Who benefits from appeals to vote? Evidence from a get-out-the-vote (GOTV) campaign in India

Get-out-the-vote (GOTV) campaigns are fairly popular as a policy instrument to increase democratic participation and voter turnout. While most GOTV drives target individual voters by nudging them to vote, little is known about the impacts of generalized group-based GOTV appeals on the electorate. A particularly popular form of GOTV messaging involves 'descriptive norm' appeals which are broadly of two types. The first is a positive frame encouraging turnout, given that peers have voted in large numbers. The second is a negative frame, which seeks higher participation as compensation for low turnout by peer voters. We exploit a unique GOTV intervention from the Indian state of Gujarat, where the independent and autonomous election administration of India made such a negatively framed descriptive norm plea to nearly 25 million voters to come out in large numbers and cast their votes in the subsequent phase of the election. We use a synthetic difference-in-differences (SDiD) econometric technique to estimate the causal effects of this GOTV appeal on voter turnout and vote shares. We find evidence of a resultant decline in voter turnout, which we support with a theoretical model and vignette-based evidence that voters attach a disutility to being told what to do and that such appeals are ineffective in credibly signaling the desirability of voting.