Abstract:
The article "Banning the Purchase of Sex Increases Cases of Rape: Evidence from Sweden" claims that Sweden's 1999 criminalization of buying sex increased rape by 44-62% (Ciacci, 2024). This finding is surprising given the flat time trend in reported rapes around the reform year. We show that errors in the paper's statistical modeling and coding produce the main result. Similar problems exist across four alternative identification strategies. In each case, measurement error or deviations between the implemented and stated methods create the illusion of an effect. Our re-analysis finds no evidence that the policy increased rape. The paper's faulty policy evaluation and top Altmetric score in the Journal of Population Economics underscore the importance of retraction.