This is a comment on this side-discussion and not on the main topic:
It turns out that it’s very difficult to explain the impressive gains in girls’ school enrollment in Bangladesh using the female school stipend program, even though that is an intuitively appealing link. Most of the gains in girls schooling can be attributed to the rapid and vast expansion of the garment sector, which opened up new economic opportunities for girls, and in turn made it more sensible for their parents to invest in their daughters’ education. So it was actually mostly a demand-side push. The details of the research are here: http://faculty.som.yale.edu/mushfiqmobarak/garments.pdf.
Easier to digest media coverage of the work here: http://economix.blogs.nytimes.com/2010/07/21/garment-factories-changing-... .
World Bank blog coverage here: http://blogs.worldbank.org/impactevaluations/stitching-together-the-futu...