Do free schools create “competitive threats”? The perceptions of neighbouring schools

A central policy claim for opening free schools in England was that these new schools would create competitive threats that incentivise nearby schools to improve. Where comparable policies have been pursued, notably charter schools in America, research has often measured competitive e ﬀ ects quantitatively. The perceptions of local actors assumed to experience competition can be overlooked. Responding, this article develops the ﬁ rst qualitative analysis of a sample of 28 schools neighbouring 14 free schools. The analysis evidences the importance of market contexts, with perceived competition mediated by supply and demand for places, choice geographies and school status hierarchies. Resulting diversity does not support simplistic assumptions nearby schools perceive competition or respond by seeking to improve. Just under half the schools studied perceived free schools created few competitive threats. Where competition was perceived, headteachers reported taking actions on promotion and national performance indicators, but rarely classroom practices. Where competition was strongest, headteachers reported intensive social selection, with free schools perceived to increase inequalities. The in ﬂ uence of school status on competitive logics of action was also revealed. Unequal material and symbolic resources a ﬀ orded by status positioning informed headteachers ’ perceptions of their relative capacity for e ﬀ ective action. Nine ideal types of logics of action are outlined, identifying how the context of competition and a school ’ s status in ﬂ uenced orientations to action. These logics are shown to be important in how intensive competition increased socio-economic inequalities by concentrating the negative e ﬀ ects of opening free schools, where these existed, in lower status schools serving more disadvantaged students.


Introduction
The free school policy introduced in England in 2010 allows for the opening of new statefunded, not-for-profit "free schools".One policy aim is to enable a wider diversity of "providers" to open and govern free schools, independent of local government (Higham, 2014a).A second aim is for free schools to meet parental demand, where this exists, for new, better or different schools, even where there is an over-supply of places (Gove, 2011).A further related aimand the focus of this articleis for free schools to create new competitive threats that incentivise neighbouring schools to improve.By "improvement", the policy is predominately concerned with a narrow focus on student attainment and progress as measured by standardised tests.The Government has made substantial claims here that attainment will improve for nearby students who do not attend free schools.In May 2015, five years after the free school policy was introduced, the then Prime Minister David Cameron (2015, p. 1) argued: free schools don't just raise the performance of their own pupilsthey raise standards in surrounding schools in the area too.
This claim rests on the assumptions that existing schools perceive new competitive threats from free schools and respond by seeking to improve.As Betts (2009) argues in the context of charter schools in America, from which the free school policy is partly borrowed, neighbouring schools may not however perceive new competition or, where they do, may choose not to or be unable to respond in ways that improve academic quality.The National Education Union (NEU, 2019, p. 1) in England argues, further, that free schools "harm neighbouring schools in areas where there is no shortage of places".Hatcher (2011, p. 489), in this journal, predicted free schools would create "probable consequent negative effects on neighbouring schools", particularly when a school lost pupils to a free school: "the consequence could be job losses by teachers and other staff, a reduction in the curriculum offer, and a spiral of decline even resulting ultimately in closure" (Hatcher, 2011, p. 500).
Such negative effects are not entirely contrary to the Government's aims to create new competitive pressures in England.The Government's impact assessments of individual free schools, made prior to their opening, collectively indicate a willingness to tolerate threats to the viability of a school, by arguing this is outweighed by potential increases in choice and pressures for improvement.There is currently little evidence however to assess these arguments.One report published by the think-tank Policy Exchange (2015, p. 1), claimed to analyse the effects of free schools on nearby schools.It argued: "the competitive effect created by a free school leads to improved academic standards in nearby underperforming schools … [with] bigger gains in higher poverty schools".These findings have been widely criticised, however, for making an "implausible claim" using an inappropriate methodology (Green, 2015, p. 1).
In the context of these debates and with over 600 free schools opened by May 2022 this article takes as its focus neighbouring school headteachers' perceptions of competitive pressure and action.The rationale for this focus is two-fold.First, in the wider international context of attempted quasi-market supply side reforms, the perceptions of neighbouring schools remain under-researched.In the case of charter schools in America, for instance, Epple et al. (2015, p. 52) argue that a predominately quantitative research literature "generally assumes we know how a competitive threat is perceived by relevant actors".Proxies of structural competition, such as distance to a charter school or the market share of charter schools have been used commonly, despite it being "difficult to establish good proxies for competitive pressure".Second, research on school staff perceptions has helped to interpret the meanings actors attribute to local competitive relations and whether, and if so how, these influence their actions (Jabbar, 2015;Levacic, 2004;Van Zanten, 2009).
The article addresses therefore two main research questions.First, in what contexts do neighbouring headteachers perceive the presence of a free school creates new competitive pressures?Second, are schools influenced to take new actions where they perceive competitive pressures and, where they say they do, what logics inform these actions?Progressing the analysis, the article concludes by considering the emerging consequences of the presence of a new free school for students locally and the related policy implications.

Quasi-markets and local status hierarchies
The formalisation of the "quasi-market" into which free schools are being opened was developed in the 1988 Educational Reform Act (ERA) but also in prior and subsequent reforms (Ball, 2009).Drawing on the economic arguments of Hayek (1944) and Freidman (1982), and with similarities to concurrent neoliberal policies in New Zealand, Chile and the USA, these reform aims included encouraging local responsiveness through school diversity, enabling greater equity through choice and improving student attainment.
The achievement of these aims has been repeatedly questioned.The effects of choice and competition on student attainment have generally been found to be weak (Allen & Burgess, 2010).There is clearer evidence that quasi-markets reproduce student inequities through socio-economic stratification (Gewirtz et al., 2006).Whether this leads to increased segregation is debated (Gerwitz et al., 1995;Gorard et al., 2003), but statefunded school diversity policies have been found to increase segregation (Gorard et al., 2013).Since 1988, there has been a growing diversity of school types, particularly relating to policies on how schools are governed.There is little evidence such diversity leads to local responsiveness or "innovation" however and rather signs of greater uniformity (Woods et al., 1998).
One set of insights into why school quasi-markets have often created these outcomes is afforded by the concept of "local status hierarchies".This is the perspective that over time choice and competition work to position a school relative to others on the basis of perceived status (Ball & Maroy, 2009).Quasi-markets do not necessarily create status hierarchies.There is a long relationship in England between a socio-economic class-stratified society and the educational system (Whitty & Power, 2015).Rather, quasi-markets are argued to give a decisive new edge to status, by stressing the consequences of losing status (Van Zanten, 2009).Three aspects of local hierarchies are important in conceiving how positioning may influence contemporary free school competition.
First, local hierarchies are rarely found to neatly reflect school quality, in terms of measures of inspection, student progress or attainment.Rather, hierarchical positions are informed by a wider range of norms, values and inequalities (Woods et al., 1998).Jabbar (2016) argues school status positions are socially defined and shaped by class and race inequities and geographies.Greany and Higham (2018) report that school leaders in England perceive status hierarchies are reproduced in relation to a variety of criteria, including: a school's location and perceived history; the socio-economic status of students; student attainment and progress; inspection judgments; and the "educational offer" of the school, including how curricular and extra-curricular activities are presented to parents.
Second, status hierarchies also inform parent and student choice.Waslander et al. (2010) argue this relates particularly to how student composition influences status.Choice patterns (expressed as preferences in England) often include middle class strategies to "escape from class 'others'" (Ball, 2013, p. 16).There are also wider class and ethnic solidarities expressed through choice (Burgess et al., 2015).Even when working class and minority ethnic parents are "active choosers" of "quality", Fjellman et al. (2019, p. 522) argue they are less likely to secure places in "good" schools due to residential patterns.Residential patterns, as relating to house prices, are also affected by school quality measures and the composition of potential peers (Gibbons, 2012).These interrelations can make a school's status hard to change and help explain how quasimarkets can work to reproduce "inequality of access to quality provision" (Gewirtz et al., 1995, p. 198).
Third, hierarchical status has been found to influence staff perceptions of competition (Jabbar, 2016).School status does not translate mechanically into perceptions or behaviour.The wider dispositions of staff are influences on their own perceived agency (Gewirtz, 2002).There is a tendency, however, for high-status schools to enjoy greater capacity for competitive action (Higham & Earley, 2013).This stems in part from the likelihood of oversubscription, which supports financial certainty and influence over admissions, while low-status schools are more likely to face undersubscription, student mobility and higher proportions of disadvantaged students (Van Zanten, 2009).For market advocates these pressures create incentives for innovation and improvement (Freidman, 1982).Yet, school quasi-markets have regularly been found to generate new uncertainties that can dampen innovation (Lubienski, 2009), in part because it is risky to be different and safer to mimic practices associated with higher status (DiMaggio and Powell 1983).
Schools still face, however, incentives to secure competitive advantages and this explains, Lubienski (2009) argues, the "'remarkable' rise of marketing" and public relations by schools as a "low risk response" to quasi-markets.Waslander et al (2010) note competition may need to exceed a certain threshold before schools respond to it, but when they do external actions such as marketing, public relations and covert selection are common.Summarising the literature on school competition, Zancajo (2020) identifies five common areas of competitive action: "Market scanning", which is used to assess a school's market positioning through information on parental choices and other schools' actions."Differentiation", used to seek to attract students through new symbolic emphases and, less commonly, changes to internal practices."Marketing", used to promote a school and manage external relationships."Academic improvement", aimed at increasing student outcomes in external tests through pedagogic changes, narrowing the curriculum and/ or teaching to the test."Student selection", with the aim of increasing the proportion of students with higher prior attainment.Zancajo (2020) argues the extent to which schools progress these different actions, as well as the rationales informing their actions, are influenced by schools' local status positionings.

Local competitive arenas and "competition schools"
Notwithstanding these patterns, how positioning in local status hierarchies influences schools' perceptions and actions is also context specific, as quasi-markets are substantially local in character.While state regulations create a framework for choice and competition, parents tend to select from relatively nearby schools with competition ensuing in "local competitive arenas" (Woods et al., 1998).The structural conditions of different local arenas, in terms of the balance between supply of and demand for places, population density and travel distances and school phase can also influence perceptions of competition and school actions (Jabbar, 2015;Zancajo, 2020).How contemporary free schools enter local "competitive arenas" and whether they are perceived to create new competitive pressures is likely therefore to be influenced by existing structural conditions and competitive interrelations locally, school status positionings and perceptions about a free school's aims.This suggests the potential of diverse local experiences.
In America, where a majority of research on new "competition schools" has been undertaken, Epple et al. (2015, p. 52) conclude charter schools have "different competitive effects in different types of environments".A systematic review by Jabbar et al. (2022) found the effects of charters on student attainment in neighbouring schools have been on average small, slightly positive, but only on the borderline of significance, with a diversity of research results reflecting both different state policies and research methodologies.In one of the few studies to research how neighbouring schools perceive competition from charters, Zimmer and Buddin (2009) surveyed c. 200 school principals in California, regardless of distance to a charter school.Over 80% reported charter schools had no competitive effects on their school.Where principals (c.100) reported students in their "local attendance area" were going to a charter school, the survey also asked whether principals made changes to school practices.80% reported no changes (in five areas: teacher pay; teacher recruitment; curriculum; instructional practices; teacher development).Zimmer and Buddin concluded charters did not create strong competitive pressures, noting how in California charters may have acted more as a "release valve" for recent demographic growth rather than creating surplus places.
Theoretically surplus places enable greater choice and competition but creating a surplus through new "competition schools" is often regarded as prohibitively expensive (Wylie, 2006).There are, therefore, several unique aspects of free schools in England that may influence local competitive pressure.First, new schools were previously allowed to open only where there was a forecasted lack of places.Free schools have been able to open where a provider shows evidence of a "demand" for places, potentially creating or intensifying local over-supply of places.Second, while preceding "supply-side" reforms involved existing schools converting to a different organisational form, free schools are entirely new schools.Since 2010, schools have been able to convert to "academy" status, leaving local authority governance to be run privately as independent state schools.Free schools have the same legal status as academies, but they are set up as new schools and by a wider range of potential providers.Free schools have been set up by parents, faith groups, charities, teachers, educational organisations and Multi-Academy Trust (MATs) (Higham, 2014b(Higham, , 2017)), with MATs driving recent growth in the number of free schools (NfER, 2018).With similarities to Charter Management Organisations in the USA, MATs govern chains of schools.MATs have been created predominately by external sponsors (from the private and third sectors) and by high status academy schools.Free schools, like academies, have certain "freedoms" that traditional state schools do not.They 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 represent new aspects of liberalisation, but a range of regulations also remain.First, free school providers are not allowed to make profit (or preach hatred or teach creationism) or charge students fees, so there remains no real market price mechanism.Free schools are funded on the same per-capita basis as other state schools, but receive startup funds, which the state provides to cover the additional costs of establishing a school, and initial capital expenditure to build or reconvert the school's estate.Second, there is no "free entry" into the quasi-market to new providers looking to establish free schools.Any interested person, group or organisation needs to apply as a proposer to central government for the right and funding to open a free school.The government therefore is the sole authoriser of free schools.Third, free schools are held accountable to the same national tests and inspections as other state schools.Fourth, free schools have to comply with the national School Admissions Code, so they cannot select students by prior attainment, exam or interview.Within these restrictions, however, free schools can set their own admissions policies, rather than have them imposed by the local authority.When setting oversubscription policies free schools can prioritise siblings, children of staff, designated feeder schools and specified catchment areas (that do not all have to be adjacent to the school).Free schools with a curriculum specialism can select 10% of students based on "aptitude" for that specialism.Designated faith free schools can select up to 50% of students based on membership of that religious faith (DfE, 2014).

Researching perceptions of competition
To answer the research questions set out above a qualitative research approach was developed.Qualitative research enables insights into events that take place, how these are influenced by local contexts and how events are perceived by participants.There are potential limitations to analysing perceptions as these include subjective judgments that can be informed by conscious and unconscious biases, for example about free schools.This can, in part, be mitigated by including perceptions of different local actors, but it is important to consider potential limitations in light of research aims.In analysis of competition, perceptions are important precisely because, as Levacic (2004:, p. 188) argued, perceptions reveal a "complex set of factors, relating to local relations between schools and the values and behavioural norms on which these are based".Perceptions can be analysed alongside data on local market structural conditions (Zancajo, 2020), but as Levacic (2004) has shown perceptions do not always correlate with measures of structural competition (Epple et al., 2015;Jabbar, 2016).
As such, "competition" was not conceived in this research as a simple cause-effect process and headteachers were not assumed to be solely rational actors working towards their own self-interest.Rather, the research drew on Ball and Maroy's (2009) and Van Zanten's (2009) concept of competitive "logics of action".Here, the rationalities underlying actors' perceptions are understood to often extend beyond costs and benefits to include feelings, values and meanings attributed to competition.Schools are also not conceived as having an explicit "competitive strategy" a researcher can simply read.Rather, an implicit logic is argued to emerge, sometimes haphazardly, as a school interacts with and interprets local structural conditions and relations, status positionings and the wider institutional environment (Van Zanten, 2009).The researcher is understood to be analysing a school's leading orientation towards a local market (Ball & Maroy, 2009).
To analyse neighbouring schools' perceptions of competition and logics of action, semi-structured interviews were undertaken with a sample of headteachers.Headteachers were considered appropriate respondents because they are often expected to analyse "consumer" and "provider" behaviours (Woods et al., 1998) and work to influence their own school's positioning (Jabbar, 2015).
A sample of free schools was initially selected.To avoid issues that may only occur in the first year after opening, the second and third years of a free school were defined as analytical years.On this basis, the first four annual waves of free schools to open were included, producing a population of approximately 175 mainstream free schools.A sample of 15 free schools was purposively selected to reflect the diversity of the population, using three criteria.First, phase of school, with these annual waves made up of approximately 50% primary, 40% secondary and 10% all-through free schools.Second, local deprivation.About a third of free schools in these waves were located in the most disadvantaged quartile of postcodes and about one in ten in the least disadvantaged quartile, as measured by The Income Deprivation Affecting Children Index (IDACI).Third, local need for new school places, as an important structural condition influencing competition.Analysis by the National Audit Office (NAO) (2013) found in the first three annual waves over 80% of primary free schools opened in districts with high or severe forecasted need.Conversely, c. 80% of secondary free schools opened in districts with no or only moderate need for places. 1  Having selected 15 free schools and a reserve set, the neighbouring school sample was identified.Aiming to gain insights into the local competitive arena, headteachers of the nearest two schools to a free school of the same phase were contacted.If one did not respond or was unwilling to participate, the third nearest school was contacted.The achieved sample included 28 neighbouring school heads, representing two of the three closest schools of 14 free schools.(In the case of the fifteenth free school, only one neighbouring headteacher was potentially willing to participate, so this case was not included as the aim was to analyse at least two local perspectives given the likelihood of different experiences and interpretations).Table 1 provides an overview of the research sample.
The interviews were transcribed in full and verbatim and coded by hand using a parallel deductive and inductive approach.Initial theoretical categories were refined through engagement with the data, enabling sub-categories to emerge from the coded data and new categories to be created (Scott & Usher, 1999).After the first cycle of coding, two further interviews were undertaken with two Local Authority officials responsible for place planning in arenas where headteachers reported conflicting accounts to provide further analytical depth.The research was approved by the Institute of Education Research Ethics Committee and adhered to the British Educational Research Association's (2018) Guidelines including by ensuring informed voluntary consent. 2

Findings
In setting out the findings, headteachers' perceptions of competition are considered first, followed by logics of action.Three thematic groupings are presented.Each grouping comprises headteachers who perceived the presence of a free school created similar competitive pressures.This enables analysis of how and why different competitive pressures were perceived.The thematic groupings are not conceived as static.Nor are they homogenous, with differences relating to context and hierarchical status.These differences are discussed further under logics of action and help explain why headteachers perceiving similar competitive pressures describe different logics of action.
In this analysis, a headteacher's perception of their school's status is used to assign schools approximate status positioning.All the headteachers perceived a status hierarchy existed locally and were reasonably clear about their school's positioning.Heads who perceived their school had high status described: "an outstanding school and over-subscribed"; "I've got a product here that's at the top of its game".Schools perceived to have middle status, were described as: "a good, improving school, striving to be popular"; "We're somewhere in the middle".A small number of heads perceived a "lower-middle" status, where: "we're not at the bottom, but not far above it".Schools perceived to have low status were described as: "we're Notes: Free school no.8 is an all-through school and the neighbours are a primary, indicated as (p), and a secondary (s).The population density is not provided for no.8 to help protect confidentiality.All other neighbouring schools have the same phase as the corresponding free school.All neighbouring school names are pseudonyms.FSM is the percentage of students eligible for free school meals at the neighbouring school (rounded to nearest 5%).The Ofsted grade relates to the neighbouring school at the time of the research.Population density reports on the lower super output area in which the neighbouring school is located, using the ONS 2011 rural-urban classification for small area geographies.To help protect the confidentiality of respondents, only the following distinction is reported: conurbation (ONS categories A1 and B1); and 'CT/TF/V' which is City and Town; Town and Fringe; and Village (ONS categories C, D and E).Competitive pressures are those perceived by headteachers, reported in three tiers of intensity, described in detail below.In the case of 2 schools marked with a *, the interviews were unable to take place as intended.The headteachers responded to the research questions in an email conversation with the author.
forever bottom-of-the-heap type of thing"; "we're the poor relation, with a lot of lower-class kids".

Perceived competitive pressure
It doesn't seem to have had much impact, so far The first of the three groupings comprised 13 of the 28 headteachers, 9 primaries and 4 secondaries.The headteachers perceived their respective free school created no or only minor competitive pressures.Their own school lost few if any students and they had been able to refill any vacant places.A number of reasons were given.First, on the structural conditions of each local market, there was evidence of prior rising local birth rates and/or recent immigration and all the heads noted existing or emerging need for new school places.Schools were also insulated from the free school in other ways.For Lipton, the free school was "very small, its catchment doesn't really overlap with ours".For Albion, the free school recruited from a wide area so "took not many students out of each school".For Kirkstone an urban geography meant that, while the free school was "very close, you have the [main] Road, which is the dividing line … It's very rare we take people from across the road".
The second reason for a lack of perceived competition was the status of the free school.Four free schools were perceived to have failed to recruit sufficient students to be full.The heads of Ibstone and Jenner reported a faith ethos free school was not full two years after opening.Ibstone's head, who was from the faith community the free school tried to serve, perceived it was "wrongly located" given the parents it sought: The emerging status of a free school, relative to a neighbouring school, was perceived to clearly influence competition.At Abbots, with historic low status, while losing a handful of students, the head perceived the free school would not affect student composition because of the socio-economic geography: I don't think, with its current situation, it will be taking any of the more high-ability students … even though it's tried to market to say it's not in [this area], it sort of is in the bog end like we are.
They told me, "He wouldn't work at the free school, they would have thrown him out by now" The second grouping comprised 5 primaries and 4 secondaries.All the headteachers reported losing students after the free school opened.For primary schools this was a decline on average of less than a fifth of one class in any year.For secondary schools, the range was wider, up to two classes.Larger declines were often a sudden shock that were temporary, where a free school opened outside local admission arrangements and a neighbour returned from the summer to a smaller than anticipated cohort.More commonly, smaller declines occurred and were sustained and perceived to be, as the head of Gladstone argued, "a direct result of the free school".
For schools with low or middle status, number declines were often accompanied by compositional changes.For schools with middle status this was perceived clearly as a threat to status positioning.Elmswood's headteacher reported an increase in the proportion of students with English as an Additional Language (EAL).The free school, with few EAL students, had been judged "outstanding" by Ofsted but Elmswood's head argued: what influenced parental choice wasn't the data, wasn't the Ofsted, but was the parents on the gate … I was told by a, if I can use the term, largely knowledgeable white middle class mum, that this school is now known as the "EAL school".
Oakham's head reported parents quickly identified "discipline as a priority" at the free school, with higher student exclusions.While Oakham focused on inclusion, the head worried about their lower-middle status school becoming "the needy community school", while the free school became the "academic school".
Reflecting on these perceived pressures, schools offered different reasons for why they were unable to re-fill places after students were lost.Two high status schools (Valeswood and Zeals) were located in areas with existing surplus places.Both expected major pressures to be experienced by a lower status school locally.The other schools were located in districts with officially forecasted need for places, but heads commonly perceived the free school had opened years before an anticipated student bulge or that there would be no bulge.For Clearview's head, this reflected intra-district variations: a need for places, that's the way it's presented, but that's not the actual reality of the situation at all.… My neighbours, they're nearly all down on reception places.… In the south of [this area] there will be a shortage of places.We are in the north.
The free schools were also perceived to be striving for students and status.All were reported to have conducted marketing campaigns with "flashy" brochures, websites and press adverts.Opening in new or refurbished buildings was perceived an "easy sales pitch".The educational ethos free schools presented was also perceived as a competitive pressure, commonly including priorities on "tradition" and an academic curriculum.The head of Clearview described the free school as "marketing themselves as very much 'pseudo public school', it's very much discipline and it's very much uniform".For Zeals's head: free schools can select through curriculum.They can offer a curriculum that excludes children, so they offer an academic curriculum with no resistant materials or food technology … So you know, it's a selection by a different way.
It's "keeping awake at night" stress, it's "is my school going to close?" stress The third grouping comprised 1 primary and 5 secondaries.Each headteacher reported that they had lostor, in the case of two high status schools, would have lostsignificant student numbers.(The cases of the two high status schools, Waddington and Yaxley, are discussed in more detail below in "logics of action".) The four other schools had low or lower-middle status.The head of Uplands reported the largest student loss: "We were averaging 160 pupils [in Year 7] before the free school, now we're down to about 90".The head of Midway reported the school "in effect had half its catchment taken away".Declines in numbers also changedor were perceived to have the potential to changecomposition.Uplands's head argued the free school: "skewed our intake significantly".Averaging a quarter of students in the lowest national prior attainment quintile before the free school, this was now c. 40% in Year 7.These changes were also perceived to threaten the viability of the educational offer.Pitswood's head described how: "at the moment if I walk into a classroom, I might just see 15 kids, which is economically not viable".
This created stress and anger.The head of Midway described how: "It hasn't killed this school.I don't want to over-exaggerate, but in terms of what we can offer and the morale.… I've never been so angry in my life before professionally".The head of Yaxley reported: "It caused me a huge amount of stress.I lost weekends.It put my marriage -I'm happy to say this on the recordmy marriage under strain.It was a very, very difficult time for me".The two high-status school heads, used to stable patterns of recruitment, were also shocked by the attitude of the Government (DfE).The head of Yaxley reported: we're a converter academy.We danced to the DfE tune.… I made that appeal [to the DfE].It was almost laughter.I couldn't believe it.But it was clear to me that free schools trumped converter academies.
One issue underlying this anger was local schools judged as "good" or "outstanding" by Ofsted already had spare places and there was no forecasted need for new places in each "local arena".Structurally there were therefore already increasing competitive pressures, which were rapidly intensified by the presence of the free school.Several free school providers also made negative comments about local schools in press interviews.The free schools' presented ethos were described as being "like a grammar", an "independent school" or a "small, but academically traditional school".For Exwick's head, this was explicitly exclusive: I very clearly see it as a class issue, you know?It's middle class versus working and it was publicly said on more than one occasion by parents of [the free school] that we don't want our children mixing with those children who live [here].

Logics of action
In the context of these perceptions, the analysis now considers the emerging actions and logics of actions headteachers reported were due to the presence of the free school.This is discussed in relation to the three thematic groupings introduced above to help reveal the different "leading orientations" of schools towards their local markets.

Am I loosing students to other schools?
In the first grouping, where few competitive pressures were perceived, eight of the thirteen heads reported taking no actions because of the free school.Kirkstone's head reported: "We have done nothing at present".The orientations of these schools were not entirely passive.Each talked about monitoring the free school's impact.The rationale for this could be mixed.For Kirkstone's head, monitoring was in part a competitive orientation, but also informed work towards sustaining a "family" of primary schools of which the free school was not a member.
Among the five other schools, with middle or lower status, heads described heightened alertness and preparedness to (try to) take "preventative" actions.For Ibstone's head, this meant working to build closer parental relations.There were several logics to these actions.The head described seeking to engage parents in a dialogue about children's learning but was clear that recruitment and retention were also being pursued simultaneously by listening to concerns, learning about "grapevine knowledge" (Ball & Vincent, 1998) and influencing "word of mouth" promotion.The headteacher also thought about how to respond if the school lost more students: If they were lower ability children I'm not going to do anything about it, I'm gonna let them go, its gonna help my standards.We're not stupid as headteachers … If I was starting to lose my middle ability or my higher ability or I was just losing children to the point that it was affecting my budget and I couldn't retain what I was doing here, then I would have to think of preventative strategies.
For Abbots's head, the existing free school created little pressure, but plans by a highstatus school to open another local free school in a new MAT were threatening.The new free school was planned for one of Abbots's few adjacent areas where "students there, usually, are middle-class, higher-ability".The free school was perceived to be "set up to attract those people".Angered by "a sense that everyone should bow to certain schools", the head complained to the headteacher responsible."She told me, 'you know, its parental choice'.I responded 'well', of course it is!But its choice being loaded by you all".

People don't realise, each child is a fair chuck of money
The nine schools in the second grouping had commonly lost a small but sustained number of students.This had funding implications.Four schools were managing funding losses by drawing on reserves and being "very cautious".The other five made cuts.The preferred approach was a range of small cuts.Gladstone cut back staff training, a book fund, intervention classes and educational trips.The rationale was to minimise costs and protect jobs, not least as there was "no neat equation" between staffing and small student declines spread across year groups.
All but one school developed new promotional activities because of the free school.Marketing already occurred, but heads described a sharper competitive focus, including "more professional" websites, adverts, "rebranded" open days and teaching into nurseries or primary schools.These could be "risky investments", with Gladstone making cuts but spending more on marketing.Such actions also reflected different logics.Lower status schools focused marketing on filling places.As Clearview's head argued: "I'm not really interested in us becoming more middle class.We want to be full".For middle status schools, there was a clearer rationale to react to compositional changes.Elmswood's head asked the Local Authority to produce a recruitment map, which showed fewer students attending from streets with higher private rental and home-ownership.The head "wanted to know where I would be targeting my leaflets".Marketing by socio-economic status was clearest at Zeals, a high status school: We've marketed much more closely this year.But we're only competing for certain parents.These are parents who make discerning choices about where their child is going to go.Those are the children who do best at school … so the marketing on that level would be: "We are an unashamedly academic school".
Selective aims did not always lead to marketing.The head of the other high status school in this grouping, Valeswood, saw marketing as "a waste of resources".Believing one school would close locally, the head planned to reduce the main Year 7 entry by one class.The rationale was to safeguard the socio-economically advantaged intake of: "an outstanding school that is no longer full.… If you have 20 spaces and you suddenly have 20 students joining you, that changes the character of a school".Valeswood retained the status of an ex-grammar school.Zeals was a newer school without traditions of privilege.The headteacher there was concerned the free school would be a "drain of the most academic children".In response, Zeals made changes to its curriculum: with different trajectories for different ability children.In short, the most able would move through subjects more quickly and then move on to other subjects.… My particular message has to be "the bright child will do extremely well here".
The majority of heads in this grouping argued they would not make substantive internal change due to the presence of a free school.Oakham's head argued: "I would never let that impact on the children's learning".Schools did take other actions, including new extra-curricular or enrichment activities, such as free music lessons, forest school provision and new after-school clubs.A common logic was to raise the school's profile through an "add-on" that responded to the free school's provision, rather than through "improvement" work.
There was also recognition of holding conflicting logics of action.Turnbrook's head argued: "the free school has not disrupted us from doing the right things", but "is an added thing that we need to think about all the time".Like the majority of schools, Turnbrook was strongly aligned with national performance indicators but, following the free school's opening, additional emphasis was given to preparing students for national tests.Rather than teaching quality, the logic here was focused more on "how your potential parents are going to see it".The school converted to academy status, arguing this helped "branding, making sure that everything looks professional and competitive".Turnbrook's head noted how existing work on literacy and formative assessment was constrained, partly because attainment interventions and academisation took priority.The free school's presence, the head argued, "presents even more issues around, you know, school improvement".

It's going to be very, very vulnerable
The heads in the third grouping all took action in response to the free school's presence, but there were clear differences in logics of action.The two high status schools made substantial changes.Waddington's head described a "major reconfiguring".This included claims to "personalised learning", with a new pastoral system, as a direct response to the free school's smaller size.Vocational qualifications were reduced while a broad academic curriculum was contrasted to the free school's narrower breath.The school converted to academy status to have "some of the powers of free schools … for our own admissions".The head of neighbouring Exwick argued Waddington went over capacity "by about 20 children, despite an agreement" not to.
Yaxley also prioritised academic achievement and reduced vocational provision in response to the free school, which the head argued was "very keen to stress that there were no vocational subjects there".Yaxley launched a new uniform, behaviour policy, "much better" liaison with primary schools and "radio adverts, local press, glossy brochures".The school developed "free transport to people who live beyond our traditional catchment" particularly around a third school.The cost was "significant", but the head argued "this is bringing additional students … ultimately they're worth more than the transport".
Both heads acknowledged their actions intensified pressures elsewhere.A Local Authority interviewee described how "there's that knock on, domino effect, where one [free] school has taken from another [Yaxley], which has then taken from another school in a bit of a food chain".Yaxley's head expressed some uneasiness, but argued further concentrating disadvantaged students into lower status schools was a predictable consequence of the DfE's willingness "to see how many [schools] float and accept that some will sink".Compared to before the free school, both schools remained similar in size and composition.Reporting "a record rise in results", Yaxley's head argued: "I resent any implication that that was because of the free school.… It's been a steady climb up".Both schools had also created a Multi-Academy Trust (MAT) as the lead school, incorporating at least one other school.
The four lower status schools all took action to try to manage the viability of the school by rationalising staffing and the curriculum.Pitswood's head reduced the number of forms in Year 7 by two and was "in the middle of a restructure of 15 teaching posts".Uplands's head made redundancies and curriculum changes.The school's Key Stage 3 curriculum was "skewing much more towards English and Maths now, simply to deal with those issues of [lower] prior attainment".At Key Stage 4 "the number of options we can offer are more limited".Before the free school, "we had 24 options.Then, we had 20.This year we're probably looking at, I don't know, 16".The cuts were particularly to vocational courses.
The heads worked to provide public reassurances to parents.Becoming smaller, each school was compositionally more homogenous and their communication targeted different audiences to the free school and higher status neighbours.Exwick predominately served white working class families and communication was informed by perceptions that choice among these families was "less about results and Ofsted" and more about friendship groups, access for parents and behaviour.This was not without apprehension.Elwick's head noted "the dangers of being branded the poor relation locally" and felt "huge pressure to be drawn into the marketing game", but concluded "there's no point in joining a game you can't win".
A consistent theme was vulnerability, to intervention, to unviability and in the case of Uplands to potential closure.Exwick's head described how, having made "a long trek" toward being judged "good" by Ofsted, the school was now "probably the weakest link in this whole system".Pitswood's head reflected on her own vulnerability as a new head appointed to a school judged "requires improvement" that then had a free school open nearby.She perceived higher status neighbours were also aggressively protecting themselves by expanding: they're at a far, far different stage of development than we are, they're opening up free schoolsit's like a candy store.… Maybe I felt threatened, but there is certainly, there's a changing climate, it's a nakedly competitive market.

Discussion
This section now considers the findings in relation to each research question, enabling an analysis of the relations between context, competitive pressure, school status and logics of action.The section concludes by discussing the emerging consequences for schools and local students and related policy implications.
The first question asked: in what contexts do neighbouring headteachers perceive the presence of a free school creates new competitive pressures?The findings showed, firstly, how competitive pressures were understood.Headteachers commonly saw competition to relate to the following perceived threats of change at their school: A loss of students, either as direct attrition from the existing student body or, more commonly, through reductions to the size of the school's main year of intake.Associated loss of per capita funding, and the extent to which this affected staffing, the resourcing of classrooms and the maintenance of the school estate.Compositional changes to the student body, with particular sensitivity to loss of students with higher prior attainment and students from wealthier and/or more aspirational families, relative to the school's existing composition.Loss of popularity, as expressed in preferences for the school and as conveyed in "grapevine knowledge" by parents about who the school was seen to serve.
The extent to which these competitive pressures were perceived by headteachers was shown to be influenced by two sets of contextual factors.The first related to local market conditions: with the following structural conditions seen to moderate the intensity of competition: Place provision, which related to whether the free school was responding to a real need for new places locally or was seen to be creating or extending a surplus of places, with the latter associated with more intensive pressure.Local geographies of recruitment, relating to whether the free school was located in an important recruitment area of a neighbouring school, which was associated with more intensive pressure, or whether local residential and travel to school patterns meant the free school was proximate but orientated towards alternate neighbourhoods.Prior competitive relations, with stronger pressures perceived where relatively more stable patterns of local competition were disrupted by a combination of increasing structural competition and the presence of the free school, so relations between schools became less predictable and prior informal agreements over student recruitment or negotiated solutions over hard-toplace students were less likely to be honoured.
The second contextual influence on the perceived intensity of competitive pressure related to the (emerging) status positioning of the free school.Status was seen clearly to have the potential to influence patterns of choice and competition.Headteachers identified the following aspects of a free school's status as being important to competitive pressure: The free school's ethos, curriculum and extra-curricular provision, and the extent to which these were seen to appeal to socio-economically advantaged families.The free school's marketing and branding, including signals about who the school sought to serve and the extent to which local schools were cast by the free school as being unsatisfactory.The free school's quality, with published external measures limited at the time of the research to an inspection by Ofsted.The free school's popularity, and whether it had admitted sufficient students to be full and oversubscribed.The free school's student composition and the extent to which it had recruited students with high prior attainment and students from wealthier and/or more aspirational families, relative to the local area.
The local contexts in which headteachers perceived new competitive pressures due to the presence of a free school shared therefore several characteristics.The free school achieved at least a middle status position, with high status associated with more intensive pressures.The free school was located in an important recruitment area of one or more neighbouring schools.There was no or little need for new places locally.The free school disrupted relatively stable patterns of recruitment and relations became more competitive.Two further characteristics can also be noted, with caveats.First, headteachers in major conurbations were less likely to perceive competitive pressures compared to regional cities, towns and rural contexts.Higher population density and overlapping recruitment areas in conurbations may have diluted a free school's perceived impact.Conurbations, however, were also more likely to experience local need for new places at the time of the research, so schools were on average more likely to be able to fill places.Second, headteachers of secondary schools were more likely to perceive competition than primaries.Secondary schools may engage in more competitive action, not least as students travel further to school.It was also true however that secondaries experienced less local need for new places at the time of the research, also increasing the potential of competition.Analysing the co-influences of urbanicity, supply of places and school phase is a potential area of future research.
The second question asked: are schools influenced to take new actions where they perceive competitive pressures and, where they say they do, what logics inform these actions?The findings showed, firstly, the types of actions headteachers reported taking due to the presence of a free school.There were clear parallels to existing literature on school competition, from which Zancajo (2020) summarised five areas of action as: "market scanning"; "differentiation"; "marketing"; "academic improvement"; and "student selection".Each of these action types were evidenced in the findings, but two additional areas of action were also identified."Structural change" occurred where headteachers sought to influence school status and resourcing by using governance reform to convert to academy status and/or to create a Multi-Academy Trust."Financial rationalisation" occurred where headteachers made changes in response to viability pressures.This ranged from making a number of small cuts to major restructurings of staffing and the curriculum.
The extent to which these actions were reported by individual headteachers was shown to be influenced by two sets of contextual factors.First, the perceived intensity of competition influenced responsive action, so greater pressure was associated with more substantive action.This was not, however, a simple linear relationship.Waslander et al (2010, p. 55) argue relations between pressure and action are non-linear, in that competition needs to "exceed a threshold before schools are likely to respond in any way".The findings here support this concept but suggest, rather than "a threshold", there can be a series of thresholds to action.In this research, the perceived loss of small numbers of students was one threshold, relating predominately to outward-facing and symbolic actions.Perceived threats to the financial viability of a school's educational offer constituted a second threshold, with actions extending to substantive curriculum, pastoral and staffing changes.
The second contextual influence on action and logics of action was the perceived status of the neighbouring school itself.Headteachers' perceptions of their own school's status closely informed their leading orientation to competition.While marketing was commonplace, for example, headteachers of low status schools tended to describe a logic of trying to recruit sufficient students to be full, while headteachers of high-status schools described logics of socially selective recruitment.Schools perceiving intensive competition often made curriculum changes but, while headteachers of low status schools described logics of rationalisation, the logics of high-status schools concerned competitive advantage.These different logics were widely perceived by headteachers to reflect the unequal material and symbolic resources afforded by a school's relative status.This was not a simple deterministic relationship, however, in which headteachers denied their own agency.Rather, as Jabbar (2016, p. 400) argues, school leaders could be understood to "exert agency and take … action to compete", but these actions were "constrained or enabled" by their school's status.
Seeking to conceptualise how actions are "constrained and enabled" by status, previous research has identified "ideal types" of logics of action.Van Zanten (2009, p. 92) argues constructing ideal types helps reveal how "different, scattered elements are linked to each other and give rise to distinct configurations", increasing "the possibility of generalising interpretations to other local contexts".Researching schools in European cities, Van Zanten presents four ideal types, relating to two variables: local market conditions (stable/unstable); and school status (high to intermediate/intermediate to low).In stable markets, with less competitive pressure and where recruitment patterns were relatively settled, higher status schools tended to a "monopolistic" ideal type, relying on reputation to remain socially selective.Lower status schools tended to be "adaptive", turning inwards and trying to adjust to the students they enrolled.In unstable markets, with stronger competition and the possibility of change in recruitment patterns, higher status schools tended to be "entrepreneurial", using promotional strategies and curriculum specialisms to influence selective recruitment.Lower status schools adopted "tactical" orientations, which were more defensive, but still sought to retain students with a "good attitude to learning".
These ideal types set out by Van Zanten find support in this research but are also extended by it.Two extensions can be identified.The first concerns the wider range of perceived competition reported, particularly intensive pressures due to the presence of a free school.This points to extending the typology of local market contexts.The second extension concerns differences between middle and low/lower middle status schools.Van Zanten used the term "low-intermediate", noting their sample contained few schools with "bad reputations" on the assumption "these schools tend to withdraw from competition" (p.87).The findings here question the idea of withdrawal and identify distinctive logics between middle and lower status schools.Incorporating these two extensions leads to an expanded set of ideal types, set out in Table 2. Eight ideal types are supported by the data (with no case in the sample relating to a "major pressures/ middle status" ideal type).
As Table 2 sets out, Van Zanten's four ideal types related most closely in this research to where moderate competitive pressures were perceived due to the presence of a free school.In these relatively unstable contexts, high status schools tended towards "entrepreneurial" logics, although there were also "active monopolistic" practices to protect an advantaged intake by reducing the student body's size.Middle status schools tended to be "tactical", finding spaces for action to defend their student composition.The distinction was lower status schools, which were best characterised by an "adaptive" logic of action.Threatened by a free school and the actions of high-status neighbour schools, headteachers tended towards adapting, for example, to becoming "the needy community school".By comparison, in more stable, less competitive contexts, where the free school had not disrupted patterns of recruitment, different logics were commonly apparent.High status schools were indeed inactively "monopolistic", displaying disinterest and a sense of being "above the fray" (Ladd & Fisk, 2003, p. 104).Middle status schools tended to be "vigilant", taking few actions, but actively scanning and alert to the fact free schools could initially struggle but then attract students and change composition.Low status schools tended to be "pre-emptive" (rather than "adaptive"), intensifying parental involvement and community engagement to deter change.Headteachers of low status schools had slightly more room for manoeuvre here, due in part to the disinterest of high-status schools and because the free school was typically of lower status and more directly comparable to their own school.
In contexts where the free school's presence created major pressures, relations were highly unstable and recruitment patterns dynamic.High status schools made curriculum and pastoral reforms indicative of social advantage, removed lower status options and inclusive practices and marketed aggressively with brand-awareness activities among young children and marketing directors on pay-roll.These investments combined with structural change, using the multi-academy trust as the "tool to hand" to seek competitive advantage through increased size, whilst avoiding compositional change at the socially selective high-status school.The characteristic logic of action was "selective expansion".By contrast, low status schools were the most constrained but also intensively active.There was little space to be "tactical", but they did not want to withdraw given the pressures perceived.Headteachers felt compelled to rationalise staffing and curriculum options, whilst reframing who their school served and how to communicate with those communities.The dominant logic was "survivalist", doing what they perceived had to be done to sustain their school's viability.
These different actions and logics were perceived by headteachers to have a range of implications for schools and students locally, including in relation to the wider policy aims of diversity, choice and improvement.In contexts where relations between schools remained relatively stable, free schools were seen variously as small, isolated schools, schools that had failed to recruit sufficient students to be considered popular but also as schools that were reverse creaming to serve disadvantaged students or offering valuable provision not deemed to be of high status.With a need for new places often reported in these contexts, there were parallels to Zimmer and Buddin's (2009) findings that Californian charter schools acted more as a "release valve" for demographic growth rather than as threatening "competition schools".
In contexts that became more unstable after a free school opened, clearer actions were reported.Additional resources were diverted to marketing and PR and extra-curricular activities.Marketing was seen to offer a "low risk option" in response to the uncertainty of "shifting enrolments, unstable budgets" (Lubienski, 2005, p. 479), but headteachers also recognised the tensions involved, particularly when funding cuts were simultaneously made to core provision.The combination of marketing and extra-curricular "add-ons" was widely understood to concern signs and signals about socio-economic (dis)advantage, increasing the potential of selective competition (Gewirtz et al., 1995).The uncertainty created by increased competition can also dampen innovation in core practices (Lubienski, 2009) and there was little evidence of teaching and learning innovations due to the presence of a free school.Rather, additional emphasis was commonly given to preparing for tests and inspections.This was closer competitively-induced compliance to external performance indicators, sometimes expressed in "gaming" (Ingram et al., 2018) or as "quick wins".Such work can potentially lead to small improvements in test scores (Cullen and Reback 2006) but was not indicative of the professional reflection and student engagement in learning argued to support "authentic" development of classroom practices (Fielding et al., 2005Wrigley, 2003).
In contexts where relations became highly unstable and dynamic after a free school opened there were additional implications for inequalities.High status, socio-economically advantaged schools displayed capacities to respond in terms of both improvement activities and socially selective actions.By contrast, it was possible to identify at least the beginnings of "a spiral of decline" among several lower status schools as they struggled with significant funding pressures in the context of a surplus of places.This was a major disruption to school life but was yet to result "ultimately in closure" (Hatcher, 2011, p. 500).Rather, heads perceived their schools would become smaller and compete with a free school that was in several cases also smaller than planned.There would be increased, duplicated capital expenditure and longer-term recurrent budget constraints negatively affecting educational provision.
There were also consequences for choice and diversity.Free schools in these highly unstable contexts were often perceived, as the head of Zeals argued, to be educating "a small number of children, selected out of the system", so that "increased choice" was seen primarily to offer opportunities for a local minority to opt for a more exclusive education.Free schools here were also seen to prioritise traditional values and an academic curriculum, so that they combined aspects of "the new and flashy", in buildings, IT and branding, with strong claims to "traditional standards" in teaching, curriculum, behaviour and uniform (Morris & Perry 2019).Neighbouring schools often responded by narrowing their own curriculum, particularly vocational and technical education (VTE) in secondary schools, despite such provision being previously seen as beneficial, especially for students at risk of disengagement.The logics of these actions varied by status, but influences included: the perceived status risks of a diverse, inclusive curriculum where a free school stressed a narrow, academic offer; new national accountability measures prioritising academic subjects; and, for secondaries, the higher costs of VTE.The combination of these patterns of curriculum standardisation and selective competition meant socio-economically disadvantaged students were seen to be further concentrated in lower-status schools rationalising curriculum options and staffing.
These locally diverse patterns of perceived free school effects on neighbouring schools were seen by headteachers to have policy implications.Firstly, there was little support for simplistic policy claims that free schools are creating a "systemic effect" for improvement in neighbouring schools (Allen & Burgess, 2010, p. 1).Not least, just under half the sample of headteachers perceived few if any competitive pressures and were unlikely to have taken new actions due to the presence of the free school.Secondly, where intensive competition was perceived, a deepening of socio-economic inequalities was occurring.This was seen to present a strong argument against opening free schools in contexts of surplus places or of authorising free schools that progress socially selective aims (Allen & Higham, 2018).
Recent policy changes have claimed to prioritise free schools where there is a need for new places and low average school quality (DfE, 2021).Actual local experiences will need further empirical analysis, but this claimed policy evolution has been argued to somewhat dull the original ideological rhetoric of free schools as "competition schools" (Julius et al., 2021).To date, however, there has been insufficient policy consideration given to the consequences of opening free schools for disadvantaged students.Government impact assessments have "tolerated" threats to the viability of schools, arguing this is outweighed by potential increases in choice and improvement.This assumes a level playing field between schools, when this analysis has shown choice and competition are strongly influenced by unequal school status positionings which incentivise selective competition.The consequence has been a concentrating of the negative effects of opening free schools, where these exist, in lower status schools serving disproportionally more disadvantaged students.

Table 1 :
Neighbouring schools research sample

Table 2 .
Logics of action.