Thread by @RiccardoCiacci

I’d like to present the newest version of my paper📜 on the effect of criminalizing the purchase of prostitution on rape. Recently, several countries, such as Sweden, Norway, Canada, Israel or France;
criminalized the purchase of prostitution without empirical evidence on (1/N)

its impacts. This paper is one of the first to provide causal evidence on the effects of this policy.

In the previous version of the article I used IV, fuzzy DID and an event study leveraging variation on fines for sex purchase to estimate the effect of the ban on rape. (2/N)

Hence, none of my results used the ban introduction per se. After attending two conferences in the last month (and receiving many great comments!💪), I added two new specifications exploiting the date in which the ban was introduced in a regression discontinuity framework. (3/N)

Namely, I use a regression discontinuity in time (RDiT) estimator and a difference in RDiT estimator. I find that this regulation increased the number of reported rapes by 44%–62%. It is encouraging to find that results are robust across estimators. (4/N)

Next, I explore several mechanisms and find supporting evidence of a change in the demand of prostitution.

Such findings might be interpreted in different ways. It might be thought they are due to a substitutive relationship between prostitution and rape. (5/N)

Likewise, it might be considered that they are due to prostitution being as “paid rape”. To this extent, further research and better data are needed. I hope this article might raise awareness on the importance of this topic. Thanks for reading! (6/N)

P.S. Writing this paper was as writing five papers🤯 since in order to build a compelling case on these findings I used five different specifications and compared results among them.🫠 (7/N)

Comments are welcome!💃 Specially if you suggest to use a different metric technique!🤣
You can find the most recent version of the manuscript below (N/N) sites.google.com/site/riccardoc…

@causalinf @jenniferdoleac @pedrohcgs @agoodmanbacon @jondr44

Great consideration by @MartinKulldlrff. If we could apply these notions also in this case it'd be great. Any paper might be criticised but first it is key to read it and understand the methodology it uses.

https://twitter.com/MartinKulldorff/status/1542570985563602947?t=JR9rpTTkWoaBxaI7wYKa2A&s=09