Six reasons why grammar schools don't help social mobility

Theresa May is determined to press on with the expansion of grammar schools, claiming that she wants to increase social mobility. But the evidence on this is clear: grammar schools do not improve social mobility. In fact, they hinder it. The broad consensus among academics highlights the consistency of the evidence.

We know three things about early selection and access for poor kids:

1.    The last great expansion of grammar schools, the 1944 Education Act, did nothing to improve the chances of poor kids accessing grammar schools. 2.    In today’s system in existing grammar areas,  the proportion of free school meals-eligible children in grammars is 3 per cent compared to 13 per cent in other schools. 3.    Even if we only look at high-performing children, only 40 per cent of them from poor homes go to grammars compared to two thirds of the rest.

So the evidence from the previous expansion of grammars and from today’s existing grammars shows clearly that early selection disproportionately disadvantages poor kids, and even those who are high achieving at primary level. Tutoring may explain some of this: more affluent parents seek to gain an advantage for their children by paying for additional teaching resources to train their child to pass the 11+. Yesterday’s report by the Sutton Trust highlights both the extent of the use of tutors and the disparities in tutoring use by family background.

Much of the current evidence compares deprived children to the rest—the bottom 15 per cent to the top 85 per cent. But what about middle-class kids? While the evidence here is limited, given that grammar school places are typically restricted to the top 25 per cent who pass their 11+, there are a large number of middle-class kids who will not get into grammar schools. Will their parents accept the secondary modern alternative?

The argument from the current government seems to be that “selection by ability” is an improvement on the current system of “selection by house prices.” But again, the evidence doesn't support that idea.

1.    Counties with grammar systems create more earnings inequality than similar counties with comprehensive systems. While top earners who grew up in a grammar area earn around 10 per cent more than top earners from comprehensive areas, low earners from grammar, or “selection by ability” areas earn 35 per cent less than low earners from similar areas with so-called “selection by house prices.” 2.    A similar pattern emerges for educational attainment: while grammar-educated children outperform similar comprehensively educated children, non-grammar children in grammar areas perform worse than similar children in comprehensive areas. 3.    Internationally, nine of the 10 best education systems in the work are comprehensive and countries with selective education systems are more segregated, in terms of socio-economic status, that those with comprehensive systems.

The Government must therefore answer two important questions to put forward a credible policy for expansion of grammars: how will the new system make access fairer where previous expansion failed? And how will the other 75 per cent who do not make it into grammar schools be guaranteed a high-quality education?