From September, state-funded infant and primary schools in England will have a legal duty to offer free meals at lunchtime to all pupils in reception, year 1 and year 2. A decision championed by deputy prime minister Nick Clegg, it has spawned a row over how prepared schools are for the change.
The Department for Education’s (DfE’s) decision to extend the provision of free school meals was informed by the findings of [a recent pilot study](https://www.gov.uk/government/publications/evaluation-of-the-free-school-meals-pilot-impact-report. All pupils in primary schools in Newham and Durham – a total of 54,000 pupils across 294 schools – were offered free school meals for two years from September 2009, and the impact of this provision on a group of these pupils’ behaviour, attainment and health was assessed.
The evaluation of this programme – co-authored with colleagues at NatCen Social Research and Bryson Purdon Social Research – found that offering free school meals to all pupils in primary schools lead to improvements in test results at Key Stage 1 (years 1 and 2) and Key Stage 2 (years 3 to 6), though the mechanisms through which this occurred are unclear.
We cannot automatically expect the new policy to provide exactly the same benefits in other areas as it did in Newham and Durham. If more pupils in other areas already take school meals or have more nutritious packed lunches, for example, then the potential benefit of universal free school meals might be smaller. But the government is presumably optimistic enough about the potential benefits to extend the scheme across the whole of England.
Covering the costs
The government has budgeted £450m for the additional provision in 2014-15, plus £150m for renovating kitchens and dining areas and transition funding of £22.5m to small schools. But will this be enough? Or might some schools be forced to pick up the tab for any shortfall?
There are two main types of costs that schools incur to offer free school meals to all infant pupils. The first is the cost associated with providing each additional meal. This might be the cost of ingredients and kitchen staff for schools with their own kitchens who produce their own meals, or the amount charged by a local meal provider to buy in an additional meal.
The second is the upfront “capital” costs, associated with buying new equipment or renovating existing kitchens to enable schools to provide more meals on site.
For the first two school terms of 2014-15, DfE have decided to reimburse schools £268 per additional newly eligible pupil. That adds up as £2.30 per meal, for 134 school days (each day of these school terms) assuming that 87% of newly eligible pupils take up school meals. In subsequent terms, funding will be based on the number of pupils who take-up free meals on the three school census days in October, January, and May.
But in the pilot study we found that the cost of each additional school meal varied between £1.90 and £2.60 across primary schools, and that was just in two local authorities. This means that some schools will receive more from the government than it costs to provide an additional school meal, while others will receive less.
These differences will probably not be random. The cost of providing school meals is likely to be systematically higher in some areas, such as where labour costs are high in areas like London. And there will be some rural ares where there is less scope for economies of scale – that is, where the cost of producing each additional meal falls the more meals are provided.
Where the cost per meal is higher than that provided by the government, there may be a greater risk that the quality of the meals will suffer or that schools will make less effort to encourage pupils to take school meals.
Schools may also face one-off (or short-term) costs of adapting school facilities. The evidence from the pilot suggested that this cost was around £2,400 per school, on average. But again this is likely to vary substantially.
There are around 17,000 primary schools in England, which implies that schools should, on average, receive sufficient funding from the £150m allocated by government, although this will be divided according to pupil numbers rather than need.
But there may be additional ongoing costs that are not covered by the funding that has been set aside. For example, schools may need to hire extra staff to help serve the additional meals at lunchtime. It therefore remains to be seen how much it might cost schools to fully deliver the policy.
Where are the benefits?
How should schools feel about the relative costs and benefits of this policy? There is evidence from the pilot that universal provision of free school meals increases academic attainment. For example the increase in attainment at Key Stage 2 was comparable to other effective education interventions such as the literacy hour, a daily time set aside for reading.
For the poorest performing children in Key Stage 1, there was a 5.3% increase in the likelihood of them reaching the expected level in maths and reading. But at this age there was no significant impact for those children already doing well. And these findings are unlikely to translate perfectly to the roll out of free school meal provision due in September.
Whether the benefits of the policy outweigh the costs and, more fundamentally, whether this is the best use of the additional funding, remains to be seen. Come September there will be a statutory duty for schools to offer free school meals to all infant pupils, forcing them to spend at least some additional money on school meals. It is an open question as to whether that money might be better spent on something else.