Even before the very first free schools had opened in September 2011, they had already made their mark as a divisive issue.

Supporters claim that they encourage innovation, raise standards, and are popular with parents. Critics say that they create school places where they are not needed, lower standards through the use of unqualified teachers, and increase social segregation. During this year’s general election campaign the Conservatives pledged to increase the rate at which they were opening. Labour, on the other hand, vowed that no new free schools would open.

There is a clear need for a more balanced understanding of the impact that the free schools programme is having - especially when we consider that the programme is estimated to cost almost £10bn in capital terms alone by 2021. Today, in a new report, the Education Policy Institute has produced one of the most detailed, impartial assessments of this recent reform of the school system.

The report analyses the coverage of free schools across England, considers whether they have helped to add capacity to areas in need of additional school places or higher quality school places, looks at whether they are popular with parents, their demographic composition and, finally, emerging findings of their impact on pupils’ attainment and progress.

Free schools’ impact

Despite the controversy and rapid expansion, free schools still represent just 2 per cent of all state-funded schools. Two-thirds of areas in England are not within reach of either a primary or secondary free school. In other words, the majority of the country has not yet seen any direct impact of the free schools programme - good or bad.

Free school growth has been greatest in the areas most in need of new school places, but we also find significant numbers in areas where there is already an excess number of places. Furthermore, the programme, particularly at secondary level, has been less successful in addressing areas of underperformance.

It is difficult for the Department for Education to meet both of these objectives. Over the coming years, the greatest need for new places will be in London and the South East, but this is where performance is already high. What is more, the free schools programme is demand-led and the DfE relies on getting applications.

If free schools are to play a role in tackling long entrenched underperformance, the DfE will have to do more to encourage applications from these areas.

Social segregation

The government is right to say that free school places have been created in areas of high disadvantage. At primary level, places have been created in the most disadvantaged areas at nearly twice the rate of the least disadvantaged.

But there’s a catch.

The proportion of pupils that are eligible for free school meals in free schools is broadly average. If anything, primary free schools have rates that are below average. This suggests that free schools are not attracting pupils from low-income backgrounds at the rate you would expect given the communities they serve. Our analysis of reception aged pupils shows that in the most disadvantaged areas, 32 per cent of children are eligible for free school meals, but this falls to 24 per cent in free schools.

While it’s too early to paint free schools plainly as a welcome or undesirable contribution to the school system, there is one thing we can be certain of: unless the free schools programme significantly shifts its focus, the government is set to encourage an increase in social segregation in our schools.

Jon Andrews is the director of School System and Performance at the Education Policy Institute.

Want to keep up with the latest education news and opinion? Follow Tes on Twitter and Instagram, and like Tes on Facebook.