This paper examines how consumers alter their behavior due to a local tax policy change aimed at dealing with the potential health hazards of sugar consumption in soda beverages. Using panel data of product purchases from university residence halls, restaurants, and retailers, we measure the consumption effects of a soda tax campaign and election in Berkeley, California. Our approach has two parts: First we use a difference-in-difference model estimating the change in soda consumed relative to the change in consumption in control product categories. Our results show that the campaign, and in particular the election, causes soda consumption to significantly drop. Second, we estimate a structural model for beverage demand as a function of attributes. We find that soda is an inelastic good, which would imply that a price increase due to a tax would not lead to a significant drop in demand. Our findings have interesting policy implications, suggesting the effects of media coverage and election outcomes on attitudes and behaviors may be larger than the effects of the soda tax itself.