From the start of this school year, every child in England whose household receives Universal Credit is entitled to free school meals. The change removes an earnings threshold that had excluded working families on low incomes, extends eligibility to more than half a million additional children, and has the potential to lift around 100,000 children out of poverty. Ministers have called it the biggest expansion of free school meals eligibility in a generation; it is certainly the largest aimed at children living in poverty.
What’s happening in England also represents one of four approaches to school food now operating in the UK, with each nation having reached a different decision about who should be entitled to free school meals and how. Researchers have long recognized the potential of this divergence: in 2010, one assessment described Scotland’s moves towards universal free school meals in the early primary years as perhaps the first case of devolution providing a “crucible” for welfare policy. Sixteen years on, that experiment is running across all four nations. Yet the evidence needed to learn from it is not being collected consistently in any nation, and in England the funding behind the new entitlement already falls short of what a meal costs.
In this article, I look at the choices the four nations navigate differently, why entitlement does not guarantee delivery, and what it would take for the UK to learn from its own experiment.
Who gets free school meals depends on where a child lives
School food followed the same basic approach across the UK for decades. The post-war settlement obliged every local authority to provide a meal to any child who wanted one at a nationally set price and in accordance with national nutritional standards, with free school meals for families with the lowest incomes.
The 1980 Education Act removed most of those guarantees—the duty to provide a meal, an established price, and a requirement to meet national nutritional standards—leaving only the means-tested free meal as an obligation. Devolution made it possible for the nations to make different decisions, and in recent years they have.
Entitlements require deciding which trade-off to make
In part this is because of the choices available for navigating a well-documented trade-off. Deciding entitlement based on income concentrates public money on families with the greatest needs, but it comes with two big challenges. One is stigma, because a free meal marks a child out as poor, and some families would rather not be identified that way. The other is incomplete take-up: entitlement depends on families knowing it exists, understanding the criteria, and completing the registration process. At each stage, some eligible children drop out of the process. A universal offer removes both of these challenges, but it means the state pays for lunches for many children whose families could afford it.
The variations reflect different views about how and why to target support
Wales has judged the challenges of targeting free meals based on income to be too great. It has implemented a universal offer, which was rolled out to every primary year group by September 2024. This means no child is identified in a stigmatizing way, and families don’t need to apply.
Northern Ireland has taken the opposite approach: households on Universal Credit qualify only if net earnings fall below £15,900, a threshold now indexed to inflation, with school lunch provision from nursery to post-primary concentrated on families with the lowest incomes.
Scotland and England fall somewhere between these two. Scotland provides universal meals from Primary 1 to 5. In Primary 6 and 7, children qualify if their families receive the Scottish Child Payment, which is open to any household on Universal Credit or other qualifying benefits, with no earnings limit. In secondary school, a means test with an earnings threshold still applies. England’s change, which took effect in September applies the same logic as Scotland’s Primary 6 and 7 route across all year groups: receipt of Universal Credit alone now qualifies a child, whereas previously an earnings ceiling of £7,400 applied on top of that. Means-testing remains, and families still need to apply, but the income threshold has gone, removing part of the main sources of incomplete take-up.
Little evidence on which approach works
With four different approaches to the same basic policy instrument operating across the UK, an obvious question might be: which of them is achieving what it set out to do? Is there higher take-up in Wales among children living in poverty because there’s no application required or stigma attached to the meal? Or is targeted spending in Northern Ireland ensuring that more children with the greatest need are being supported?
In 2014, England decided to provide universal meals for infants following pilots in Durham and Newham that ran from 2009 to 2011. The evaluation of those pilots found that pupils made four to eight weeks’ more progress in English and maths than similar children elsewhere, with vegetable consumption increasing at lunchtime and take-up increasing among low-income families. The evaluation provided the government with evidence that the policy would work before rolling it out to every child in English state-funded schools in reception years 1 and 2.
Only Scotland is checking compliance
At the moment, very little official monitoring of school meals is happening in the UK. Only Scotland consistently checks whether schools meet the required standards through specialist health and nutrition inspectors. A 2022 comparative review of school food across the four nations concluded that there is “no consistent assessment, monitoring or reporting of compliance with the standards for school food”. England has committed to a national system for monitoring compliance from September 2027, to be developed with the Food Standards Agency, and has yet to publish the details.
Where researchers have assessed this independently, findings suggest that compliance cannot be assumed: in England, only one in four schools was known to be compliant with the standards, and 64% of the calories in school meals come from ultra-processed foods. More recent evidence is no more reassuring, confirming earlier estimations of compliance: in August, the food charity Bite Back analysed the published menus of nearly 400 English secondary schools and found that 76% contained at least one confirmed breach of the standards.
Even full compliance would mean different things
Comparison between the nations is even harder because the systems differ not only in what they provide but also in what they measure. Standards are not alike. For example, Scotland and Wales regulate the nutrient content of meals, whereas England and Northern Ireland regulate only the types of food served. Therefore, full compliance would mean something different in each nation.
Academic studies occasionally bridge the gap: one recent evaluation compared children’s diets in England and Scotland under their different offers. But these are one-off efforts. And with Wales only recently completing its rollout and England expanding provision this September, it will be some time before any future assessment can demonstrate that the approach achieves its goals. Without standards that can be compared and routine monitoring of them, none of the nations is really in a position to learn from the others’ experiences.
The challenge of turning entitlements into meals
There is an even more basic question than which design works best: whether the funding for the entitlement is enough to meet the costs of producing the meals. Delivery rests on government funding schools or local authorities at a set amount per meal or per eligible pupil. School kitchens must produce a meal that meets the national standards for that amount. But this doesn’t mean the grant covers the actual costs of ingredients, catering staff, wages, energy, etc. And when funding falls short of costs, a school can do one of three things: cover the shortfall from its broader budget (meaning trade-offs elsewhere), reduce the quality of the meal, or pass the loss on to a caterer.
England’s expansion is funded below the cost of a meal
Under England’s expansion, funding is £505 per newly eligible pupil per year, equivalent to about £2.66 per meal. Independent research estimates that the actual cost of producing a meal that meets the requirements is £3.16 on average, and more than £4 per meal in small schools due to fixed costs for the kitchen and cooking staff that are spread across fewer meals.
Modelling projects that schools in England will face a shortfall of more than £310 million in 2026/27 across existing and newly eligible pupils. Analysis of the expansion suggests the shortfall’s expansion share is around £47.5 million. Revised school food standards, proposed in April and confirmed in September, will raise the requirements for a compliant meal from autumn 2027, which at current funding rates would likely widen the shortfall even further.
The devolved nations have met the same constraint in different ways
In Wales, the government has raised the per-meal rate twice since 2022: from £2.90 to £3.20, then to £3.40, as food, energy, and labour costs have risen. In an interim evaluation published in February, providers warned that these increases lag inflation, kitchens and dining halls are running at capacity, and catering staff are carrying heavier workloads without higher pay.
In Scotland, funding constraints stopped a planned expansion in 2024, with the government abandoning its commitment to extend universal meals to Primaries 6 and 7 and instead targeting provision at families receiving the Scottish Child Payment.
Northern Ireland has not expanded provision, but costs have risen there too. Its system differs from the other nations: school meals are provided centrally by the Education Authority, which reports that a meal now costs £4.28 to produce, while families who pay are charged £3.10, a price raised this January from £2.60, where it had been frozen since 2017, and still well below cost. The difference has to be met by the education budget, which faced a shortfall of around £300 million last year.
England is attempting the largest expansion of the four, benefiting more than half a million additional children, at a funding rate below what Wales pays for a meal, and below what independent research says a meal costs to make.
Reading the results
The free school meals expansion that began in England this September deserves to succeed: half a million children will be better off if it does. But success depends in large part on a funding rate already set below the cost of a meal and a set of standards that nobody is consistently checking.
The same is broadly true across the UK. Only Scotland has put any of this in legislation, and even there the central duty of the Good Food Nation Act is not yet in force. Everywhere else, entitlements, funding rates, and monitoring rely on decisions that can be revisited every year. No government is obliged to collect the evidence on whether its approach is working.
Part of the difficulty is down to how success is measured. Entitlement is a means to an end with children eating meals that meet the standards set, every school day, without stigma or administrative headaches. Giving children the chance to eat that food requires uptake among eligible children, compliance with the standards set for the meals served, and funding that covers the costs of providing them. None of the four nations currently collects data on all three aspects.
Other changes are needed, too, such as adjustments for small schools, and regularly reviewing costs. Compliance needs to be checked by someone with authority to act on what they find, as Scotland’s inspectors are doing. England’s planned system will need the same authority. Each nation should publish data on take-up, compliance, and costs on a comparable basis so others can learn from it. Without something like this, it’s hard to see how any nation could show that its approach achieves its goals. Four different approaches to feeding children at school are being implemented across the UK. Someone should be reading the results, ideally the governments themselves using an agreed set of indicators. Until they do, who will?
I write about the future of food and the connections between our food systems, the environment and public health. Sign up for my newsletter.
You can also read more here.
