Opening a New Supermarket Solves Only 10% of Nutrition Inequality
Bringing healthy food supply into low-income neighborhoods closes only about 10% of nutrition inequality; the remaining 90% is driven by demand — and demand can be changed by taxes and culture.
The video won't play here. Listen to the audio instead:
The argument · tap a timestamp to hear it
Put a supermarket there and you solve only one-tenth of the problem
The six authors (three economists, the rest from marketing, data science, and real estate) conclude that giving low-income households the same products and prices as high-income households — i.e., eliminating food deserts — would close only about 10% of nutrition inequality; the remaining 90% is driven by demand-side differences. This directly refutes the policy assumption that increasing healthy food supply can significantly reduce nutrition inequality. In other words, the problem isn't whether there's salad on the shelf, but whether people will buy it.
— StephenThey pieced together a national diet map
The study links four datasets: the Nielsen Home Scan shopping panel covering 61,000 households, sales data from 35,000 supermarkets (about 40% of U.S. grocery purchases), surveys on nutrition knowledge and preferences, and the opening dates and precise locations of 6,721 new supermarkets across the U.S. between 2004 and 2016, plus data on other retail outlets in each zip code. Stephen says that compared to this dataset, all previous research looks trivial — this is what the "big data revolution" should look like.
— StephenWhen a new supermarket opens, people just switch stores
The causal design is clean: if this were a supply problem, then when a new supermarket finally opens in a neighborhood, residents should flock to it and start buying fruits and vegetables. They don't. What people do is switch from one supermarket to another that's closer, with essentially no change in what they buy. One reason is that they already drive an average of about five miles to buy groceries — the severity of food deserts is not as high as imagined, because people are already used to driving rather than walking to buy food.
— StephenTaste comes from family, not from supermarkets
Angela quotes her daughter Anya's question: when, where, and how does a person learn to eat nutritiously? The answer points to the family and community you were born into. Stephen himself grew up in a "radically free-range organic" family, making bread and yogurt at home, growing vegetables and raising chickens, and he says he was lucky to be raised with a taste for nutritious food. Conversely, if habits are already unhealthy and habits are extremely stubborn, then a fresh supermarket next door can change very little.
— AngelaCalorie labels make people eat more
Mistaking the problem for a supply problem leads policy astray. New York City required chain fast-food restaurants to label calories on burgers, and studies found people didn't eat less — sometimes they ate more — because 600 calories for a $2 burger feels like a bargain, "more bang for your buck." Another paper found that Starbucks' voluntary calorie disclosure sometimes drove customers away (especially men), and they often went to Dunkin' Donuts instead.
— AngelaComplaints about supply shortages are mostly demand shortages
The show solicited examples on Twitter of things "treated as supply shortages but actually demand shortages," and got a flood of replies: fact-based news (there's plenty, just no one reads it), bipartisan politicians (everyone complains there are none, but no one wants their representative to compromise), trade schools (portrayed as low-income and unstable, so consumers avoid them, even though skilled trades are severely short-staffed), sleep (people complain they don't get enough, but actually they actively choose other activities). The common thread: blaming an external supplier is more comfortable than admitting you don't want it.
— AngelaSugar taxes really can change behavior
Angela wanted to buy a can of Diet Coke at a corner bakery, was told it was $4.29, and said she wouldn't buy a second can — the sugar tax worked on her. She compares the mechanism to tobacco: warning labels and those ads have an effect, but many believe taxes are the single biggest factor in reducing smoking demand. The paper's authors and many other economists recommend sugar taxes to combat our widespread preference for sugar, because people respond to costs and benefits.
— AngelaCultural change is harder than policy, but it has happened
Besides the two other channels — making people richer and getting them through college — there's cultural change. Smoking now carries shame and a sense of "I shouldn't do this" in many communities, and watching Mad Men shows this wasn't always the case in America. Angela points out that cultural icons in various fields — the successful people in movies and music — mostly eat well, exercise, and take care of themselves. The question is how to make these role models more visible without making people feel judged, because another universal rule of human nature is not wanting to feel bad about yourself.
— AngelaIn their own words · checked verbatim
exposing low-income households to the same products and prices available to high-income households, in other words, ameliorating the food desert, reduces nutritional inequality by only about 10%, while the remaining 90% is driven by differences in demand
Stephen2:05
What people are doing is they're shifting from one supermarket to a different supermarket, and they will go to the closer supermarket.
Stephen6:19
if you try to solve policy, assuming that the problem is on one side of the equation, but it's on the other, you're probably not going to address that
Angela11:34
all of us human beings, you and me included, would have some motivation to think, well, it's not my fault and it's not my problem
Angela15:46
if you could remove one thing from the American diet and just solve 80% of the problem overnight, it would be sugar
Angela17:47
if the problem of generally poor nutrition were caused by low supply that would be amazing because supply is pretty fixable
Angela22:56
I have to say that habits are hard to break, but not impossible.
Angela27:04
Right. That demand really is dynamic. Demand can change and it's not easy, but it's possible.
Angela28:07
Figures
| Reduction in nutrition inequality from eliminating food deserts | About 10% | 2:05 |
| Contribution of demand-side differences to nutrition inequality | 90% | 2:05 |
| Households covered by Nielsen Home Scan | 61,000 | 4:08 |
| Supermarkets covered by sales data | 35,000 | 4:08 |
| Share of U.S. grocery purchases covered by sales data | About 40% | 4:08 |
| New supermarkets opened in the U.S. from 2004 to 2016 | 6,721 | 4:08 |
| Average driving distance to the supermarket | About 5 miles | 6:19 |
| Price of a can of Diet Coke at the corner bakery | $4.29 | 15:46 |
| Calories in a Big Mac | About 550 | 30:30 |
| Average U.S. Big Mac price in January 2022 | $5.81 | 30:30 |
Glossary
- food desert
- A neighborhood lacking supermarkets and fresh food, with only convenience stores and fast-food outlets.
- external locus of control
- The psychological tendency to attribute outcomes to external circumstances rather than one's own choices.
- Nielsen Home Scan
- A representative survey panel tracking households' actual grocery purchases.
How to listen
People working in consumer, retail, public policy, or behavior-change products; anyone who wants to see how the assumption that "if supply is there, behavior will change" fails in real data.
The listener voicemail at 20:53-22:56 — repetitive and low on information.