What Would Happen If Charter School Availability Were Expanded Greatly? Part 2

Chris Walters, who I first mentioned in this post, gave his job market talk at Berkeley yesterday. Roughly speaking, his story is that although some of the lottery evidence in Boston suggests that some Charter Schools can substantially increase test scores, selection is a large concern. He puts some structure on the demand for charter schools and comes up with a surprising finding – the students who’d gain the most from attending charter schools appear not to want to attend charter schools on average. One way of thinking about this is to ask how far people would be willing to drive to go to a good charter school. It seems like those willing to drive the furthest distances are not the students who’d benefit the most from charters, in fact they appear to benefit the least. Thus, in simulations (in which public school quality is held constant – which is a strong assumption that was first pointed out by @Modeledbehavior), he finds that large expansions of charter schools may leave empty desks. He notes that a lack of information (or heterogeneous charter school quality that makes it hard to figure out which ones are good and which ones aren’t great) about the potential gains of some charter schools may affect students preferences, but this underwhelming demand from those who’d gain most is what he is seeing based on the data he has from Boston about who is apply and attending charter schools and a model of demand for charter schools.

About these ads