Philly soda tax study sees sales dip

A new study bolsters evidence soda taxes can reduce sales, but whether they influence health remains unclear.

The research found Philadelphia's 2017 tax led to a 38% decline in sugary soda and diet drink sales that year, even when taking into account an increase in sales in neighboring towns.

The results echo previous studies in Philadelphia, Berkeley and other places imposing beverage taxes, although the sales decline was larger than in some previous research. Unlike studies in Berkeley and Mexico, nontaxed drink sales didn't increase during the study, "suggesting consumers were not substituting with these drinks in Philadelphia," the researchers said.

The new results were published Tuesday in the Journal of the American Medical Association. Bloomberg Philanthropies, a charitable group supporting anti-obesity measures including soda taxes, paid for the study but otherwise had no role in the research.

While Americans' soda intake has declined in recent years, physician groups including the AMA support soda taxes as a way to combat obesity. The beverage industry and many merchants have fought back, with court cases and ballot initiatives, with mixed success. Last year, Pennsylvania's highest court upheld Philadelphia's tax, which is levied on distributors.

Philadelphia's tax has generated more than $130 million, which is paying for free preschool programs and other community services. The study has no data on beverage consumption or obesity rates, but lead author Dr. Christina Roberto of the University of Pennsylvania said the sales decline suggests Philadelphians are drinking less soda and is a promising sign.

"I think it's well on its way to accomplishing its goal" of funding preschool programs and other city projects, said Roberto, an assistant professor of medical ethics and health policy. "It has this side benefit of reducing purchasing of these unhealthy drinks. It's win-win news for public health."

The new results are based on sales at nearly 300 supermarkets, big box stores and pharmacies in Philadelphia and its Pennsylvania suburbs. The researchers also looked at Baltimore, which doesn't have a soda tax, for comparison purposes.

They tracked sales starting in January 2016, the year before the 1.5-cents-per-ounce tax was imposed, through December 2017. Per-ounce soda prices in Philadelphia increased by 1.56 cents at pharmacies, .87 cents at big box stories, and .65 cents at supermarkets. Sales volume decreased by 1.3 billion ounces in Philadelphia, or about 50 percent, but that dropped to 38% when counting a 308 million-per-ounce sales volume increase in bordering communities, where prices were nearly unchanged. Soda sales prices edged up, and soda sales volume dipped slightly in Baltimore.

Despite uncertainty over the potential health benefits, a journal editorial stated there's enough supporting evidence "to move forward with adoption of taxes while continuing to monitor outcomes." The editorial authors include Dr. James Krieger, a public health physician and executive director of Healthy Food America, an advocacy group supporting soda taxes and other measures aimed at curbing America's junk food habit.

William Dermody, a spokesman for the American Beverage Association, said the study conflicts with other evidence suggesting a bigger effect from cross-border shopping for untaxed drinks outside of Philadelphia. U.S. drink-makers are reducing sugar in their products and creating smaller bottles and cans -- steps he said make more sense than taxes to help Americans cut their sugar intake.

Soda taxes "hurt working families, small local business and their employees," he said.

The tax has become a flashpoint in the run-up to the May 21 Philadelphia mayoral primary. Mayor Jim Kenney's two Democratic challengers have derided the tax as regressive and are calling for it to be repealed.

In a statement, the mayor's office said the study is "entirely supportive" of the administration's stance on the tax, and added it doesn't support industry claims the tax has led to job losses.