It Was Never About the Cereal
By September, 680,000 children in England were getting a free breakfast at school. The trial that got the policy this far found something odd: the children who never went to the club did better too. Which suggests the thing doing the work isn't the food.
By the end of September, around 680,000 children in England were sitting down to a free breakfast at school. In April it was 300,000. The plan is to reach all 17,000 primaries by 2028.
Almost all of the argument about this has been about food. Whether children are arriving hungry, whether toast and cereal counts as a proper meal, whether feeding children is a school's job at all.
The trial that got the policy this far points somewhere else entirely. It found that children who never went to the breakfast club did better too.
That trial ran a decade ago. The Education Endowment Foundation gave around 100 primary schools in poorer areas the support to run a free breakfast club open to every child, then compared them with similar schools carrying on as usual. About 8,600 children were involved.
Year 2 children in the breakfast club schools finished the year roughly two months ahead in reading, writing and maths. Absence fell, by something like 26 fewer half-days across a class of thirty. Teachers reported better behaviour.
And the gains were not limited to the children actually eating the toast. The evaluators concluded that the clubs helped children who did not attend, by improving the classroom those children walked into. That is a peculiar thing for a bowl of cereal to manage.
When the EEF reviewed the wider research on breakfast schemes in 2024, it came to a similar place. Across the studies, free breakfast had either no effect on attendance or a small one. Where things did improve, eating breakfast alone did not seem to be what mattered.
What a breakfast club really changes is the shape of the morning. Children arrive twenty minutes early instead of five minutes late. The register isn't interrupted. Nobody walks in at twenty past nine still in their coat, pulling everyone's attention with them. The first lesson starts on time, in a room full of children who have already settled.
The code printed in the original report did not match the code that had been used, so neither the reviewers nor the EEF spotted it at the time. Strictly speaking the study is no longer a randomised controlled trial. Its rating has been knocked down from high security to moderate-to-high.
That does not make the findings worthless. It does mean holding the two months loosely rather than quoting it as settled fact. It is also unusual for an organisation to reopen its own best-known study and say so publicly, which is worth more than most people give it credit for.
That is a childcare answer. A free half hour at 8am is the difference between making an early shift and not making it, and families are choosing schools accordingly.
The universal part matters too, in a way that is easy to miss. Sixty per cent of parents said they are more likely to use help when it is offered to everyone. Thirty-eight per cent of single parents said they feel guilty taking up free support, against 28 per cent of parents with a partner. A club anyone can walk into is a club nobody has to qualify for.
When the Department for Education sent researchers to watch the early clubs and talk to parents, the descriptions were strikingly ordinary. Calmer houses. Less shouting about shoes. Children picking their own breakfast, which one parent thought was the real prize: as a child, she pointed out, you don't get to choose very much.
Of the 750 schools in the first wave, around 80 dropped out and had to be replaced, in many cases because the money did not stretch. An unhurried twenty minutes sounds free. It costs an adult being paid to stand in a hall at eight in the morning, five days a week.
None of which means the food is beside the point. Magic Breakfast's 2025 survey found more than a third of teachers see hungry children arriving every day, and nearly half said it had got worse over the year. Morning hunger is real and it is not improving.
But if the thing quietly doing the work is the calm start rather than the cereal, that changes what schools should protect when the budget gets tight. And it is the one part a family can borrow. Most of us cannot do anything about school funding. We can look at what happens in our own kitchen between half seven and half eight, and notice that the difference between a rushed exit and an early one was never about breakfast at all.
Almost all of the argument about this has been about food. Whether children are arriving hungry, whether toast and cereal counts as a proper meal, whether feeding children is a school's job at all.
The trial that got the policy this far points somewhere else entirely. It found that children who never went to the breakfast club did better too.
That trial ran a decade ago. The Education Endowment Foundation gave around 100 primary schools in poorer areas the support to run a free breakfast club open to every child, then compared them with similar schools carrying on as usual. About 8,600 children were involved.
Year 2 children in the breakfast club schools finished the year roughly two months ahead in reading, writing and maths. Absence fell, by something like 26 fewer half-days across a class of thirty. Teachers reported better behaviour.
The Bit That Doesn't Fit
If the food were doing the work, the Year 6 children should have gained as well. They didn't. The attainment effect showed up at the end of Key Stage 1 and nowhere else.And the gains were not limited to the children actually eating the toast. The evaluators concluded that the clubs helped children who did not attend, by improving the classroom those children walked into. That is a peculiar thing for a bowl of cereal to manage.
When the EEF reviewed the wider research on breakfast schemes in 2024, it came to a similar place. Across the studies, free breakfast had either no effect on attendance or a small one. Where things did improve, eating breakfast alone did not seem to be what mattered.
What a breakfast club really changes is the shape of the morning. Children arrive twenty minutes early instead of five minutes late. The register isn't interrupted. Nobody walks in at twenty past nine still in their coat, pulling everyone's attention with them. The first lesson starts on time, in a room full of children who have already settled.
The Flaw Worth Knowing About
There is a complication, and it is a fairly big one. The EEF has since had to re-publish that evaluation, because an error came to light in how schools were sorted into the two groups. The sorting was not actually random. Inner city schools, London ones especially, were more likely to end up in the breakfast club group, which means the comparison was never quite fair.The code printed in the original report did not match the code that had been used, so neither the reviewers nor the EEF spotted it at the time. Strictly speaking the study is no longer a randomised controlled trial. Its rating has been knocked down from high security to moderate-to-high.
That does not make the findings worthless. It does mean holding the two months loosely rather than quoting it as settled fact. It is also unusual for an organisation to reopen its own best-known study and say so publicly, which is worth more than most people give it credit for.
What Families Are Actually Getting
Ask parents using the clubs and the answer has less to do with nutrition than the policy debate assumes. The government puts the saving at up to £450 a year and around 95 hours of reclaimed morning time. Nearly half of parents, 45 per cent, now say they would prioritise a school that offers one.That is a childcare answer. A free half hour at 8am is the difference between making an early shift and not making it, and families are choosing schools accordingly.
The universal part matters too, in a way that is easy to miss. Sixty per cent of parents said they are more likely to use help when it is offered to everyone. Thirty-eight per cent of single parents said they feel guilty taking up free support, against 28 per cent of parents with a partner. A club anyone can walk into is a club nobody has to qualify for.
When the Department for Education sent researchers to watch the early clubs and talk to parents, the descriptions were strikingly ordinary. Calmer houses. Less shouting about shoes. Children picking their own breakfast, which one parent thought was the real prize: as a child, she pointed out, you don't get to choose very much.
Easier to Announce Than to Run
The harder truth is that schools are struggling with it. In a DfE survey of 770 primary leaders this April, three quarters said free breakfast clubs were not financially viable for their school. Seven in ten pointed to staffing. Half said they did not have the space.Of the 750 schools in the first wave, around 80 dropped out and had to be replaced, in many cases because the money did not stretch. An unhurried twenty minutes sounds free. It costs an adult being paid to stand in a hall at eight in the morning, five days a week.
None of which means the food is beside the point. Magic Breakfast's 2025 survey found more than a third of teachers see hungry children arriving every day, and nearly half said it had got worse over the year. Morning hunger is real and it is not improving.
But if the thing quietly doing the work is the calm start rather than the cereal, that changes what schools should protect when the budget gets tight. And it is the one part a family can borrow. Most of us cannot do anything about school funding. We can look at what happens in our own kitchen between half seven and half eight, and notice that the difference between a rushed exit and an early one was never about breakfast at all.
Sources & Further Reading