Why do voters often fail to reject corrupt or criminal politicians? In this paper, I argue that in settings where government institutions are weak and corruption is widespread, criminal politicians can gain control over state resources and use their delivery as a mechanism to buy voter support. To test this theory, I examine the causal effects of electing criminal politicians on India’s largest rural workforce program in the state of West Bengal during the 2011 to 2020 period. Using a regression discontinuity design, I find that the election of a criminal politician leads to a drop in the number of completed projects and a rise in work allocation. This effect is more pronounced for legislators who seek re-elections, are accused of serious criminal allegations, and contest from non-reserved constituencies. The results further indicate that criminal politicians spend a significantly larger portion of the funds on the labor dimension of the program rather than on materials. These findings suggest that criminal politicians use the program to deliver clientelistic public goods to their constituents. This could explain why voters might be willing to support such candidates, despite the criminal allegations against them.

Does the politicians family connections affect the level of effort they exert once elected to public office? Economic theory predicts that political power is often unequally distributed, where certain individual enjoy an electoral advantage over others. One prominent example of this phenomenon is political dynasties, where candidates belonging to political families are persistently elected to political office. While the existing literature highlights why dynastic politicians maintain political power, the consequences of electing such politicians is relatively unknown. In this paper, I examine if dynastic politicians under perform in terms of their effort once elected to political office in the context of India, where members of prominent families often hold political office for generations. I use two main proxy measurements to measure political effort: (i) A field experiment to investigate whether the dynastic ties of the legislator affects their response to citizens’ request about the lack of regular water supply and (ii) the amounts of funds utlilized under the constituency development scheme available to each politician in their constituency. Using a regression discontinuity design, I estimate the effect of electing a dynastic politician on politician effort at the constituency level for all Indian state assembly elections held during the period of 2018 to 2023.