Schools are allowed to use lotteries for places to make the allocation fairer and avoid discrimination. It's a topic that's been discussed here before - and in my papers in Philosophy and JAP. It seems, however, that Ed Balls is in the news again, complaining about the 'destabilising' effects that lotteries have.
I'd like to see some support for his claim that this is bad for the welfare of the students involved. As someone who went to the same secondary schools as only three of my 63 primary school peers, I didn't find the process too destabilising. Moreover, even if there's an acknowledged cost on this point, it would have to be set against improvements achieved in social justice and exposing students to diversity (no longer are middle class and working class children de facto segregated), which could outweigh any losses by making greater positive contributions.
I can't, of course, say that this is the case any more than Ed Balls can - but simply making assertions or conjectures about the effects of a lottery isn't a helpful basis for informed public policy. We need both more empirical research and a sensible public debate.