The Effect of Age-Specific Sex Ratios on Crime: Instrumental Variable Estimates from India

Using data from 1961 to 2001, the paper shows the impact on crime of two age-specific sex ratios corresponding to pre-marital (ages 10 to 16) and marriageable (ages 20 to 26) age groups in India. To deal with the endogeneity of sex ratios, the paper has used an Instrumental Variable (IV) strategy that exploits district level variation in historical area under wheat-rice cultivation, and time-variation in relative producer prices of wheat-rice. The paper finds that an increase in 10-16 age sex ratio by one female per 1000 males leads to a 1.5% decline in violent crime, and a decline of almost 1% for both non-violent and property crimes. The results are not robust to alternate specifications for the effect of sex ratio in the 20-26 age group. These estimates suggest that the imbalance in the sex ratio in India between 1961 and 2001 have resulted in a 28.5% increase in violent crimes and 21% increase in non-violent and property crimes.