The Regressive Effects of Direct Democracy: Evidence from Property Tax Assessors

Abstract

What effect does increasing opportunities for participation have on policy outcomes? While formal models predict progressivity (Acemoglu and Robinson 2006; Besley and Coate 2003), empirically, there are large biases in who actually participates in politics (Verba, Schlozman and Brady 1995). Exploiting a New York law requiring the appointment of tax assessors in towns with over 10,000 residents, I show that a particular form of direct democracy – the direct election of tax assessors – makes local tax policy more regressive.