By January of this year, seven American cities had implemented so-called soda taxes of one to two cents per ounce on sugary beverages. And just about every week there’s a new study out on whether or not they’ve been effective at improving public health outcomes and generating revenue for localities. Verdict? That depends a lot on which side of the intervention debate you favor.

One thing we do know, however, is that if the beverage industry has its way, there will be no more of these taxes.

In recent years, as evidence has mounted that drinking sugar-sweetened beverages is directly linked to diabetes, obesity, and other chronic, diet-related conditions, beverage companies have tapped into Big Tobacco’s playbook: deploy lobbyists who can convince lawmakers to enact preemption laws that stop mayors from enacting exactly this kind of local legislation.

And how do the lobbyists do it? Not by diminishing the health outcomes, but instead, by raising a whole host of other issues. Beverage company lobbyists say, for example, that soda taxes are paternalistic, and restrict consumer choice. They’ve also called the taxes “regressive” because they disproportionately affect low-income and middle-class Americans.

While that all might be true, a group of economists says that’s looking at the issue through the wrong lens. Rather than ask whether or not they work, they say, we should unite all the various debates around soda taxes, and evaluate the “overall economic rationale” for imposing them.

On Monday, Hunt Allcott of NYU, Benjamin Lockwood from the University of Pennsylvania, and Dmitry Taubinsky from the University of California, Berkeley  released a new analysis of soda taxes, showing that they generate “net benefits” for society—that is, their benefits outweigh the drawbacks. More than that, they found, those state-level preemptions are particularly harmful.

Here’s how they came to their conclusion: The group added up the cost to society of health effects related to drinking sugary beverages, the value of the “enjoyment” that people derive from them, and the financial impact that taxes have on the poor. What they found was that implementing an “optimal” soda tax of 1.5 cents per ounce would ultimately outweigh all of those negative effects.

“While there is considerable uncertainty in these optimal tax estimates, the optimal tax is not zero and may be higher than the levels in most U.S. cities to date,” they write.

How did the economists arrive at that particular number? We already know that drinking soda and other sugary beverages is linked to diabetes, obesity, and heart disease. And those conditions can result in medical bills that are paid, in part, by taxpayers and insurance companies. Over 10 years, according to studies published by Health Affairs in 2012 and the American Journal of Preventive Medicine in 2015, soda consumption can result in health problems that rack up between $17 and $23 billion in insurance costs. Those costs are known, in economist-speak, as “externalities,” or the costs absorbed by others for a person’s individual behavior. Cigarettes, we might say, have their own externalities (cancer caused by second-hand smoke, for instance).

That’s why, the group says, concerns over the “regressivity” of a soda tax are exaggerated. It’s true that the financial burden of sales taxes are disproportionately felt by the poor, as some soda tax opponents, including Vermont’s Senator Bernie Sanders, have pointed out. But by the same logic, they could also benefit from greater reductions in heart disease and diabetes.

Because soda taxes are an overall good, Alcott implores lawmakers to stop passing statewide bans on soda taxes, as they’ve done in Arizona, California, Michigan, and Washington. Those bans, Alcott says, aren’t economically justifiable, and harm cities. More than that, he thinks that states should go ahead and implement their own taxes. That would be a preemption of a different kind.

There’s a simple explanation: a net beneficial policy at the state level would cover more people than one at a more local level. And here’s a real-life example: In Berkeley, California and Philadelphia, Pennsylvania, there has been ongoing evidence of “leakage,” or shoppers heading to outlying suburbs to buy their soda tax-free. In California, perhaps, a state-wide soda tax would prevent that from happening. But it wouldn’t do much to stem the tide of shoppers leaving Philadelphia for cheap drinks which, as Tamar Haspel from The Washington Post has pointed out, borders New Jersey.

There’s an analogy there, too. I’m from Chicago, a big city that’s situated near several states with different taxes and rules. Time was, my dad told me, he’d drive an hour to Indiana to buy cheap cigarettes. He doesn’t do that anymore, but he still drives over to the Hoosier State to buy fireworks, which aren’t sold in Illinois.

The report concludes with a suite of policy recommendations for lawmakers. Besides the suggestion for action in statehouses, the group suggests that taxes should be levied per grams of sugar not per ounces of liquid, and that diet drinks and fruit juice should be included in the tax if lawmakers can’t prove that the “additional vitamins and nutrients” in those beverages offset the negative effects of the sugars.

You can read the report in its entirety here.