The London effect did not just happen without hard work

Chris Husbands
For several years, the outstanding success of London Challenge has been a beacon for school improvers across the nation and beyond. The marked improvement in the performance of London secondary schools in the decade after 2002 has been a clear indication that school systems can be significantly improved for all young people, given commitment, imagination, investment and collaboration.
London schools significantly out-perform schools across England and the best of London’s boroughs – Camden, Tower Hamlets, Hackney – perform outstandingly well. In recent months, this narrative has been unpicked. The Institute for Fiscal Studies argued that the success of London’s secondaries was an illusion caused by earlier improvements in primary schools.
Now, in a more direct assault, Simon Burgess from the University of Bristol argues that it’s not the schools’ success of schools at all: London’s schools do well simply because of their pupil composition. Burgess concludes that “pupil progress on standard measures is significantly higher than the rest of England, 9.8% of a standard deviation. This is entirely accounted for by ethnic composition”. London schools, for Burgess, are not good; they are just lucky.
Burgess’s argument takes us straight to one of the longest lasting and most debated questions in social science: are social changes driven by structure or agency? For those who believe that the answer lies in structure, outcomes are largely pre-determined. To understand any social phenomenon, understand the social context, and unpack the deep social drivers which shape institutions. This view leads directly to Basil Bernstein’s famous dictum that ‘schools cannot compensate for society’. Education outcomes are largely read off from social patterns: class, gender, and ethnic differences in performance ­– which are the result of complex interactions. To improve a school, transform its intake, whilst the most important thing any society can do to improve education is to attend to wider social factors: income distribution, income inequality, standards of health care and so on.
Those who believe in agency take a different view: deprivation is not destiny, as the former Education Secretary, Michael Gove put it. No one is a prisoner of circumstances. Determined actions, by committed and driven individuals, can bring about change. Class, gender, ethnic influences are there but can be overcome, especially by determined individuals. To improve education outcomes, change the way teachers teach and schools work. In education, this view leads to interventions like the KIPP programme, or Teach First (which is explicit about breaking the link between material circumstances and outcomes). Education, on this view, can compensate for society – indeed, it has to.
The differences between structure-based and agency-based approaches are not usually reconcilable by reference to evidence. They are different ways of looking at the world. Burgess’s evidence is strong that structure – the ethnic composition of London schools – helps to explain their strong performance.
But if he is wholly right, how to explain the much poorer performance of schools in Leicester and Bradford? He does argue that Birmingham schools also perform well, using the same analysis of demographic characteristics. But parallels between London and Birmingham aren’t just about structure; they also involve agency. Tim Brighouse was formerly Chief Education Officer in Birmingham and then Commissioner for Schools in London and David Woods was Deputy to Tim in Birmingham and a successor in London.
Then there are those of us, including me, who argue that what needs to be explained is not just London schools’ absolute level of performance, but the rate of improvement: London performance improved sharply after 2002 – more rapidly than the rate of demographic change. Moreover, as Chris Cook pointed out, although London has a number of advantages, including demography and a well-qualified and well-trained teaching force, even its poor white British children, defined as those eligible for free school meals, “can expect to beat poor white children outside London by one grade in three subjects”
The structure/agency debate has persisted for two centuries. I rarely take the risk of quoting Marx in this blog, but in the Eighteenth Brumaire of Louis Napoleon, he got it right: “Men [he really meant ‘people’] make their own history, but not of their own free will; not under circumstances they themselves have chosen but under the given and inherited circumstances with which they are directly confronted”.
Contemporary social theorists have tended to look at the intimate relationship between structures and agency – Anthony Giddens’ developed the concept of ‘structuration’ to describe their interplay. There’s no doubt that the changing make-up of schools in London was a factor: it made some things more possible, for example mobilising community engagement with improvement. LKMCo’s report observed that improvement was more marked in some London boroughs than others. But it’s not enough.
Even Burgess acknowledges (pp 11 and 15 of his paper) that there are measures on which there is a London premium, and which are not wholly accounted for by the composition of schools. It’s the relationship between structure and agency that matters; getting interventions right in relation to circumstance. Put differently, the London effect did just not happen without hard work.