This paper documents a reverse gender gap in secondary schooling outcomes
in Bangladesh drawing upon several rounds of nationally representative household survey data. In
terms of enrolment status and years of schooling completed, boys are found to lag behind girls in
the rural as well as in the urban area. Within the urban sample, the gender gap is widest in the
non-metropolitan area. These findings are robust to extensive control for demand and supply-side
determinants of schooling and remain unchanged even when we use a within household estimator.
We consider one hypothesis, namely gender-differentiated response to a conditional cash transfer
program to reconcile the findings of this reverse gender gap.