Independent research unit Education Datalab reported last year that: “For pupils who were FSM (free school meals)-eligible, on almost every occasion the school census has been taken, their attainment, relative to the national average, has actually been falling.”

This is a cold hard fact. Let it sink in. During eight years of the pupil premium project, more than £10bn has been spent on it: and yet the policy has floundered, the attainment “gap” simply hasn’t narrowed.

Why hasn’t it worked? A significant part of the answer for me lies in the research I’ve flagged up recently. Schools can only influence around only 20 per cent of a child’s educational attainment. To put it another way, to see pupil premium attainment rise by more than 20 per cent of a grade per student on average would be a miracle of science and genetics. Their home environment and the impact of parenting far exceeds any effect that a teacher can have. Within that context, some pupil premium projects within schools, although well-meaning, are high-expense and low-impact.

Apart from not working, an alarming side-effect of pupil premium has been the temptation for schools and teachers to do some questionable stuff. Given that Ofsted inspectors are told to “take particular account of the progress made by disadvantaged pupils from their starting points”, this is understandable. Teachers have been asked questions such as: “What are you doing differently for your pupil premium students?”, a question that should be “What is your expectation of all students, regardless of their background or parents financial status?”

Pupil premium rethink

One teacher who contacted me on Twitter recently told me: “We have to mark pupil premium student’s books first and put extra comments for them to show how to improve, we give intervention to PP only and “closing the gap” is seen as number one. We were told on 1 September that if something works for a pupil premium student, keep it for them only.”

Not only is it failing, it is adding to an already extreme accountability nexus.

We have to have a radical rethink - £2.5bn per year is a lot of money. There are a number of options I’d like to share from my humble position at the chalkface: 

Option 1 - An emphasis on parenting and home environment 

The importance of schools helping parents to enhance home learning environments is widely recognised. I can’t count the number of times I’ve heard a teacher say: “I had parents evening but no one I wanted to turned up!” We need to try and test different methods at increasing parental involvement with school. One way of doing that is in ensuring they have the tools and knowledge to support their child. 

One example of a successful programme that a couple hundred million pounds could enable might be the Home Instruction for Parents of Preschool Youngsters (HIPPY), which already runs in several countries. It aims to improve the home literacy environment, as well as to enhance the parent-child relationship, to prepare children for school. Aimed at parents of 3- to 5-year-olds, it combines 30 sessions of home visiting and community centre-based support over two years, with daily activities at home. It has been tested in nine countries and studies have shown a positive impact on child adaptation and readiness for school. This is just one programme that could be combined with others.

Figures show that more than 580,000 children known to be eligible for free school meals had not reached the government’s definition of a good level of development at the age of 5. By the time students received their GCSE results, around 32 per cent of the variace in performance can be predicted based on indicators observed at, or before, the age of 5. In other words, the younger the better when it comes to narrowing the gap. With that in mind, whatever the various interventions planned for schools and families, whether they are universal or not, should surely be much more weighted on the early years and on what happens outside the classroom than they are now.

Option 2 - Focus on recruitment, retention and training of teachers

I’d say keep 20 per cent of the £2.5 billion to maintain a “pupil premium lite”. The remainder of the money, let’s call it £2 billion, should be ploughed into the biggest challenge facing the education of all students - recruiting and retaining good teachers. For example, the employment of 9,000 NQTs could suffice. It could provide a £500 annual pay increase for every teacher in the country. Alternatively, targeting teachers in the first five years by paying them a retention bonus of £1,000 each year they stay in the classroom would be achievable. Of course, the primary reason teachers enter and leave the profession isn’t money, it is excessive workload. The £2 billion per year could lower class sizes, increase teaching assistant support and resource schools properly. John Hattie’s hugely respected Visible Learning ranks teacher credibility as seventh in his list of influences on student achievement. Imagine what could be achieved by retaining and continually developing our best teachers.

Above all, teachers in deprived areas should be given extra time away from the classroom to develop their practice. Any teacher who works in one of these areas should be given extra PPA allowance, smaller class sizes and an extra half day per week to plan as well as an extensive CPD package, including a permanent mentor. In cities such as London, they should also be given an extra housing subsidy. Spend the £2 billion on that and let’s see what happens to the achievement of all students in these areas, including those on pupil premium. We know that the wellbeing and efficacy of teachers directly impacts student achievement.

One thing is for sure, the pupil premium surely cannot remain unchallenged. Reform is of paramount importance. Start moving on it, policymakers.

Thomas Rogers is a teacher who runs rogershistory.com and tweets @RogersHistory

For more columns by Tom, view his back catalogue

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