New research by Professor Birgitta Rabe and Dr Angus Holford on school breakfast schemes is published today, 13 July. The study, funded by the Nuffield Foundation, looked at the impacts of free universal breakfast schemes, operated in disadvantaged primary schools between 2006 and 2024, with the aim of informing the national roll-out of Free Breakfast Clubs in England.
Take-up by provision mode
- Schemes are delivered as Breakfast Clubs (in school halls before the day starts), Classroom provision (during arrival time), Grab and Go (taken while passing through school), or a combination (Mixed provision).
- Take-up is highest for Classroom provision (84%) and lowest for Breakfast Clubs (28%), with Grab and Go and Mixed in between. FSM-eligible children make up a similar share of participants across all modes (around 49%).
Absence
- Breakfast schemes very slightly reduced absence rates (0.13 days per child per year) and persistent absence (one child in 240 no longer persistently absent).
Children’s emotional and behavioural difficulties
- Breakfast schemes reduced emotional and behavioural difficulties, measured by whether statements such as them being “easily distracted” of “often loses temper” being “Not true”, “somewhat true”, or “certainly true”.
- The benefit was concentrated entirely in boys, and driven by reduced hyperactivity and conduct problems.
Child development and attainment
- Breakfast schemes very slightly improved teacher-assessed attainment at Reception (age 5), measured by the Early Years Foundation Stage Profile.
- Breakfast schemes resulted in approx. 1 week of additional progress in KS2 (age 11) Reading and Maths tests, and this did not differ between girls and boys. The effect is roughly half the benefit we previously for found for Universal Free School Meals at lunchtime (https://doi.org/10.5526/misoc-2024-001).
Children’s bodyweight
- Breakfast schemes raised obesity prevalence by 0.25 and 0.36 percentage points among Reception and Year 6 children respectively. This is equivalent to one additional child living with obesity for every 400 Reception children and 280 Year 6 children. These impacts are about half of the size of the beneficial effect of Universal Free School Meals at lunchtime (https://doi.org/10.5526/misoc-2024-001).
- Detrimental effects are mainly driven by Classroom provision. Breakfast Clubs, Mixed and Grab-and-Go showed no significant increases in obesity prevalence; though Breakfast Clubs did raise BMI among Year 6 children.
Families’ subjective financial situation
- Breakfast schemes do not affect mothers’ and fathers’ subjective financial situation or their expenditure on groceries or takeaways.
Authors’ main messages
- Free breakfast schemes in disadvantaged schools raise children’s educational attainment in Reading and Maths by a modest but meaningful amount by the end of primary school. This suggests that breakfast schemes can form part of a wider policy toolkit for raising attainment among disadvantaged pupils.
- Breakfast provision may be a useful complement to dedicated behavioural interventions in schools, which are expensive and hard to scale. Girls do not show the direct wellbeing gains seen for boys; though their KS2 improvements suggest they may too benefit from a calmer classroom environment.
- Prioritising rollout of breakfast schemes to more deprived schools is well-founded on educational grounds. Less is known about what the impacts will be in less disadvantaged schools.
- Adverse effects on bodyweight are a concern. Classroom provision, the mode with the highest take-up, produces the largest increase in obesity – consistent with breakfast becoming a social default even for children who have already eaten at home. Delivery models should target genuine nutritional need without promoting over-consumption.
- Revised school food standards are expected to raise the nutritional quality of meals, but standards will need to be accompanied by monitoring and staff training to be effective.
- Evidence on modes of breakfast provision shows that Mixed provision, which offers children a choice of how to access breakfast, consistently outperforms a single delivery mode across attainment, development, and bodyweight outcomes. The national Breakfast Club programme should consider whether it can accommodate flexible delivery within schools, rather than prescribing a uniform before school format.
The study uses school-level data on the availability of breakfast schemes in England by three major providers: Magic Breakfast, the Greggs Foundation, and the National School Breakfast Programme and link these data to population or nationally representative datasets on children’s or school-level characteristics and outcomes: National Pupil Database, National Child Measurement Programme, and Understanding Society, the UK Household Longitudinal Study (UKHLS).
The research uses a staggered difference-in-difference design to compare changes in outcomes for earlier adopters with later adopters. This enables the researchers to calculate the difference between observed outcomes for children exposed to a breakfast scheme, with a prediction for what would have happened had they not (yet) introduced the scheme. The estimates are ‘intention-to-treat’ (ITT), reflecting the effect of universal free breakfast being available rather than the effect of individual take-up. See more detailed results and methodology in the full project report, Impact of school breakfast programmes in primary schools in England.
About this analysis
The Nuffield Foundation has funded this project, but the views expressed are those of the authors and not necessarily the Foundation. Website www.nuffieldfoundation.org. Bluesky: @nuffieldfoundation.org, LinkedIn: Nuffield Foundation. We also acknowledge support from the Economic and Social Research Council (ESRC) through the ESRC Research Centre on Micro-Social Change (UKRI215). We thank members of our Advisory Group, Magic Breakfast, the Greggs Foundation and the DfE for sharing their data on breakfast provision. This work uses data provided by individuals and collected by the NHS as part of their care and support. Specifically, this work uses data from the National Child Measurement Programme, supplied by NHS England’s Health and Social Care Information Centre, part of the Government Statistical Service. This work also uses data from the Department for Education’s National Pupil Database, carried out in the Secure Research Service, part of the Office for National Statistics (ONS). Statistical data from ONS is Crown Copyright, and copyright of the statistical results may not be assigned. The use of the ONS or NHS England statistical data in this work does not imply the endorsement or quality assurance of the ONS or NHS England in relation to the interpretation or analysis of the statistical data. This work uses research datasets which may not exactly reproduce National Statistics aggregates. Cite as: Holford, A., Rabe, B., (2026), ‘Free Breakfast schemes in disadvantaged schools have mixed impacts’. MiSoC Explainer 2026:002.