Pornography and social ills: Evidence from the early 1990S

Beginning from the hypothesis that private post office boxes accommodate consumption of pornographic magazines by lowering some aspects of the cost (risk of social stigmatization) associated with the purchasing of such items, I demonstrate that a positive correlation between the abundance of such boxes and the subscription rate to Penthouse magazine across markets in the United States can be observed. I then proceed to estimate the effect of pornography on violent sex crimes and family instability, with and without using P.O. Box availability as an instrumental variable. Results suggest that unobservable population characteristics severely bias upward the estimated harmfulness of adult magazines. My OLS estimates imply, like several previous studies, that consumption of pornography contributes to both higher frequencies of rapes and divorces. When instrumental variables are employed, however, the correlation between rapes and pornography turns negative while the statistical significance of the coefficient for pornography on the rate of divorces disappears.