As much as Americans like their soda, they dislike paying an unwanted tax even more. When, in 2010, the state of Washington implemented a tax on soda, people who opposed the tax were more likely to reduce their soda consumption than those who supported it, new research finds. This ironic twist could be good news for policy makers seeking to curb unhealthy behaviours, but also undercuts the revenue generating potential of this type of tax.

In the wake of the Great Recession, Washington faced a budget shortfall, and it chose to implement a soda tax to help cover the gap. The measure was always intended to be temporary, but state residents voted to repeal the tax on a popular ballot after just a few months following implementation. At one-sixth of a cent per ounce, the soda tax added only about 12 cents to the price of a six pack of soda. But that was enough to affect consumer behaviour, according to research by Daniel Goetz, an assistant professor of marketing at University of Toronto.

Unlike many soda taxes, Washington’s was positioned as a budgetary measure, rather than a health one. The state’s then-Governor Christine Gregoire was the driving force behind the policy, which was implemented and repealed over the course of just a few months in 2010. The soda tax was proposed during a special state legislative session in April, took effect in July, and was voted down in a public ballot in November. Approximately 60 per cent of public ballot voters chose to repeal the tax, and while it did not turn out to be a lasting a policy measure, it did create a natural experiment in how consumers respond to a policy they dislike.

“Because this was a statewide measure, we had access to detailed voting data for individual precincts. That allowed us to determine which parts of the state supported the tax, and which ones really hated it,” says Goetz. “No other U.S. soda tax has covered an entire state, and a city-level tax just does not provide enough  data on variation in support. But with state-wide information about how people voted, we were able to tease apart how their views on the tax changed their shopping behaviour.”

Goetz correlated voting results with data on retail purchasing from the Nielsen IQ data set, a rich trove of information about U.S. consumer spending and sentiment maintained by the University of Chicago’s Kilts Center for Marketing. Importantly, the tax was applied to the shelf price of soda — the marked price that consumers see before they buy a product — so it was obvious that prices had gone up. This affects shopping behaviour more than when a tax is applied at the register, and the shift in purchasing was significant. In an article published in Marketing Science, Goetz found that stores in areas that opposed the tax had a 53 per cent greater reduction in soda sales than stores in areas that voted to keep the tax.

And while soda taxes have been particularly polarizing in the U.S., Goetz views the results as being less about what was being taxed, and more about how people respond to a policy they disagree with. That has implications for governments considering measures that could be unpopular.

“How much people dislike a tax matters,” says Goetz. “It is important to get a sense of the level of opposition, and not just to understand whether it could be overturned. Once a tax is in place, it could also generate less revenue than expected because people are motivated to change their shopping behaviour to rebel against it.”

The findings align with previous research that has shown income tax evasion is more common among people who don’t agree with a government’s policies, and could also hold lessons about how tariffs will be received by U.S. voters who oppose them. Just as tariffs are strongly associated with President Donald Trump’s politics, Washington state’s soda tax was tied to the political brand of its Democratic then-Governor Christine Gregoire. And like the soda tax, tariffs are built into a product’s shelf price.

“People who oppose Trump are in a similar situation to residents of Washington state who did not vote for Governor Gregoire,” says Goetz. “They might perceive price increases from tariffs to be especially unfair, and want to do what they can to ensure that the unfair policy doesn’t work very well to raise revenue.  And that could actually drive people to buy more domestic products as they seek to avoid buying tariffed goods.”