Local Taxes and Economic Voting: Evidence from City Ballot Measures

Do voters punish local politicians for raising taxes? In California, proposed tax increases must be approved via local ballot measures. Despite the fact that voters themselves are responsible for tax increases, incumbent politicians might still pay a price if voters aren't happy with how city officials spend the new revenue or if voters mistakenly blame them for the tax hike. Using a regression discontinuity design that exploits the narrow passage of local tax initiatives, I find that incumbents don’t generally suffer a penalty when voters raise taxes, with the notable exception of business taxes. I uncover suggestive evidence of a mechanism: with most types of increases, cities do not actually generate more revenue from taxes in subsequent years, suggesting that politicians might strategically be adjusting their revenue generating activities to avoid alienating voters. But after new business taxes are passed, cities do raise significantly more revenue over the next four years, and incumbents subsequently suffer an electoral penalty the next time they run for office. This research adds to a growing body of literature that studies ``attribution errors'' in economic voting and also suggests that cities might be even more fiscally constrained than previous literature assumes if newly authorized taxes generally fail to increase overall revenues.