Free schools arenewstate-funded schools in England. They have been opened by non-state actors who apply to central Government for the right and funding to set up and govern a not-for-profit school. Free schools can disapply the National Curriculum, do not have to adhere to national teachers' pay and conditions and can set the length of their school day. These ‘freedoms’ were reflected in the name ‘free school’. Partly borrowed from the Swedish free schools (Friskolar) policy, the Government in England has associated the term ‘free’ with an argument that free schools “aren’t run by the local council. They have more control over how they do things” (DfE, undated). A central policy argument for opening free schools in England is that they will create new competitive pressures for improvement in neighbouring schools, thereby “forc[ing] existing schools to up their game” (DfE 2010: 57). This argument contains several assumptions about how school choice and competition operate. As Betts (2009) argued in the case of Charter Schools in America – from which free school policy is partly borrowed – the assumptions are that: free schools will compete well in terms of academic quality; parents will express a strong preference for higher quality schools; existing schools losing students or status to free schools will (be able to) respond by improving academic quality. There are numerous ways, however, in which this “chain of causation” can break down (ibid: 197). Free schools may not offer better quality environments. Parents may not prioritise or be able to recognise academic quality. Existing schools may not perceive new competition or, where they do, may not (be able to) respond in ways that improve quality or equity. Little change or even deterioration in student outcomes could result.

As of June 2022 there are over 600 free schools open (which represent the vast majority of new schools opened in England since 2010). Many more schools are neighbouring schools to these new free schools.  This paper draws on a qualitative case study of the competitive effects of free schools on their neighbours, from a bigger mixed methods study, the aims of which were to: 

1. Test for the presence of free school competitive effects on student outcomes in neighbouring schools.
2. Identify the mechanisms through which potential free school competitive effects are manifested, by analysing whether free schools compete well in terms of quality, whether parental preferences for local schools change with a free school opening and whether existing schools respond by changing their practices.  

These aims require attention to the complexities of choice and competition across local markets. As such the main research question this paper is addressing is:

1. How are choice and competition manifested in local markets in which a free school opens? To what extent do local structural conditions, a free school’s aims and the local status of neighbouring schools influence perceived competition and action-taking?

Policy makers assume free schools create efficient competition, yet competition due to a free school’s presence works through a mix of mechanisms including selective competition. This has influenced the actions schools take, the distribution of improvement and deterioration and the patterns of social segregation.

A key lesson from the free school experiment is for policy makers to recognise the potential of selective competition and the outcomes this can create. This paper is relevant for an international audience interested in how the free schools experiment in England has played out in relation to choice and competition and what this might mean for other education systems that operate within the context of quasi-market supply side reforms.

The case study included 9 cases which allowed for an appropriate range of school types and local contexts to be included. Following our wider project’s neighbourhood definition, the boundaries of each case study were defined as a free school and nine closet schools of the same phase. We followed a convenience sampling approach. This drew on survey responses, where respondents were asked whether they would be willing to participate in a follow up interview. Invitations to participate were made to the headteacher of the free school and neighbouring schools that had not participated in the survey. The achieved sample, including the number of participating neighbouring schools in each case, is summarised in Table 7.1 below. Table 7.1: Achieved case study sample Cluster Case Phase Forecasted need prior to opening Participating neighbouring schools 1 A Primary Surplus 3 B Primary Shortfall 2 2 C Primary Surplus 4 D Primary Shortfall 3 3 E Secondary Surplus 3 F All-through Surplus 5 H Secondary Shortfall 4 4 G Secondary Shortfall 3 J Secondary Surplus 1 A common set of research procedures in each case supported comparative cross-case analysis combining two data collection methods, documentary analysis and semi-structured interviews. We interviewed the headteacher of the participating free school and the headteachers of participating neighbouring schools. The aims of the interviews were to understand the headteacher’s perceptions and experiences of competition locally; the schools own competitive actions and logics of those actions; wider relations with other local schools, including potential collaboration; and reflections on the wider consequences of the free school opening for local students. The data was coded by hand and analysed thematically through a parallel inductive and deductive approach, using the initial codes of: context; structural conditions; free school origins and ethos; student recruitment; perceived impacts of the free school; responsive actions; logics of action; and local consequences. Apriori codes were refined and added to through engagement with the data. On the basis of this thematic analysis we wrote individual reports for each case study to enable cross-case analysis. The cross-case analysis identified 4 clusters. Local cases were clustered on the basis of similarities in: i) their contexts and structural conditions; and ii) free schools aims and ethos. We were then able to analyse the extent to which these factors influenced perceived competition and, in turn, any action-taking.

First, the analysis highlights factors influencing the intensity of perceived competition, due to the presence of a free school. Local structural conditions were shown to be important, including both the extent of residential segregation and the balance of supply and demand for places. Declining rolls and increasing surplus places increased the perceived intensity of competition and the impacts of the free school. Second, the analysis demonstrated several foci of competition. In Cluster 1 competition was over student numbers and funding. In Clusters 2 and 3, it extended to social selection. This was influenced by the free school’s marketing, negative stereotyping of neighbours and recruitment practices perceived to cream, crop or exclude students. Third, action-taking in response to a free school’s presence was common, although not universal, and was influenced by perceived impact. Marketing and promotion were widespread. Sometimes this combined with new extra-curricular activities, particularly in middle and high-status schools, highlighting their use in signifying status. Differentiation was also identified, where schools used messaging to (seek to) restate the legitimacy of their provision. Fourth, while headteachers’ logics of action were often context-specific, there was a clear difference between high and low status. High status schools had locally advantaged intakes, likelihood of historic oversubscription and greater financial security. Their heads were less likely to report negative free school impacts and perceived greater capacity for action. Their dispositions towards action did vary by context, reflecting a distinction made by Van Zanten (2009). Where heads perceived their intake remained relatively stable, they tended towards a “monopolistic” logic, relying on an existing reputation to remain socially selective. Where heads perceived stronger competition, they tended towards a “entrepreneurial” logic, using promotional, differentiation and recruitment strategies to sustain an advantaged intake.

Betts, J. (2009) The Competitive Effects of Charter Schools on Traditional Public Schools, in Berends, M., Springer, M., and Ballou, D. (Eds) Handbook of Research on School Choice. New York: Routledge. DfE (Department for Education). (2010) The Importance of Teaching. London: TSO. DfE (Department for Education). (Undated) https://www.get-information-schools.service.gov.uk/glossary Van Zanten, A. (2009) Competitive arenas and schools' logics of action : a European comparison. Compare, 39(1): 85-98.