My commitment to transparency is improving over time as I build transparency in at earlier stages in my work. I have made my .do files (Stata code) publicly available accompanying this paper, alongside the datasets I have control over (which is only a subset of all of the datasets, most of which are owned by others but which should be publicly available). To get this to run, please follow the following steps:
1. Download the datasets described in the paper (most of which I cannot provide access to myself)
2. Change the 0_master.do file at the top so that the globals reflect your own file locations

The analysis should then run - but if you experience any problems, please do not hesitate to get in touch and I will be happy to help.

Ben Baumberg Geiger
26/9/2017