the Swedish case, 22 interviewees participated in the study. Nine were managers/officials at the municipal administration level: five heads of education department (‘Head’) and four administrators in other positions, for example, a quality manager or project leader (‘Official’). Four politicians were also interviewed: three local authority education committee chairs (‘Chair’) and one education committee member. In addition, principals at eight public upper secondary schools were interviewed (‘Principal’). The principals’ primary tasks are at the school level, but they are also key people in enacting both municipal and state education policies, which are the dimensions of their work being considered here.

Although our original focus was on different levels of education systems (lower and upper secondary), the comparative approach enables us to analyse the dynamics of governing local school markets. Overall, our interviewees represent the overall authority to provide basic education, including financing, implementation and evaluation. To comprehend the dynamics in full, we also explored the main steering documents from each municipality in both countries. These are crucial documents as the responsibility for organizing and operating schooling is devolved to municipalities.

The interview data were organized thematically in accordance with crucial issues of a three-dimensional framework for analysing contingency, such as how the municipalities comprehend and respond to policies (interpretation and translation), the impact of (de)regulations and norms (market logics of school choice, inequality), the importance and dynamics of various contexts (local school markets), and organizational practices and structures.

Two devolved contexts for school choice

In terms of the political situation, municipalities ² are vital actors in implementation – but also, due to their considerable autonomy, in production – of education policy in the Nordic context. The concept of municipality implies various aspects. For example, it constitutes a geographic district, an arena for democratic decision making at the local level, and local community public offices.

The contemporary model of governance in Finland and Sweden is a result of altering political situations. Around 1990, a severe financial crisis, combined with neoliberal influences, constituted a political situation that opened up political possibilities for school choice. Several reforms occurred throughout the 1980s and 1990s: in broad terms, they implied decentralization, management by objectives and results, and school choice reforms. Currently, the state sets the objectives and the providers of education implement the national guidelines. Typically, resource allocation, the role of the employer of school staff, school organization, and school evaluation are issues under their jurisdiction as providers of education.

Finland

Finnish local government has been described as operating ‘outside’ the sphere of central government (Green et al., 1999; Temmes et al., 2002). The lack of traditional control mechanisms is manifest in the actual scope of local autonomy. Unlike Sweden, for instance, Finland has never had a tradition of mandatory national testing, nor are school inspections and school league tables in use (Eurydice, 2004; Varjo et al., 2013). The trajectories of deregulation and decentralization have altered local policies and practices concerning admission to and selection for basic education. The 1999 Basic Education Act (Law 628/1998) only obliges municipalities to assign each child of elementary school age to ‘a neighbourhood school’; the legislation does not refer to ‘school districts’ any more. The notion of a neighbourhood school means that children are obliged to attend a designated school defined in terms of proximity and local conditions. Thus, municipalities are authorized to develop distinctive policies and practices in order to allocate children to their neighbourhood schools in an equitable manner (Seppänen et al., 2012; Varjo et al., 2014).

The Basic Education Act (Law 628/1998) enabled parents to choose between schools. Municipalities, through their elected education boards, have been given powers to decide on the allocation of lesson hours in all the schools under their jurisdiction. Schools have started ‘taking profiles’ (see Ylonen 2009: 42–43), that is, offering specialization in particular subjects in the curriculum or placing an emphasis on a few more general themes (the environment or communication, for instance). These ‘classes with a special emphasis’ (painotetun opetuksen ryhmät) function as separate streams within regular municipal schools. They have more lessons (for instance, in music, sport, science, languages, or creative arts) than the National Core Curriculum requires. In large urban municipalities, 30–40% of pupils are selected for a class with special emphases (Varjo et al., 2016).

As a result, educational diversity inside the traditionally homogeneous national curriculum has increased since the 1990s. However, because of the strictly limited number of private schools, parental choice takes place within the publicly funded and governed comprehensive system. Finnish private schools are mostly schools with a specific religious or pedagogical emphasis. According to the OECD definition, they are government-dependent private schools, that is, institutions that receive more than 50% of their funding from government agencies (Musset, 2012). In 2009, 96% of Finnish comprehensive schools were owned by municipalities (Kumpulainen, 2011).

Sweden

School choice reforms in Sweden were part of several substantial reforms introduced during an intensive school reform period at the beginning of the 1990s (Government Bill 1991/92: 95; 1992/93: 230) that greatly influenced power relationships. These reforms resulted in a shift in the Swedish school system towards more decentralization (Lubienski, 2009; SOU, 2014). The transition from a long tradition of social democratic education policy to a school system strongly influenced by market-liberal and new public management ideas, and the consequences of this shift, have been described in several studies (Arnesen and Lundahl, 2006; Bunar, 2010b; Erixon Arreman and Holm, 2011; Gustafsson et al., 2016; Holm and Lundström, 2011; Lund, 2008; Lundahl et al., 2013; Lundström and Holm, 2011; Vlachos, 2011). However, the declining outcomes and discontent with the consequences of school reforms have led to a change from ‘decentralized management by objectives’ to ‘centralized performance-based management’ (SOU, 2014: 5, 30).

In Sweden, school choice is a nationally promoted policy that involves various market actors. The majority (85%) of the independent schools are owned by national or international limited companies (Vlachos, 2011). The proportion of students in independent upper secondary schools expanded from 2% in the 1992/93 school year to 26% in 2014/15 (SNAE, 2016). However, opportunities for school choice vary in accordance with geographic and demographic conditions. For example, 50% of the students in large cities, but only 3% of those in sparsely populated areas, went to independent schools during the 2015/16 school year. There is also substantial variation between socio-economic groups regarding distribution of independent schools: 55% of the pupils in the highest socio-economic decile go to independent compulsory school, compared with the 5% in the lowest socio-economic decile (Gustafsson et al., 2016).

The Education Act (SFS 2010:800) applies to both public and independent upper secondary schools and states that the requirements and economic conditions should be the same for all providers. But municipalities have wider responsibilities, such as guaranteeing that all youths in the municipality have the opportunity to enrol in upper secondary education, irrespective of whether they choose independent schools or schools in other municipalities. The municipalities fund education via taxation, but cannot influence how the independent schools allocate the resources. Even if the Act states that municipalities have the right to obtain information and engage in dialogue with independent schools, municipalities’ control over independent schools can be described as limited or incomplete. These are at least partly outside the realm of municipal jurisdiction.

Parents can choose any lower secondary school they like in the municipality and the municipalities should satisfy these requests, unless they crowd out children who live near the school. In principle independent schools are not allowed to choose students but there are queues for popular schools. In such cases, the selection should follow at least one out of three principles: date of application, priority of siblings, and proximity. All students that are eligible for upper secondary school can choose any school they choose in the country and if there is a queue, grades are decisive for the ranking. The providers of education are responsible for the admission. Municipalities often cooperate with each other and also include the independent schools in the local admission systems. Students within these districts are given a higher priority for admission than those from other municipalities if there are insufficient places. In all, the voucher is crucial in the system as it follows the student and constitutes the main funding of a school.

Recent reports on declining school achievement, equity, and teachers’ working conditions, as well as unsatisfactory local governance, pose questions regarding the role of municipalities as school providers (OECD, 2016; SNAE, 2011). A state evaluation (SOU, 2014: 5, 29) claims that the decentralization of school responsibility to the municipalities in 1991, the so-called municipalization, was a ‘failure’, emphasizing dilemmas concerning the municipal school governance role and articulating the tension between equity and marketization:

The independent school reform and the new school choice opportunities created competition between schools that partly changed the conditions for the municipalities and also for state governance. School choice reforms have resulted in increased segregation between schools and have consequently complicated the municipal task of achieving equivalence. (SOU, 2014: 5, 377; authors’ translation)

The delicate art of governing local education markets

The political situations in Finland and Sweden changed somewhat during the 1980s and 1990s. In Finland, emerging school choice policies were merged into a new public management-influenced ‘classic public administration paradigm’ (Homburg et al., 2007), whereas in Sweden they took more explicitly market-oriented forms, including the introduction of a considerable private school sector, for instance. Due to disparities in the political situations in the two countries, the emerging political possibilities concerning school choice evolved in a strikingly different manner in Sweden and Finland. First, issues concerning the elements of local education markets – privatization, competition between schools, and the evaluation of schools – were politicized in dissimilar ways. Secondly, the diversification of schools was also politicized differently. Therefore, the political Spielraum for municipal education authorities – the ways in which they were able exploit existing situations and opportunities – has become narrower in Sweden, especially in terms of controlling the social costs of school choice.

The elements of local education markets

According to Varjo et al. (2016), while promoting the social benefits of school choice, Sweden has vigorously endorsed individuality and freedom of enterprise. The notion of parental choice as a choice between public and independent schools has certainly enabled economic activity within the free school sector. The Swedish attempt to break the state monopoly in the provision of education can be interpreted as a focused objective to restructure organizational forms of the educational system.

The Swedish municipalities are influenced by the strength and outcomes of competition and the kind and numbers of schools that exist in local markets, but also the relationships between the municipal and the independent schools. Furthermore, the municipalities are affected by how students act in local school markets. Most interviewees mentioned the difficulties of foreseeing rapid shifts in local markets, and they must be very flexible. According to one principal: ‘we need to be prepared all the time’.

The interviews indicate that municipal and independent schools coexist and interact in local school markets, and some describe these dynamics in positive terms. The most obvious positive example of collaborative outcome is the organization of joint upper secondary school admissions between regions, including the independent schools in each area. Another example is the municipalities’ agreements to offer certain programmes for each other’s youths, such as the case of a large vocational programme in one municipality which the neighbouring municipality has agreed not to offer.

According to the Education Act, the municipalities have a right to scrutinize the independent schools’ work to some extent. This is carried out, apparently quite smoothly, by the municipalities via one or several meetings and visits per year. Some of the interviewees were dissatisfied with the independent schools’ lack of transparency: ‘It is only now that the municipality has realized that we actually have no control of what the independent schools do’ (Official). Another interviewee thinks it is a democratic problem that the independent schools are publicly funded but largely act independent of the local democratic processes as corporate legislation applies to the independent schools that are private companies:

It is a political dilemma because as a politician you want to show your voters that ‘I want to change, I want to develop’. But you have no influence on half of the market. (Chair)

However, at the school level there is hardly any cooperation between the public and independent schools and they largely constitute quite separate worlds. For example, a principal exclaimed: ‘Why should we collaborate if we are competing?’. Apparently, the market logic contributes to promoting competition rather than cooperation. However, some of the interviewees nuanced the description by recalling that the independent schools are in fact not a uniform group, but rather are a diverse group of schools with different owners and differing in size, age, characteristics, traditions, and so on. This means that the municipality must address the various stakeholders, some of whom are perceived as being competitors while others are viewed more as partners.

Choice and competition can also be understood as mechanisms to increase the overall quality of the education system (see Chubb and Moe, 1990, for instance). Higher quality was a crucial argument for the introduction of school choice in Sweden (Government Bill, 1992/93), and the parental right to choose between schools has become an established policy. Consequently, devolution and privatization have stressed evaluation and inspection. Market logic and bureaucratic logic share an interest in having accessible information on school/student performance (Apple, 2004), which in Sweden has resulted in a substantial growth of evaluation systems that lay emphasis on the data of school-based learning outcomes (Lindgren et al., 2016). It is a trend that largely reduces the definition of quality and student achievement to publicly available, easily measurable knowledge (Lundström, 2017).

One way to control the preconditions of school choice is to control information and its public availability. This is crucial in Finland, because there are no national exams for the whole age cohort during the compulsory education years, and neither governmental organizations nor the mass media publish school league tables (Varjo et al., 2013). Municipalities produce, at least occasionally, a variety of school-based evaluations, but with strictly limited transparency. All the interviewees shared the sentiment that in order to prevent local league tables, test results and other school-specific performance indicators must remain both confidential and for administrative purposes only (see Kauko and Varjo, 2008). Another issue is the use and availability of data within the municipal politico-administrative system. To avoid any ‘information leaks’, politicians have been excluded. Curiously, even the ones elected to a position of trust (like the municipal Board of Education) do not personally feel the need for this type of data.

As a member of the [municipal] Board of Education, I don’t expect the local education authorities to deliver a map of the weakest schools in our city to the Board meeting. That would just not be clever. (Politician)

In this respect, the Spielraum for local education authorities is different in Sweden, where publication of performance data is part of new public management governance and ‘explicitness’ is important, which according to Green (2011) implies distrust of the teaching profession. The explicitness has the double function of holding staff accountable for results and serving as the basis for school choice, which assumes well-informed customers.

When contrasted with the political options in Sweden, by and large those for Finnish local education authorities appear to be quite dissimilar. The relations between the actors within local education markets are strictly mediated though local education authorities and the governance they exercise. As was the case with the Swedish interviewees, Finnish interviewees considered the quality and attractiveness of local schools to be the key factor that could form a virtuous or vicious circle within the neighbourhood. In this perception, choice was taken as a crucial element of the dynamics of socio-spatial segregation, able to accelerate unwanted processes, like residential segregation and unbalanced housing markets.

The fundamental ethos, regardless of your political allegiance, is equal educational opportunities. My view is that in Finland we have a strong consensus on the comprehensive school as a vehicle for promoting equality. (Head)

Nevertheless, at least to a certain extent, the Finnish interviewees recognized school choice as a means of screening and sorting the potential and abilities of individuals, loosely along the theory of human capital (Becker, 1964). However, these conceptions were neither precisely nor thoroughly articulated.

In this municipality, we want classes with a special emphasis; that has been the political will of the majority. There are good things behind this: the urge to support pupils’ different capabilities, and so on. (Official)

The provision of basic education in Finland is to the greatest extent a local matter. Furthermore, unlike in Sweden, the social benefits of school choice are unquestionably expected to occur within the comprehensive system, without reference to school markets as such. Arguably, decentralization and a strong, shared ethos of the universal comprehensive school have jointly constructed local contexts that are extremely concerned to recognize and control the social costs of school choice, while simultaneously leaving the conceivable social benefits aside (Varjo and Kalalahti, 2015).

Within-school homogeneity and diversification of education system

To share equal levels of quality, the social composition of pupils at each school cannot be too different from each other; within-school heterogeneity is considered to be one of the objectives of universalist education politics (Kalalahti et al., 2015). If within-school homogeneity occurs, schools start to differentiate themselves from each other, and eventually the school system becomes less universal. Our analysis indicates the ways in which local education markets feed within-school homogeneity, how diversification is politicized, and how local education authorities are equipped for politicking with the issue in Sweden and Finland.

The geographic location of schools in the local school market is a crucial aspect of school choice. In all the Swedish municipalities in our study, there is a pattern for upper secondary students to move from the periphery to the centre. This implies that students commute from suburbs to city centres and from sparsely populated municipalities to larger towns, in line with previous studies of national trends (SNAE, 2012).

Young people are attracted to the cities. For them it is OK to commute to town, but very few from the town commute to the provincial areas. … The surrounding municipalities have been hit hard. (Principal)

To reverse these student movements, the local authorities in one municipality have decided to try a new strategy for solving recruitment problems at one of their suburban schools with the most students of non-Swedish ethnicity. The school is struggling with low results in achievement tests and a bad reputation. The trend is that many of the high-achieving students in the neighbourhood leave and apply to more popular schools in the city centre. However, many of these students tend to come back to the suburb school after a while, so-called ‘returners’, as they felt alienated and excluded as immigrants at the city schools. To avoid closure, the municipality has decided to change the profile and promote a more academic orientation and recruit students mainly from the local neighbourhood. The previous vision of attracting applicants from other districts has now been abandoned. The strategy is to offer suburban youth educational opportunities in their neighbourhood. According to the Head, the assumed alternative is that the school would be closed and that many local youths would end up not completing upper secondary education.

So, ultimately there is no segregation, because the alternative – no local school at all – would mean that many young people wouldn’t get any [upper secondary] education, and then they would really be segregated. So, it might look like segregation, but it is exactly the opposite. (Head)

The Head’s reasoning illustrates the clash between marketization and equity and that it is a highly problematic task to participate in a competitive school market in a socially disadvantaged context. Several interviewees stressed that school status, reputation and ‘brand’ have become more important than ever. These findings confirm other research, for example Bunar and Ambrose (2016: 11) have illustrated how school choice has become ‘a competition between locations and between social groups over the symbolic power of legitimate naming and labelling’.

In broadest terms, the emerging significance of school status also concerns Finland (see Kosunen, 2014). Nevertheless, the notion of choice and the conceivable consequences – diversification of local education systems in specific and socio-spatial segregation in general – are different. Regardless of their position and political allegiances, all the Finnish interviewees, at least to some extent, shared the premise that one’s right to choose automatically means that someone else is inevitably left behind. Hence, according the Finnish sentiment, choice is something to be treated with caution and to be managed by public authorities, not markets. It can be argued that the prejudice concerning the correlation between choice and segregation is an unquestioned article of faith among Finnish local education authorities.

It is as simple as that. If all well-educated and active parents get together, everybody else will be shut out. That’s exactly what happens in music classes, you know. (Head)

In the Finnish interviews, the controlled diversification of school and limited options for choice were considered to be an artifice for keeping parents satisfied with their municipality and preventing them from appealing against the placement of their child. Moreover, some explicit reasons that could be considered as ‘social’ (that is, siblings or friends at preferred school) were also seen as legitimate reasons for school choice.

We can take health-related reasons into consideration, and issues concerning siblings at the same school, as well. This is very humane; I think they must be taken into consideration. (Head)

School differentiation was also discussed in terms of pedagogical practices, as an opportunity to enrich teaching in a certain subject and take individual aptitude into account more thoroughly. Pedagogical practices are understood to be a vehicle for conscious and continuous school-based developmental work. The emphasis on the development of quality, commonly articulated in the technical terms of learning outcomes and attracting of pupils, was obviously an issue that particularly concerns schools located in deprived neighbourhoods.

If we put up a class with a special emphasis in a school located in a demanding neighbourhood, we try to ensure that parents who are generally interested in their child’s hobbies and upbringing stay in the neighbourhood and don’t send their child to other schools, at least during grades 1-6, perhaps in grades 7-9, as well. (Official)

There was also consensus among the Swedish interviewees that free school choice and increased competition between schools imply stronger targeting of specific categories of students, resulting in more endogenously homogenous schools. For example, they describe the distribution of academic and vocational programmes between and within schools (i.e. in different buildings) as a delicate issue, as the programmes vary in status and distribution strategies contribute to student sorting.

I’m of the opinion that schooling is society building, so students of various nationalities and classes must meet to create understanding and dynamics. Research demonstrates that if high- and low-performing students meet, they lift each other. (Chair)

Municipalities have tried various reorganizations to make all municipal schools attractive, but this has been a problematic process. For example, in one case, moving a high-status programme to a school with a poor reputation resulted in students choosing that programme at an independent school instead.

In addition, the findings from Sweden indicate that the location of introductory programmes might affect inclusion, as the students in those programmes constitute a kind of outsider group as they are ineligible to choose the ordinary, national programmes. The inclusion strategies vary between the four municipalities investigated in the extent to which these introductory programmes are in separate or the same schools as the national programmes, and in whether the introductory classes are homogenous or include other students as well. Two of the municipalities in our study use a model in which introductory students are mixed with other students, i.e. they are included in ordinary classes.

We try to organize their studies close to the ordinary classes because we know that is what the students want … and then we try to phase them in so that they are seamlessly included in the regular programme … It is my personal intention, always, to make them feel as equal as possible to the others. (Principal)

You shouldn’t gather all these students in one place … They get labelled, they will be singled out and stigmatized. They are not in a normal context – that’s negative from an attainment perspective. No, I think it’s important to spread them out. (Chair)

In Sweden, municipalities are expected to use compensatory resource distribution among their schools to achieve equivalence. However, the Schools Inspectorate (2014) reported that it works unsatisfactorily: the various models are often not evaluated, the compensatory distribution is sometimes too small to have a significant impact, and resource allocation needs to be combined with broad social approaches to counteract segregation.

In the more strictly governed Finnish system, to maintain similar quality levels and evenly distributed options for a limited amount of choice in all local schools throughout the municipal area, conscious measures are taken. The ‘policy of equalizing’ (Varjo and Kalalahti, 2015) means the equal and principled allocation of resources within the municipal provision of comprehensive education in the formation of teaching groups. For example, if a school applies for a guaranteed attraction such as a music class, more ‘demanding’ or ‘resource consuming’ obligations might be imposed on the school, such as special education groups or preparatory instruction for immigrant pupils.

. . . when talking about special education or preparatory instruction for immigrant pupils, we have the idea or philosophy that each and every school should do their fair share regarding these matters. We don’t allow free riders, so to say. (Official)

The policy of equalizing is also present in efforts to govern admission policies; local education authorities use a wide variety of means to ensure the heterogeneity of schools and classes. They modify admission policies through geographical admission districts, set limits for selectivity and encourage schools to draw pupils from neighbourhood areas with the use of an incentive bonus, for instance. According to the Finnish interviewees, fair allocation of resources requires large amounts of information. Governing local education markets by data demands a wide variety of statistics, including knowledge of population socio-spatial segregation, learning results and the heterogeneity of schools and classes, measured by socio-economic indicators, as well as ratios of pupils in special education and in classes with a special emphasis.

Concluding remarks

Despite the supranational trends of marketization, choice and competition, our analysis has disclosed the different ways in which national legislation has authorized and obliged local authorities to govern the provision of education in neighbouring Nordic countries, Finland and Sweden.

The political situations in which local education markets occur are dissimilar in the two countries. In Finland, school choice takes place within the public school system, strictly provided and governed by public authorities, through classes with a special emphasis, which are the Finnish mechanism for exercising parental choice (Seppänen et al., 2015). In general, these kinds of premises justify actions that one might see as bureaucratic and non-competitive (Anttiroiko et al., 2005). The marketization of basic education (in its literal sense) has not fully occurred in Finland – despite the emerging practices of endogenous privatization, which is largely carried out within a ‘classic public administration paradigm’ (Homburg et al., 2007).

In contrast, the choice occurs both within and between the private and public sectors in Sweden and all schools, both public and private, compete for pupils and funding (Varjo et al., 2016). The number of independent schools − and proportion of students involved − has risen markedly, driven by the rather unique ability of independent school owners to extract profits. The favourable conditions for privatization have led to the establishment of an increasing number of large independent school public limited companies. The consequences of following market logic are a relatively uncontrolled strengthening of the internal homogenization of schools and external segregation between schools in terms of socio-economic background and ethnicity (Varjo et al., 2016). The establishment of a school market is followed by the growth of performance data, focusing on measurable knowledge.

In both countries, choice is a social practice that simply cannot be denied in modern times – the attraction of choice is compelling, indeed. It might be based on ‘temptations of a big city’ available for young people, or public authorities’ considerations of ‘social reasons’ or ‘basic rights’. The Swedish attempt to ‘break the state monopoly in the provision of basic education’ (Varjo et al., 2016: 81) can be interpreted as a more focused and principled objective to restructure the forms of the provision of education in favour of choice and competition.

There are well-reported, differentiation-related downsides of choice, such as increased inter-school variance in student achievement (Bernelius and Kauppinen, 2011; Östh et al., 2013). Holmlund and her colleagues (2014) claim that in the Swedish case, this problem is caused mainly by residential segregation but also by the increased number of independent schools. Trumberg (2011) concludes that school choice results in schools that are internally more homogenous and externally more segregated in terms of socio-economic background and ethnicity. Our findings confirm Bunar and Ambrose’s (2016: 37) conclusion that ‘what seems to matter most to parents is the ethnic and social structure of a school, something that largely reflects the ethnic and social structure of the geographical space in which the school is located’. Arguably, the double bond between the all-encompassing compulsion of choice and its downsides – articulated in terms of segregation, inequality and exclusion in our data– is one of the determinant factors concerning the provision of education at the local level.

Under these conditions, endogenous, intra-school heterogeneity becomes more important. There was consensus among the interviewees in both countries that school choice and increased competition between schools imply stronger targeting of specific categories of students (different sorts of learners, social classes and ethnic backgrounds, for instance), resulting in relatively homogenous schools. Our analysis reveals that from the perspective of local education authorities, schools that are too homogenous are problematic. If schools become endogenously too homogenic, the variation between schools will grow, and that could be considered to be a problem in terms of equality.

The municipal jurisdictions concerning the possible reconstructive and preventive actions are dissimilar in Sweden and Finland. In terms of securing the endogenous, within-school heterogeneity, an apparent municipal control deficit has occurred in Sweden. All in all, the political Spielraum for local education authorities – the ways in which they are able exploit existing situations and possibilities – has become narrower in Sweden. The owners and governing bodies of independent schools are not public authorities, and hence they are neither obliged nor authorized to act on social issues like segregation. Although the local education board governs the municipality’s educational issues, its decisions apply only to the students in municipal schools, which constitutes a democratic accountability problem.

In the Finnish context, in legal terms, municipalities are well authorized to govern the tasks under their jurisdiction, like the provision of basic education. Despite the jurisdictive potentials and tradition of governance at local level, a group of ‘failing schools’ has emerged in Helsinki, the national capital. A gradual differentiation in learning results measured in PISA studies and its intertwinement with socio-spatial segregation has been noticed in Finnish education policy discourse. Arguably, despite its wide range of applications in the past, the ‘classic public administration paradigm’ (Homburg et al., 2007) might not be adequate for governing local education markets in altering political situations.