Recently Ed Balls, Schools Secretary, re-announced - probably for the third or fourth time - money for one-to-one tuition for pupils falling behind at school. It's a policy close to the Government's heart, and it's easy to see why. Politically, it's very appealing: who could argue with the logic of extending the benefits wealthier parents can buy for their children to those from hard-working families unable to afford it? In policy terms, it's an important plank in the Government's personalisation agenda.
Perhaps that's why this expensive policy hasn't been given the critical scrutiny it deserves. Just how effective will it be? Could the money be better spent? There are a number of issues with the small print.
First, like many education initiatives in the past decade, it is well-meaning but poorly evidenced. Much of it will be focused on "catch-up" tuition for those failing to make the required progress by the end of each key stage, particularly at the end of primary school and the start of secondary.
But for many children, the failure to progress starts much earlier than the year before they sit their Sats. It is important not to label children as slow learners right from the start; they should be given the opportunity to make progress. But if a child is falling behind after a year of whole-class teaching in literacy and numeracy, it's too late to wait until they are nearing the end of primary school before intervening.
Programmes for struggling readers based on one-to-one tuition are more successful the younger they start: the most effective reading tuition programmes, such as Success for All and Reading Recovery, are aimed at six or seven-year-olds.
And it's not just any tuition that will do. A comprehensive review of the evidence on programmes for struggling readers by the Institute for Effective Education has found that some, despite being expensive, have very little effect. The evidence on the strongest programmes is much better than that on the Government's own tuition pilots.
Moreover, the review found that the gains from tuition have to be consolidated with improvements to whole-class teaching - requiring greater investment in training and continuous professional development for primary school teachers.
Second, this is a classic case of top-down "initiativitis", so sadly familiar to heads. Schools are given ring-fenced funding for two to three years to roll out an initiative; money is then diverted to a new initiative in the expectation that schools will continue to fund the old one; then without that money, schools struggle to deliver it. This means that sometimes a very effective intervention - such as the introduction of learning mentors - is lost.
This is not a sustainable way to reform education. If the Government really wants to improve outcomes for children falling behind, it needs to stop paying lip service to evidence-based policy, and really throw its weight behind it. And it needs to step back. Heads - not Whitehall bureaucrats - are best placed to assess the needs of their pupils.
The costs of poor literacy and numeracy are immense. However, the issue will not be properly addressed until the Government empowers schools to be as discerning as the affluent parents who can afford to buy tuition for their own children.
Sonia Sodha, Head of the capabilities programme at think tank Demos.