This paper investigates the effect of a policy measure that gives secondary schools additional resources for low-ability pupils. Schools are free in deciding how to spend the additional money. I use a nonparametric bounds analysis to estimate upper and lower bounds on the effect of additional funds on exam results of pupils. First I investigate what can be concluded without imposing assumptions; next I layer weak nonparametric assumptions to tighten the bounds. The tightest bounds show that the additional school funds significantly increase the probability that a pupil passes the exam and it significantly increases math and language test scores.
school resources nonparametric bounds education spillovers