Voter Preferences, Electoral Promises, and the Composition of Public Spending

Abstract

We examine whether public spending misallocations can be traced to voter demand characteristics such as political and interpersonal trust, or risk and time preferences. A model of optimal choice under public spending tradeoffs provides testable hypotheses suitable for individual-level evidence. The data come from an original survey in which we presented voters in seven Latin American countries with choices between different types of public spending. Respondents reporting higher mistrust or higher impatience are more likely to choose transfers over public goods; respondents with higher impatience are also more likely to choose current spending over public investment. These patterns appear in two salient policy areas, education and security, and are robust to plausible alternative explanations. Randomized experiments designed to increase demand for public investment have the intended impacts, while also indicating that political mistrust and high impatience attenuate the estimated treatment effects.