Co-Location and Crime Reporting: Does a Salient Crime Impact Subsequent Crime Reporting in

The scale of the protests, and the intensity of the public’s opprobrium of the police and the administration of Delhi, was unprecedented. [...] First, we calculate the difference-in-difference (DID) statistic, which is the post- 15 intervention difference between the actual and the synthetic minus the pre-intervention difference between the actual and the synthetic; we then calculate the DID rank, which is the ranking of the absolute value of the magnitude of the DID of the treatment state (colocation) against all the placebo (co-located. [...] The interpretation of a large DID is that the difference in the post- intervention outcome between the actual and the synthetic is much larger compared to the same during pre-intervention. [...] We also calculate the p-value of the DID statistics and the donor probability – the probability of obtaining a DID as large as that of Delhi’s if one were to assign the intervention at random in the data. [...] In columns 1-4 of panel A of Table 4, we present the pre-intervention characteristics of the synthetic (collocated in Delhi) of each outcome along with the characteristics of the actual Delhi.