Faced with rising obesity and diabetes, health authorities have toyed with the idea of taxing unhealthy foods. There are many arguments for and against the idea, not all of which can be scientifically tested. However, at least we now have a good answer to whether sales are affected.

Philadelphia’s 1.5 cents per ounce (53 cents per liter) tax on sweet drinks started on January 1, 2017. The first study of its effects compared sales in the following year with the previous one, and found a billion fewer ounces of drink were sold – equivalent to 83 million cans. Perhaps a few people stocked up in December 2016, but that couldn't account for more than a tiny fraction of the 38 percent fall in sales observed across 291 retailers in and around Philadelphia. Stores just outside the city limits increased sales; within the city itself, the reduction was 51 percent.

The Journal of the American Medical Association reports Baltimore, which was used as a control, had minimal drink sales reductions over the same period.

Opponents of the tax question whether it is the government's job to use taxes to dissuade people from activities they almost certainly already know are unhealthy. They also note that, as a very cheap luxury, sweetened drinks are consumed more heavily by people on low incomes, whose disadvantage may be increased by the tax.

There are further questions about the merits of taxing only drinks, rather than foods, and whether it was appropriate that Philadelphia included artificial sweeteners, whose health effects remain heavily debated.

Dr Christina Roberto of the University of Pennsylvania, the study's first author, expressed no such doubts. "Taxing sugar-sweetened beverages is one of the most effective policy strategies to reduce the purchase of these unhealthy drinks. It is a public health no-brainer and a policy win-win," she said in a statement.

Six other cities have now passed similar taxes, with a study of a smaller tax in Berkeley finding a 10 percent reduction in overall sales, but a big decrease among lower-income residents. Connecticut and Colorado are considering taking the tax statewide. Besides the potential health benefits, Philadelphia is using the money raised to fund childcare, community schools, and recreation facilities, targeting communities where sugar consumption is highest.

Intriguingly, while the pharmacies surveyed passed on the full price increase to customers, supermarkets and mass merchandisers swallowed around half the cost themselves, yet it was supermarkets that saw the biggest fall in sales. In an earlier study, Roberto and colleagues found the tax had no measurable effect on employment.

A reduction in sales is certainly necessary, but as an accompanying editorial noted, "it is still unclear if these taxes improve health outcomes.” Evidence of that is likely to take much longer.