The OECD and the nation-state: an interdependent but ambivalent relationship

ABSTRACT

The aim of this article is to explore the interrelationships between a national government and the Organisation for Economic Co-operation and Development (OECD) in terms of the power and legitimization of education reforms, here based on the reciprocal dependence of the two actors. Although the current case study only analyses the shifting relationships between Sweden and the OECD, the assumption is that the case is quite representative of the ambiguities of power in the relationships between the OECD and its smaller member countries and the Nordic countries in particular. To explore the relationship between the Swedish national government and OECD, we make use of discursive institutionalism as a general framework. For a closer understanding of the standpoints of the OECD and Sweden on different matters and at different times, two explanation logics are used. The discourses we found characterizing the different kinds of relationships over time are a discourse of opposition, a discourse of mentorship and a discourse of subordination. We argue that there are both ideational and institutional reasons affecting the conditions for cooperation on educational policy. Although institutional reasons strengthen the maintenance of the cooperation, ideational reasons both challenge and develop the nature of the collaboration between the OECD and Sweden.

KEYWORDS:

• Education policy
• discursive institutionalism
• Nordic countries
• logics of explanation

Introduction

The OECD represents a transgovernmental type of network (TGN) based on an internationalization of the public sector in terms of strategic networks for states to extend their public authority beyond national borders to achieve common outcomes (Stone, 2020). With transgovernmental organizations is here meant international governmental organizations, as for example the Organization for Economic Co-operation and Development (OECD), and the European Union (EU). Both these organizations are examples of organizations that strive to formulate common education policies in cooperation with their member states. According to the OECD, the role of an organization is to ‘work on establishing evidence-based international standards’ in education by providing ‘a unique forum and knowledge hub for data and analysis, exchange of experiences, best-practise sharing, and advice on public policies’ (OECD, 2022). As we can see from the OECD’s description, the OECD does not provide policy suggestions to its member countries from a direct top-down perspective. Instead, the OECD uses more sophisticated tools in a mix of cooperation, evaluation and advice. The central question in this study is, therefore, the following: What different kinds of relationships do this palette of means of influence open up between the OECD as a TGN and an individual nation-state?

The purpose of the present article is to explore the interrelationships between a national government and the OECD in terms of the power and legitimization of education reforms, here based on the reciprocal dependence of the two actors. Although the current case study only analyses the shifting relationships between Sweden and the OECD, the assumption is that the case is quite representative of the ambiguities of power in the relationships between the OECD and its smaller member countries, the Nordic countries in particular. The research questions are as follows: ‘How does the national government use the authoritative power of the OECD to legitimise its policy?’ ‘How does the OECD use the Swedish education policy to substantiate its policy proposals?’

Nation-states participate in the transnational work of setting up new frameworks of international standards and serve as places for negotiating transnational policies and national adaptations. For nation-states, it could be helpful to refer to transnational evaluations and policy efforts for the legitimation of reforms (Steiner-Khamsi, 2004). Thus, the question of who is contributing with what and who is receiving what in this collaboration is not always clear. A result of the greater demands on transnational cooperation is an increased interdependence between transnational and national arenas (Wahlström & Nordin, 2022). In the present article, we explore the reciprocal relations between Sweden as a nation-state and the OECD as an organization for transnational policy, here regarding the nature of the relations, congruent or opposing policy agendas and the type of disputed reforms; in looking at this, we start with the Swedish school reform from 1991.

Cooperation within the OECD: a brief background

Today, the discourses that frame national policy texts are not only limited to national policy spaces; they are also part of policies emanating from international and transnational organizations like the OECD. The ‘transnationalisation of policy’ refers to the international transfer networks consisting of international organizations, state and nonstate actors based on soft governance, mainly in terms of convergences of norms (Stone, 2004). Education policy remains a state responsibility, but the state’s work takes place in both national and international spaces. According to Rizvi and Lingard (2010), the relationships between nation-states and transnational organizations are asymmetrical. The extent of asymmetry depends on the political and economic resources of the nation-state. Thus, an analysis of current policy processes needs to include analyses of contemporary states and their relationships to transnational organizations. In the case of Sweden, the OECD is the main transnational coactor for considerations about education policy. Rizvi and Lingard argued that the OECD’s ‘advocacy for the politics of deregulation and privatisation has been based squarely on its ideological beliefs about the role of the state, free trade and individual enterprise’ (2010, p. 39). As will be shown in the present study, this does not mean that the OECD without question supports the efforts of decentralization and privatization in individual countries. On the contrary, the OECD has criticized Sweden for these two points more than once.

The OECD was founded in Paris in 1961 by the 20 countries that, since 1948, had participated in the OEEC’s¹ work to coordinate the US Marshall Plan for Europe; among these countries, Sweden has been a member of the OECD since its inception (Swedish Government Office, 2023). The OECD has been groundbreaking in disseminating policy through international teams of experts who examine and assess the public performance of nation-states with the aim of helping the reviewed state improve its policymaking, adopt best practices and achieve established transnational standards and principles (Stone, 2020).

Within the policy field of education, the OECD has expanded its role as a policy actor from the mid-1990s onwards, mainly by shaping an infrastructure for collecting and comparing statistical data in education and through the organization’s capacity to influence the content of current educational reforms through collaboration with key policy actors (Sellar & Lingard, 2013). Therefore, the OECD plays a key role today in formulating what characterizes an effective national education system or what it means to be a competent learner or an excellent teacher (Sorensen et al., 2021). Woodward (2009) developed a four-dimensional model to illustrate how the OECD operates. The first dimension, cognitive governance, represents the belief that liberal democracy and market economy are the means that best serve collective solutions. The second dimension is normative governance, which places the OECD as a central actor for cooperation and shared knowledge for the member states to achieve common goals in national and international politics. The OECD has no legal or financial resources to influence national politics. Instead, the OECD uses its authoritative and moral resources, which include high competence, expert knowledge and a reputation of being an ‘impartial observer of global events’ (Woodward, 2009, p. 67). A third dimension is the legal instrument of the OECD, in accordance with the OECD Acts, and the fourth dimension is termed palliative governance because of the OECD’s role in dealing with complex or neglected problems (Woodward, 2009). For the present study, it is the soft power in the cognitive and normative dimensions of OECD governance that is central.

From the start, the OECD education policy has been characterized by tension between promoting the growth and freedom of the individual and developing the prosperity of society as a whole. For the OECD, the use of comparative statistics in particular has been an effective tool in the policy field of education (Ydesen & Grek, 2020). Approximately 1,600 Swedish experts and delegates visit the OECD every year (Swedish Government Office, 2023), illustrating the scope of ‘political work’ that results in the establishment of certain educational values and conclusions, while other alternatives are given a more subordinate role in a continuous cycle of institutional self-understanding (Ydesen & Grek, 2020). The statistical comparisons, especially the OECD’s Programme for International Student Assessment (PISA), have implied that countries seek references in other countries’ school systems in terms of ‘projection’, which means that the important thing is not to fundamentally understand the other school system but instead to attribute to this other school organization characteristics that one strives for in one’s own national school system. The statistics and standardization of education help reduce the complexity of real-life education systems and thus facilitate policymaking based on ‘rational’ arguments from international comparisons (Waldow, 2019), which can explain parts of the OECD’s attractiveness as an arena for transnational education policy.

As highlighted by Ydesen and Grek (2020), to fully understand the inner complexity of transnational organizations like the OECD, there is a need to explore the interdependence and interplay between individual nations and international bureaucracies and policy actors. According to Ydesen (2019), it is obvious from the origin and character of the organization of the OECD that economic goals and values are central when it comes to comparisons of efficiency and different aspects of productivity outcomes of the national school systems. Nationally based merits, such as those presented in PISA, are now important currencies in the global economic market (Pettersson et al., 2017). Thus, a high ranking in international knowledge assessments is attractive for nation-states in and of itself, not only for the OECD’s comparisons on what factors that ‘works’. Moreover, what keeps the member states together is still a shared value based on market economy (Ydesen, 2019) and, at least so far, liberal democracy (Woodward, 2009).

Education is viewed as a tool for developing both a market economy and liberal democracy. It is, however, an empirical question how asymmetries between nation-states and the OECD should be understood regarding the role of the different actors in the tension between economic and democratic considerations and between national and transnational policy solutions over time and for different countries.

A theoretical framework

To explore the relationship between the Swedish national government and the OECD, we use discursive institutionalism as a general framework. Discursive institutionalism focuses on the ways ideas are conveyed by people through discourses to understand how ideas lead to collective action that both maintains and changes institutions. Discursive interactions take form through discussions, deliberations, contestations and negotiations (Schmidt, 2011). Thus, to explore policy discourses means to examine what is said, by whom, in what contexts and with what purpose. Schmidt (2008) distinguished between coordinative and communicative discourses. In this context, coordinative discourses are important in the policy sphere for individuals and groups to create, elaborate and justify common policies. These policies then become contested in the public debate when they are presented to the public through communicative discourses, typically in the political sphere.

Discursive institutionalism focuses on ideas as the substantive content of discourse (Schmidt, 2011). These ideas can be understood at different levels. First, there are normative ideas forming a kind of general, philosophical understanding in society regarding what problems the society needs to handle and, in a broad sense, in what ideational direction possible solutions should be developed. The approach and content that together – and often unarticulated – shape the general philosophy of these public ideas in a society at a specific time can be termed normative background ideas. Normative ideas indicate what development and changes are preferable and what shifts should be prevented from a moral perspective. Second, there are programmatic policy ideas, representing more articulated and coherent policies that function as consciously formulated frames of references or overarching programmes that form the boundaries of different policies. At a third level, there is the local level where the policy solutions proposed by policymakers at the programmatic level are enacted and carried out. Both the policy ideas at the programmatic level and policy solutions at the local level can be understood as cognitive foreground ideas because they constitute the result of conscious ideas and formulations that imply a certain way of acting. In this way, cognitive ideas represent solutions to problems to be handled at the local level, being framed by the principles and norms underpinning programmatic ideas at the second level (Schmidt, 2008).

In the present study, the term coordinative discourses is used for the formations of discourses, where we argue that certain policy is becoming established, mainly at the level of ‘public philosophies’ (Schmidt, 2008, p. 306), a level that also can be termed the ‘societal level’ (Wahlström & Sundberg, 2018). Communicative discourses are represented on both a philosophical and programmatic level when policy is expressed and communicated to the public and stakeholders in terms of actual policy proposals.

Whereas discursive institutionalism is helpful in identifying competing discourses with different normative backgrounds, it has less to say about the explanations for why these discourses arise. Therefore, we use Parsons’ (Parsons, 2007) four logics of explanation to understand the standpoints of the OECD and Sweden on different matters and at different times. The typology of explanation comprises structural, institutional, psychological and ideational logics of explanation. First, structural explanations can be found in the intersection between what people do as a function of their position and the physical and material circumstances surrounding this position. Second, institutional explanations concern what people do as a function of their position in relation to institutional traditions, social norms and institutional rules. Third, psychological explanations are based on an assumption that people, at a general level, act as a function of cognitive or affective elements in the way of thinking. Finally, ideational explanations understand actions as related to peoples’ interpreting the world through certain ideational elements and norms. Although structural and institutional claims are logics of position, ideational and psychological claims are logics of interpretation (Parsons, 2007).

Method and analysis

The data consist of five reports from the OECD published between the years 1992 and 2021 and five SOUs ² [the State Public Investigations] between the years 2007 and 2022. The OECD reports have been selected due to their central roles in the formal communication between the OECD and Sweden in relation to Sweden’s two latest fundamental school reforms. The 1992 document is written against the backdrop of Sweden decentralizing its compulsory school system and opening up for market solutions in the school sector in the early 1990’s and where the OECD expressed strong opinions about the path that the Swedish politicians chose to take. The other OECD reports display the formal communication in relation to the last fundamental school reform in 2011 and the incremental reforms that followed with a strong focus on Sweden’s PISA results from 2000 onwards. The SOUs, which are similar to the Anglo-Saxon tradition of green papers, here represent domestic discourses and policy initiatives related to the 2011 school reform and the issues raised by the OECD. Although the empirical material does not enable us to give a comprehensive picture of the communication taking place between the OECD and Sweden during the time period studied, we argue that the centrality of the documents makes it possible for us to capture their main discursive elements.

The method applied is a discourse analysis of education policy documents from the OECD and the SOUs. The analytical approach has been inspired by critical discourse analysis (CDA) (Fairclough, 2001, 2010). CDA in accordance with Fairclough (2010) fits well with discursive institutionalism for mainly two reasons. First, CDA in accordance with Fairclough recognizes that discourse is constructed at the intersection of language and social reality, which means that some discursive constructions become possible while others are not. Secondly, by using Fairclough’s three stages of analysis, the focus is on the final stage, the explanation. To be able to explain why certain policy discourses emerge or become dominant during certain periods of time is in accordance with Schmidt’s (2008, 2011) purpose of discursive institutionalism.

The focus in the first stage of the analysis is on the text with an analysis the genre, formulation and content (description). In this step, characteristic arguments, terms and concepts are identified. In the second stage, the analysis concerns how the text has been used and interpreted in different social and political contexts. The focus in this step is on the interpretation and use of a particular text in terms of intertextuality. The third stage of the analysis is the explanatory phase, where we analyse the relationships between discourse practice (interpretation) and social practice. In the explanatory phase, the aim is to explain how the interpretation of texts relates to different social realities that contribute to the normative and cognitive understandings of a certain policy issue and why these understandings are constantly changing. The explanatory factor in the analyses is supplemented by Parsons’s (2007) logic of explanation regarding ideational and institutional explanations.

OECD and Sweden: contingent relations and positions over time

In this section, we describe three different discourses that emerge in the documents, representing shifting relationships between Sweden and the OECD; the descriptive and interpretive phases of the analysis resulted in three demarcated discourses. The discourses we have found characterizing the different kinds of relationships over time are a discourse of opposition, a discourse of mentorship and a discourse of subordination.

A discourse of opposition

In 1991, Sweden implemented a reform of the Swedish school system that can be termed a fundamental reform (Baek et al., 2018). The reform was fundamental in the sense that it represented a rupture in the governance of the Swedish schools, from a state control of the Swedish school system regarding financial resources, operational responsibility and employer responsibility to a decentralized governance of the schools where municipalities and private school owners took over the three main responsibility areas from the state. Simultaneously, the state shifted its governance to goal and result management with performance-based governing at all levels.

The 1991 Swedish school reform represented a third-order reform, with a shift in the following three components of policy variables: the overarching goals that guide a certain policy, the policy instruments used and the settings of these instruments (Hall, 1993). With reference to this third-order reform of the Swedish school system, the OECD (1992) wrote a report. The OECD stated, ‘Sweden has established a reputation of being at the forefront of educational policy thinking and in relating the development of its education to the country’s broader political, social and economic context’ (1992, p. 4). The OECD also noted that the reform took place at a time when Sweden was getting closer to the European Union (EU) and when a financial crisis hit Sweden hard as an export-dependent country, a scenario that ‘was constantly present throughout our discussions during our two-week visit to Sweden’ (OECD, 1992, p. 5). However, the OECD (1992), for its part, instead emphasized a changed political scene, one characterized by a new conservative-led coalition government as the main cause for the fundamental reform. The OECD wondered what the deficiencies were that should be remedied by the new, decentralized school system based on competition and free school choice. Furthermore, the OECD asked, ‘What precise measures are envisaged to support school improvement and stimulate innovation in teaching and learning which at present seems to be conspicuously absent in schools?’ (1992, p. 58). The OECD was also sceptical of ‘the government’s emphasis on the academic versus the social role of schools’ and asked if ‘the exclusive accent on judging performance by reference to subject-matter achievement’ would not lead to a diminution of nonacademic elements in school curricula and if not a ‘diminution of the socialising role of schools tend to reinforce the educational disadvantage of students from poorer socio-economic backgrounds?’ (OECD, 1992, p. 58). In this report on the Swedish 1991 School Reform, the OECD (1992) was consistently critical and questioning about the Swedish reform.

In the 1992 OECD review of education policy in Sweden, a clash of normative background ideas became visible. The reforms following a market-oriented logic to promote effectiveness, flexibility and, thereby, competitiveness in a global market were criticized by the OECD for being too far-reaching. The OECD questioned the ideological rationale behind the new policy, ‘in particular the application of the market concepts of choice and competition’ (p. 57), asking ‘how the search for diversity … can contribute to improving quality’ and how diversity and decentralization together can ensure the achievement of national standards and ‘sustain the objective of equality among social groups and geographical regions’ (p. 57). Moreover, the OECD questioned whether policy instruments like equality and differentiated resource allocation would be ‘sufficient instruments in enabling government to steer the performance of the system, and can such steering in effect be meaningful in the absence of a vision for the longer-term development of education, on which government at present seems to be agnostic?’ (OECD, 1992, p. 57). These challenging statements from the OECD are important to pay attention to because they illustrate that, although current research claims that the market-oriented governing system of new public management (NPM) was largely launched by the OECD (Gunter et al., 2016; Rizvi & Lingard, 2010; Stone, 2004), there was no sign of enthusiasm or praise for such a system in this first 1992 OECD report on the market-based school reform in Sweden. The OECD also criticized Sweden for abandoning its slow policy process using pilot studies and being in favour of a more politicized policy process. Although Sweden in the early 1990s transformed along managerial principles to maintain competitiveness in a global market, the OECD argued that Sweden should slow down and maintain central control over its educational institutions. Ironically, the OECD also warned not to rely too much on quantifiable outcomes from educational evaluations:

The pitfalls which surrounds all educational evaluation – the danger of undue reliance on quantifiable outcomes as against the quality of the educational process, of treating evaluation as an external objective distinct from the pedagogical value of self evaluation. … (1992, p. 16)

At this time, Sweden found itself in opposition to the OECD in trying to carry out reforms that today are perceived as integral parts of the OECD’s agenda. The Swedish reforms supported market-oriented principles in line with international policy, with a specific emphasis on freedom of choice regarding schools and education.

Around Two Decades Later

In the 2015 OECD report on the declining Swedish PISA results, the OECD again found itself in need of expressing disappointment because of the Swedish reluctance to follow their advice. The OECD acknowledged that, since the reforms in 2011, Sweden did a lot to strengthen external control of the school system. However, the OECD (2015) noted simultaneously that some of the measures taken have, to a lesser degree, responded to the OECD suggestions and that some have not responded at all. Thus, the OECD noted that the Swedish government did not follow its recommendations blindly but seemed to incorporate or translate some while ignoring others. However, this tactic was not considered a sign of national responsibility but rather a deviation from a prescribed way forward and a problem that must be dealt with.

A few years later, parts of the public in Sweden and the OECD once more were in a conflictual position regarding the launch of the 2018 PISA results, which attracted a lot of attention in Sweden when presented in December 2019. This especially the case because it emerged that a large part of the students who were selected to take the test had been excluded. Exclusion here refers to students who had been selected to write the exams but were later excluded from participating because of disabilities or lack of language skills. In PISA 2018, the percentage of excluded students in Sweden amounted to 11.1%, which can be compared with the OECD’s limit for a permissible individual exclusion of 5% based on disabilities or deficiencies in the assessment language. However, the OECD approved the number of excluded students based on the argument from the Swedish National Agency for Education (NAE), referring to the large migration of refugees above all during 2015. A vivid debate on the reliability of the Swedish results in the 2018 PISA test followed (Wahlström, 2023).

The Swedish Minister of Education and Research commissioned the OECD to investigate if the Swedish procedure around the 2018 PISA assessment was in accordance with the rules in its report. The OECD (2021) stated that although Sweden’s exclusion rate (11.09%) was exceeding the exclusion rate limit of 5%, Sweden’s data were deemed to be acceptable by the PISA Data Adjudication Group, for the reason of a large ad temporary increase in immigrant and refugee flows … (OECD, 2021, p. 3). The OECD argued that the high exclusion rate could be explained by the language factor and the fact that the migrant and refugee students had not yet achieved sufficient knowledge in the language of instruction (OECD, 2021). However, the Education Committee in Parliament was not satisfied with the government’s initiative to commission the OECD to carry out a review of the Swedish implementation of the PISA test in 2018 (Swedish Parliament, 2020). In addition, the Swedish National Audit Office (NAO) questioned whether the OECD was a suitable candidate to carry out the assignment because the OECD had already approved Sweden’s share of excluded students once. Furthermore, the NAO felt that the OECD lacked knowledge of the Swedish school system for new arrivals. The NAO’s assessment was ‘that the [OECD] report is based on incorrect assumptions and cannot be used to assess the reasonableness of the high exclusions for Sweden in the 2018 PISA survey’ (NAO, 2021, p. 57).

The discourse of opposition can be seen as conflictual and rebellious in nature. The discourse reveals a partnership out of step, where different actors and positions ended up in conflict. Instead of a helpful mentor, the OECD can be regarded as a biased and unreliable actor who cannot be expected to admit any potentially wrong decisions. Instead of being seen as an independent expert, the OECD was perceived as ignorant when it came to Sweden’s national school system and how it works.

A discourse of mentorship

In 2014, when the OECD visited Stockholm in an assignment from the Swedish government to evaluate the Swedish school system to reverse the trend of declining PISA results, it becomes clear that the OECD has understood its role as an expert organization providing external, neutral conclusions to which the national government should follow (Wahlström & Nordin, 2022). Thus, it can be designated as an asymmetric relationship that both parties agree on. In the document informing about the process of the OECD review of the Swedish school system, the temporal aspect was stressed, emphasizing Sweden’s urgent need for support.

In the introduction of the OECD report (OECD, 2015) presenting the final review results, Andreas Schleicher set the stage for the interdependent relationship between the OECD and Sweden, depicting the OECD as a provider of hope for a nation in despair, a mentor and a helpful friend:

Across the world, Sweden was once seen by many as a model for high-quality education, and it possesses many of the ingredients to become that again. Among these is the unwavering commitment of its citizens and policy-makers from across the political spectrum to do whatever it takes to provide all children with the knowledge, skills and values that they need to succeed in tomorrow’s world. The OECD is there to help Sweden rise to that challenge. (OECD, 2015, p. 3)

The quote reveals an implied competitive rationale based on the idea that we are all engaged in a global competition where we compete with our educational systems, schools and students. However, as in the case of Sweden, the OECD was there to help and provide hope reminding worried and anxious nations that success is possible if a country is prepared to accept the help offered by the OECD. The OECD functioned as a mentor, pushing and inspiring towards success. Looking into the final report of the 2018 PISA result where Sweden performed better, the strategy seems to be the same: first giving positive feedback on what was positive and then suggesting improvements in areas where the performance was still weak. In the report, the OECD stated that ‘Sweden showed an improving trend in all three subjects between 2012 and 2018, reversing earlier declines in mean performance’ (2019, p. 132). Then, the OECD continued to reveal areas of improvement.

As for the OECD’s evaluation of the Swedish school system (OECD, 2015), one of the areas of improvement had to do with the establishment of a national professional programme for teachers and school leaders. It was a proposal that the School Commission took note of, and it proposed that a national professional programme should be established to promote the professional development of teachers and school leaders and school development while creating the general conditions for capacity building in the school system (SOU 2017:35). An underlying assumption was that such a programme should strengthen teaching in Swedish schools and contribute to positive effects on students’ knowledge results. This led to another public inquiry into how the foundations of a national professional programme for school staff could be designed. In the official report on professional development (SOU 2018:17), this investigation continued along the lines established by the OECD and the School Commission by proposing a structure for a professional programme as a cohesive system for both teachers’ and principals’ professional development and to use this competence strategically.

Another area of improvement for Sweden to follow up on from the advice given by the OECD in its 2015 evaluation is the official report on an equivalent school (SOU 2020:28), which had the same investigator as for the report on a national professional programme for school (SOU 2018:17). The OECD (2015) pointed out that Swedish schools needed a changed governance, including a changed structure for financing. The OECD believed that the state needed to take back parts of the control over the governing of schools from which the state withdrew during the decentralization reform of the 1990s. The OECD believed that the reform at the time, taking away the state County School Boards in connection with the decentralization of the school, left a void behind. The OECD questioned that each municipality had the administrative capacity, competence and governing and control systems required to offer high-quality schooling. Therefore, the OECD claimed that there is a need to re-establish a state presence at the regional level to work systematically to maintain and develop the quality and equivalence of the Swedish school system (SOU 2020:28).

The official report on a more equivalent school (SOU 2020:28) presented its proposals in line with the advice given by the OECD regarding strengthening the state presence in the governing of the school system both nationally and regionally. The investigator suggested that the state should introduce a more systematic evaluation of the school’s results. The report also proposed a state regional organization to support the school organizers in their work of improving the knowledge results at those schools showing deficiencies in the students’ results. Moreover, the official inquiry proposed a joint choice of school where all guardians can participate and where all schools, both municipal and independent, could be selected. This latter reform was also in line with the OECD’s (2015) view of the reforms needed in the Swedish school system (SOU 2020:28).

A third official report building on the advice given in the 2015 OECD report and final report from the School Commission (SOU 2017:35) was an investigation into a state governing of the school (SOU 2022:53). The OECD noted that, during the 1990s, Sweden went from being one of the most centralized school systems to being one of the most decentralized. A large part of the responsibility for education was then transferred from the state to municipalities and private school organizers. Here, the OECD believed that this division of roles was problematic because it opened up ambiguities at several levels in the chain of governance. Above all, there was a lack of clarity in the division of responsibilities between the state and local levels. The OECD (2015) concluded that there was an imbalance between increased local autonomy and possibilities for accountability. The official report (SOU 2022:53) presented two possible lines of future development for the Swedish school system. In the first proposal, the consequences of a complete state takeover of the Swedish school were investigated in terms of both resources and employer responsibility. In the second proposal, a strengthened state governing was presented, including state funding of the school, while the school organizers retained the employer responsibility at a local level. In both proposals, the proposed state reforms had to do with recentralizing power and control in a highly decentralized and marketized school system.

In this discourse, Sweden accepted the OECD as a mentor and built its national reform agenda based on the advice given by the OECD. In this sense, this discourse can be seen as harmonious and functions along a rationale of mutual responsibility. As for the content of this discourse, the OECD tried to help Sweden balance between two normative background ideas. The first one could be described as equality through central government, here in line with the traditional Swedish social-democratic welfare state, and the other as effectiveness and competitiveness through decentralization in line with the NPM that started to influence public administration in Sweden at the end of the 1980s.

A discourse of subordination

The OECD stated in its 2015 report that Sweden has to ‘to”shift from a culture of compliance to one that strengthens accountability” OECD (2015), p. 172). Furthermore, the OECD stated that the expectations of student performance need to be raised, that accountability arrangements must be strengthened and that more focus must be paid to continuous improvement. A central part in this movement away from a culture of compliance was, according to the OECD, responsibility through accountability and evaluations. In the 2015 OECD report reviewing the Swedish school system, the OECD questioned the number of educational reforms in Sweden from the 1990s onwards and the ambiguity that the decentralization and diversification of the school system had implied. The OECD emphasized the need to strengthen the external control system, highlighting the importance of the introduction of an independent school inspectorate in 2008 to strengthen vertical accountability within the system. However, despite the introduction of an independent school inspectorate, new governing documents, and the state taking a firmer grip on the decentralized Swedish school, the OECD did not believe that the reforms were enough.

The OECD (2015) emphasized the use of assessments and evaluations as the road to success and stated that ‘promoting evaluation and assessment is clearly in the interest of students and their families, educational practitioners and education systems’ (OECD, 2015, p. 52). Although Sweden underwent a period of fundamental reforms, the OECD described them as ‘somewhat piecemeal, and simply too few, considering the serious situation of the Swedish school system’ (OECD, 2015, p. 55). The OECD tried to paint a picture that legitimizes the OECD to request Sweden to act in accordance with its guidance.

Although several Swedish investigations on education referred to the OECD, not least to results on PISA tests, there was a particularly clear connection between the OECD and the School Commission’s report ‘Gathering for school’ (SOU 2017:35). The proposals from the School Commission should be based on the final recommendations from the 2015 OECD report evaluating the Swedish school system. What distinguished this investigation from others was precisely that the government had not given an ‘open’ mandate to the School Commission to present proposals for changes in the Swedish school system as it usually does. This open mandate instead was assigned to the OECD, while the School Commission’s mission was restricted to further investigating the proposals that the OECD (2015) presented. The School Commission (SOU 2017:35) noted that its problem description for the Swedish school was in accordance with the view of the OECD (2015). The challenges for the Swedish school system, as reported by the School Commission, were, among other things, failing capacity and responsibility among many school organizers, too weak compensatory resource distribution, a lack of conditions for professional development and a school segregation, leading to quality differences between schools (SOU 2017:35). However, the solutions to the problems were not entirely in agreement between the OECD’s evaluation and School Commission’s proposal. Although some proposals from the OECD were embraced by the School Commission, others were downplayed or even rejected. It might be fair to say that the School Commission partly placed a different emphasis on their proposals compared with the OECD, rather than offering quite other solutions (Wahlström & Nordin, 2022; Wahlström, 2020).

Overall, however, SOU 2017:35 is a good example of when a nation-state in need of advice because of troublesome circumstances places itself in a subordinated position in relation to the OECD in what could be described as a highly asymmetric relationship. When the Swedish government gave the task to the commission to present proposals directly based on OECD’s suggestions in the 2015 OECD report, it deviated from a well-established Swedish custom of giving appointed commissions open mandates, although with some overarching directives. However, the only directives given by the government this time were that the commission should follow up on the suggestions by the OECD. In doing so the Swedish government recognizes the OECD as the primary source authority placing itself in a subordinate position. This approach also places the commission in a subordinated position since its work to a higher degree than what is usually the case is determined by ready-made proposals, although the result shows that the Commission made use of its autonomy and did not blindly follow the suggestions by the OECD.

In this discourse, the OECD can be regarded as a neutral, international expert who, at least to some extent, takes over governmental duties placing the government in a subordinate position. With its knowledge and transnational overview the OECD can guide Sweden in dealing with problems in the school system. In this discourse, there is not really room for opposition; instead, it is mainly about compliance.

A concluding discussion

The results have revealed an interdependent relationship between Sweden as a nation-state and the OECD as a transnational organization, here characterized by varying degrees of influence and power when it comes to the legitimization of educational reforms. As the present study has revealed, the relationships between the OECD and the Swedish government are both contradicting and shifting as they appear in the Swedish State Public Investigations (‘SOUs’). The same political parties that emphasized the OECD as a world-leading and neutral educational expert when in government (Wahlström & Nordin, 2022), a few years later argued that the OECD is partisan and ignorant when in opposition (Wahlström, 2023). The OECD, on its part, has been moving between an international, independent expert role, offering its advice to Sweden, and an annoyed patriarchal role, when the advice offered has not been considered sufficiently taken into account. In this concluding section, we strive to understand the mutually dependent – but over time ambivalent – nature of the relationships between the OECD and Sweden in terms of ideational and institutional explanations (Parsons, 2007).

The three identified discourses concern the interdependent relationships between the OECD and Sweden. The discourse of opposition, represents a communicative discourse in which the OECD and Sweden either ignore the other party’s arguments or formulate competing arguments in conflict with the otherwise agreed-upon roles. The second discourse, the discourse of mentorship, represents a coordinative discourse in which the OECD and Sweden have mutually affirmed and strengthened each other without questioning each other’s roles or legitimacy. The OECD has assumed the role of the leading actor. This is a ‘consensus discourse’ that usually places the two parties in a harmonious relationship with each other. Finally, the third reciprocal discourse is represented by a communicative discourse of subordination, with the OECD as the communicative agenda-setter. This third discourse takes the discourse of opposition one step further, placing Sweden in a role of subordination, where Sweden is expected to unilaterally implement the OECD’s proposals.

Ideational explanations

Ideational reasons are founded on the logic of interpretation, which means that the claims are particular and interpretative in nature. Ideational claims assume that certain actions are based on certain interpretations rather than on direct objective rationality. Thus, ideational reasons include the occurrence of ideational elements and a clarification of how these ideational elements have oriented actions in a certain way (Parsons, 2007). A normative background idea in the logics of transnational organizations like the OECD is that nation-states alone cannot develop the necessary capacity of policymaking in education to cope with global competition. The global trend places the OECD in a superior role in relation to national governments and policymakers because of its international expertise in education policy, here based on the overview that comparisons between different national school systems offer (Meyer & Benavot, 2013; Rizvi & Lingard, 2010; Stone, 2020). As Barnett and Finnemore (1999) argued, international organizations can become autonomous sites of authority through their classification and organization of information and knowledge, a process bound up with power. Related to the classification and categorization of knowledge is the ability to fix the meanings of core concepts and relevant actions. Having established knowledge, rules and norms, the international organization disseminates its ideas about ‘good’ political solutions. From this aspect, the OECD can be considered to act based on ideational reasons.

The transnational ideas from the OECD were challenged by the decentralization and privatization reforms of education in Sweden in 1991 and 1992. In the OECD report from 1992, the OECD appeared to be an advocate of education from a mainly societal and social perspective, while Sweden emerged as a proponent of free school choice, competition and accountability in a management by goals and results implemented by the government. At this time, the OECD and Sweden based their interpretations on the meaning of school education on different ideational norms and, thus, preferred different directions of action.

In the OECD, the Nordic countries, especially Norway and Sweden, had a leading role during the 1960s and 1970s (Eide, 1990). The ‘Nordic model’ was characterized by a compulsory education with no differentiation until the age of 16 and equal quality between schools because of an economic redistribution that ensured equal standards of resources. However, by the mid-1980s, conservative US delegates dominated the OECD’s education discussions on the theme of the struggle against communism and religious fundamentalism (Eide, 1990, p. 48). The influence of the Nordic countries became strongly reduced, even if the ‘Nordic model’ still had the potential to function as a desirable model, not least for the Mediterranean countries. Towards the end of the 1980s, the OECD’s interest again turned towards the question of education and economic growth. The social aspect of education (e.g. the risk of the ‘2/3 society’) and progressive education were back on the OECD agenda (Eide, 1990).

This brief retrospect has illustrated how ideational aspects affect the OECD’s agenda based on the ideational norms represented by the OECD delegates and their governments, and the same is true for the policy agenda in each country. Thus, in the report from 1992, the OECD criticized Sweden for its 1991 policy agenda of decentralization and privatization, asking what problem this reform was supposed to solve. The OECD’s questions about the social and economic potential of education, largely in line with ‘a Nordic model’, were in sharp contrast to Sweden’s reform of a decentralized, results-driven school system with a free and publicly funded school choice. The ideational basis of the Swedish reform was, by the time, mainly domestic, with very few references to foreign experiences, in the State Public Investigations from the 1980s (SOU 2007:75). As noted in the SOU (2007:75), it was only later in the 1990s that the OECD became a proponent for NPM and national reforms began to be perceived as elements in a common international reform movement. Hence, if Sweden was ideologically out of step with the OECD in the first years of the 1990s, its ideas converged a few years later. However, the OECD remained critical of Sweden’s decentralization of responsibility for the school for the following two decades, as well as to the wide framework Sweden offers private organizers of the school (OECD, 2015).

The OECD, as a policy organization, cannot base its legitimation solely on the member states’ funding and on providing expertise. The organization also depends on a direct involvement from the member states (Centeno, 2021). For direct involvement in common projects, the OECD depended on an adaptation of norms from the individual nation-state to the transnational norms of the OECD. The OECD has a certain advantage in the shaping of desired norms because of its blending of comparable, policy-oriented knowledge data with what is considered reliable guidance for policymaking. Taken together, this blending makes the OECD a ‘boundary organisation’ (Grek, 2014). Close cooperation between the OECD and member states requires a common agenda setting for joint goals and projects. The agenda setting, in turn, needs to be based on a collective understanding of ideational norms of education, an agreement of deficiencies in the national school systems and adequate policy concepts and desirable actions for solutions. This ideational influence from the OECD on Sweden as a member state becomes evident in the discourse of subordination when the Swedish government commissions the OECD to conduct an evaluation of the Swedish school system. The commissioned evaluation meant an open mandate for the OECD to point out the improvement measures that Sweden needed to implement in schools. At the same time, the subsequent Swedish Public State Investigation was requested to base its proposals on the OECD’s recommendations.

It is not always the national government that finds itself in opposition to the OECD because of its different ideational outlooks. In the case of ‘the 11%’, here related to the effectuation of PISA 2018 and the proportion of students who have been allowed to refrain from participating in the PISA test because of difficulties in mastering the assessment language, it was not the Swedish government that was in opposition to the OECD. Instead, it was some political parties – the same parties that previously had regarded the OECD as an independent expert in the assignment to evaluate the Swedish school system (Wahlström & Nordin, 2022) – parts of the media and parts of the public that ended up on different sides in an ideationally based struggle. This example shows that the nation-state can also use the OECD to discredit the national government in domestic political controversies on issues of a particularly contentious nature, in this case the issues of migration and integration (Wahlström, 2023).

Institutional reasons

Institutional reasons are based on logics of position, meaning that people’s actions within institutions are guided in certain directions. Such reasons understand action as the unintended consequences of the rules and conventions of social institutions. First, for an action to be institutional reasoned, institutional positions need to be linked to action. Second, these actions need to be explained as an institutional response to contingencies (Fairclough, 2010). The establishment of the EU may unintentionally have led to the OECD gaining a strong position as an education expert in Sweden. The combination of hard and soft governance in the EU may have led to a fear that too much national self-determination is transferred to the EU. Education is a national concern, but the mere suspicion that the EU would exercise supranational power in areas of national responsibility may imply that the OECD, which does not have the same power of regulation, was perceived as a ‘safer’ cooperation partner for the individual country. For the same reason, collaboration with the OECD may be considered easier to communicate to the public without being questioned in terms of influence and self-determination. As Nordin (2012) has shown, the influence of the EU on education policy in Sweden usually can be understood in terms of ‘silent borrowing’ (see also Waldow, 2009). Therefore, an institutional explanation is that, in the presence of two contemporary transnational organizations, the EU and the OECD, the OECD was perceived as a neutral authority on education and as a seemingly equal partner and knowledge coordinator, sharing the interests of the member states.

The discourse of mentorship characterized by Sweden’s acceptance of a need to be guided by the OECD because of the organization’s international outlook and expertise can also be explained by institutional reasons. The OECD is a formal intergovernmental organization, which is supposed to work to improve the economic conditions of the member states with the help of transnational (Stone, 2004), rather than supranational, agreements. From this perspective, it makes sense for a country like Sweden to accept the OECD as a mentor who reinforces the legitimacy of the government’s proposals.

The interdependent but ambivalent relationship between the OECD and Sweden

The analysis in the present study has revealed that the main tools for OECD governance are cognitive governance (Woodward, 2009), representing the belief that liberal democracy and market economy are the means that best serve economic prosperity for the member states and normative governance (Woodward, 2009), where the OECD becomes the central actor to create cooperation and shared knowledge as a basis for educational policy guidelines (Barnett & Finnemore, 1999; Sellar & Lingard, 2013; Ydesen, 2019). However, to also understand the character of the relationships between the OECD and individual member states, there is a need to explore the interdependence between the transnational organization and individual countries, as in the case of Sweden (Ydesen & Grek, 2020).

The present analysis has demonstrated that the relations between the OECD and Sweden are of a long-term nature and are based on a long history. Basically, the relationship is characterized by mutual respect and the recognition of each other’s dependence, which does not mean that interdependence always runs smoothly. In the present article, we have argued that there are both ideational and institutional reasons affecting the conditions for cooperation on educational policy. Although institutional reasons strengthen the maintenance of cooperation, ideational reasons both challenge and develop the nature of the collaboration between the OECD and Sweden (Schmidt, 2008, 2011; Parsons, 2007).

Acknowledgments

We are grateful to the Norwegian research project POLNET for creating an inspiring and innovative research environment with a policy focus on the Nordic countries. The project was chaired by Kirsten Sivesind in collaboration with Gita Steiner-Khamsi and Berit Karseth.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. The Organisation for European Economic Cooperation (OEEC).

2. Statens Offentliga Utredningar (SOU).