Punishment and crime: Impact of death sentences on local crime rates in India

Abstract:

Does the severity of punishment deter crime when the probability of conviction is low? We answer this question in the context of India, a country which retains the death penalty but where conviction rates are low. We link a new database of 2,097 death sentences passed by district-level trial courts from 2000-2021 to district-level crime records to estimate whether the issuing of a death sentencing event in a district affects crime in subsequent years. By using sentencing data rather than data on executions, as is common in much of the literature, we are able to generate a considerable amount of sub-national variation in sentencing events, which are arguably more likely to be reported in the local press and affect local crime rates, compared to executions that take place rarely and many years into the future. We find no deterrent effect of sentencing events on murder or rape. Our results are robust to the use of estimators that correct for heterogeneous treatment effects and with a matched sample that compares observably similar districts to one another; nor do we find any evidence of spillovers to neighbouring districts. While we cannot directly measure the reporting of sentencing events in the local media, we find no evidence of heterogeneity in treatment effects by proxies of the salience of sentencing events. On whether the deterrent effect varies by state capacity, we do find a higher impact on murder in districts with relatively greater police strength, but not in districts with higher court capacity or murder conviction rates. Taken together, our results suggest that increasing the rate of capital sentencing will not reduce the crimes it targets through deterrence.

Website:  Ashwini Deshpande - Ashoka University