Giacomo De Giorgi discussion of: Religious symbols in schools

Abstract

This is an important and timely (as of April 2022) paper on a charged political issue: the banning of the Islamic veil in French schools. The authors employ a well-established and appropriate identification strategy to study whether banning the veil has any effects on educational outcomes. They find that such a ban has a positive effect on high-school graduation, an effect of about 10%. In Table 1, the authors present the results of a difference-in-differences strategy using a standard cohort approach. I suggested, and as far as I know this suggestion is partly taken into account in the revised manuscript, that the main specification should actually be a triple difference, with the third difference coming from a gender difference, i.e. males are not affected directly by the policy so that the extra difference would take care of other residual unobservables at the religion-cohort level. This suggestion is also motivated by the smooth increase in graduation for females of Muslim background and the fact that males of Muslim background appear not to be affected by the policy as per the original Figure 1 (Panels A and B).