Consumption Responses to an Unpopular Policy: Evidence from a Short-lived Soda Tax

Public policies that restrict or intervene in consumer choices are often controversial. For instance, the compliance rate of COVID-19 pandemic social distancing rules varied dramatically across cities and states, and these policies even backfired among some consumers who strongly disagreed with them. Motivated by these controversies, we investigate whether such rebellious behavior happens in the context of a soda tax. More specifically, we hypothesize that tax opponents will decrease their soda consumption by more than tax supporters, in order to avoid the tax burden. We use precinct-level voting data for a soda tax repeal in Washington State to create a novel measure of store-level tax opposition. Combining this measure with price and quantity data from a large sample of grocery stores, we show that in response to the short-lived Washington State soda tax, stores frequented by tax opponents experience a 60\% greater reduction in quantity sold of taxed beverages compared to stores frequented by tax supporters, even though the tax pass-through is uniform across all stores. Our additional analysis supports that tax opponents' reaction is mainly driven by perceived unfairness, rather than conservative ideology. If this behavioral response is not taken into account, the optimal tax rate may be set too high, resulting in much lower tax revenue. We also derive conditions under which policymakers (firms) find it worthwhile to spend to persuade consumers to support (oppose) the tax.