The problem is- it would take a lot of money to really equalise opportunity. And perhaps the upto £90bn that Leurig hypothesises might have better returns at the age of 3 or 4. We should be talking about the future of Surestart as much about secondary schools.
But, given that we are talking about secondary schools, what should a Lib Dem policy look like? Well, we should look at what actually does help equalise opportunity (note: not outcome). The first thing to realise is that selection is unfair, as the party have no unequivocally unfair (with a supposedly right-wing leader!). If you want to go to an academically selective school, by all means spend your money in the private sector or tutor your kid privately (normally the New Labour state school bypass).
Why is it unfair? Well, outcomes are not the result of just natural ability. If you run a factory, part of the best indicators of what your profit level will be, is how much capital an entrepreneur puts in. Equally, a middle class kid (full disclosure: like me. No class war here.) will have more behind him than a kid with no books in his house and a parent on welfare. Even at 11, grammar schools didn't select the 'brightest' kids, they selected those with the most natural advantages.
So, once we eliminate 'selection', what is wrong with the comprehensive system as currently constituted? Well, the answer is that it isn't very comprehensive. You see, it isn't just the private sector that has school fees, the state sector does too. But you pay them to the estate agent rather than to the school bursar.
So, what does determine educational outcomes? Well, if you run a basic regression analysis on the OECD's PISA test results, the answer is that peer effects, parents' education attainment, and parents income, are the three biggest determinates of pupil outcomes. Therefore a pupil premium would work primarily as a price signal to counteract the natural instinct of schools to select the plummiest pupils in order to get a shortcut to the best results. That way, it would mean that rich and poor would genuinely share a classroom. And that, as Robert Puttnam and the Joseph Rowntree Trust have pointed out, would have some major sociological benefits as well as educational ones.
Therefore, I would suggest, we need both more money for schools with pupils from poor backgrounds, and also a voucher system or a lottery so that we have genuinely mixed classrooms. More money is a start, but as educational economists like Eric Hanushek have pointed out, money isn't the total answer either. Thankfully, with thoughtful columns like Tim Leunig's, we'll get closer to finding a policy that truly fulfils our ideals.