Parents know better : sorting on match effects in primary school

I show that parents select schools by considering attributes of the student-school match that improve the learning outcomes beyond average school quality. Using the centralized algorithm for offers to primary school in London, I compare the achievement of students who are as good as randomly enroled in schools ranked differently in their application. Enroling at the most-preferred school versus an institution ranked lower increases achievement by 0.10 SD beyond school value-added among students with similar characteristics. Only a small part of the match effects of parental choice can be explained by student's characteristics such as gender, ability, or socioeconomic status.