Soda taxes are an example of sin taxes levied on items that hurt public health or other aspects of society. These are also generally per-unit taxes—Mexico, for example, in 2013 started charging consumers an extra 1 peso per ounce for sugary drinks. Berkeley, California, followed in 2014 with a tax of 1 cent per ounce.
Policy makers may want to levy more such per-unit taxes, also known as specific taxes, suggests research by University of Toronto’s Kory Kroft, University of Calgary’s Jean-William P. Laliberté, the Bank of Mexico’s René Leal Vizcaíno, and Chicago Booth’s Matthew J. Notowidigdo. They find that these per-unit taxes are more efficient at promoting public welfare than the alternative.
For more than a century, research has indicated that an ad valorem tax, levied on the basis of the assessed value of an item, as is done for a real-estate or sales tax, raises more money and leads to lower consumer prices than a specific tax, such as a soda tax but also a federal or state per-gallon excise tax on gasoline.
Governments rely on both ad valorem and specific taxes as major sources of revenue, and so far the field has been tilted toward sales and other ad valorem levies. US states collected $336 billion in 2019 through ad valorem taxes and $77 billion from specific taxes. But taxes also affect consumer demand for the underlying products and retailers’ decisions about how to set prices. So from the consumer perspective, which is better?
With this question in mind, the researchers analyzed a 2006–14 sampling of Nielsen Retail Scanner Data hosted by Chicago Booth’s Kilts Center for Marketing. The data covered the top 20 percent of product categories, amounting to 11 million observations from 3,822 grocery stores in 543 counties across the US. The researchers then used variation in sales taxes by state, county, and product category to develop formulas that account for the efficiency of each tax regimen.
By comparing food and nonfood pricing across jurisdictions with varying sales taxes, the researchers find that retailers did not absorb either specific taxes or sales taxes. Instead, they passed the levies directly to consumers. Also, when sales taxes rose on a limited group of products, retailers responded by increasing prices on all products—while hiking prices even more for the specific goods subject to the sales-tax bump.
To measure the effect of sales taxes on consumer demand, the researchers compared product sales across stores, holding pretax product pricing steady by replacing individual store-level prices with average national charges. This allowed them to develop an index of quantity demand on the basis of common prices, isolating the demand change caused by variances in the sales-tax charge. They find that sales taxes—in addition to reducing consumer demand for products—also resulted in a reduction in the variety of products available to consumers, equal to about one-third of the effect of sales taxes on demand.
The finding that specific taxes are more efficient at the margin than ad valorem taxes suggests that more US policy makers should consider using specific taxes on products such as soda as one of the tools for raising revenue.