Last week's Economist carried a remarkable article about education policy. It looked at the idea of allowing independently managed but state-funded schools, called charter schools in the United States and academies here in Britain, an idea the newspaper has long supported. It reveals weaknesses in the policy. These weaknesses are very revealing.
Independent management of schools has always been more popular with policy wonks than with educationalists. Nevertheless studies were commissioned to show that charter schools and academies performed better than ordinary state-managed schools, usually based on exam results. This is referred to evidence-based policymaking. And it drives me mad. The effectiveness of such an approach depends on the questions you ask the evidence to answer - a process that always entails risks. Schools policy shows two major ones: looking at the wrong question, and making success criteria too narrow.
What's the problem? Research is now showing that, now that the policy is becoming more mainstream, these independent schools generally perform no better, and often worse, than state managed schools, especially in middle-ranking schools. Worse, there is some evidence from the US that where improved exam results have been achieved, they have not actually improved the life-chances of the pupils. This turn of events was entirely predictable. But whereas the Economist innocently suggests that the policy needs to be tweaked to address these weaknesses, the rest of us must ask whether it was ever a good idea to make the policy mainstream, rather than just applied to a few schools to ginger up innovation.
The first problem is that the policy, and the evidence used to support it, was addressing the wrong thing. Changing management arrangements does not change the way schools are actually run. You can set your school up as an academy and carry on just the same way as before. And indeed many schools in England have done just that, so it hardly surprising that their performance is unchanged. The government's current drive to make all schools into academies falls precisely into this trap.; schools do the minimum to comply, and so change nothing important. So what was the point?
And it is no secret what really does make a difference in schools, based on countless studies, often reported by the Economist. That is the quality of school leadership and the quality of teaching. This trumps, money, quality of facilities, and even class sizes. To be fair, policymakers advocate independent management because they think it is the quickest way to fix leadership and teaching (where it challenges the stranglehold of teachers' unions). But they won't necessarily do anything of the kind, and it is possible to replicate their changes to leadership and teaching in state-managed schools, once politicians understand what is to be done. This is what happened London, for example, before the academies programme got going. I could show you some truly wonderfully state-run schools near where I live - and others who are on their way there. It is really hard to see how turning them into academies is going to help.
The problem is that political structures are something that politicians and policy wonks find it easy to talk about and comfortable to fiddle with. I have seen something similar in business management - senior managers playing with organisation structures before they have really understood what the real issues are and how they are best to be fixed. So school independence has now become the chief aim of political policy in English education, and the subject of huge amounts of political effort, which will have little direct effect on the quality of education - and could even harm it.
And public policy journalists, including those at the Economist, have egged the politicians on. They often wrote approvingly of Michael Gove, the British Education Secretary (whose remit is actually just England) from 2010 to 2014, who led the recent drive to academies obsessively, in spite of plenty of evidence that he was barking up the wrong tree. Economist articles are anonymous, but I am sure that Anne McElvoy is at the root of this. She seems to care more about her public media profile than the quality of her journalism.
The second issue I pointed to was that of narrow performance measures. This is secondary in the Economist article, but it has been a huge problem in education policy, and in many other areas too. In education the issue is a focus on test and exam results. The article reports concerns that better results achieved in charter schools do not feed through into the employment market (for example in how much pupils are paid in employment). Their focus has become too narrow on improving the scores, at the expense of life-skills. It is even reported that charter school advocates do not send their own children there. The same weakness has been alleged for English academies, though I am not sure how true this is. The system of Ofsted inspections makes England less vulnerable - the inspectors look at broader issues, especially when conferring the coveted "Outstanding" label. Now a broad education, done well, is not incompatible with excellent test results - the pupils use their improved life skills to improve their learning, and in a highly sustainable way - but it takes top-quality leadership to appreciate this and weather the short-term costs. It remains tempting to short cut this hard road by narrowing the focus.
In fact policymakers should be thinking much harder about the best way of preparing young people for later life, and of recruiting and training top quality teachers and school leaders. And not engage in silly think-tanker debates, for example as to whether schools focus too much on teaching skills rather than knowledge, as Mr Gove was prone to engage in.
And as for school management, no doubt the old ways of state direction of state-funded schools will re-emerge in a new guise. But will politicians and their advisers ever learn the lessons for policymaking, evidence-based or otherwise? Aim directly at the key drivers of success, not just the management structures. And use numerical measures with extreme caution.