Combating low completion rates in Nordic welfare states: policy design in Norway and Sweden

ABSTRACT

Low completion rate in upper secondary education is seen as a big problem in the Nordic countries. School failure has shown to dramatically increase the risks for unemployment and labour market exclusion with severe consequences for both society and the young person. This paper analyses national policy measures to combat low upper secondary education completion rates in Norway and Sweden, often regarded as representing a social democratic welfare model and a universalistic transition regime. The analysis demonstrates that although this issue has received extensive political attention, the two countries display somewhat different policy designs. The Norwegian approach is proactive and targeted while the Swedish policy is more general and directed towards reforming organisational structures in upper-secondary education. In sum, our analysis demonstrates that national governance structures shape and influence policy design in the context of an increasingly diversified Nordic social democratic welfare state regime.

KEYWORDS:

• Upper secondary school completion
• education policy
• social democratic welfare state
• universalistic transition regime
• Norway
• Sweden

Introduction and aim

Low completion rates in upper secondary education are acknowledged as a considerable challenge in the Nordic countries (Markussen 2010). In Norway, the completion rates¹ have varied between 67 and 74.5 per cent since 2000 (Statistics Norway 2018), in Sweden the corresponding proportion is 63–65 per cent since 2011.² In both countries, low completion rates constitute a challenge both for individuals and for society. School failure has been shown to dramatically increase the risks for unemployment and labour market exclusion (Bäckman et al. 2015). Furthermore, the labour market misses these young people’s potential knowledge and competence (Arntsen and Grøgaard 2012; Kauppinen et al. 2014). Falch et al. (2010) claim that the most important factor for completion is the average grade level achieved in elementary school, implying variations in qualifications for completing upper secondary school. Moreover, psychological distress among adolescents has proved to be a decisive factor for upper secondary school dropout (Esch et al. 2014; Brekke and Reisel 2017). Several studies have also pointed to the importance of school-related factors: neglect of early warning signs and lacking support (Elffers 2012; Lamote et al. 2013; Lundahl et al. 2017), as well as high levels of bullying and truancy (Balfanz and Byrnes 2013; Cornell et al. 2013). Thus, contexts, such as those provided by the local government, which is responsible for upper secondary schooling are important (Borge, Falch, and Strøm 2011). The problems of dropout and low completion rates are the focus of several OECD and EU initiatives and programmes (European Commission 2014: OECD 2014; 2015) and are placed high on national policy agendas, although with different national and local foci and policy designs. Combining strong universalistic features with extensive local discretion and New Public Management (NPM) reforms, the welfare and education policy context of the Nordic countries is interesting to study in this respect. The aim of this article is to analyse and compare Norwegian and Swedish policy designs for raising the level of students who complete upper secondary education by preventing early school leaving and providing opportunities for returning to school and graduation.

This article rests on several previous research projects led by the authors and includes a host of empirical data from interviews, national surveys, statistical data and intensive studies in selected schools. In addition, official documents, policies and statistics were analysed.³

In the next section, we give a contextual background to the two national contexts, before outlining our analytical framework and approach to study policy design in more detail. We then move on the policy responses to combat non-completion in the Norwegian and the Swedish settings, with a particular focus on developments during the last two decades. This is followed by a comparison of the national policy designs and a concluding discussion.

Context and analytical framework

Norway and Sweden are social democratic welfare states that historically have shared certain economic, cultural, social and political traits (Esping-Andersen 1990). They have been characterised by a comprehensive state, strong citizen rights and universal welfare arrangements. In line with this regime, the Nordic education systems have promoted social inclusion by securing equal access to education for all, comprehensive public schools and an emphasis on democratic values, community and equality (Aasen 2003; Arnesen and Lundahl 2006). Since the Second World War, the education systems have been important in building the universal social democratic welfare state that is considered an important explanatory factor of more or less universal or targeted state intervention in educational transitions. Approaches to youth unemployment, disadvantage and design of transition policies (Walther 2006) are essential to understanding how different states attempt to combat low completion rates.

Previous research indicates that policies are mediated through national institutional contexts that have both common and diverging features (Helgøy and Homme 2006; Bjurstrøm and Christensen 2017). Both Norway and Sweden are decentralised welfare states because of the prominent roles local and regional governments play in the political systems (Baldersheim et al. 2017). Under terms of a negative delimitation of power, local governments in Norway and Sweden have authority to engage in a wide range of discretionary activities, albeit subject to control by the national governments. Thus, local governments have not been passive recipients and implementers of national policies; they have also been active in developing welfare practices, including education. Compared to most European countries, the level of local autonomy is high in both countries and slightly higher in Sweden than Norway (Ladner, Keuffer, and Baldersheim 2016).

The responsibility of the local government levels differ when it comes to education (Lundahl 2012; Helgøy and Homme 2013). In Norway, school ownership is divided between municipalities (primary and lower secondary schools) and county municipalities (upper secondary schools). In Sweden, schools have either municipal or private owners, and the municipality is responsible for both primary and secondary public schools. Twenty-six per cent of the students in Sweden are enrolled in private upper secondary schools (Skolverket 2017a) and in Norway eight per cent (Statistics Norway 2017a).

Regarding government institutions and state reforms, Norway has followed an incremental sector-based strategy from NPM-inspired policies in the 1990s and 2000, whereas reforms adapted more recently have been more post-NPM oriented (Bjurstrøm and Christensen 2017), also in education (Helgøy and Homme 2016). Sweden is the Nordic country that has followed the most radical reform path since the early 1990s. In the 1990s, NPM inspired changes were introduced to emulate managerial practice common in private corporations, such as managing by objectives and results, customer orientation, and contracting out. Neoliberal ideas of choice and competition have had a significant impact in Swedish education where a veritable school market has been created (Lundahl et al. 2013). Norway has to a greater extent stuck to local government production of local services (Sivesind 2016).

Thus, although classified within the same welfare state regime and sharing values of social inclusion and equity in education, Norway and Sweden differ when it comes to governing structures and educational marketisation. At the same time, upper secondary school non-completion is seen as a threat to social inclusion and equality in both countries, and we thus set out to analyse how they deal with this challenge, and whether they differ or converge in their policy design. Being characterised as both social democratic welfare states and as universalistic transition regimes (Walther 2006), we would expect strong state intervention and a focus on education and activation in the policies to combat low school completion rates. However, we would also expect that the different governing structures would impact the design and type of policy measures adopted in each national setting.

Transition regimes and policy design

We assume that different structures of employment, welfare, education and training result in different policy designs that also reflect the main expectations from society towards young people. According to Waltheŕs model of transition regimes (2006), the concept of youth in a universalistic transition context, exemplified by the Nordic welfare states, has an orientation towards individuals’ personal development and citizenship. As young people are expected to be either in education or employment, youth unemployment is to be avoided and second chances to be promoted. The policy measures are directed towards allowing access and developing young people’s orientation towards regular and recognised options. The transition policies in the universalistic transition regime to target early school leavers mainly addresses education and different forms of activation ensuring that young people are involved in education and welfare options within the system’s framework (2006).

We turn to theories on policy design as a way to analytically distinguish national policy measures within the universalistic transition regime. Relying on Bèland (2005), we define policies as solutions to ameliorate perceived problems, i.e. solutions to improve the situation for young people who have left, or are at risk of leaving, school before completing upper secondary education. Bobrow and Dryzek (1987) point to three core elements in policy design: clarifying values for developing political alternatives, characterising the context, and defining the target group. Actors/stakeholders, implementation structures, the social construction of target groups, regulations and measures are included in the policy design (Schneider and Ingram 1997). Policy design comprises characteristics of the policy goals and means as well as the relation between them (cf. Howlett 2014). We see the relation between goals and means as dynamic and socially constructed. Consequently, it is important to consider the context of the policy formation process, its changing environments, clarity of mandate, disparate values and interpretations.

In analysing national policies to combat upper secondary non-completion in Norway and Sweden, we first investigate how the policies are regulated within the two countries: what characterises the education governing system and how is upper secondary education regulated within the legal frameworks? As discussed above, we expect to find differences in education governing structures due to marketisation and restructuring. But we also, in line with Walther (2006), anticipate that the comprehensive school systems in Norway and in Sweden include national frameworks, which set standards and at the same time are flexible to allow for individual plans. Second, we turn to the problem definition: how is early school leaving constructed as a problem, what are the specific goals that politicians want to achieve, and what stakeholders are involved in policy formation? For instance, what policy outcomes are considered successful, such as completion of secondary education or waged labour? In a universalistic transition regime we would expect that non-completion is constructed firstly, as an individual problem limiting the young person’s development and possibilities for choosing her/his biography. It can also be constructed as a societal problem; education is regarded as vital for realising the young people as resources for social development and economic growth, and thus for the maintenance of the welfare state. Moreover, we expect the political goals to aim at getting school leavers back to school to finish their education or to enter employment. We also expect the state would be a main stakeholder, together with other public education and welfare actors. Third, we analyse the policy solutions to goal attainment by identifying the definition of the target group, the content of the policy measures, and the implementation structures. Are the policies directed towards all pupils in primary and secondary education, or are they limited to upper secondary pupils, pupils in vocational programmes, or pupils who are performing below a certain level? Are the measures universal or targeted, national or local? What are the responsibilities of school owners and different government levels in the policy implementation process? Although the focus of transition policies in the universal transition regime is assumed to be mainly on education and activation, we anticipate that many different policy solutions are available to the politicians, dependent on the specification of target groups, and responsibilities of the different stakeholders in the implementation process.

Policy design in the Norwegian setting

Policy regulation

Despite the unitary education system and the role of education in building the universal social democratic welfare state, local autonomy is a core value in Norway. Over time, municipalities have received various degrees of responsibility for basic education within the state legal framework (Homme 2008). Moreover, the responsibility for lower and upper secondary education is divided between the municipality and the county municipality, making the collaboration between these two school administrations a crucial matter. During the 1990s, both lower and upper secondary education underwent reforms (Helgøy and Homme 2006). Since 1992 the Municipality Act decentralised responsibility for primary and lower secondary education to the municipalities. This reform was followed up by curriculum reforms in 1997 and 2006, which consigned the input-driven governance model more or less to history. The 1994 upper secondary school reform gave all children a legally established right to upper secondary education within five years after finishing lower secondary school. In 2018 the so-called ‘youth right’ was extended to nine years.⁴ Currently, almost every youth (98 per cent in 2016; Statistics Norway 2017b) enters an upper secondary programme track. The pupils can choose among four academic and eight vocational programmes. Traditionally, the education system has included a prescriptive national curriculum, standardised teacher training and extensive regulatory legislation. In the past two decades, there has been general political agreement on the introduction of several NPM elements into education (Solhaug 2011; Helgøy and Homme 2016). However, marketisation elements such as privatisation, public–private partnerships, subcontracting and school choice have been introduced only to a limited extent.

Problem definition

The model of a social democratic welfare state presupposes a high employment rate in order to sustain the high standard of welfare services. Demographic ageing and need for readjustment from an economy highly based on oil production reinforce this point as well as underline the need for a skilled workforce. A dropout rate of 30 per cent represents a huge threat to the maintenance of the welfare state. Promoting a skilled workforce and enhancing the completion rate of upper secondary education have become political priorities since 2003. The 1994 statutory right to upper secondary education for all young people was followed by a significant increase of students successfully completing secondary school. At the same time, it became obvious that a considerable group of pupils did not complete school. To combat non-completion, it became vital to the government to identify the reasons for dropping out, in order to implement accurate policy measures.

The two centre-left coalition governments (2005–2013) led by Prime Minister Stoltenberg from the social democratic Labour Party put early school leaving on the political agenda and pointed to a strengthening of strategies to prevent school dropout in their policy platform for 2009–2013. Moreover, government officials dominated the policy formation process. Several national policy programmes have been launched, with somewhat different definitions of what causes non-completion. Two ‘trends' characterise the problem definitions: one focusing on lack of basic skills in numeracy and literacy, and the second on lack of motivation and psychological distress. However, the significance of each of the trends has varied over time. A White Paper, Quality in Education (Ministry of Education and Research 2007–2008) drew attention towards early intervention by stating that pupils dropout because of lack of skills. In addition, the document pointed out that the lack of motivation and the shortage of apprenticeship places were prevalent factors. The Green Paper Diversity and Mastering (NOU 2010:7) underlined lack of basic skills as a core reason for early school leaving. The Green Paper expressed a concern that the number of low-performing pupils had increased and tended to dropout or take longer time to complete upper secondary education. Hence, the government decided to increase basic skills and to motivate pupils who were tired of school or had dropped out. The ideas of basic skills and early intervention were further developed in the White Paper Motivation – Mastering – Possibilities – Lower Secondary school (Ministry of Education and Research 2011). Since autumn 2013, a right wing coalition government led by Prime Minister Solberg from the conservative Right Party is still in power. During the past few years, the government has signalled a more diverse understanding of the early school leaving problem. Lack of basic skills and motivation are still seen as relevant but in addition, psychological distress among pupils has been promoted as a contributing factor in research as well as in the policy documents and strategies (Ministry of Education and Research 2015; The Norwegian Government Ministries 2017). The shift in government represented also a shift of stakeholders in the policy formation process towards decentralisation. The state, represented by the Ministry for Education, still formulated the national goals, but responsibilities for defining measures were devolved to the county municipalities and municipalities (Homme and Hope 2017).

Policy measures

What was the content of the policies that aimed to intervene early, motivate pupils, raise the level of basic skills, and provide an increase of apprenticeship places? Since 2000, a few national strategies were implemented. First, in 2003, the national government introduced an evaluation of an anti-dropout programme for upper secondary education (Satsing mot frafall, 2003–2006), consisting of numerous measures at county and school levels as a part of a broader national anti-poverty campaign. Another strategy, prioritising the teaching of basic skills in writing, reading and mathematics, was part of the larger 2006 Knowledge Promotion Reform. In 2011, the government introduced two major programmes: The New Possibilities (NP) Transition Project and the NP Follow-up Project. The Transition Project included intensive tutoring in basic skills for low-performing pupils in lower secondary education (Helgøy and Homme 2013). In addition, the policy introduced measures aimed at supporting pupils in the transition phase between lower and upper secondary school, such as career guidance and different efforts to motivate upper secondary pupils to complete. The Follow-up Project included training models that combined work experience and upper secondary school learning objectives (Sletten, Bakken, and Andersen 2015). While the Transition Project was directed to 15–16-year-olds in lower secondary school, the target group of the NP Follow-up Project was pupils in upper secondary school, age 16–21, who were on the brink of quitting school or already had left.

The Government introduced the NP strategy relatively rapidly in 2011 through a top-down implementation process, which presupposed a systematic cooperation between the county municipality and municipality. Moreover, the policy underlined that the county municipality was responsible for easing the transition phase between lower and upper secondary school. Several networks were established to ensure the transition phase between lower and upper secondary school: at national level (county municipality administrators) and at county level (administrators and lower and upper secondary school teachers). The organisation of responsibilities in the different counties appeared unclear and the programme NP strategy coordinators tended to have variable legitimacy and influence at both county and school level (Helgøy and Homme 2013).

The experiences from the NP strategy have shown that there is little or no evidence of positive effects regarding improved student performance (Lødding and Holen 2013; Huitfeldt, Kirkebøen, and Rønning 2016). In general, the implementation tended to be temporary and not very well integrated in schools’ overall activities (Holen and Lødding 2012; Helgøy and Homme 2013; Rønning, Hodgson, and Tomlinson 2013). The continuation of the efforts was therefore at risk. The programme ran until the end of 2013 and was partly integrated into the subsequent national programme introduced by the current government (2013–2019), i.e. the Strategy for Enhancing Completion (SEC).

The goal of the SEC programme (2014–2016) was twofold: to prevent ESL and to retrieve students who had already dropped out. Accordingly, the target groups were youths between 16 and 21 either in danger of leaving, or who had already left, school. The NP national networks were continued in the new programme. The networks were assigned to make the local efforts more systematic and evidence-based. Moreover, they were expected to contribute experience-based knowledge in systematic trials (RCTs) of local measures. Consequently, the programme aimed to bring forward good experiences and to learn from former policy programmes and local practices on the one hand, and from the trials of new measures and their effects on the other.

The specific framework for enhancing completion consisted of six steps:

• (1). Transition from lower to upper secondary education: Counties and municipalities were supposed to collaborate in identifying and preparing students in need of special support;

• (2). Starting upper secondary education: mapping out and planning adjusted teaching and psychosocial support;

• (3). Following up through the first year: motivating and compensating measures to prepare for the second year;

• (4). Following up throughout the second year: creating a foundation for choosing a career;

• (5). Transition from second to third year to ease completion education;

• (6). Completing by giving a realistic assessment on how to complete successfully.

A ministry directive regulating school absence was implemented in autumn 2016. Absence from school was seen as a risk factor leading to dropout, so, the education authorities decided to limit the number of permitted undocumented absences to 10%. This means that students with higher undocumented absences will not pass. According to the evaluation of the directive it has succeeded in decreasing absence and truancy, although students more often have to contact physicians in order to document absence.

Of special significance for this paper is the important dimension of governing and organisation principles. The SEC programme relied on experiences of the organisation and implementation of earlier policy initiatives. Responsibility to implement locally adjusted measures were devolved to the county municipalities and municipalities. However, the ministry applied typically soft steering mechanisms like stronger coordination and spreading of results among the counties and municipalities. Compared to the former top-down strategies the central actors were the same, but with different roles and responsibilities. Increased local responsibility and autonomy were balanced against a supportive and coordinating state.

Policy design in the Swedish setting

Policy regulation

As in Norway, Sweden has a long history of extensive local autonomy, and the 290 municipalities play a central role in welfare delivery, including upper-secondary education and related measures to improve completion rates (Lundahl and Olofsson 2014). The 1990s were a period of extensive reform, resulting in far-reaching decentralisation and marketisation. Several political decisions in the early 1990s opened up for school choice and introduced a tax-funded school voucher and privately owned independent (‘free') schools. In addition, municipalities took over all employer responsibilities from the state and a new funding system allowing municipal allocation of school budgets was introduced. A new curriculum was implemented to fortify this general direction of local autonomy and freedom (Jarl and Rönnberg 2015). In Sweden, almost all young adults enter upper secondary education (students normally aged 16–19), and even if it is not compulsory, completed upper secondary education is considered a necessity for future employment and inclusion in society. Upper secondary education is a key stage and passage in the education system, situated between compulsory education and later transitions to further studies and/or work. As such, it is exposed to and expected to fulfil a range of demands, which sometimes even conflict, making it an important and controversial policy field. Some of the controversies are rendered visible in the different upper secondary education reforms that have been launched during the last three decades (Lundahl et al. 2010).

A major reform of upper secondary education was decided in 1991, targeting, among other things, the scope and function of upper secondary education. All programmes, academic and vocational, were to grant eligibility for higher education, largely following an earlier tradition stressing both individual security, economic growth and a drive to avoid early and strong stratification of students. A new upper secondary reform, Gy11, was initiated by the non-socialist government, in office during 2006–2014. This reform constituted a significant break with the former policy direction (Lundahl et al. 2010). Out of 18 three-year national programmes, the six academically oriented programmes provide eligibility for higher education, but the 12 vocational programmes do not, unless optional courses are added. Five introductory programmes target students who are not eligible for the national programmes because they have not completed lower secondary education (Loeb and Wass 2015). Both municipalities and independent free school owners have to adhere to the national legislation, curriculum as well as programme plans and syllabi. The state operates an extensive and intensified school inspection scheme in which all municipalities, private school owners and upper-secondary schools are subjected to regular supervision and quality assessments. National testing of all upper secondary students is an additional centralised means of control currently employed by the state (Jarl and Rönnberg 2015).

Problem definition

Low completion rates have been central concerns for governments from both the political left and centre-right. However, the framing and attempted solutions to improve completion have differed. The separation or integration of academic and vocational tracks within the comprehensive upper secondary school organisation has been a matter of long-standing political conflict. Since the early 1970s, it has influenced how problems and solutions are defined. Swedish vocational education and training (VET) have been clearly school-based, with little or no apprenticeship training. The upper secondary reform in the early 1990s meant a prolongation of the vocational programmes, the introduction of a certain common core of courses and general eligibility for higher education. This line of policy was strongly contested by the non-socialist parties, arguing that vocational programmes had become too ‘academic', which was negative for school-tired young people, and resulted in increasing dropout (Carlbaum 2012). Following this framing of the problem, the non-socialist coalition government presented Gy11 as a means to increase employability and improve completion rates, by creating a stronger division between vocational and academic tracks and by promoting apprenticeship training (Lundahl et al. 2010). The results have however been meagre; the proportion of students who attend vocational programmes has gone down considerably, with potential labour market effects such as shortages of educated labour in vocational professions; and the completion rates have not risen to the levels expected. Recently the School Commission, appointed by the Social Democratic-Green Party government in office since 2014, proposed to reverse this policy, with the vocational programmes again providing general eligibility for higher education, as before Gy11. This illustrates how attempts to tackle low completion rates display party political differences and that definition of problems and their solutions display political dividing lines. It also exemplifies how measures to resolve the problem give rise to additional problems, such as low attendance in vocational programmes, and spark pressure to address new problems that emerge as a result of the attempted solutions (cf. Helms Jørgensen, Järvinen, and Lundahl 2019).

At the same time, there are also similarities in the political problem descriptions. Thus, they tend to focus on the young people rather than on structural and institutional factors, such as the characteristics of the labour market (Helms Jørgensen, Järvinen, and Lundahl 2019). A target group paid increasing policy attention to is young people neither in employment, nor in education or training (‘NEET’). Furthermore, Sweden has considerably larger proportions of migrant youth than the other Nordic countries, who, on average, have lower completion rates and higher unemployment levels than Swedish-born young people (Skolverket 2017b; Lundahl and Lindblad 2018). Students with special needs constitute yet another target group considered to be at risk of non-completion. Similar to Norway, being tired and unmotivated for studies constitutes a common problem characterisation (Lundahl and Olofsson 2014). VET is seen as a way to getting these youths back on the track, but also as a problem, since the completion rates from VET are generally lower than from the academic ones (Bäckman et al. 2015).

Policy measures

In general, measures to combat low completion rates are dispersed and the implementation structures involve several different actors and stakeholders at different levels, which are not unified in any common national strategies, as in Norway. The policy measures intended to prevent and manage school non-completion emanate both from a national frame legislation and from measures that are initiated and implemented within the far-reaching local municipal autonomy and discretion (Lundahl 2012). Thus, up to the age of 20, young people at risk of failing to complete secondary education are offered a seemingly standardised set of measures, but the local variations are considerable (Skolinspektionen 2016). Young people aged 20 or more who lack complete compulsory and/or upper secondary education, are referred to adult education. This means that ‘youth education' ends considerably earlier than in Denmark, Norway and Finland, but the consequences of this early cut-off is seldom or never discussed by politicians or researchers. Rather few national policy measures specifically target non-completers and students at risk of not completing their education at lower and upper secondary level. For instance, the Education Act (SFS 2010:800) asserts schools’, and notably the headmasters’ obligation to see to it that all students get an individual/action plan and enough support to reach the curriculum goals, thereby constructing universal target groups rather than selective ones.

Special needs support should normally be given in the ordinary classroom, but in reality, it also takes place separately, addressing individual students or small groups (Takala and Ahl 2014). Evaluations and research have indicated that Swedish education has major difficulties living up to the intentions of the Education Act to ensure that every student should get enough support to succeed in school (Lundahl et al. 2017). The decentralised Swedish education system has no nationally managed programmes addressing special needs students, and the steering documents are rather vague with regard to such support (Albinsson et al. 2016). In contrast to, for example, Finland (Jahnukainen and Itkonen 2016), Sweden also lacks a national strategy for the provision of special education.

The so-called activity responsibility means that municipalities have to keep themselves informed about young people 16–20 years of age not in education or work and offer them activities that may encourage them to continue studying or get a job. They are commonly offered a place at one of the introduction programmes, with an aim of giving individually adapted preparation for upper secondary education and/or work establishment. The local design and funding of the introduction programmes however vary considerably (Loeb and Wass 2015).

In the 2010s, Sweden has had a very high proportion of newly arrived refugees, many of them unaccompanied minors, but there are still few measures explicitly targeting migrant students. These students are normally included in the ordinary classes directly or a short time after the arrival, and they are part of the general support system described above. In addition, they are entitled to certain instruction and study help in their mother tongue. At the upper secondary level, recently arrived young people are normally offered a place at the language introduction programme (Lundahl and Lindblad 2018).

Since the mid-nineties, Swedish national and local measures to address school non-completion are increasingly influenced by supranational bodies and organisations and in particular the OECD and EU, for instance through youth policies and funding of youth projects. The so-called flagship initiative of the EU, ‘Youth on the Move' (European Commission 2014), that inter alia includes efforts to reduce the proportion of early school leavers, constitutes a recent example. The OECD has identified improved completion rates as an area that should be a policy priority for Sweden (OECD 2017) and these issues are high on the national political agenda. The recent Upper Secondary School Commission had the enrolment and completion of upper secondary education as the very focus of its investigation (SOU 2016:77). The commission proposed a range of measures to make all youth enrol and complete upper secondary education, but the proposal has resulted in few decisions so far.

Comparing policy designs

Based on the findings in the previous sections, we now turn to discuss similarities and differences in the policy designs in the Norwegian and Swedish settings, using the models of welfare regimes and transitions regimes as broad frames of reference (Esping-Andersen 1990; Walther 2006). Regulation: The first aspect of policy design concerns characteristics of the education governing system and how upper secondary education is legally regulated. Generally, Nordic education polices tend to deviate from other OECD countries, but neoliberal influences prevail and reduce the differences (Arnesen and Lundahl 2006; Lundahl 2012; Lundahl et al. 2013). The school systems in Norway and in Sweden both include national frameworks which set standards and at the same time are flexible to allow for individual plans (c.f. Arnesen and Lundahl 2006; Walther 2006). Both countries have left a regime of regulatory input-driven centralised steering systems, even if an output-driven control system ensures continued state influence. For example, the reforms of national curricula mean that detailed prescriptions have been replaced by broad goals. Both countries have introduced NPM elements into education, even though Sweden has gone further than Norway by introducing far-reaching marketisation. Tax-funded school vouchers, free school choice and a notable number of private schools, as well as an even stronger local autonomy, signify a weaker, or rather more market-oriented state regulation compared to Norway.

Turning to the problem definition, how the problem is constructed, and which specific goals politicians want to achieve, we find both similarities and differences between the two countries. In Norway, early school leaving is constructed as an important and urgent societal problem. Education is regarded as vital for realising the need for skills for readjustment of the labour market and the maintenance of the welfare state. In line with Walther’s typology of the universalistic transition regime, we find that youth unemployment is seen as politically unacceptable. In Norway, both left and right-wing parties agree that a 30 per cent dropout rate conflicts with the principle of aspiring to full employment. Furthermore, in this way of defining the problem, individual personal development for the youth becomes subordinate to structural societal challenges. The Norwegian Government has been eager to identify evidence-based factors that reduce the dropout rate and which can be politically manipulated. Two main factors have been pointed out: lack of basic skills and psychological distress, the latter of which has been given increased weight recently. However, the increased role of stakeholders at different governing levels has resulted in a variety of problem definitions in different local settings. In Sweden, low completion rates have been regarded as an urgent problem by both the political left and centre-right governments. Described as a threat to the welfare state in Norway, it is more cautiously formulated in Sweden as something that may have detrimental economic and social consequences, both from the individual and societal point of view. Especially, but not only, the political right and centre have associated the problem of dropout with overly far-reaching ‘academization’ of the vocational programmes and lack of apprenticeship training for allegedly unmotivated young people. The problem formulation also often relates to the high proportion of migrant youth in Sweden.

In analysing the policy solutions (c.f. Table 1), we identified the definition of the target group, the content of the policy measures, and the implementation structures. Both Norway and Sweden have developed and implemented policy measures to combat low completion rates that imply a general approach. All children and young people in primary and secondary education are in principle considered included in the target group, due to goals of early intervention and completion. However, where Swedish general measures tend to focus on all young people (up to 20) in risk of dropping out, Norway focuses more specifically on different target sub-groups – currently youth in need of strengthening basic skills and youth that suffer from psychological stress. In addition, pupils who have finished their first two years of vocational training but have not received apprenticeship training places are of particular concern. Hence a distinct difference between the two countries is that Sweden has a more general approach, while Norway has both a general and a targeted approach, which influences the content of policy measures. Norway has a national policy dedicated to combatting low completion rates, whereas Sweden combats low completion rates by integrating such efforts in general structural education reforms. Typically, while Norway implements policy measures of raising basic skills to increase the completion rates for the most vulnerable groups of pupils, Swedish reforms have, in general, targeted the organisation and structure of upper secondary education. However, targeted measures are by no means absent in Sweden. The five introduction programmes which target young people who are not eligible for a three-year national programme in upper secondary school are an important example; the municipal responsibility to stay informed and actively offer NEETs activities – mostly education – is another. Similar to Norway, one has attempted to increase the number of apprenticeship places by state subsidies.

Table 1. Policy solutions to combat non-completion: policy design in Norway and Sweden

	Norway	Sweden
Target group	Youth in need of basic skills Youth with psychological distress (targeted approach).	All young people up to 20 at risk of dropping out (general approach).
Policy content	Targeted policy programmes for early intervention, transition from lower to upper secondary education, strengthening basic skills and social support. Increase of apprenticeship places.	Overall structure of upper secondary school introduction programmes. Measures based on individual action plans. Keeping track of NEETs and offering activities in order to re-enter and complete secondary education.
Implementation structures	National strategies – local implementation (top-down). Local measures – local implementation (bottom up).	Not a unified national strategy, but national legislation stress the right for special support for each student etc. Public and private actors arrange and implement local measures.

The implementation structures in Norway and Sweden display both similarities and differences. The local level has devolved responsibilities for goal achievements in both countries. This implies a certain amount of local discretion and autonomy to influence policy content and the way the policy is carried out. Hence, both countries are characterised by governing structures that allow for bottom-up processes. However, even though Norway promotes such processes, the implementation structure is two-sided and includes certain top-down characteristics. The national strategies in Norway all imply local implementation of national policy programmes. However, the current ‘Strategy for Enhanced Completion' encourages the local authorities to identify local completion challenges and formulate adequate policy solutions. After a 15-year period of extensive decentralisation and marketisation, there have been increased efforts from the Swedish state to tame and steer the market, e.g. sharper school inspections, stricter regulation of the work of private schools and more restrictive establishment policies. The responsibilities of the principals have increased in the latest Act of Education, not least in order to discover early signs of students having problems reaching the educational goals, and to initiate adequate measures to prevent school failure. Still however, the local discretion is extensive, and the market characteristics tend to promote rather than delimit local differences. Table 1 sums up the main characteristics in Norway and Sweden in terms of policy solutions.

Conclusion

This article aimed to analyse and compare national policy designs to improve upper secondary education completion rates. A major conclusion is that Swedish and Norwegian policies intended to combat upper secondary non-completion are designed both to be included in a wider framework of general education policies and as targeted policies and programmes, but the Norwegian approach is more proactive and targeted than its Swedish counterpart. In Sweden, decentralisation and marketisation limit the space for state intervention, and leave extensive responsibilities to public and private school owners and to local education authorities to develop and implement measures to improve completion rates, with much local variation as a result. There may be several explanations for the diverging policy designs in this respect. Swedish politicians are still cautious to reduce the far-reaching local discretion that was enabled by the decentralisation and deregulation reforms of the 1990s. Even if there have been some proposals to strengthen state governing over education for example by reintroducing more targeted funding, they have not been realised hitherto; management by objectives and results and frame regulation still prevail to large extent. The changed policy landscape in Sweden, including large private companies in the school market that actively work to protect their interests, may render re-centralisation efforts more difficult. Also, and related to this, it is doubtful if there is enough of political consensus in Sweden to decide on a national programme to combat early school leaving and incomplete upper secondary education. In Norway, a stronger sense of urgency seems to have contributed such consensus, and the political muscles have been stronger than in Sweden.

Waltheŕs (2006) notion of transition regimes, including a universalistic transition regime, exemplified by the Nordic countries, did not touch upon policy differences within the regimes; he instead wanted to characterise the ‘Gestalt’ of the regime (2006, 125). Our analysis however highlights the need to do such intra-regime comparisons, and illuminates the considerable span that may exist, even when we look at countries as similar as Norway and Sweden. Finally, we argue that the findings in this paper underline the remaining significance of national policymaking and national policy design in times when external pressures for action and convergence are preeminent, not least from the OECD (2017). Our analysis of school completion policies has shown how certain national regulatory frames and governance structures continue to shape and influence policy design in the context of (increasingly diversified) Nordic social democratic welfare state regimes.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Ingrid Helgøy is Excecutive Vice President and a Research Professor at NORCE Norwegian Research Centre, Department for Social Science. Her main research interests concern public policy, education policy, multi-level governance, implementation of reforms and marketisation.

Anne Homme is Senior Researcher at NORCE Norwegian Research Centre, Department for Social Science. Her research interests concern public management and administration, in particular, multilevel governance in education, comparative education policy, and administrative reform and institutional change in education. Homme is the leader of the research project ‘Lost in Transition. Governance, management and Organisation of Policy Programmes to Improve Completion of Upper Secondary Education’ funded by the Norwegian Research Council.

Lisbeth Lundahl is Professor at the Department of Applied Educational Science at Umeå University, Sweden and visiting professor at the University of Turku. Her research interests concern education politics, youth politics and the school-to-work transitions of young people. She is the leader of the research programme ‘Education policy and young people’s transitions’ at Umeå University, and a team leader in the Nordic Centre ‘Social Justice through Education’.

Linda Rönnberg is Associate Professor in Political Science and Senior Lecturer at the Department of Applied Educational Science, Umeå University. She is also a Senior Research Fellow at the Centre for Research on Lifelong Learning and Education (CELE) at the University of Turku. Her main research interests concern education governance, politics and policy with a focus on inspection, evaluation and marketisation/privatisation.

ORCID

Ingrid Helgøy http://orcid.org/0000-0003-0332-2163

Anne Homme http://orcid.org/0000-0001-6237-2350

Lisbeth Lundahl http://orcid.org/0000-0002-5791-081X

Linda Rönnberg http://orcid.org/0000-0002-0209-558X

Notes

1. The percentage of pupils that have passed all subjects at all the course levels and are eligible for an upper secondary school diploma or a vocational certificate within five years after finishing lower secondary school (Statistics Norway 2018).

2. 2011 was the first intake after a curriculum change including changed grading and examination rules. Official statistics present percentages of pupils who receive a diploma from upper secondary school within 3, 4 and 5 years after starting upper secondary school (Skolverket 2018).

3. Norway: ‘Evaluation of the New Possibilities Strategy – the Transition Project’ comprised of studies of six lower and six upper secondary schools and six school owners (three municipalities and three county municipalities). Interviews were conducted with school owner’s representatives, school leaders, and teachers, in both 2012 (71 informants) and 2013 (52 informants). The study was funded by the Ministry of Education. The ‘Lost in Transition’ project included a questionnaire to all county municipalities and to the municipalities where the county municipality administrations are located, interviews at national government level and case studies in four counties, plus written documents and interviews with school owner’s representatives, professionals at school level, and pupils. The total number of interviewees were 152. The project was funded by The Norwegian Research Council (RCN).

Sweden: ‘Upper secondary school as a market’ included two questionnaires to principals and career counsellors in 60 municipalities, and intensive studies of eight case schools, including interviews with students, teachers, principals and career counsellors, in all 156 respondents. ‘Unsafe transitions’: (1) repeated in-depth interviews with 100 young adults in 23 municipalities, representing widely varying local contexts; (2) interviews with 45 local politicians and leading officials in the same municipalities; 2 national surveys to all 290 municipalities; analysis based on data from PLACE database. Both projects were funded by the Swedish Research Council.

4. The ‘adult right’ to upper secondary education is directed towards young people over 25 without upper secondary education.