Marketization on export: Representations of the Swedish free school model in English media

Introduction

One central element in many attempts to modernize Europe’s welfare states is marketization, including the implementation of business-like conditions and ideals in the public sector, expressed by terms such as new public management (NPM). By introducing mechanisms such as consumer choice and competition between different welfare providers, the public sector is orienting itself towards the functions of a market—or, more correctly, a quasi-market (Le Grand and Bartlett, 1993; Lubienski, 2009). Looking at education, there are good reasons to speak about a growing ‘edu-business’ (Ball, 2007, 2012; Mahony et al., 2004), with an extensive (and lucrative) involvement of private actors in education activities and educational provision. The involvement of private providers can take many forms, such as ‘through advice, consultation, evaluation, philanthropy, partnerships, representation, programme delivery and other outsourcing’ (Ball, 2009: 96). Taken together, these developments have opened the way for new actors in both policy-making and in the provision of services, creating complex networks and interactions that blur traditional distinctions between private and public (Ball, 2007, 2008; Lingard et al., 2013; Pierre and Peters, 2000).

Business and policy increasingly intertwine and these interlinkages reach far beyond the nation state. National borders do not bind reforms or ideas aiming at modernization. They travel and are learned, borrowed and brokered across nation states (Lawn, 2011). These developments show that ideas and practices related to welfare sector privatization in both its wider and narrower meanings, including how they diffuse and travel, are urgent objects of study (Meseguer, 2009; Schmitt, 2014).

From an international perspective, the Swedish education system is viewed as being extensively affected by marketization. A series of reforms implemented in the 1990s actually turned the Swedish school system into ‘one of the world’s most liberal public education systems’ (Blomqvist, 2004: 148; Lundahl et al., 2013). The Swedish system of free school choice has received international attention: ‘When it comes to choice, Milton Friedman would be more at home in Stockholm than in Washington, DC’ (The Economist, 2013). The Swedish system of school choice has been transformed into a ‘commodity’ to be ‘exported’ to other countries. Not only have the reform ideas seemed to travel, but companies that operate Swedish free schools have been on the move as well. Swedish free school chains, large for-profit stock companies that own several schools, have set up new schools abroad (Erixon Arreman and Holm, 2011).

In fact, ‘social democratic’ Sweden initiated and implemented choice reforms that attracted the interest of ‘liberal’ England (Baggesen Klitgaard, 2008). In the campaign leading up to the UK’s general election in 2010, the Conservative Party pushed the Swedish model of free schools hard:

The country that provides the closest model for what we wish to do is Sweden. Over the past fifteen years, Sweden has introduced a new system that has allowed the creation of many new high quality state schools that are independent from political control (…) And there is evidence already that what has worked in Sweden can work here. (Conservatives, 2007: 16)

Scholars have sought to examine the extensive Swedish marketization in education and to compare it to England’s (Allen, 2010; Baggesen Klitgaard, 2008; Hicks, 2015; West, 2014; Wiborg, 2010). However, this research has neither focused on the acclaimed political policy transfer from Sweden to England, nor on the role of the media in this transfer. By acknowledging the mediatization of policy and politics and linking it to notions of policy transfer and borrowing of policy ideas from Sweden to England, this paper hopes to contribute to a discussion about educational privatization and marketization in a mediatized, globalized world where ideas travel and are mediated across national borders. By studying how English media framed and portrayed the Swedish free school ‘export’ from 2008 to 2014, this paper aims to describe and discuss how a market-oriented policy idea, the Swedish system of free schools, is represented as it travels across national contexts.

The next two sections outline the theoretical approach and how the empirical material was gathered and analysed. After a brief description of the Swedish case of marketization in education, the rest of the paper presents the English media’s representations of the Swedish free school model in chronological order. A concluding discussion highlights the main findings of the study.

Policy as transferred, borrowed and mediatized

This research draws on perspectives related to notions of policy transfer, borrowing and learning (Benson and Jordan, 2011; Dolowitz, 2009; Dolowitz and Marsh, 1996; McCann and Ward, 2013; Meseguer, 2009; Ozga and Jones, 2006; Phillips and Ochs, 2003), studying issues such as why policies spread from one government to the next and why governments turn to other nations when designing their policies. The questions of whether, how and why voluntary policy transfer takes place have attracted a lot of scholarly interest and constitute important objects of empirical study and general theorization in several academic disciplines. The abundant literature has explored these issues under different conceptualizations, for example, policy transfer, diffusion, borrowing and learning. Meseguer (2009) cites a review that concluded that cross-national policy learning had been ‘overtheorized and underapplied’ (Meseguer, 2009: 14, citing Benson and Howlett, 1992). Another review article reports over 1000 academic articles published on policy diffusion (Shipan and Volden, 2012: 1).

A debated definition of policy transfer from Dolowitz and Marsh (1996: 344) refers to ‘a process in which knowledge about policies, administrative arrangements, institutions etc. in one time and/or place is used in the development of policies, administrative arrangements and institutions in another time and/or place’. Policy transfer intimately relates to notions of knowledge and learning (Alexiadou, 2014; Benson and Jordan, 2011; Lange and Alexiadou, 2010; Lingard, 2010; Meseguer, 2009). It is important to acknowledge that policies do not just transfer from one place to another, but transform as they travel from one location to another. Policies are ultimately

…assemblages of parts of the near and far, of fixed and mobile pieces of expertise, regulation, institutional capacities, etc. that are brought together in particular ways and for particular interests and purposes… An assemblage is always in the process of coming together and being territorialised just as it is always also potentially pulling apart and being de-territorialised. (McCann and Ward, 2013: 328)

Bringing such pieces together means constructing, moving and brokering policy knowledge. As pointed out by Ball (2012), the growing edu-business involves private consultancies, policy entrepreneurs and other actors in the making and implementation of policy. Policies travel and are carried by such policy actors for many reasons, and the issue of legitimation should not be underestimated (Waldow, 2009). The political legitimation of policy is vital for any government when launching and implementing policy and in so doing the media play a crucial role.

The relationship between the media and policy-making has been widely recognized, including studies of education policy (Anderson, 2007; Gewirtz et al., 2004, Levin, 2004; Lingard and Rawolle, 2004; Rawolle, 2010; Thomas, 2009; c.f. Barnard-Wills, 2011; Skorkjær Binderkrantz, 2012). The concept of mediatization attempts to highlight how ‘the media acquire greater authority to define social reality and condition patterns of social interaction’ by acknowledging that the media are

…a part of the very fabric of particular social and cultural spheres (…) a semi-independent institution that provides a nexus between other cultural and social institutions, as well as interpretative frameworks for our understanding of society as a whole, and that constitutes a common arena for public discussion. (Hjarvard, 2013: 3)

However, and even so, ‘the problems facing the conceptualization of mediatization require both a broader empirical base, as well as more thorough theorization as to the mechanisms and forces associated with the process’ (Rawolle, 2010: 25).

The mediatization of politics (Strömbäck, 2008) restricts how knowledge is exchanged, transferred, brokered and, importantly, mirrored in the public eye. Media provide interpretative frames for how to understand policy and politics and logic of the media is an essential part of how citizens, voters or ‘welfare consumers’ perceive and understand political life, as it sets up forms of communication according to a certain format set by the media (Altheide and Snow, 1979). Media framing ‘present[s] information in a way that resonates with existing schemas among their audience’ (Scheufele and Tewksbury, 2007: 12). Framing becomes a part of the very message construction rather than merely accounting for events and processes. Media outlets thus actively contribute to constructions of meanings (Thomas, 2009). Media reports and condensed forms of ‘packaging and selling information’ tend to follow a certain logic. In this dramaturgy, aspects such as angling and framing play a crucial role via mechanisms such as selection, framing, simplifying, exaggeration and enhancing drama and conflict. This logic is important to the media–policy relationship and to the study of media as a conduit of education policy (Rönnberg et al., 2013).

In general, there are several ‘cross-field effects’ (Lingard and Rawolle, 2004) between media, journalism, policy and politics. Strategic media use, such as spin, also contributes to ‘making up’ actual policy: mediated presentation of policy becomes as important as the policy itself: ‘Certain policies require the demonstration of progress and success and that this in itself becomes an intrinsic feature of the policies’ (Gewirtz et al., 2004: 327). This paper argues that aspects such as these need to be taken into account when analysing and discussing the mediatized context of policy transfer and borrowing. Even if ‘hitherto understudied by comparative education researchers’ (Takayama et al., 2013: 309), mass media coverage and the images conveyed are ‘important for how educational references to “elsewhere” are inserted into domestic educational reform discourses’ (Waldow et al., 2014: 305) and for understanding how ‘reference societies’ are constructed (Schriewer and Martinez, 2004; Sellar and Lingard, 2013).

Empirical material and sources

The English free school agenda has its own prehistory. The process of setting up academies—sponsored grant-maintained schools—as forbears of free schools is, of course, relevant to a wider understanding of the issue and its origin (Walford, 2014a). However, given the limitations of this paper, the focus is on the explicit free school discussion related to the handling of the issue before and after the 2010 general election. This was when the new Conservative–Liberal Democrat coalition government introduced free schools in England with explicit reference to the Swedish free school model (Walford, 2014b).

The English media climate appears increasingly sensationalist in a highly competitive market and public enquiries (such as Leveson) have indeed added to this (Baxter and Rönnberg, 2014). Printed news usually falls into three categories: broadsheets, with quality papers such as the Guardian, Daily Telegraph and Times; mid-market tabloids, such as the Daily Express and Daily Mail; and popular tabloids like the Daily Mirror, Star and Sun. The categories differ considerably, not only in their content, approach and style of reporting, but also in their readership (Evans et al., 2014).

The material in this paper is collected from the media storage database Factiva, including about 30,000 sources worldwide, such as newspapers, journals and radio transcripts, in more than 30 languages. Factiva includes all main English national and local newspapers and magazines, with or without a paywall. The database was searched systematically using the search terms free school, free schools, Sweden and education. The searches were limited to 2007–2014. In total, they generated over 700 hits. The reporting gained momentum in the election year of 2010, producing almost 300 articles, 11 of which were front-page stories. The selection of articles for further analysis involved several steps. Initially, I scanned the title and excerpts of all full-text articles that included the search terms. This scan made it clear that not all hits generated within the database were relevant for the study; some were from newspapers from other countries than England, for instance. Secondly, I produced full-text reports of the remaining articles for each year. Reading these full-text reports further reduced the number of articles included in the study. Many articles mentioned Sweden, but not in the context of free schools—the articles could be about school meals, health or a different topic. Even so, more than 400 articles remained for the next and final step—a qualitative content analysis (Bergström and Boréus, 2005). It focused on reading and coding the texts according to a set of thematic questions, namely (a) how the Swedish free school model and its export is represented, (b) what angles are chosen and highlighted, (c) who gets to speak and (d) what the claims are as well as the alleged evidence for those claims. Out of these, 45 articles were actually quoted in this paper (see the reference list), but the analysis was based on the much larger number of articles and was conducted chronologically.

The selection process aimed at extracting relevant material for studying how the Swedish free school model is represented in the English media and the format and outlet of that material was not a part of the selection criteria. Almost half of the initial hits from the Factiva database came from the broadsheet segment with quality newspapers, and their share increased in the final selection and comprised the main bulk of the material included in the study. Along with these broadsheets, the final sample also represented mid-market press well, but comparably few popular tabloids were present. The final selection also included a few local newspapers, press items from news agencies, weekly and monthly newspapers and magazines like The Economist and the Spectator, as well as a few transcripts from TV and radio.

Market orientation in Swedish education

Sweden has traditionally been regarded as a universalistic, social democratic welfare state regime (Esping-Andersen, 1990) and even if this welfare state regime typology has been subjected to much criticism over the years, it has been argued that we ‘should expect the liberal welfare state of America to be more likely to introduce vouchers than social democratic Sweden, as liberal regimes are more liable to allow for market-produced patterns of social stratification’ (Baggesen Klitgaard, 2008: 481).

The Social Democratic Party holds a prominent position in Swedish political history. This party has dominated the political scene in the past, leaving a distinct mark on national politics and policy, including education, often with a focus on equality and education as a means for social cohesion. Even so, the nonsocialist opposition has been active in educational issues and has also initiated significant reforms when in office. After winning the elections in 2006 and 2010, the nonsocialist alliance in government initiated and implemented a far-reaching educational reform agenda. However, the free school policy and profound marketization of Swedish education started much earlier. Both Social Democratic and nonsocialist governments have promoted and sustained it.

In the late 1980s and early 1990s, several education reforms were launched under different governments. Decentralization, deregulation and marketization were the main guiding principles. It can be argued that ‘decentralization paved the way, and was a necessary step, for the external privatization to come’ (Lundahl et al., 2013: 503). For instance, all employer responsibilities were transferred from the state to the municipalities; a new funding system was introduced; a new curriculum, grading system and syllabi were implemented; and a system of free school choice was set up with a tax-funded voucher that accompanied the student. A Social Democratic minority government initiated the reforms. From 1991 to 1994, a nonsocialist minority coalition intensified and carried out the reforms. The reforms significantly restructured the system, from one of the most centralist to one of the most decentralist education systems in the West (Dobbins, 2014).

Swedish parents can choose any school for their child free of charge, public as well as private—but still tax-funded—free schools. The education providers are allowed to extract profit. The number of students attending free schools has increased dramatically, particularly during the last decade. At present, about 25% of all students in the upper secondary level attend a free school and the corresponding figure for compulsory school is about 10% (Skolverket, 2013). In addition, the distribution of ownership of independent schools has changed during the last few decades. In the 1990s, foundations and other non-profit organizations were the most common owners. However, the share of independent schools run as stock companies by profit-making corporations spiked to about half of all free schools at the compulsory level and 85% at the upper secondary level (Vlachos, 2011). There is steep competition using different forms of marketing to attract students. This is costly in monetary terms and in the time head teachers and teachers have to spend on such marketing. Demographic changes will make competition even tougher as the total number of students decreases, causing the closure of large public schools and even bankruptcy of school companies (Hartman, 2011; Lundahl et al., 2014).

In their 2014 election campaign, the Left party strongly advocated against allowing free schools to extract profit. This remains a debate. Other parties have also raised concerns but have not been willing to redesign the Swedish free school model. A government commission from 2013, which excluded the Left, united several political parties behind suggestions to improve the system. The commission saw increased control, inspection and scrutiny to maintain high quality, including an even larger role played by the inspectorate, as the way to move forward. The commission approved the basic principles of the free school model (Jarl and Rönnberg, 2015; SOU 2013:56).

The coming sections analyse English media reporting of the Swedish free school model in three time periods. The concluding discussion then addresses the themes of the analysis and their implications for the role of the media in policy transfer processes—and the spread of global edu-businesses.

Media representations up until the 2010 election

Early material tended to display an amazed, but not yet overly critical, image of the Swedish free school model. Several articles appeared curious about it. They both sustained and challenged the stereotypical image of Sweden: The Economist (12 June 2008) noted that ‘big state, social-democratic Sweden seems an odd place for a free-market revolution’, and Prospect Magazine (26 March 2009) noticed that ‘interestingly, the normally egalitarian Swedes don’t seem too bothered that the chains make profits’. Even if the Conservatives were in the driver’s seat moving towards the Swedish model, early reports also emphasized that there was not much party-based opposition. Instead, political parties claimed to be ‘fighting to see who can come up with the most Swedish system’ (The Guardian, 14 April 2008a). All parties saw Sweden as the ‘flavour of the month by parties anxious to improve standards…all political parties have adopted something from the Swedish in their education plans’ (The Guardian, 14 April 2008b).

The issue of profit was common in the media coverage throughout the period. Most media did not believe this particular feature of the Swedish reforms to work well in England, but a Swedish actor receiving considerable media attention in the matter claimed otherwise. As one of the architects of the Swedish reforms, former government adviser Anders Hultin said the system needed profit to work. After helping put the Swedish free school model in place in the early 1990s, he went on to found the for-profit chain Kunskapsskolan that later also established themselves in England with several academies. He had a strong voice in the debate, talking about the need to allow profit and the success of the Swedish model. In an article in The Spectator, Hultin described the policy as a grassroots revolution in Sweden, claiming that ‘for-profit chains have greatly helped social mobility’ (The Spectator, 3 October 2009).

While some Swedish interviewees raised concerns, others described the model in very positive terms, such as ‘an immense success’ (Swedish MEP Gunnar Hökmark, the Moderates, The Guardian, 17 April 2008) or as ‘flattering and, OK, maybe we deserve it’ (Swedish Teacher Union representative quoted in The Guardian, 14 April 2008b). It was claimed as ‘cited evidence that children in these schools get higher grades’ (Press Association, 22 March 2009).

Media reports highlighting disputes then gradually increased. A local Tory chairman attracted attention by questioning his party leader’s ‘flag-ship for Swedish style free schools’ (The Observer, 5 April 2009) by saying the Swedes were moving in an opposite direction. The Observer mentioned a report from Sweden of increasing costs for local authorities, supporting Schools Secretary Ed Balls’ argument, whilst his Conservative counterpart, Michael Gove, maintained that ‘every robust academic study… has shown that resources are used more effectively and standards rise more quickly as a result’ (The Observer, 23 August 2009). Headlines such as ‘Swedish warning: Do not repeat our free school errors’ (Times Educational Supplement, 23 March 2010) framed the issue as contested with severe, unwanted consequences. The disputes—as represented in the media—occurred on several levels. The dramaturgy of exaggeration and conflict concerned the policy in general among politicians, and was mainly portrayed as a battle between Mr Gove and Mr Balls.

In the electoral campaign, the free school model and Swedish experiences assumed a central position in the media debate, even though ‘normally educational news in Sweden would be of slightly less interest to the British electorate than the cotton harvest in Kazakhstan’ (Times Educational Supplement, 12 March 2010). The media interviewed several Swedish informants, including researchers, leading officials and unionists. One particular voice receiving attention was the Swedish director general of the Swedish national educational agency Skolverket, Per Thullberg. He made an appearance on BBC Newsnight and claimed that results were not overly positive. ¹ This story then travelled and made its way into several media that cited the statements of this Swedish top administrator. However, the so-called evidence was increasingly difficult to interpret, for both politicians and journalists, as the following quote illustrates: ‘Sweden was a focal point of the debate, with Mr. Balls quoting various pieces of evidence arguing against their claimed success and Mr Gove cited an equal amount in favor’ (Times Educational Supplement, 19 March 2010).

The closer to the general election, the more polarized was the treatment of the Tory free school model. Mr Gove often had to defend his Swedish advocacy. Despite growing concerns about the evidence to support the presumed Swedish success, Gove’s quotes consistently and repeatedly asserted better attainment and parental power as drivers behind the reform, admitting that he is ‘utterly shameless in wanting to borrow the ideas that have driven up standards in other countries’ (Express on Sunday, 20 June 2010).

Media representations after the 2010 election

Now in office, the Conservative-led coalition could start to implement free school reform. Overall, the media material contained far fewer positive statements than at the beginning of the studied period, and headlines such as ‘Free schools come with a heavy price’ (Sunday Mirror, 30 May 2010) or ‘a great policy disaster’ (Western Morning News, 29 May 2010) are just a few examples. The less positive Swedish experiences were put more and more in the limelight and this framing included claims about the Swedes themselves backing away from the policy. The media also favoured reports on the Swedish decline in international and national test rankings, whether segregation had increased and the claimed popularity of the reforms in Sweden.

In the second half of 2010, reports began to present the English free school model as based not only on Sweden, but also on the US charter school model. Thus, they toned down the Swedish contribution in the forthcoming reforms, arguing the best parts from the USA and Swedish experiences were blended to make the English model. The reports emphasized international league tables and tests, making the declining Swedish achievements noticeable. At the end of 2010, it was clear that the release of the Programme for International Student Assessment (PISA) results influenced the ways the media represented Sweden. This included how they framed the way the Conservatives dealt with the inspiration they took from Sweden:

The Government is understood to have seen the latest unpublished Pisa results for 2009 before the white paper was completed. It makes only one mention of Sweden, which partly inspired ministers’ flagship ‘free schools’ policy, suggesting that the country may have slipped in the rankings. (Times Educational Supplement, 3 December 2010)

A few days later, this quote from the opinion pages amply illustrated both tendencies in the studied media texts in a quite straightforward manner:

This is the same Sweden that was namechecked in just about every education speech by a Tory politician until recently as the inspiration for their reform agenda. The free schools policy… was heavily lifted from the supposed Scandinavian paradise, not as the Conservatives would now have you believe, exclusively from the charter school movement in the US, a nation with a rather better Pisa than one might have expected. (Times Educational Supplement, 10 December 2010)

The role of and results from the Swedish free schools continued to be debated and, in a comment signed by Michael Gove, the Swedish success was reiterated despite the PISA decline and the notion that

…some sceptics fear that successful free schools will leave hollowed-out schools in their wake, but international experience shows that the dynamics do not work like that. In Sweden, free schools have helped drive up standards in neighbouring schools. As the OECD points out, two of the most successful countries in Pisa—Hong Kong and Singapore—are among those with the highest levels of school competition. (Mr Gove in Times Educational Supplement, 17 December 2010)

However, the reasons underlying the Swedish PISA decline were also disputed and notably few Swedish voices were heard in the English press in the wake of the release of the PISA results in late 2010.

Debate continued on the issue of profit. Advocates claimed different forms of evidence as to why the profit incentive is crucial to Swedish success, such as in this letter to the editor:

For free schools to deliver reform to deprived parts of the country, Mr Gove should look again to Sweden, the programme’s inspiration. Profit-making free schools have incentivised the establishment of good schools in poor areas and revolutionised Sweden’s education system. An Adam Smith Institute report earlier this year showed that the profit motive did not harm the quality of education in schools. If free schools are to become the revolution that they once seemed to be, the power of profit must be unleashed. (The Times, 21 June 2011)

But, sceptics continued to draw attention to the poor results, segregation and other unwanted effects, and that the Swedes themselves were starting to oppose the for-profit aspects of the reform. Critical voices also came from the English teacher unions, and the Labour Shadow Education Secretary remained one of the most cited opponents of the free school system. The media reports repeatedly noted the toning down of the system’s Swedish origin for strategic reasons: ‘Last year, Sweden was the model for reform. The government barely mentions Sweden these days, not since it emerged that its free schools produce marginally improved results, but increased social segregation. Now the emphasis has shifted to America’ (The Guardian, 29 November 2011). In addition, some took the opportunity to say that the once cherished ‘Nordic Nirvana’ (The Times, 19 January 2011) may not exist anymore, if it ever did:

To put it bluntly, Sweden is not a socialist, welfare state paradise of equals because it is not socialist; its welfare state is in some ways tougher than ours; it is not a paradise; nor are the Swedes as equal as assumed. In fact, take any popularly accepted belief about Sweden and it is probably wrong. (The Spectator, 9 July 2011)

In early September 2011, The Guardian reported that ‘In Sweden, a country known for its free schools, a state school programme consists of playing World of Warcraft—from home’ (The Guardian, 7 September 2011). A few days later, The Observer took a trip to Malmö. The angle was that in ‘the land that pioneered free schools, the education backlash is gathering pace’. In Sweden, as The Observer noted, a just-released research report had questioned the market forces at work in social welfare and education (Hartman, 2011). The reader also received a tour of a free school and a talk with researchers and the education minister, as well as one of the founders of a school chain. The reporter was surprised to observe the contrasts: two schools, located just a few minutes apart, displayed quite different realities. The less popular school was described as follows:

It was so noisy that I thought it must be break time. ‘Students here, they don’t have to do every task if they can show that they know it,’ a teacher said. Much of the learning at the 32 schools in Sweden run by the company is done alone by students, using an online system, with one-on-one guidance from teachers once a week, interspersed with lectures in classes of up to 60 students. If students prefer to play cards and chat all day, it’s up to them…Peter Connee, who runs ProCivitas, argued that segregation was an unavoidable side-effect of the system. ‘Fifteen years ago in Sweden, we had segregation based on where you live, now it’s based on ambition and ability’. Osterman also doesn’t believe it’s necessarily a bad thing. ‘We are becoming a school for ambitious immigrants,’ he said. But as I was leaving his school, one of his students, Mohammed Mahmoud, put it differently. ‘This is a school for criminals,’ he declared, to laughter. ‘Nobody’s working in this school, because no one here has any future’. (The Observer, 11 September 2011)

Media representations from 2012 onwards

The reporting on Swedish free schools during 2012 and the first half of 2013 revolves around a few themes. Several articles considered Swedish companies running schools in England, such as how Kunskapsskolan ‘attempts to import its radical approach’ (Times Educational Supplement, 10 February 2012). Barbara Bergström, founder of the Internationella Engelska Skolan (IES), met with top-level politicians: ‘Gove has told Bergström that he would like to see her manage a growing number in this country under the free schools programme’ (The Sunday Times, 1 January 2012). The fallen glory of the Swedish free school model was another angle in the reports, in which the international league tables continue to be important sources of information.

The recurring issue of profit was described as ‘politically toxic’ (The Guardian, 30 May 2012). Different teacher union representatives were visible in the profit discussions, and they turned to Sweden for one argument in particular: ‘It is well known that in Sweden one of the major ways in which schools make profit is by employing non-qualified staff to do the jobs of teachers’ (NUT general secretary is quoted in The Guardian, 30 May 2012). The Leveson inquiry fuelled the for-profit issue even further as Gove had hinted that he might allow for-profit companies to operate schools. The Labour Shadow Education Secretary took his chance to do battle with his political opponent by claiming first-hand experience from Sweden: ‘I returned yesterday from Sweden…there are real concerns about what has happened with profit making schools. One of the biggest is that it allows companies to run a free school for a period of time and then sell it on at a profit’ (M2 Presswire, 31 May 2012). However, the for-profit arguers are active as well; for instance Toby Young, English journalist and writer, claiming that for-profit schools ‘were the key to success of educational reform in Sweden’ (The Times, 18 July 2012). Another high-profile for-profit advocate was Sir Michael Barber, who ‘personally argued in favour’ of for-profit, even if the Blair administration he was advising took no action along those lines (Times Educational Supplement, 10 August 2012). Hultin was active again in making his case about the necessity of the for-profit motive to get the free school system working (Management Today, 1 September 2012). Proponents such as Fraser Nelson in the Daily Telegraph argued the dangers of not including it:

Oddly, this is not controversial in socialistic Sweden but it is beyond the pale in supposedly capitalist Britain, where Nick Clegg vetoed any expansion of IES [The Swedish school chain Internationella Engelska Skolan, author’s remark] as soon as he found out about it. So the Swedes have an education industry, but Britain—which has far better schools—does not. (The Daily Telegraph, 18 April 2013)

Sweden is represented as disappearing even more from the Conservative rhetoric: ‘Free schools in Sweden—anyone heard Gove mention them lately?’ (The Guardian, 3 November 2012). In addition, the Swedish drop in the international league tables formed a central line of the argument pursued by the Labour Shadow Education Secretary, with arguments such as ‘Michael Gove says Swedish-style free schools are the answers to raising standards but Sweden is 15 places lower then England in this league table’ (M2 Presswire, 27 November 2012).

In 2013, two particular events framed the free school issue in the English press. First, there was a bankruptcy: ‘One of Sweden’s largest free school operators has announced it will shut down, leaving hundreds of students stranded and raising questions about how far the UK should follow the country’s education model’ (The Guardian, 1 June 2013). Swedish politicians warned England:

Ibrahim Baylan, the education spokesman for Sweden’s opposition Social Democratic party, said the closures should come as a warning to the UK not to slavishly adopt the Swedish model, where private companies can set up profit-making free schools…Before you do something like this you have to really really think about how you set up the system,’ he said. ‘Nobody could foresee that so many private equity companies would be in our school system as we have today’. (The Guardian, 1 June 2013)

Mr Lofven said that countries should be ‘very cautious’ about allowing for-profit schools to be established. ‘Look at what happened here. It has developed into something which is almost uncontrollable,’ he said. ‘We see today schools going bankrupt, so youth that had started their three-year upper-secondary school education find that after two years their school is gone’. The school system has developed into the Wild West in Sweden. And it’s not a good system’. (Times Educational Supplement, 19 July 2013)

Sweden’s education minister, Jan Björklund, was reported to have plans to change the regulation of the free school sector at the time of this dramatic event (Times Educational Supplement, 30 August 2013). One of the central actors was once again Hultin, now employed as the chief executive for JB Education, who said,

Bankruptcy shows that the system is working. ‘From a general perspective I think that this is what the system is about,’ he said. ‘The alternative would be for bad schools to be protected in some way and not to do anything about their problems, and that would be very, very difficult for students. (Times Educational Supplement, 28 June 2013)

The second event in late 2013 was the release of the PISA 2012 results. Sweden had dropped significantly. As The Guardian puts it: ‘Sweden’s star falls’ (The Guardian, 4 December 2013), summing up the rise and fall of the Swedish model as a legitimate source of policy inspiration:

In 2008 Gove told the Tory conference that Sweden’s school reforms would be introduced if he was in government—and in 2010 promptly did so, with the advent of free schools. A few years later and Sweden’s star has dimmed. The 2012 Pisa results show Sweden’s exam results falling abruptly across all three measures of reading, maths and science—with the country recording the largest drop in maths performance over 10 years. (The Guardian, 4 December 2013)

In March 2014, Ofsted judged a school operated by the Swedish chain IES inadequate and in need of special measures (Eastern Daily Press, 10 March 2014). This resulted in media reactions: ‘for many on the left and right, it was always likely to represent something more than just a school providing an education to the young people of Brandon’ (Eastern Daily Press, 15 September 2014). Ofsted declared the school fit for purpose a few months later, but the debate continued.

Contrasted with the early reports from 2008 and 2009, the following quote illustrates the quite different political language surrounding the Swedish free school model, with PISA results as sources of legitimation, used as evidence to support claims from both the political left and right:

Mr Hunt [Labour’s Shadow Education Secretary, author’s remark] attacked Mr Gove for espousing Sweden as a model for Britain’s schools system, pointing out that its performance has dramatically nosedived since the last Pisa study. He said: ‘In 2008 you informed the Daily Mail, your journal of choice, that we have seen the future in Sweden and it works. Will you today confirm that this is no longer the case? In fact no other country has fallen so abruptly as Sweden in maths over a ten-year period… Mr Gove also brushed aside criticism of his approach regarding Sweden… ‘It is the case that in Sweden results have unfortunately slid. But as I mentioned in my remarks earlier what we need to do is not just grant greater autonomy, as they have to school leaders in Singapore and in Hong Kong, in South Korea and elsewhere, we also need a more rigorous system of accountability’. (Press Association, 3 December 2013)

Concluding discussion

This paper examined representations of the Swedish Free School policy in English media to illuminate policy-borrowing processes and the role played by the media in them. It also exposed the role of actors representing private sector interests in shaping national education policy agendas. As pointed out by Ball (2012), the configurations of the welfare state in the wake of marketization and privatization have engaged new actors in creating and carrying out education policy. This study has illustrated how certain actors become carriers of different policy ideas reaching across national borders. They even become policy entrepreneurs, as they are acting as retailers; they help market policies and the companies that will deliver them to the educational ‘consumers’. They export not only ideas, but also their services, in the form of educational content and practices. The Kunskapsskolan school chain and one of its founders are telling examples. After setting up the free school system in Sweden, Hultin started a company and he went to England to market the policy idea and also to establish the company on the English market, to ‘prove that their approach can work in England—something that could make the model more marketable elsewhere’ (Thanet Extra, 8 June 2008), consistently arguing for the Swedish for-profit model. As Kunskapsskolan and IES managed their media coverage and maintained visibility, media acts as a conduit of their activities in policy marketing and retailing.

The media reports include numerous references to meetings and contacts among political and business central actors in these processes, such as representatives from Swedish education companies visiting top-level English politicians and English politicians visiting Sweden. The media is willing to report on these interactions and thus help shape the representation for the wider public. In these mediated Swedish–English policy actor interactions, a particular form of ‘policy tourism’ (Whitty, 2012) is visible, holding implications for what policy is launched and how it is legitimized, perceived and represented, also in its mediated form. The actors’ media display, both as travelling policy tourists and as retailers, shape and reshape what is perceived and thought of. ‘Policy is not only remoulded when it is adopted in a new setting, but the mobilising of policy, as a socio-spatial, power-laden process, often involves change along the way, as policies are interpreted and reinterpreted by various actors’ (McCann and Ward, 2013: 328–329).

Focusing on the role of the media in processes of policy transfer, the empirical study shows that the media convey certain representations of the Swedish free school model. Important media filters are in play, such as the angles chosen, which actors are allowed to speak and what ‘evidence’ manages to get into English reporting. Selection and angling is ever-present. The reports frame recurring themes in similar ways and stories travel as press items to several media outlets. In the electoral campaign reporting, exaggeration, conflict and drama are used as techniques to attract attention, and this style of reporting continues after the election as well. As a Swedish reader I note that there are numerous misconceptions in the English reporting portrayed as facts. Names, places or descriptions of historical accounts are sometimes inaccurate, and such misconceptions travel to other articles and media outlets without being corrected or opposed.

The study illustrates the importance of acknowledging media as a format of communication in the process of policy transfer by providing certain communicative frames of reference during policy transfer. The media has done so here by simultaneously reinforcing and challenging national stereotypes and by treating international assessments as incontrovertible ‘evidence’. Policy transfer is part of national political legitimation strategies. Politicians frequently refer to learning from (perceived) good examples from abroad, which can be quite politically rewarding under certain circumstances. The stereotypes associated with the Swedish egalitarian, ‘socialist paradise’ and other descriptions found in media link the profoundly marketized system of free schools to a perception of an egalitarian, prosperous, welfare-oriented society. The policy becomes associated with and connected to a nation that in its stereotypical appearance should reassure sceptics that market forces are not uncontrollably unleashed. This serves as a ‘political shelter’ and convenient frame for the policy that can divert potential opposition. The role played by national stereotypes becomes important for how the policy is represented and also understood by the wider public, including what cognitive schemes it applies to for the readers. However, at the same time the contradictions in this stereotyping trigger media attention: Sweden does not seem to fit the stereotypical image, and challenging the stereotype, for instance via the common media angle about profit-making, means ‘taking the Eden out of Sweden’ (Times Educational Supplement, 3 April 2009). Media both reinforce and challenge this national stereotype.

When processes of political legitimation, aided by the mediated images of national stereotypes, coincide with the ‘governing by international league tables’ due to the declining Swedish PISA performance, two dominant dramaturgies of media framing merge. International assessments displayed in league tables are conveniently fitting a preferred format by the media by displaying a logic and dramaturgy that easily distinguishes winners and losers, competition, the sense of ‘crisis’ and subsequent attribution of blame. The media framing of international assessments as indisputable evidence invites both the political left and right to back their claims with such ‘evidence’, thereby contributing to the reinforcement of such media framing. Ultimately, it turns what was initially described as political ‘Swedophilia’ (Prospect Magazine, 26 April 2009) into something of an education policy Swedophobia, as other countries such as the USA start to serve as policy-legitimizing ‘reference societies’ (Schriewer and Martinez, 2004; c.f. Sellar and Lingard, 2013). The media reports on international assessments are connected to media priming, that is, ‘the changes in the standards that people use to make political evaluations’ (Lynegar and Kinder, in Scheufele and Tewksbury, 2007: 11). PISA results thus forcefully enter the media representation of the free school issue and function as such priming, via a certain logic of media framing that arguably influences both whether and how the public think about a particular educational policy issue.

Finally, this paper has argued for the importance of looking at media representations of the lender from the borrower’s point of view, emphasizing the role played by the media in shaping certain policy imageries and (re)presenting them to the public. Such processes of mediatization, including simultaneous national stereotypicalization and (de)legitimation, are relevant for understanding the assemblage of education policy in other systems penetrated by attempts of modernization via marketization in Europe and beyond.