This paper estimates the causal impact of electing criminally accused politicians and their nature of criminality on crime in India. We exploit the quasi-random variation in the outcome of close elections between candidates with and without criminal accusations to instrument the share of constituencies in the district won by criminally accused leaders. We find that a standard deviation increase in the share of criminally accused leaders in institutionally weaker states leads to a 4.3 percent increase in crime in districts, including crimes against women. The effect is more pronounced when the leaders are accused of serious crimes, indicating that seriously accused leaders have a detrimental impact on society.