I have elaborated, in the previous four chapters, two central criteria for evaluating any set of educational institutions. Education policy should aim at ensuring that every child has a real opportunity to become an autonomous person, and should aim at rough equality of educational opportunity. Equal educational opportunity requires at least the following three things: the quality of the educational inputs in the school system should not reflect the level of wealth of the parents; they should not reflect the decision-making ability of the parents; and children with disabilities should get substantially greater educational resources than children without disabilities.
In this chapter I propose to look at the best evidence available with respect to two quite different choice schemes, and ask of each how it measures up against these criteria. Though none of the evidence is conclusive, it appears that none of them is exemplary with respect to what really matters. But this fact does not, in itself, impugn school choice. There are two reasons for this. The first is that how well a scheme does with respect to some criterion cannot be decisive against it or for it because we have to compare the scheme both with what could be expected if the scheme it replaced had continued, and also with what could reasonably have been expected from alternative feasible reforms. A choice scheme might do very badly with respect to equality, but still represent an improvement on the performance of the previous arrangements, and also be superior to the other reforms which were rejected in favour of it. In that case the absolute weakness of the scheme does not justify its rejection, since it is superior to other options. One consideration I shall advance in favour of choice in the US context—but not, I should emphasize, in the UK context—is that it actually stimulates a dynamic towards equality by helping to shift the bulk of funding away from the local level and toward the state level, where it is politically much more difficult to defend the existing profound inequalities of funding.