The Philadelphia City Council has voted to become the second city in the United States to impose a tax on the sale of particular types of sweetened beverages. The tax applies to sugared soda, diet soda, sports drinks and more, while excluding drinks that are more than half milk or fruit, as well as drinks to which sugar is added such as coffee. The tax will be 1.5 cents per ounce, amounting to 18 cents per standard size can of soda or $1 per two-liter bottle.

Public health advocates often propose taxes on sugary drinks, colloquially known as “soda taxes,” as a means of improving public health outcomes. They argue that such beverages disproportionately cause obesity and that consumers of sugary beverages impose external costs on others through higher medical costs associated with obesity.

The evidence supporting the disproportionate effect of sugar beverages on obesity is not powerful. An article in Obesity Review concluded, “The current evidence does not demonstrate conclusively that nutritively sweetened beverage consumption has uniquely contributed to obesity or that reducing NSB consumption will reduce BMI levels in general.” 

And the externalities of the obese also appear to be minimal. “The existing literature … suggests that obese people on average do bear the costs and benefits of their eating and exercise habits.”

But for purposes of discussion assume that consumption of such beverages does result in obesity and its health effects, which, in turn, create costs for others. Are the taxes a good corrective?

An article in Regulation by California Polytechnic State University professors Michael Marlow and Alden Shiers examined the economics of soda taxes. They find several problems. First, the tax isn’t on obesity and thus taxes the soda consumption of the majority who are non-obese. If obesity creates additional health care costs, the first best solution would be for health insurers to price health insurance for the obese higher and then those affected would have incentive to alter all behavior that leads to excessive weight and not just soda consumption. But under the Affordable Care Act, such pricing is not allowed. Second, the level of taxation enacted by Philadelphia, while significant in a percentage sense, will have very little effect on weight. Every percentage point of tax increase is estimated to result in a body mass index decrease of only .003 points. Coca-Cola on sale at my supermarket is $2.50 for 12 cans of 12 ounces or 1.7 cents per ounce. The tax of 1.5 cents per ounce is about an 86 percent tax on the price and would decrease BMI by about .26 points (86.4 X .003), “a trivial effect given that obesity is defined as a body mass index of at least 30.” Finally, the tax encourages substitution of other beverages that are taxed less. Notably, the tax on beer in Pennsylvania stands at only 9 cents per 12-pack, about 4 percent of the new soda tax in Philadelphia.

There are some positives in the Philadelphia case. The program was not sold to voters as a public health measure, but rather as a means of raising new tax monies. The discussion of the tax and the public spending for which the revenues would be used was explicit. And the tax is a consumption tax rather than a tax on the rich or corporations. To be sure, the tax is on a very narrow consumption base and thus is distortionary, but at least the tax is visible. The voters will see the tax and the public services that result and can make an informed decision in the next election about the tax and its uses.

Cato research assistant Nick Zaiac contributed to this post.