Local excise taxes, sticky prices, and spillovers: evidence from Berkeley’s soda tax

This paper evaluates the price and consumption effects of the first municipal soda tax imposed in the United States. Using high-resolution scanner data and data-driven approaches to select comparison units for counterfactual analysis, we estimate the tax has no effect on prices or consumption at drugstores, but increases supermarket prices of some soda products, constituting a minority of soda consumption. We estimate UPC-level pass through rates and find that there is significant heterogeneity across UPCs, much of which is explained by brand and size; average UPC-level pass through estimates in the supermarket range between 19% and 23%. We find limited evidence of reduced supermarket purchases of soda in the taxed jurisdiction. Half of these reduced purchases are substituted to just outside the taxed jurisdiction. Retailers’ limited price responses are attributed to the localness of the tax; other research studying the Philadelphia soda tax has demonstrated more substantial pass-through in this much larger jurisdiction.