Tory ideology ignores evidence that selection undermines the prospects of the poorestIt is one of the worst kept secrets in Westminster that education secretary Justine Greening is not the biggest supporter of the policy that is now the social mobility "flagship" of Theresa May's government - expanding the number of grammar schools.
Greening must be aware of the clear UK and international evidence that selective education both fails to raise overall standards, and undermines the prospects of poor children. Education Policy Institute researchers last year analysed the government's own schools data and drew two key conclusions. First, that almost no children on free school meals get into grammar schools - a risible 4,000 out of more than eight million pupils in the whole of England. Second, that although there is a small benefit for pupils who are admitted to selective schools, this is offset by the worse results for other pupils in areas with a significant number of grammar places.
Greening is nevertheless doing her best to support the prime minister's ill-judged policy. Last week she made a speech in which she sought to divert attention away from the poorest quarter of children, to a so-called "ordinary working families" group (previously JAMs - Just About Managing). This more affluent group, argued Greening, does far better in securing access to grammar schools. The education secretary is quite right - but only because these children are already doing better at school.
But if these OWFs are doing relatively well, why is the government talking so much about them? It is the poorest quarter of children who really lag behind. Two-thirds of these poorest children already fail to secure the modest benchmark of five C grade GCSEs, including English and maths. That "failure" rate will rocket to 80% when the new, more challenging GCSE standard is introduced this year, so it makes no sense educationally to move the emphasis away from these poor children. So what is driving this new government narrative? It is tempting to conclude politics, not education.
The poorest children are very unlikely to gain from any solution involving a selection test at age 11. By then, 60% of the disadvantaged gap has already emerged - meaning these children are on average 10 months of learning behind their peers. To give these children a chance, the government needs to improve the quality of early years education, increase the number of excellent primary schools in poor areas, attract and develop more high quality teachers, and protect pupil premium funding from the coming budget squeeze.
Nor are quotas likely to be the answer. They would be highly controversial, and to make a material difference they would require grammars to admit poor pupils of much lower performance. Do we really want classes of lower-attaining, predominantly poor children taught in a segregated stream?
Greening is loyally trying to make the best of a policy that all the evidence suggests should be buried. Scarce time spent trying to make the unworkable work is diverting her from the important efforts she is making to improve teacher training and target the areas of greatest need.
Education policy should be driven by evidence. It is surely now sensible for Downing Street to drop the grammar plans from the Queen's speech. If not, Greening, who could be a rather good education secretary, will waste her time and talents on a policy which is likely to make our country a more unequal place.
David Laws was schools minister from 2012 to 2015 and is executive chairman of the Education Policy Institute