E59: Hunt Allcott on the Optimum Soda Tax
Today’s guest, Dr. Hunt Allcott, had two recent papers with colleagues Benjamin Lockwood and Dmitry Taubinsky, on whether soda taxes are effective, and how an optimal soda tax might be established. They were published by the National Bureau of Economic Research. These are important papers at an important time, given all the activity around the world on soda taxes.
What are the key ingredients for calculating the economically optimal soda tax?
That’s a great question, and thank you so much for having me on the program. I’m a big fan of the program and I’m a big fan of the work that you’re doing.
The key ingredients for calculating the economically optimal soda tax, I would say, are actually the key ingredients for thinking about the optimal policy in any number of domains. And so, there’s a subfield called public economics that has a framework for thinking about this, and it is a utilitarian framework. It starts with the idea that everybody is maximizing their happiness, possibly making mistakes in doing so, but they’re trying to maximize their happiness or utility, and there is a benevolent policy maker or a government that’s trying to do well by its citizens, but it’s trying to maximize this sum of happiness across all people, and you might call that social welfare.
So the way that we think about lots of economic policy problems, setting optimal taxes on sugary drinks, but also tobacco taxes or income taxes or disability insurance, is from this perspective of a benevolent policy maker that’s trying to maximize social welfare.
And so, what does that framework specifically then say, or what are the key ingredients for an optimal soda tax? I think that the two key ingredients are, really, what share of health costs that are caused by soda consumption, which are those costs that are borne by other people, and then, what you might call internality, are we making some mistake? Are we failing to maximize our own happiness by drinking too much soda, possibly because we don’t know how bad soda is for us, or because we have self-control problems?
Building on those two key ingredients, you, in general, want impose a corrective tax on tobacco or sugary drinks or carbon pollution or anything else, if and only if the consumption generates externalities or internalities. We, in our paper, calculate what we think is the economically optimal tax rate and, for a nationwide tax, our calculation is that somewhere between one and two cents per ounce actually it would be the optimal nationwide tax rate. And so, there’s a sense in which the cities are actually pretty close to the optimal tax that we would set nationwide.
And these calculations are speculative. There’s a whole industry of economists and public health folks that are thinking about what would be the optimal tax on carbon pollution or the optimal tax on tobacco, and this is an early attempt at setting that optimal tax in the context of soda. And so we’ve done our best, and I think there’s more work to be done there to try to figure out what the optimal amount is.
You may have noticed that I was a little bit careful in distinguishing between the optimal nationwide tax versus the taxes that are being set in different cities, and that’s actually a key distinction. Most of the taxes, in fact, all of the explicit soda taxes that we have are at the city or sometimes the county level in the U.S., and when that’s been done, there are some papers that document what economists call “leakage,” in other words, that when you tax soda in Philadelphia, people, when they’re shopping outside of town, start buying more soda outside of town. Similar things tend to be happening in Berkeley and I imagine they’re happening in other cities that have soda taxes.
This leakage reduces the effectiveness of the tax. You know, it’s basically, you’re taxing one good, which is soda in Philadelphia, and demand for another harmful good, soda outside of Philadelphia, rises. And that means that the optimal level of the taxes lower, and the effectiveness of a tax, at any given rate, is lower. And so, it does mean that the optimal city level tax, taking into account leakage, is certainly lower than the optimal nationwide tax that you would set.
And what are the ways in which soda tax design can be improved?
The most common structure of soda taxes in the U.S. is that you would tax it one cent or two cents per ounce of sugary drinks. Now of course, sugary drinks vary in the amount of sugar that they contain. Soda pop has more sugar per ounce than some other types of SSBs. Sometimes ice tea, for example, have less sugar per ounce, but it really is that sugar that’s causing the harm to our health. And so, we would advocate, and some countries have started to do this, that you would set a soda tax that scales with the amount of sugar in the drink, not the amount of liquid that comes with the sugar. And in fact, our team is working on a paper to sort of quantify that point, and other people have made that point as well. So that’s one example, and there are others.
And given this work, do you think soda taxes are effective public health policy?
Yes, absolutely. Absolutely. So one is that it would be cause consumers to, even if they like some kind of sugary drink, it causes consumers to substitute the less sugary drink, so that’s good for their health. And the other benefit is, as you stay on the production side, it will induce faster changes in the product. But it certainly seems like the benefits of a tax are not just that price goes up so demand goes down, but indeed what people infer, what information people start to sense that they’re getting from their government about the types of things that are being packed.
And as an example of this, you may know that there’s a paper by Alex Rees-Jones and Kyle Rozema that is circulating recently, that looks at cigarette taxes, and they look at cities that debated, but in the end, did not pass cigarette taxes. So there was no impact in these places on cigarette prices, yet there was a lot of coverage about the health harms from cigarettes, and there seems to be some evidence that people’s cigarette consumption in these areas actually went down in response to these debates. And so I think that’s clear evidence that the benefits of taxation are not just through the price channel, but they’re also through the sort of public debate and information channel.
You also have another line of work on food deserts, and have published impressive work on that topic. How much effort do you believe we should put into combating food deserts by subsidizing grocery stores in underserved neighborhoods?
It’s a great question, and yes it is a something that we’ve been interested in. We’ve just had a paper accepted at the Quarterly Journal of Economics that touches on this issue. And we got interested in this because we’ve been reading about food deserts and reading about nutritional inequality, just the idea that low income folks aren’t able to eat as healthfully as high income folk, and then also the fact that low income neighborhoods often have more convenience stores and fewer healthy grocery stores with fresh produce. This sort of correlation, this connection between lack of healthy supply and lack of demand for healthy groceries might’ve been sort of a causal thing, where the food desert is causing the lack of demand for healthy groceries.
And so, we were interested to try to find out if that was actually true. When new grocery stores open up, does that actually impact, does that actually cause people nearby to buy healthier groceries? And so, we looked at this, we gathered Nielsen Homescan data, and we combined that with data on all the grocery stores that have entered in the United States over the last 15 years or so, and it turns out that when grocery stores enter in the food deserts, they don’t cause people to eat healthier food. So what does happen is that people shift their shopping towards the new supermarket, but they don’t buy healthier foods.
And I think this is an important fact, and when we think about why we’d want to subsidize grocery stores in underserved areas, I think that this is an important fact to take into account. It doesn’t seem to have much of an impact on healthy eating.
Explore Other Podcasts:
Other Food Industry Behavior & Marketing Podcasts:More
Other Food Policy Podcasts:More
Other Soda Taxes Podcasts:More
Hunt Allcott is a principal researcher at Microsoft Research, an associate professor of economics at New York University, a research associate at the National Bureau of Economic Research, and a co-editor of the Journal of Public Economics.