The paper you cite on the benefits of choice in education is utterly lamentable. It starts from the presumption that for most kids another, equally good school is just a short drive away. I live in Northumberland. We have kids here who have to stay in a hall of residence during the week because the only school, for them,is a very long drive away. Choice is often not an option in rural communities - and the same goes in healthcare and a dozen other public services. Leaving aside that difficult detail however, the underlying assumption is that choice will drive up performance,based on a presumption of rational actors making rational decisions in a climate of universal knowledge about the factors that make up a good school. Bluntly, if you believe that, you'll believe anything. Markets(and a choice based education system is a kind of market) function on asymmetries of information, with the profit going to the person who has the most useful stock of information. Choice does not deliver improved performance; it delivers variation, but variation does not, on and of itself, imply improved performance.