A review of the evidence on declining pupil numbers in the English state school system

September 2026

Simon Uttley

Director, Koin-Ed.org

Abstract

Falling school rolls have become one of the defining structural problems of the English state school system. After a decade and a half in which the central task of school place planning was to accommodate growth, primary rolls peaked in 2019 and have declined since, and secondary rolls have now begun to follow. This paper reviews the available evidence on the causes, scale, consequences and possible remedies of the phenomenon. It draws on official statistics, National Audit Office and parliamentary scrutiny, think-tank analysis and the international peer-reviewed literature on school closure and consolidation. Three conclusions emerge. First, the root cause is demographic rather than educational: a sustained fall in fertility, compounded in particular localities by internal migration driven by housing costs and by a sharp reduction in net international migration, has removed children from the system faster than official projections anticipated. Second, the scale of the adjustment is larger and arriving sooner than the sector has planned for, with the Department for Education now projecting 726,000 fewer pupils by 2031 than were on roll in January 2026. Third, the harm arises less from the demographic change itself than from the way the system is currently configured to absorb it. Per-pupil funding transmits falling numbers directly into falling budgets, while school costs are substantially fixed; the mechanisms intended to cushion this transmission are too small and too narrowly drawn to do so; and responsibility for adjustment is dispersed across local authorities, academy trusts and individual schools without a national strategy or an adequate evidence base to guide it. The consequence is that decisions with lasting educational effects, including staffing reductions, curriculum narrowing, the adoption of mixed-age classes and ultimately closure, are being taken for financial reasons and often too late. The paper argues for reform of the funding architecture, a published national strategy supported by commissioned research, strengthened local place-planning powers, systematic reuse of surplus estate, and an explicit protocol for managing closures where they prove unavoidable. It further argues that the primary experience should be treated as a rehearsal for the secondary phase, where the same adjustment will shortly be required at greater cost.

Keywords: falling rolls; pupil projections; school funding; school closure; place planning; educational disadvantage; England

1. Introduction

For most of the past two decades the central problem of school place planning in England has been shortage. Between 2009/10 and 2024/25 the number of pupils in state-funded primary and secondary schools rose by 844,000, or twelve per cent, and the sector created a net 1.3 million additional school places to accommodate them (National Audit Office [NAO], 2026). Capital programmes, basic need allocations and the free schools programme were all configured around expansion, and the statutory framework within which local authorities operate, in place since 1944, obliges them to secure sufficient places rather than to manage a surplus of them.

That period has ended. Primary rolls in England peaked in 2019 and have fallen every year since. Secondary rolls, buoyed for a decade by the passage of the 2008 to 2012 birth cohorts through the system, plateaued between 2024 and 2025 and have now begun to decline. The Department for Education’s 2026 projections showed a state-funded school population of 7,809,000 in January 2026 falling to 7,082,000 by 2031, a reduction of 726,000 pupils or 9.3 per cent in five years (Department for Education [DfE], 2026c). This is not a temporary trough of the kind the system has absorbed before. It reflects a sustained decline in the number of children being born in England and Wales, which reached its lowest level in almost half a century in 2025 (Office for National Statistics [ONS], 2026). As a school leader with over twenty years’ experience running maintained secondary schools in England, I have seen and lived through the experience of managing this challenge.

The policy significance of this demographic change lies in the design of the system that must absorb it. Approximately 90 per cent of school revenue funding is allocated according to pupil numbers, so a fall in rolls translates almost directly into a fall in income (NAO, 2026). School costs behave quite differently. Around three-quarters of expenditure is staffing, and a school that loses a handful of pupils from each year group loses income without being able to remove a corresponding fraction of a teacher, a caretaker or a boiler. The result is a slow compression of budgets that is felt first in the discretionary margins of school life and eventually in the viability of institutions themselves.

Scrutiny of the issue has intensified. In April 2026 the NAO reported that, despite primary numbers having fallen since 2018/19, the Department for Education had provided the sector with limited support to manage the resulting risks to educational quality and value for money. As this paper is written, in September 2026 the House of Lords Public Services Committee, concluding a short inquiry, wrote to ministers to say that the Government had failed to grasp what was required and was treating a demographic and place-based challenge as though it were narrowly an education one (House of Lords Public Services Committee, 2026). Both bodies identified the same gap: an absence of national strategy, guidance and evidence to inform decisions that are being taken locally under financial duress.

This paper sets out to answer four questions. What is the root of the problem? What is its scale? What is its impact on schools, pupils and communities? And what can and should be done about it? It proceeds by examining the demographic and socio-economic drivers of declining rolls, quantifying the extent and distribution of the decline, analysing the financial and educational transmission mechanisms through which it affects schools, and evaluating the options available to policymakers.

2. Approach and scope

The paper is a desk-based synthesis of published evidence. Its principal quantitative sources are the Department for Education’s national pupil projections and school capacity data, the Office for National Statistics’ birth and fertility statistics, and the NAO’s 2026 value-for-money examination of the school place system, which incorporates a survey of 76 of the 153 English local authorities responsible for place planning. Qualitative and sectoral evidence is drawn from the written and oral evidence submitted to the House of Lords Public Services Committee’s 2026 inquiry into falling primary school rolls, and from analyses published by the Education Policy Institute, the National Foundation for Educational Research, the Institute for Fiscal Studies, London Councils and the Nuffield Foundation. Causal claims about the effects of school closure and consolidation on pupils rely on the international peer-reviewed literature, chiefly from the United States and Northern Europe, since the English evidence base is at present limited.

The scope is England, and within England the emphasis falls on mainstream state-funded primary schools, which have experienced the decline first and most acutely. Where the argument extends to the secondary phase this is made explicit. Independent schools, further education and early years provision fall outside the analysis except where they bear directly on the state school population. Comparisons with Wales, Scotland and Northern Ireland are used illustratively rather than systematically, since school organisation is a devolved matter and the institutional arrangements differ materially.

Two limitations should be registered at the outset. The first is that pupil projections are estimates conditioned on assumptions about fertility and migration that have repeatedly proved optimistic, so the figures presented here should be read as a central case rather than a forecast. The second is that the causal evidence on how falling rolls affect children is sparse. Much of what follows about educational impact rests on plausible mechanisms, practitioner testimony and analogous international findings rather than on direct English estimates. This evidential weakness is itself one of the paper’s findings.

3. The root of the problem

Falling rolls in England are commonly attributed to the falling birth rate, and this is correct as far as it goes. It does not go far enough. The national decline is demographic, but its local distribution, which is what determines whether any given school becomes unviable, is shaped by migration, housing costs, the historic pattern of capacity creation and the structure of the school system itself. Four drivers can usefully be distinguished.

3.1 Fertility decline

The fundamental cause is a sustained fall in fertility. The total fertility rate for England and Wales has been below the replacement level of approximately 2.1 children per woman since 1973, but the decline has accelerated markedly since 2012. Provisional figures put the rate at 1.39 in 2025, down from 1.41 in 2024 and 1.42 in 2023, each of which was in turn a record low (ONS, 2025, 2026). There were 585,396 live births in England and Wales in 2025, a fall from 594,677 the previous year and the lowest annual figure since 1977 (ONS, 2026).

Two features of this decline matter for schools. The first is its persistence. The birth rate rose for more than a decade to a peak in 2012 and has fallen almost continuously since, which means that the cohorts now entering reception are structurally smaller than those that preceded them and that this will remain true for the foreseeable future (DfE, 2026c). The second is that official projections have repeatedly assumed a stabilisation that has not occurred. Sibieta (2025) observes that successive Office for National Statistics forecasts over the past decade have anticipated fertility levelling off, and that it has instead continued to fall, so the realised decline in child numbers has consistently exceeded the projected one. The Department for Education has responded by shortening its projection horizon, first from ten years to eight and then to five, an implicit acknowledgement that the uncertainty has grown rather than diminished (NAO, 2026).

Behind the aggregate rate lies a shift in the timing of childbearing. The standardised mean age of mothers in England and Wales reached 31.1 years in 2025 and of fathers 34.0 years, compared with 26.4 and 29.5 respectively in 1975 (ONS, 2026). Postponement compresses the years available for childbearing and reduces completed family size; it also makes the short-run relationship between the adult population and the number of births weaker and harder to project. The most pronounced reductions in age-specific fertility have been among women in their late twenties (ONS, 2025), which is the group whose behaviour most strongly determines reception intakes five years hence.

3.2 Internal migration, housing costs and the cost of living

Fertility explains the national trajectory but not the geography. The Education Policy Institute’s analysis of London primary pupil movements found that the capital’s decline could not be accounted for by births alone. Around 17 per cent of pupils who started reception in London in 2012/13 had left the city by Year 6; for the 2017/18 cohort the figure rose to 20 per cent, with the East of England and the South East absorbing the largest inflows (Andrews & Wielar, 2025). The authors attribute this to the cost of living, housing pressures and regional differences in provision, which together shape where young families are able to settle.

Evidence to the Public Services Committee described the same mechanism operating beyond London. Local authorities and school leaders reported families relocating from expensive urban areas to more affordable rural and coastal ones, producing population change in both origin and destination (House of Lords Public Services Committee, 2026). The counties bordering London have seen rolls hold up or rise even as inner London boroughs have emptied. This makes the problem asymmetric: the areas losing pupils lose funding immediately, while the areas gaining them must create capacity, and the funding lag means that receiving schools carry the cost of additional pupils before the money arrives.

A related complication is seasonal and in-year mobility, which is particularly acute in coastal areas where families move during the academic year. Because per-pupil funding is driven by the October census, a school that admits a child in January may not be funded for that child until the following financial year, and the school the child has left retains the funding in the interim. The mobility factor within the national funding formula is intended to address this, but witnesses to the committee considered its effect limited (House of Lords Public Services Committee, 2026).

3.3 International migration

The most significant recent development, and one that post-dates much of the published analysis on falling rolls, is the contraction of net international migration. The Department for Education’s 2026 projections note that provisional estimates for the year ending December 2025 showed long-term net migration down by almost half compared with the previous year, following a similar halving in the year to June 2024, and identify this as a likely contributor to the fall recorded in the January 2026 school census (DfE, 2026c). Migration affects the school population through two channels: directly, through the arrival of school-age children, which has historically been a modest effect; and indirectly and more powerfully, through the number of women of childbearing age resident in the country, since fertility rates among women born outside the United Kingdom have been higher than among those born within it. In 2025, 40.2 per cent of live births in England and Wales had at least one parent born outside the United Kingdom, and 34.6 per cent had a mother born outside it (ONS, 2026).

The consequence is that a policy variable largely external to education has become a principal determinant of future school rolls. The Office for National Statistics’ mid-2024-based population projections, which underpin the Department’s current pupil projections, assume both lower long-term migration and lower future fertility than their 2022-based predecessors (DfE, 2026c). This is the main reason the projected decline has deepened between successive publications, and it introduces a source of volatility that local place planners cannot reasonably be expected to anticipate.

3.4 The legacy of the expansion decade

Falling rolls are a function of supply as well as demand. The capacity created during the expansion years did not disappear when the cohorts shrank, and in some places, it exceeded what was needed even then. The NAO reported that the Department had estimated in 2017 that just over half of the places created by free schools opening between 2015 and 2021 would generate spare local capacity, and that free schools could be approved on grounds of quality or parental choice in areas where places were already sufficient (NAO, 2026). Between January 2010 and July 2025, 625 free schools opened and 51 closed. The Department has since tightened its criteria, confirming in March 2026 that it would cancel 12 planned mainstream free schools and was minded to cancel a further 13, with 19 proceeding.

The mismatch is visible in reception admission numbers. Reception places peaked in 2019 at 711,000, having risen from 662,500 in 2013; between 2017 and 2025 they fell by about two per cent, from 707,700 to 696,900. Over the same period the number of reception pupils fell by roughly twelve per cent, from 660,500 to 580,500 (NAO, 2026). Schools have reduced their published admission numbers, but considerably more slowly than demand has fallen, and every place that remains published but unfilled is a place that generates no income while continuing to carry a share of the school’s fixed costs.

3.5 Structural fragmentation and the quasi-market

The final driver is institutional. Responsibility for the number of school places in England is divided. Local authorities hold the statutory duty to secure sufficient places for children in their area, but 52 per cent of mainstream schools are academies whose admission numbers are set by their trusts, and until recently local authorities could not require an academy to reduce its intake (NAO, 2026). Of the 76 local authorities responding to the NAO’s survey, 88 per cent identified their statutory powers as one of the three main challenges they faced, and 97 per cent said they would recommend that those powers be increased.

This fragmentation interacts with parental choice to concentrate the decline. When the overall pupil population falls, parents have more choice, and they exercise it in favour of schools with better inspection outcomes. In 2024/25, schools rated inadequate had 26 per cent of their places unfilled and those rated requires improvement 22 per cent, compared with 13 per cent in good schools and 6 per cent in outstanding ones (NAO, 2026). Falling rolls therefore do not fall evenly; they hit schools that were already struggling hardest. Witnesses to the Public Services Committee described the resulting dynamic as self-reinforcing: fewer pupils produce smaller budgets, smaller budgets produce staffing losses, staffing losses depress quality and reputation, and reputation drives further falls in admissions (House of Lords Public Services Committee, 2026). Sibieta (2025) similarly notes evidence of larger falls in maintained schools than in academies and in lower-rated schools than in higher-rated ones.

4. The scale of the problem

4.1 The national picture

Between 2018/19 and 2024/25 national demand for primary school places fell by three per cent (NAO, 2026). At the time the NAO reported, the Department’s 2025 projections indicated a further seven per cent fall in primary numbers between 2025 and 2030, equivalent to about 300,000 pupils, before the Public Services Committee’s inquiry.

The 2026 projections, published in July 2026, revise that picture substantially downwards. The nursery and primary population is now projected to fall from 4,424,000 in 2026 to 3,951,000 in 2031, a reduction of 473,000 pupils or 10.7 per cent. The revised 2030 figure of 4,021,000 represents a 10.7 per cent fall from 2025, where the previous model had projected 6.7 per cent, and is 184,000 pupils lower than the earlier estimate for the same year. Across all state-funded schools, the 2026 model projects 7,217,000 pupils in 2030 against the 2025 model’s 7,504,000, a downgrade of 286,000 in the space of a single year (DfE, 2026c). The January 2026 school census had already recorded 112,000 fewer pupils, a fall of 1.2 per cent on the previous year, across all school types.

Table 1 sets out the current projections. Two features are worth noting. The first is the size of the annual reductions in the primary phase, exceeding 100,000 pupils in each of 2027 and 2028. The second is that special school numbers move in the opposite direction, having risen 4.8 per cent to 178,000 in the 2026 census, so the composition of the falling population is shifting towards higher need even as its total shrinks.

Table 1

Actual (2026) and projected pupil numbers by school type, England (thousands)

School type	2026	2027	2028	2029	2030	2031
State-funded nursery and primary	4,424	4,321	4,211	4,111	4,021	3,951
State-funded secondary	3,216	3,179	3,140	3,085	3,029	2,967
State-funded special	153	155	154	153	150	148
Alternative provision	16	17	17	16	16	16
Total state-funded schools	7,809	7,671	7,522	7,365	7,217	7,082

Note. Figures are full-time equivalent and cover ages up to 15 unless otherwise specified. Adapted from Department for Education (2026c).

4.2 Unfilled places

The most direct measure of the mismatch between supply and demand is the proportion of places left unfilled. Between 2009/10 and 2018/19 this figure was stable, ranging between nine and eleven per cent. It has risen every year since, reaching 14 per cent of all primary places in 2024/25 (NAO, 2026). The distribution is heavily skewed: 37 per cent of primary schools had 15 per cent or more of their places unfilled, while 15 per cent had between five and ten per cent unfilled.

There is no agreed benchmark for what level of surplus is appropriate. A 2007 target of six per cent no longer applies, and the Department’s funding allocations assume an operating margin of two per cent to allow for parental choice and in-year movement. Local authorities interviewed by the NAO variously regarded five per cent, five to ten per cent, or no fixed figure as sensible. The absence of a shared parameter matters because it removes any trigger for intervention: without a view on how much surplus is too much, neither the Department nor local authorities have a defensible point at which to act.

4.3 Geographical variation

The national averages conceal very wide local variation, and it is at local level that viability is determined. Between 2018/19 and 2024/25, 84 per cent of local authorities saw primary numbers fall, but 16 per cent saw them rise (NAO, 2026). Looking forward, 136 local authorities, or 89 per cent, forecast a decline between 2024/25 and 2029/30. Critically, 66 per cent of those forecasting an overall decline also expect increases in at least one of their smaller planning areas, and 28 per cent of planning areas nationally forecast growth. Lincolnshire, for example, forecasts a three per cent overall fall while expecting rising numbers in 44 per cent of its 72 planning areas. Local authorities must therefore create capacity in some places while removing it in others, often within the same authority and the same budget.

London has been the epicentre. Andrews and Wielar (2025) found that nine of the ten local authorities with the largest falls in primary numbers over the preceding five years were London boroughs, with Westminster recording a decline of 15.9 per cent between 2020/21 and 2024/25; Redcar and Cleveland was the only authority outside the capital in the top ten. Islington, Lambeth and Southwark are expected to see falls of between 14 and 20 per cent over the following five years. Sibieta (2025) records a nine per cent fall in London primary numbers between 2016 and 2025 against one per cent for England as a whole, with most inner London boroughs down by between ten and twenty per cent.

London is no longer the only affected region. London Councils (2026) reports that, for the first time, forecast declines in Year 7 places across the capital exceed those in reception, at 3.8 per cent against 2.5 per cent between 2025/26 and 2029/30, rising to 7.6 and 6.4 per cent respectively in inner London. Evidence to the Public Services Committee identified pressure in Brighton and Hove, parts of the North East, Lincolnshire and Northumberland, and a survey by the National Governance Association found that more than a quarter of primary schools cited falling rolls among their principal concerns, with concentrations in London and the South West (House of Lords Public Services Committee, 2026).

4.4 The approaching secondary wave

The large birth cohorts of 2008 to 2012 have now passed through the primary phase and are working their way out of secondary schools. The Department’s 2026 projections show the secondary population, having plateaued between 2024 and 2025, falling from 3,216,000 in 2026 to 2,967,000 in 2031, a reduction of 249,000. The revised 2030 figure represents a 6.3 per cent fall from 2025 where the previous model had projected 3 per cent (DfE, 2026c).

This matters for two reasons. First, secondary schools have higher fixed costs and more specialised staffing, so the loss of a given proportion of pupils threatens curriculum breadth more directly: a school that can no longer sustain a physics teacher or a modern languages department cannot easily replace that provision. Second, and more importantly for policy, the primary experience of the past seven years constitutes a natural experiment whose lessons have not been systematically captured. The Public Services Committee’s closing recommendation was that the Government should use those lessons to plan for the secondary phase, and it asked ministers to state what work was under way and where the risk sat on the departmental risk register (House of Lords Public Services Committee, 2026). The Department had added the risk of the system failing to respond adequately to demographic change to its register only in 2024, having concluded that its existing arrangements for managing demographic risk were unsatisfactory (NAO, 2026).

5. The impact on schools, pupils and communities

5.1 The funding mechanics of decline

The mechanism by which falling rolls harm schools is straightforward. Around 90 per cent of primary school funding is allocated on a per-pupil basis, rising to 93.5 per cent in 2025-26, and the schools block of the Dedicated Schools Grant was approximately 92.4 per cent pupil-led in 2026-27 (DfE, 2025; NAO, 2026). Income therefore falls with numbers. Costs do not. In 2023-24, 76 per cent of school expenditure was staffing, 14 per cent premises and occupation, and 11 per cent supplies and services (NAO, 2026). Utilities, administration, estates maintenance and leadership costs are largely invariant to marginal changes in roll, and teaching costs can only be reduced by removing whole classes. Steve Rollett of the Confederation of School Trusts put the arithmetic to the Public Services Committee in terms that have been widely quoted: a school losing five pupils loses several thousand pounds, but “you cannot cut a teacher in half” (House of Lords Public Services Committee, 2026).

The NAO quantified the aggregate effect. Applying the 2026-27 minimum per-pupil funding rate of £5,115 to the Department’s projected reductions, it estimated that primary schools would receive £288 million less in 2027 on the basis of 56,300 fewer pupils, a further £410 million less in 2028 on 80,200 fewer, and £334 million less in 2029 on 65,300 fewer (NAO, 2026). These are illustrative figures rather than forecasts of budget reductions, since they take no account of wider funding decisions, including the £2 billion real-terms increase in the core schools budget between 2023-24 and 2028-29 announced at the 2025 Spending Review. They nonetheless indicate the order of adjustment required. Since the Treasury allocates the core schools budget with reference to pupil numbers, a smaller cohort does not automatically leave the Department with a surplus to redistribute.

At institutional level the effect can be abrupt. The NAO’s modelling of a representative primary school with 875 pupils and 40 teachers indicates that a fall in occupancy from 98 per cent to 93 per cent is sufficient, absent any action, to move the school from an in-year surplus to an in-year deficit, with the deficit widening as occupancy falls further (NAO, 2026). Departmental modelling in 2023 suggested that between 33 and 55 per cent of primary schools could face financial risk by 2030 in the absence of mitigating action, although subsequent work concluded that there was no single tipping point and that risk rises progressively.

Schools enter this adjustment from an uneven starting position. In 2024-25, 51 per cent of local-authority-maintained schools reported an in-year deficit, while 82 per cent held surplus revenue reserves; maintained schools collectively held net reserves of £1.6 billion, within which £380 million of deficits were concentrated, and academy trusts held £4.2 billion including £73 million of cumulative deficits (NAO, 2026). Reserves buy time, but only once. Survey evidence from the National Foundation for Educational Research found that 21 per cent of primary leaders reported an in-year deficit in 2024/25 that they had no reserves to fund, and that 71 per cent anticipated an in-year deficit in 2025/26 (Julius & Henndel Lopes, 2025).

These pressures are not occurring in isolation. Witnesses to the Public Services Committee pointed to inflation, rising utility costs and national insurance contributions, pay awards funded partly from core budgets, growing demand for special educational needs and mental health support, and the withdrawal of the physical education and sport premium, announced after schools had set their budgets (House of Lords Public Services Committee, 2026). Falling rolls reduce the capacity of schools to absorb any of these.

5.2 The national funding formula and its protections

The national funding formula contains three mechanisms of relevance. Funding is allocated on a lagged basis, using the previous October’s census, which gives schools roughly a year’s notice before a fall in numbers affects income and time to reorganise staffing. Every school receives a lump sum, set at £152,700, which is not driven by pupil numbers and is proportionately more valuable to small schools. A sparsity factor delivers up to £58,600 to eligible primary schools facing the particular costs of rural isolation. A mobility factor makes some allowance for pupils arriving outside the census window (DfE, 2025; House of Lords Public Services Committee, 2026).

Witnesses accepted that these provide some relief while maintaining that they do not solve the problem. The lag defers the reduction rather than preventing it, and offers no protection at all against a sustained decline in which each year’s smaller cohort is followed by a smaller one still. The lump sum and sparsity factor raise the floor beneath small schools without arresting the annual erosion of the pupil-led element. Nothing in the formula prevents a school’s total budget from falling year on year. The Public Services Committee concluded that the formula in isolation is not sufficiently protecting schools experiencing sustained decline, and noted that the Department has no plans to review it (House of Lords Public Services Committee, 2026).

Stakeholders proposed three kinds of additional protection: a damping mechanism to limit the year-on-year reduction in any individual school’s budget; funding for a defined number of unfilled places where those places serve a system purpose; and greater recognition within the formula of fixed costs and of the complexity of need within a school’s intake. Each carries a distributional cost, since the schools block is a fixed envelope and protecting one school necessarily means allocating less to another. That trade-off has not been publicly examined.

5.3 Falling rolls funding

Outside the formula, the Department operates a dedicated falling rolls fund. For 2025-26 it allocated £141,890 for each eligible area within a local authority experiencing a fall of ten per cent or more in pupil numbers between two census years, provided the places are expected to be needed again within a short period. Some £155 million was allocated to local authorities for growth and falling rolls combined through the Dedicated Schools Grant in 2026-27 (House of Lords Public Services Committee, 2026), but the falling rolls element accounted for just under £27 million in 2025-26 (NAO, 2026).

Two criticisms recur. The first is scale. Twenty-seven million pounds is small relative to an annual per-pupil loss estimated in the hundreds of millions, and witnesses described it as insufficient to reflect the pressures schools face. The second is eligibility. A ten per cent fall between two census years is a demanding threshold, and the requirement that the places be needed again within a short horizon excludes precisely those schools facing a permanent structural decline, which are the schools in greatest difficulty. The fund is designed for temporary dips in a system that is no longer experiencing temporary dips. The Public Services Committee concluded that in its current form the fund is too limited and too restrictive to offer meaningful support, and asked whether the Department would reform or supplement it to provide genuine transitional funding (House of Lords Public Services Committee, 2026).

Some local authorities have stepped into the gap. Camden operates a £2.1 million local falling rolls safety fund intended to help schools avoid sudden redundancies (House of Lords Public Services Committee, 2026), and Sheffield maintains a policy fund of £200,000 with criteria requiring that numbers are expected to recover within three years (Sheffield City Council, 2025). Such arrangements depend on local authority finances that are themselves under strain; the London Borough of Waltham Forest told the committee that schools were entering deficit while the authority lacked the resources to support them. In my own local authority, Reading Borough Council reported a projected surplus of 12.9 per cent of primary places across the borough in autumn 2025, set against a planning assumption of a minimum 5 per cent surplus retained to absorb in-year admissions (Reading Borough Council, 2025).

5.4 Staffing reductions and curriculum narrowing

Because staffing is where the money is, staffing is where the adjustment falls. Research by the National Association of Head Teachers found that eight in ten school leaders anticipated reducing the number of teaching assistants or teaching assistant hours, and that 71 per cent were taking on additional teaching responsibilities themselves to fill gaps (House of Lords Public Services Committee, 2026). The Department’s own survey of school and college leaders found that reducing non-teaching staff was the most common response among primary schools experiencing a fall in numbers, reported by 59 per cent of respondents, and NFER survey work found that 41 per cent of primary leaders with concerns about falling numbers had reduced staff (NAO, 2026). Schools are also sharing senior leadership across sites and, in around seven in ten cases, revising their marketing in an attempt to attract pupils from neighbouring schools.

The second casualty is enrichment. Half of primary leaders reported reducing extracurricular activities or out-of-class interventions, and the Government has acknowledged that decisions on staffing capacity could affect curriculum breadth and reduce enrichment opportunities (House of Lords Public Services Committee, 2026). The effect compounds: the staff who ran the after-school clubs, the choir or the reading intervention are the same staff whose posts are removed, so schools lose both classroom capacity and the wider offer at once. Since participation in enrichment is already stratified by family income, the withdrawal of school-based provision falls hardest on the children least able to buy it elsewhere.

5.5 Mixed-age classes

Where a cohort is too small to fill a single-form entry, schools combine year groups. NFER survey evidence indicates that around a quarter of primary schools already use mixed-age classes, that 28 per cent of leaders had increased their use over the preceding two years, and that around 30 per cent of those concerned about falling rolls were looking to increase them further (House of Lords Public Services Committee, 2026; NAO, 2026). Evidence to the committee described schools accommodating as many as four year groups in one class.

The difficulty is not that mixed-age teaching is inherently inferior. It is that England has almost no recent evidence on whether it is. NFER’s research director told the committee that there had been essentially no new work on mixed-age classes in England for a quarter of a century, that initial teacher training does not prepare teachers for it, and that the accountability system and available support are configured around single-age classes even though mixed-age classes are already common (House of Lords Public Services Committee, 2026). The Confederation of School Trusts noted that the arrangement is least well suited to older year groups, where parts of the curriculum assume a level of maturity. Nuffield-funded research is under way but is not due to report until July 2027.

The consequence is that a pedagogical change affecting a growing share of primary pupils is being made for financial reasons, without an evidence base, without adapted training and without guidance. The Association of School and College Leaders characterised the shift as driven by budgets rather than by what is best for pupils. Whatever the eventual research finds, the sequencing is the wrong way round.

5.6 Special educational needs: pressure and opportunity

Falling rolls interact with rising special educational needs in two opposing directions. The pressure is immediate. As overall numbers fall while the number of children requiring additional support rises, pupils with complex needs form a growing proportion of a shrinking cohort. In Camden, the absolute number of children with special educational needs grew by ten per cent between 2018 and 2023 while their proportion of the school population grew by 17 per cent; the borough’s high needs block overspent by £6.3 million in one year, exhausting historical reserves and creating a £1.1 million deficit (House of Lords Public Services Committee, 2026). Because schools must fund the first £6,000 of additional support per pupil from their core budget, rising need draws directly on a budget that falling rolls are already reducing. Schools also reported meeting needs from existing budgets for children who arrive without any prior assessment, while lengthy statutory processes run their course.

This collides with policy. The Government’s white paper, Every child achieving and thriving, sets out an intention to move more special educational needs provision into mainstream settings, supported by £4 billion over three years and by at least £3.7 billion of capital between 2025-26 and 2029-30 to create 60,000 specialist places and expand units within mainstream schools (DfE, 2026a; NAO, 2026). Delivery depends disproportionately on teaching assistants, who are precisely the staff schools are shedding. Lincolnshire County Council told the committee that reducing teaching assistant support was the only route to a balanced budget, and one submission summarised the tension by observing that the policy ambition of inclusion may collide with the financial reality of lower pupil numbers (House of Lords Public Services Committee, 2026).

The opportunity is equally real. Spare capacity is exactly what an inclusion agenda requires: rooms for resource bases, speech and language provision and sensory spaces, and smaller cohorts permitting small-group teaching. Local authorities described placing specialist resource provision into mainstream classrooms where mainstream demand had receded, allowing children to remain within their own communities rather than travelling to special schools, with positive early feedback. The Catholic Education Service described a Gloucestershire trust that repurposed underused space so that children who would otherwise have been in specialist settings or at home could attend mainstream classes for part of the day (House of Lords Public Services Committee, 2026). Whether falling rolls become a constraint on inclusion or an enabler of it depends almost entirely on whether the transition is funded.

5.7 Closure, merger and community

Where mitigation is exhausted, schools close or merge. Primary closures in England fluctuated between 2010 and 2024, from 20 in 2010 to a low of seven in 2016 and a peak of 25 in 2024, with proportionately more in London (NAO, 2026). Fagan and Bradbury (2026) count nearly 100 permanent closures across England in the five years to 2026 and find that the schools that closed were disproportionately smaller schools and schools in deprived areas. NFER analysis identified five closures and five mergers in 2024 alone within the ten London authorities worst affected by declining rolls (Iocco et al., 2025).

Local authorities regard closure as a last resort and find it very difficult to execute. Only five per cent of those surveyed by the NAO described closing a school as easy or very easy, and 11 per cent said the same of merging schools, against 36 per cent for downsizing and 45 per cent for repurposing space (see Table 2). Statutory guidance imposes a presumption against the closure of rural schools, and requires proposers to consider medium and long-term requirements, alternative options, community impact including transport times, and the effect on pupil outcomes, particularly for disadvantaged children.

Table 2

Approaches to managing falling pupil numbers and local authorities’ assessment of their difficulty

Approach	Principal considerations	Rated “easy” or “very easy” (%)
Closing schools	Prescribed statutory processes, differing by school type; maintained school closures require local authority consultation, academy closures require the agreement of the Secretary of State.	5
Combining or merging schools	May involve closing one school and expanding another to absorb displaced pupils.	11
Downsizing schools	Reducing the published admission number (subject to a six-week consultation), reducing teaching and non-teaching staff, cutting non-essential programmes, and adopting mixed-age classes.	36
Repurposing space	Flexible reuse of surplus accommodation, most commonly for school-based nurseries and specialist provision for pupils with SEND.	45

Note. Percentages refer to the proportion of 76 responding local authorities describing each approach as easy or very easy. Adapted from National Audit Office (2026).

Closure has costs that fall outside the education budget. The most immediate is transport. Where a rural school closes, the next school may be a considerable distance away and poorly served by public transport, so savings on the school budget reappear as home-to-school transport costs on the local authority’s. Because the two budgets are held separately, the authority making the decision does not necessarily internalise the cost of it (House of Lords Public Services Committee, 2026). There is also a cost to children that no budget records. Witnesses described the compression of children’s social lives by long journeys: the inability to go to a friend’s house after school, or to mix with children living nearby.

The community dimension is well documented in the research literature. Small rural schools function as meeting places and as institutions around which local action is organised, and closure removes a site of association that is rarely replaced (Bagley & Hillyard, 2014). Comparative work has found that the erosion of local social capital following closure is measurable and persistent (Kłoczko-Gajewska, 2020). Schools North East told the committee that in rural, coastal and disadvantaged areas the primary school is a vital community asset whose closure weakens cohesion, increases travel distances and reduces access to wider services. The general secretary of the Association of School and College Leaders placed it in a sequence: communities have already lost the pub and the post office, and the primary school is next (House of Lords Public Services Committee, 2026).

What closure does to the children who experience it is less clear, and the English evidence is almost non-existent. The international literature is mixed and context-dependent. Several studies find short-run reductions in attainment attributable to disruption (Beuchert et al., 2018; Brummet, 2014; Engberg et al., 2012), with the Danish evidence indicating that adverse effects are most pronounced for pupils exposed to outright closure and for those initially enrolled in small schools, and that they attenuate within a few years. Other studies find no average effect on displaced pupils (Steinberg & MacDonald, 2019; Larsson Taghizadeh, 2019). The consistent finding is conditional: gains occur where pupils move to substantially higher-performing schools, and losses occur where they do not (Brummet, 2014; Engberg et al., 2012; Steinberg & MacDonald, 2019). Steinberg and MacDonald (2019) additionally found that attainment fell in receiving schools as the share of displaced pupils rose, that absences increased among displaced pupils, and that both absences and suspensions rose with the distance pupils had to travel to their new school. Tieken and Auldridge-Reveles (2019) argue that the literature has under-examined the distributional question of who bears these costs.

The policy implication is precise. Closure is not intrinsically harmful and may improve outcomes where children move to demonstrably better provision and are not concentrated in numbers that destabilise the receiving school. It is harmful where children are moved laterally, in large groups, over long distances, without additional resource for the receiving school. The English system is currently organised in a way that makes the second outcome more likely than the first, because closure decisions are triggered by financial exhaustion rather than by an assessment of where children would be better educated.

5.8 The distribution of harm

The most serious finding in the evidence is that falling rolls are regressive. In 2024/25, 19 per cent of places were unfilled in the most deprived schools compared with eight per cent in the least deprived (NAO, 2026). NFER analysis found that primary schools with the highest rates of free school meal eligibility saw pupil numbers fall by 7.3 per cent between 2018-19 and 2025-26, against 3.3 per cent among the least disadvantaged schools, and that the schools losing most pupils had also seen free school meal eligibility rise fastest.

The mechanism is the interaction of choice with quality signals described in section 3.5. As surplus capacity opens up, families with the resources and information to exercise choice move towards higher-rated schools, leaving lower-rated schools, which serve more disadvantaged intakes, to absorb the decline. Those schools then lose funding, cut staff, narrow provision and become less attractive, and the cycle repeats. Fagan and Bradbury (2026) find the pattern in the closure data: it is smaller schools and schools in deprived areas that have closed. Evidence to the Public Services Committee described this as approaching a structural bias, and noted that it runs counter to the Department’s stated objective of narrowing the disadvantage gap (House of Lords Public Services Committee, 2026).

The NAO put the same point in value-for-money terms while noting the ambiguity: if lower-performing schools shrink and close, more children may end up educated in better-rated schools, which would be a gain. But in the short term the schools least able to withstand financial shock are absorbing the largest ones, and the Department has carried out no research or assessment of the consequences for different groups of children, including disadvantaged children (NAO, 2026). Its recommendation that the Department consider the wider impact of falling numbers on educational quality and outcomes remains, at the time of writing, an open commitment rather than a completed piece of work.

6. What can and should be done

The demographic trajectory is fixed for at least a decade. The children who will fill reception classes in 2031 have already been born, or will be born to a population whose fertility behaviour is well established. Policy cannot change the number of pupils. It can change how the reduction is distributed, how quickly it is transmitted into budgets, whether the resulting decisions are made deliberately or under duress, and whether the capacity released is wasted or used. Seven areas of action follow from the analysis above.

6.1 Reform the funding architecture

The immediate requirement is to break the automatic and complete transmission of falling rolls into falling budgets. Three reforms merit consideration.

• A damping mechanism within the national funding formula limiting the year-on-year reduction in an individual school’s budget, analogous to the minimum funding guarantee that operates on a per-pupil basis. This would give schools a predictable glide path rather than a series of annual shocks, and would allow staffing reductions to be achieved through attrition rather than redundancy.
• A substantially enlarged and redesigned transition fund replacing the current falling rolls fund. The existing threshold of a ten per cent fall between two census years, coupled with a requirement that places be needed again within a short period, excludes schools in structural decline, which is now the typical case. A fund that supported multi-year adjustment, with eligibility based on the trajectory of a school’s budget rather than on the expectation of recovery, would address the actual problem. Its scale would need to bear some relation to the losses involved, which the NAO puts in the hundreds of millions annually.
• Greater recognition within the formula of fixed costs and of the composition of need. The lump sum and sparsity factor already acknowledge that some costs are invariant to roll; the case for increasing their weight strengthens as average school size falls. Separately, the £6,000 notional contribution that schools must make towards each pupil with additional needs becomes progressively harder to sustain as the proportion of such pupils in a shrinking cohort rises.

These are redistributive rather than additive proposals unless new money accompanies them, and that should be stated plainly. Within a fixed schools block, protecting schools with falling rolls means slower growth for schools with stable or rising ones. The alternative, which is what is currently happening, is that the adjustment is made anyway, but through unplanned deficits, staff losses and closures. Sibieta (2025) frames the underlying choice accurately: falling rolls give government the option of banking the savings or of holding funding constant and allowing per-pupil spending to rise. That choice should be made explicitly rather than by default.

6.2 Publish a national strategy and commission the missing evidence

The most consistent criticism across the evidence is the absence of national direction. The NAO found that the Department had not set expectations for when the sector should respond, had no framework for determining how much surplus capacity is appropriate, and had not shared insights from previous domestic or international experience. Almost a third of local authorities were satisfied with departmental support only to a small or very small extent, and 28 per cent had no formal plan for managing school places at all (NAO, 2026). Witnesses to the Public Services Committee described a lack of basic strategy and warned that government was sleepwalking into a problem it had long recognised (House of Lords Public Services Committee, 2026).

The Department has begun to move. Its February 2026 estates strategy commits to a local decision-making framework for the use of mainstream school space through demographic change, due in autumn 2026, and to pathfinder pilots testing the flexible use of surplus space (DfE, 2026b; NAO, 2026). The Public Services Committee welcomed this while expressing a specific reservation: the framework appears oriented towards how to repurpose space for special educational needs and community uses, and not towards the in-class decisions that most schools will actually face, or the point at which merger and closure should be contemplated. The committee argued that comprehensive guidance should cover evidence-based practice on mixed-age classes, staffing and curriculum changes; should take account of consequences for other local authority budgets, notably home-to-school transport; should help stakeholders take more significant decisions at the right moment rather than as a last resort; and should be framed around minimising the effect on children’s outcomes and wellbeing (House of Lords Public Services Committee, 2026).

Guidance of that kind requires evidence that does not yet exist. Four gaps are conspicuous: the effect of mixed-age classes on attainment in the English context; the effect of school closure on displaced pupils in England, as distinct from the United States and Scandinavia; the differential effect of falling rolls on disadvantaged pupils; and the conditions under which repurposing succeeds. The Nuffield Foundation is funding work on the first three, with the University of Reading examining permanent closures and NFER examining mixed-age classes, but the mixed-age research does not report until July 2027 (Nuffield Foundation, 2026). Decisions are being taken now. The Department should commission or accelerate work to inform them, set out when it will report, and say how the findings will feed into guidance.

6.3 Strengthen and clarify local place-planning powers

The statutory framework was built for expansion. Local authorities have a duty to secure sufficient places but no explicit duty regarding unfilled ones, and it is unclear how the existing duty applies to surplus (NAO, 2026). They cannot direct academies to reduce admission numbers, yet they are accountable for the coherence of provision across an area in which 52 per cent of schools are academies.

The Children’s Wellbeing and Schools Act 2026 addresses part of this. Section 61 introduces duties on mainstream state schools and local authorities to co-operate on admissions functions, and the Act empowers the Schools Adjudicator, where it upholds an objection to a published admission number, to specify what that number should be. Objections may now be made where a number is increased or left unchanged, whereas previously only proposed reductions could be challenged (House of Lords Public Services Committee, 2026). The change matters because a school increasing its intake during a period of local decline exports the consequences to its neighbours without bearing them itself.

Three cautions attach. First, the provisions create a route by which admissions at popular schools could be constrained in order to sustain less popular ones, which the Confederation of School Trusts argued would not serve children’s interests; the Government has responded by requiring that decisions take account of parental preference and quality, and departmental officials indicated that a high-performing school should be reduced only where the alternative would be clearly detrimental to children elsewhere. Second, local authorities are both providers of maintained schools and arbiters of overall provision, which several submissions identified as a conflict of interest. Third, the Act received Royal Assent only in April 2026, key provisions are not yet in force, and the admissions code will not be finalised until September 2027, so its effects are untested. The Department should commit now to evaluating whether the co-operation duty and the adjudicator power deliver effective place planning, and what they do to parental choice and to standards.

A further clarification is needed on the sufficiency duty itself. If, as the Department maintains, the duty extends to reducing or repurposing capacity that is no longer required, that should be stated in guidance directed at the whole sector rather than inferred from documents aimed at academies contemplating significant change. Local authorities also need a parameter for acceptable surplus. Restoring a national benchmark would be crude given local variation, but a published range, with an expectation that authorities justify departures from it, would give the Department a trigger for support or challenge and would give authorities a defensible basis for difficult conversations.

6.4 Treat surplus space as an asset

Surplus capacity is a liability on a school’s budget but an asset in the hands of a system that has other unmet needs. The most developed use is special educational needs provision. Repurposing was the response local authorities found easiest, with 45 per cent describing it as easy or very easy and 92 per cent reporting that they had done it (NAO, 2026), and it aligns with the inclusion ambitions of the schools white paper. Resource bases, inclusion hubs and specialist units housed in mainstream schools allow children to be educated closer to home, at lower cost than special school placement, with reduced transport spending.

School-based nurseries are the second use. The Department has committed £400 million between 2025-26 and 2029-30 to create or expand nurseries, with grants of up to £150,000 per school; the first phase awarded £37 million to 300 primary schools creating 6,000 places, and in March 2026 a further £45 million supported 331 schools to create 6,000 more, with over a third in the most disadvantaged areas (NAO, 2026). Around one in four primary leaders concerned about demographic decline had introduced nursery provision since 2022/23. Nurseries generate income and connect schools with families before reception, which may support later recruitment. The obvious caveat is that where demand for primary places is falling because there are fewer young children, demand for nursery places will eventually fall for the same reason.

Beyond education, schools are physically and socially central to their communities, and evidence to the committee identified opportunities in breakfast clubs, family hubs, youth services and intergenerational provision (House of Lords Public Services Committee, 2026). The Department’s estates strategy anticipates piloting the use of surplus school space for wider government priorities, including family hubs and neighbourhood health facilities (DfE, 2026b; NAO, 2026). Realising this requires attention to three practical obstacles: the upfront capital cost of conversion, which schools in deficit cannot meet; safeguarding arrangements that make partner organisations reluctant to co-locate; and the fact that local authorities can only persuade, not direct, academies to host provision the authority has identified as needed.

A final consideration is reversibility. Benchmarking by two local authorities in 2025 put the average cost of creating a primary place in a new school at £22,890 (NAO, 2026), and the Department’s own institutional memory includes the experience of the 1990s and 2000s, when places removed in response to falling rolls had to be recreated at considerable expense when numbers recovered. Repurposing that is straightforwardly reversible should be preferred to disposal, and any sale of publicly funded school land, which requires the Secretary of State’s approval, should be tested against a long-horizon assessment rather than an immediate receipt.

6.5 Manage closures deliberately where they are necessary

Some schools will close. The Government has acknowledged that where a school has become so small that it is damaging the education of its pupils, closure may be the right outcome (House of Lords Public Services Committee, 2026). The objection raised in evidence is not to closure as such but to the manner of it: schools shrinking at random, or closing by default once reserves are exhausted and staff have left, which is the worst point at which to move children. The National Education Union framed the alternative as a planned reduction in school size guided by local need and evidence rather than closures forced by budgets.

The international evidence outlined in section 5.7 supports a specific protocol. Where closure is judged necessary, displaced pupils should be moved to schools that are demonstrably better than the one they leave, since this is the condition under which the evidence finds gains rather than losses (Brummet, 2014; Engberg et al., 2012; Steinberg & MacDonald, 2019). They should be distributed in modest numbers across receiving schools rather than transferred as a block, because attainment in receiving schools deteriorates as the share of displaced pupils rises (Steinberg & MacDonald, 2019). Travel distance should be minimised, given the association between distance and both absence and exclusion. Receiving schools should be resourced for the transition rather than expected to absorb it. And the decision should be made while the closing school still has functioning staff and leadership, so that transition can be managed rather than improvised.

The presumption against closing rural schools should be retained, but it should be accompanied by a genuine appraisal that internalises the costs currently sitting outside the education budget. Home-to-school transport is the clearest example: an appraisal that counts the saving on the schools budget but not the recurring cost to the transport budget, still less the cost to children of an additional hour on a bus each day, will systematically over-recommend closure in rural areas. Bringing those budgets into a single assessment is an administrative change with substantive consequences.

6.6 Treat the workforce effect as an opportunity

Falling rolls arrive alongside a persistent teacher recruitment and retention problem. Initial teacher training targets have been missed repeatedly, particularly at secondary level, where 62 per cent of the target was met in 2024-25 against 88 per cent at primary (Institute for Government, 2025). Departmental modelling suggested a surplus of around 1,200 primary teachers in 2025/26 (NAO, 2026).

There is an obvious complementarity here that the system is not currently exploiting. Falling primary rolls could be used to improve teaching quality and reduce workload rather than simply to reduce headcount, and the Nuffield Foundation has argued that the demographic decline can be treated as an opportunity to improve teaching and learning if decisions about recruitment and retention are made deliberately (Nuffield Foundation, 2026). Options include retraining experienced primary teachers to work across phases or in specialist roles, redeploying capacity into the inclusion agenda where the white paper creates demand for expertise, and adjusting initial teacher training numbers by phase rather than allowing the adjustment to occur through unplanned exits. The risk of inaction is that the system loses experienced teachers through redundancy in the primary phase while continuing to under-recruit in the secondary phase, and then has to recruit them back.

6.7 Prepare for the secondary phase now

The final recommendation is the most straightforward and the most neglected. Secondary rolls have started to decline and will fall by around 249,000 by 2031 (DfE, 2026c). Every difficulty documented in the primary phase will recur, with the additional complication that secondary curriculum breadth depends on subject specialists whose posts cannot be sustained below a certain cohort size, and that secondary school organisation is harder to reverse.

The Public Services Committee asked what work is under way and where the risk sits on the departmental risk register (House of Lords Public Services Committee, 2026). The value of answering that question honestly is that the primary phase has already generated the evidence. Which mitigations preserved quality and which did not; which local authorities planned well and which were overtaken; what repurposing succeeded; what closures cost. None of this has been systematically collected. The NAO found that good practice was developing locally but was not being shared, that the Department does not collate the narrative commentary local authorities already provide in their capacity returns, and that it is not within the remit of its regional place-planning advisers to assess what a good plan looks like (NAO, 2026). Establishing a mechanism to gather and disseminate this learning is inexpensive, and it is the difference between meeting the secondary decline having learned something and meeting it having learned nothing.

7. Limitations

Three limitations qualify the analysis. First, it relies on projections whose central assumptions have recently proved unstable. The downward revision between the 2025 and 2026 pupil projection models, amounting to 286,000 pupils for 2030 alone, illustrates how quickly the picture changes when migration and fertility assumptions are updated (DfE, 2026c). Variant projections indicate that migration assumptions now move the primary-age population by more than fertility assumptions do over a five-year horizon, which makes school place planning unusually sensitive to immigration policy.

Second, the causal evidence on educational impact is imported. The studies relied upon in section 5.7 examine urban American districts and Scandinavian consolidation programmes, where the closures were often motivated by low performance rather than by demography, where school systems are organised differently, and where distances and transport arrangements differ from those in rural England. The direction of the conditional finding, that outcomes depend on the quality of the receiving school and the scale of displacement, is likely to travel; the magnitudes are not.

Third, much of the sectoral evidence is drawn from surveys of school leaders and from submissions to a parliamentary inquiry. Such evidence is valuable in identifying mechanisms and in describing conditions, but respondents are self-selecting and have an interest in the outcome. Where survey figures are cited here, the base is generally small: the NAO surveyed 76 of 153 local authorities, and NFER’s omnibus surveys draw on a few hundred school leaders. The figures should be read as indicative of direction rather than as precise estimates.

8. Conclusion

Falling rolls are the predictable consequence of a fertility decline that has been visible in the birth data since 2012 and which has been reinforced since 2024 by a sharp contraction in net international migration. There is nothing surprising about the arrival of smaller cohorts in reception classes, and nothing that education policy could have done to prevent it. What education policy determines is how the system absorbs the change.

On that measure the record is poor. Primary numbers have been falling for seven years. The Department for Education added the risk of failing to respond adequately to demographic change to its risk register only in 2024, has not assessed the effect of falling rolls on educational quality, has not examined the consequences for disadvantaged children, has not established what level of surplus capacity is appropriate, has not collected or shared the good practice developing locally, and has not reviewed a funding formula that transmits demographic decline into school budgets without meaningful protection. The mechanism intended to cushion the transition is a fund of £27 million against annual losses estimated in the hundreds of millions, with eligibility criteria that exclude schools in structural decline. Meanwhile schools are cutting teaching assistants, withdrawing enrichment, merging year groups into classes for which their teachers were not trained and about which England holds almost no evidence, and closing when their reserves run out.

The distributional pattern is the most troubling element. Unfilled places, roll reductions and closures are all concentrated in schools serving the most disadvantaged communities, because parental choice in a slackening market operates as a sorting mechanism. A demographic change that is neutral in origin is being converted, by the design of the system absorbing it, into a widening of the gap that policy is elsewhere committed to closing.

None of this is irretrievable. Smaller cohorts genuinely do create opportunities: space for the inclusive provision the white paper envisages, room for nurseries and community services in buildings that already sit at the centre of their neighbourhoods, and a chance to improve teaching quality rather than merely to reduce headcount. Realising any of it requires the Government to do what both the National Audit Office and the House of Lords Public Services Committee have asked of it: to treat falling rolls as a demographic and place-based challenge across public services rather than a budgeting problem for individual schools, to fund the transition rather than only the destination, to build the evidence base before rather than after the decisions are taken, and to apply what the primary phase has taught before the secondary phase repeats it. The children currently in Year 6 will be in the secondary system when the same pressures arrive there. The time available to prepare is known precisely, and it is short.

References

Andrews, J., & Wielar, L. (2025). So long, London: An analysis of London primary pupil movements. Education Policy Institute. https://epi.org.uk/publications-and-research/so-long-london-an-analysis-of-london-primary-pupil-movements/

Bagley, C., & Hillyard, S. (2014). Rural schools, social capital and the Big Society: A theoretical and empirical exposition. British Educational Research Journal, 40(1), 63–78. https://doi.org/10.1002/berj.3026

Beuchert, L., Humlum, M. K., Nielsen, H. S., & Smith, N. (2018). The short-term effects of school consolidation on student achievement: Evidence of disruption? Economics of Education Review, 65, 31–47. https://doi.org/10.1016/j.econedurev.2018.05.004

Brummet, Q. (2014). The effect of school closings on student achievement. Journal of Public Economics, 119, 108–124. https://doi.org/10.1016/j.jpubeco.2014.06.010

Children’s Wellbeing and Schools Act 2026, c. 21 (UK). https://www.legislation.gov.uk/ukpga/2026/21

Claude.ai sifting and sorting

Department for Education. (2025). The national funding formula for schools and high needs: 2026 to 2027. https://www.gov.uk/government/publications

Department for Education. (2026a). Every child achieving and thriving (CP 1508). https://www.gov.uk/government/publications/every-child-achieving-and-thriving

Department for Education. (2026b). Education estates strategy: A decade of national renewal (CP 1501). https://www.gov.uk/government/publications/education-estates-strategy

Department for Education. (2026c). National pupil projections: Reporting year 2026. Explore Education Statistics. https://explore-education-statistics.service.gov.uk/find-statistics/national-pupil-projections/2026

Engberg, J., Gill, B., Zamarro, G., & Zimmer, R. (2012). Closing schools in a shrinking district: Do student outcomes depend on which schools are closed? Journal of Urban Economics, 71(2), 189–203. https://doi.org/10.1016/j.jue.2011.10.001

Fagan, E., & Bradbury, A. (2026). Falling pupil numbers and school closures: Setting a research agenda for a new era of precarity. British Educational Research Journal. Advance online publication. https://doi.org/10.1002/berj.70193

House of Lords Public Services Committee. (2026, September 2). Letter from Lord Bradley, Chair of the Public Services Committee, to Georgia Gould OBE MP, Minister of State (Education): Falling primary school rolls. UK Parliament. https://committees.parliament.uk/publications/54664/documents/303780/default/

Institute for Government. (2025). Performance tracker 2025: Schools. https://www.instituteforgovernment.org.uk/publication/performance-tracker-2025/schools

Iocco, M. P., Kuhn, L., & Julius, J. (2025, October 23). Pupil numbers and school finances: What do we know about how demographic change is impacting on schools? National Foundation for Educational Research. https://www.nfer.ac.uk/blogs/pupil-numbers-and-school-finances-what-do-we-know-about-how-demographic-change-is-impacting-on-schools/

Julius, J., & Henndel Lopes, G. (2025, September 18). Senior leaders under pressure as pupil numbers fall. National Foundation for Educational Research. https://www.nfer.ac.uk/blogs/senior-leaders-under-pressure-as-pupil-numbers-fall/

Kłoczko-Gajewska, A. (2020). Long-term impact of closing rural schools on local social capital: A multiple-case study from Poland. European Countryside, 12(4), 598–617. https://doi.org/10.2478/euco-2020-0031

Larsson Taghizadeh, J. (2019). Effects of school closures on displaced students and future cohorts in Sweden 2000–2016 (Working Paper 2019:4). Institute for Evaluation of Labour Market and Education Policy. https://www.ifau.se/globalassets/pdf/se/2019/wp2019-04-effects-of-school-closures-on-displaced-students.pdf

London Councils. (2026, February 2). Sharp decline in London pupil numbers puts school standards at risk. https://www.londoncouncils.gov.uk/news-and-press-releases/2026/sharp-decline-london-pupil-numbers-puts-school-standards-risk

National Audit Office. (2026). Responding to changing demand for school places (HC 1818). https://www.nao.org.uk/reports/responding-to-changing-demand-for-school-places/

Nuffield Foundation. (2026, June 4). Why falling birth rates mean big changes for England’s schools. https://www.nuffieldfoundation.org/news/why-falling-birth-rates-mean-big-changes-for-englands-schools

Office for National Statistics. (2025). Births in England and Wales: 2024 (refreshed populations). https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/livebirths

Office for National Statistics. (2026). Births in England and Wales: 2025. https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/livebirths/bulletins/birthsummarytablesenglandandwales/2025

Reading Borough Council. (2025). School admission arrangements consultation 2027/28 [Report to the Adult Social Care, Children’s Services and Education Committee, 15 October 2025]. https://democracy.reading.gov.uk/documents/s37186/Committee%20Report%20School%20Admissions%20Oct%2025%20amended.pdf

Sheffield City Council. (2025). Falling rolls policy 2025–26 for all Sheffield primary schools. https://www.sheffield.gov.uk/sites/default/files/2025-06/falling_rolls_policy_2025-2026.pdf

Sibieta, L. (2025). Demographic change and schools across the UK: Lessons from history. Institute for Fiscal Studies. https://ifs.org.uk/publications/demographic-change-and-schools-across-uk-lessons-history

Steinberg, M. P., & MacDonald, J. M. (2019). The effects of closing urban schools on students’ academic and behavioral outcomes: Evidence from Philadelphia. Economics of Education Review, 69, 25–60. https://doi.org/10.1016/j.econedurev.2018.12.005

Tieken, M. C., & Auldridge-Reveles, T. R. (2019). Rethinking the school closure research: School closure as spatial injustice. Review of Educational Research, 89(6), 917–953. https://doi.org/10.3102/0034654319877151