03

Our identification strategy would be undermined if, for instance, the family planning
policies in place 16-25 years ago were bundled with other policies that affected crime with
an approximate 16-25 years lag. As is well known, the same period saw transformative
market-oriented reforms, and it is conceivable that past “reform mindedness” impacted both
sex ratios and crime. To address this concern, we include as a covariate a measure of the
openness of the province (foreign direct investments, export and import, as share of GDP),
and our main result remains robust. Moreover, if the relationship between crime and sex
ratios were spurious, the sex ratio of 10-15 years old would also be able to predict crime
rates, which we do not find.
Another concern is that the data on violent and property crime reflect province-level “fight
crime” drives rather than underlying criminal activity. To address this concern, we look at
corruption. A general law and order zeal would affect corruption statistics as well; while
there is little reason for the sex ratio of young adults to affect corruption rates