Who gets targeted for vote-buying? Evidence from an augmented list experiment in Turkey

Abstract

Understanding the dynamics of vote-buying is essential to improve accountability of elections in developing democracies. While list experiments are useful for attenuating social desirability bias associated with measuring vote-buying, they are not conducive to multivariate analyses, and the question of what types of individuals are targeted is left inadequately explored. We overcome this limitation by combining a population-based list experiment with an estimator (LISTIT) that allows for multivariate analyses in an efficient manner. Our analysis suggests that in the 2011 parliamentary elections of Turkey over one-third of the electorate was targeted for vote-buying, which is more than double the proportion willing to admit when asked directly. Additionally, we find that strong partisans of the ruling Justice and Development Party (AKP), less-educated individuals, and urban residents are significantly more likely to be targeted for vote-buying. We present compelling evidence for the hypotheses that parties target their core supporters and socio-economically vulnerable individuals. The strength of our evidence derives from the use of original data on vote-buying that has been collected in an unobtrusive manner and analyzed at the level of individuals.