In the Riverdale neighborhood on the far South Side of Chicago, there’s a store with a sign out front that advertises a “full line of groceries.” Shoppers can pick up boxes of cereal and flavored oatmeal for breakfast, ramen for lunch, and sloppy joe seasonings or bottled tomato sauce for dinner. But there are no fresh fruits or vegetables here, nor at a store a few blocks west with a sign in the parking lot that promises “Village Groceries.” There are, however, plenty of packaged desserts and snacks. The nearest conventional grocery store is more than a mile away.

The community is widely considered a food desert, an area lacking in easily available fresh fruits, vegetables, and other whole foods. The description—which refers to both low-income city neighborhoods and rural areas—generally applies in the United States to census tracts that as of 2019 were home to about 40 million people, or nearly 13 percent of the US population, according to one definition.

Policymakers have long been concerned about the effect of these low-income and low-access areas on nutrition and diet-related conditions such as obesity, diabetes, cardiovascular disease, anxiety, and depression. But is the problem the supply of fresh food or the demand for it? That’s a long-standing question. “Policies to encourage the supply of affordable and nutritious food . . . will not affect residents’ health if they do not change their food-purchasing behavior,” observed the US Department of Agriculture’s Michele Ver Ploeg in a 2010 article. Nearly 15 years later, policymakers continue to focus on supply. This April, the governor of Illinois announced grants of up to $2.4 million “to encourage the establishment of new grocery stores in USDA-defined food deserts.”

If the demand explanation is significant, it points to a much tougher policy challenge: how to change consumers’ shopping habits to improve diets. And that challenge goes well beyond low-income communities. Even in well-stocked supermarkets, many American shoppers still choose to load up on ultraprocessed foods such as sweetened cereals, packaged snacks, industrialized breads, and microwavable dinners. Ultraprocessed food makes up close to 60 percent of dietary energy consumption in the United States, the most among higher-income economies, according to a 2024 BMJ review of multiple studies by a team of international researchers led by Deakin University’s Melissa M. Lane. The comparable figure in Australia is 42 percent; in Italy, it’s just 10 percent. Less than 10 percent of American adolescents and adults eat enough fruits or vegetables, according to the Centers for Disease Control and Prevention.

Where healthy food is hard to find

By one measure, about 40 million Americans (representing 13 percent of the population) live in low-income census tracts where many people have scarce access to healthy foods. These tracts can be found in both rural and urban areas across the country.

This American exceptionalism has a high cost for the US health system. In the US, obesity (which afflicted four in 10 Americans in 2020, according to the CDC’s National Health and Nutrition Examination Survey) costs about $170 billion per year, the CDC estimates.

Some academic research is adding rigorous data to the evidence that demand is, in fact, a crucial piece of this puzzle. In which case, what can policymakers do to improve diets in Riverdale and across the country?

The ultraprocessed problem

Consumers are barraged with mixed messages about food—what is healthy, what isn’t, which fad diet they should follow today (low-carb, low-fat, high-fat, carnivore) and which is no longer in favor. Most nutritionists say there is a place for all foods—including fries, frozen pizza, and ice cream—in any balanced diet. Eating hot dogs and chips at your Fourth of July barbecue is not a problem for most, and no one should be expected to replace a birthday cake with a bowl of kale. Even having nightly dessert after a balanced dinner does not have to pose a problem for your health.

However, most also agree that diets on the whole should lean more toward foods that are fresh, such as raw produce and legumes, or only lightly processed, which can include choices such as canned beans, Greek yogurt, olive oil, and nut butter. And they should lean away from ultraprocessed foods, many of which contain high amounts of artificial flavorings, colorings, sweeteners, and preservatives.

Nutritional value within this category can vary, but the biggest offenders contain “often chemically manipulated cheap ingredients such as modified starches, sugars, oils, fats, and protein isolates, with little if any whole food added,” per an editorial by Carlos A. Monteiro, Eurídice Martínez-Steele, and Geoffrey Cannon from the University of São Paulo, published alongside the BMJ review. Research led by University of Michigan’s Ashley N. Gearhardt and also published in the BMJ finds that items such as chips, cookies, sugary cereals, and packaged rolls might be classified as addictive substances, just like alcohol and cigarettes.

Diets that lean too far toward tasty but bad-for-you foods have a high cost for patients as well as health systems. The BMJ review finds that diets high in ultraprocessed foods are associated with an increased risk of type 2 diabetes, cardiovascular disease, obesity, anxiety, and depression. The researchers note that, broadly, more controlled trials would be needed to better establish causality between this type of diet and outcomes related to these serious health problems, but, says Lane, “setting up trials testing the effect of long-term exposure to diets rich in ultraprocessed foods on hard disease endpoints such as cardiovascular disease or cancer won’t be possible, for obvious ethical reasons.”

This issue of diet is of particular policy concern in poorer communities, both urban and rural, where lack of car ownership, among other factors, can keep people from quality grocery stores. A 2023 study by five American Cancer Society researchers finds that lack of accessibility to healthy foods, in terms of distance and transportation, was associated with lower life expectancies at birth.

Many policymakers have focused on access. The USDA in 2022 partnered with Reinvestment Fund, a mission-driven financial institution, to invest nearly $23 million in improving access to healthy foods—and this April announced another $40 million in grants “to provide financing and technical assistance to food retailers in underserved communities.” Ridding the country of food deserts was also a key part of first lady Michelle Obama’s Let’s Move! initiative. In 2011, speaking at a mayor’s summit in Chicago, she noted, “It’s not that people don’t know or don’t want to do the right thing; they just have to have access to the foods that they know will make their families healthier.”

The importance of demand

But some research casts doubt on this line of thinking. A research team that included Chicago Booth’s Jean-Pierre Dubé studied grocery purchases across 60,000 households and about 35,000 stores throughout the US from 2004 to 2015, when more than 1,000 grocery stores opened in areas meeting the government definition of food deserts. The data confirm what shoppers have previously indicated on self-reported surveys: there is a large nutrition gap between the wealthiest and poorest households.

But placing a supermarket in a food desert did not have a substantial effect on food choices. The researchers cite data from the 2009 National Household Travel Survey to note that people who lived in food deserts still shopped predominantly in supermarkets offering healthy products, even though it required traveling to retailers outside of their area. These residents overall traveled an average of 7 miles, and those in households earning less than $25,000 a year traveled an average of 5 miles each way to buy groceries and other items. Even low-income residents of a food desert without a car traveled an average of 2 miles to shop at a grocery store. (According to the USDA Economic Research Service, of the almost 40 million people living in census tracts defined as low-income and low-access in 2019, just over half of them have did have some access to a food store.)

Diversity in diets

A review of nearly 100 studies across countries finds that in the United States and United Kingdom, more than half of the energy intake of the average individual was derived from ultraprocessed foods.

“People do start shopping in a new supermarket when it’s closer to them,” Dubé says. However, “the composition of their shopping basket does not change, and they do not start buying healthier items.”

The research notes a disparity in food preferences between higher- and lower-income consumers. As household income rises, the willingness to pay for healthy macronutrients such as protein increased—while it decreased for unhealthy macronutrients such as saturated fat. Relative to those with lower incomes, households making more than $70,000 per year were willing to pay almost double for vegetables and nearly three times more for fruit, the researchers estimate. By contrast, households making less than $25,000 per year had a higher relative preference for sugar and saturated fats. “All income groups value healthy groceries, but the highest-income group is willing to pay the most, making healthy eating a normal good,” they write.

In economics, a normal good refers to something people buy and consume more of as their income rises. “From a policy perspective, determining that good nutrients are ‘normal goods’ confirms that eating healthy is a preference of the wealthy,” says Dubé.

So if the point of the government initiatives to eliminate food deserts is to inspire healthier eating among lower-income consumers, opening supermarkets is not the answer, he argues. Rather, 90 percent of the problem, his research finds, is due not to supply but to food preference—and addressing that means confronting habits shaped and reinforced by poverty. If you aren’t used to eating fresh produce, availability won’t necessarily make you want it. And if you’re working long hours, with time, money, and energy in low supply, you might well reach for a prepackaged meal rather than cook.

The suggestion that demand, not access, is the major factor is perceived by some people as out of touch. Dubé recalls a 2010 nutrition conference at Booth at which a scholar from the National Institutes of Health advocated for investing millions of dollars to get fresh produce into bodegas in areas that lacked healthful food choices. “I asked the question about whether local consumers actually want fresh produce,” he says. “I was accused of being elitist.”

But Dion Dawson, the founder of Dion’s Chicago Dream—a nonprofit food service he started in 2020 that aims to “improve health outcomes and center health equity” by providing free fresh fruits and vegetables to underserved communities in the city—says he agrees the demand issue is an important factor. “I tell people all the time, even what we’re seeing in our in-house data collection and evaluation is there’s a lack of education because there’s a lack of familiarity with healthy food,” he says. “A lot of times, you have hands-on learners who never had a consistent relationship with healthy fruits and vegetables.”

A carrot to buy a carrot

What would change that? For an economist, the answer is incentives. If someone had an economic reason to buy a banana—such as a discount—that could change the equation and create an attention effect. “Let’s say you have a habit that was developed through the process of living in a food desert,” says Booth’s Alex Imas. “All of a sudden, you go to the store and there’s a bunch of bananas; you don’t even look at them because you’re not used to buying them.” However, he says, an economic incentive would be “almost like a nudge, like, ‘Hey, maybe you should be considering these bananas.’”

Government policies could do more to help lower-income communities in this realm, suggests the research by Dubé and his coauthors. They estimate what would happen if federally funded food benefits, often known as food stamps, could be used only for healthier foods—and find that it would be a more effective and economical way to reduce the nutrition gap.

A diet lacking in proper nutrition can lessen lifetime earnings and aggravate inequality.

A study conducted by Andy Brownback from the University of Arkansas, Imas, and University of Oregon’s Michael Kuhn notes that the US government–backed Supplemental Nutrition Assistance Program tends to focus more on food security—that is, the availability of any food—than on nutritional quality. A different research project, led by Simone A. French from the University of Minnesota, finds that scores in the USDA’s Healthy Eating Index were actually lower, and empty-calorie purchases higher, among low-income households enrolled in SNAP as compared with both higher-income households and poor households not enrolled in SNAP.

The research by Brownback, Imas, and Kuhn indicates that instead of making a change that imposes restrictions on shoppers, benefit programs might become more effective in encouraging healthy eating if they gave consumers an incentive to purchase fruits and vegetables and whole foods, as well as the time to consider their choices. From March 2018 to April 2019, they partnered with market-research platform Field Agent on a study that involved low-income consumers across the US, obtaining access to information about behavior before, during, and after grocery shopping trips. The study had two parts and a total of five treatment groups, with about 150 people in each group. All participants had a gross household income of less than 185 percent of the federal poverty line. (The cutoff was $25,750 annually for a four-person household in 2019.)

The researchers focused on three key questions, all related to how offering targeted subsidies could more effectively lead to healthy food choices. Would it help to give shoppers an active choice over the subsidies? What about imposing a waiting period to encourage or allow shoppers to deliberate over their subsidy choices? Finally, should shoppers have to make their subsidy choices before going shopping, rather than when already in the store?

Shoppers in the treatment groups were either randomly assigned to receive a 30 percent subsidy for healthful food only (in this case, fresh or frozen produce, or canned vegetables with no added salt) or to be given a choice between using the subsidy for the vegetables and fruits or for less-healthy baked goods. A control group received no subsidies.

All shoppers were surveyed on their typical food choices and preferences, and the majority indicated that they wanted to eat more vegetables and fruits. The price of these items was a key factor as to why they weren’t doing so. Additionally, notes Imas, “people want to be healthy, but they have something in front of them that tastes good or is cheap . . . and they discount their goal by too much and give in to temptation in the present.”

The treatment groups were given subsidy choices either before they went shopping or during the shopping trip. Once they made their choices, they could not change their minds. Someone who chose to apply the subsidy to healthy foods was not made to buy them, but the discount didn’t apply for other purchases.

The researchers hypothesized that if shoppers had to apply their subsidy before the shopping trip, they would be in a better position to focus more on their future selves—that is, the version of themselves that would benefit in the long run from a better diet. Shoppers went on multiple shopping trips during the research period and provided receipts to the researchers. They also kept 24-hour food diaries.

Give shoppers a choice and a pause

The results suggest that it is indeed effective to subsidize healthy foods, particularly for consumers who are more price sensitive. Offering a subsidy for only fruits and vegetables increased spending on these items by 124 percent relative to a control group. These results are similar to those of the USDA’s Healthy Incentives Pilot program, conducted between 2011 and 2012, in which some SNAP participants received a subsidy for purchasing fruits and vegetables.

However, the two behavioral interventions—giving shoppers a choice of whether to apply the subsidy to healthy foods, and having them make that choice before going shopping—had an even more dramatic effect. They pushed spending on healthy foods even higher, ultimately creating a nearly 200 percent boost compared with the control group.

This may seem counterintuitive. After all, if consumers have the choice to save money on either carrots or cupcakes, why wouldn’t more of them choose to save on tasty cupcakes? However, says Imas, the results indicate that allowing for agency, rather than taking an approach that could be seen as paternalistic and condescending, is key. “People just don’t like being told to do something,” Imas says—or at least they’d rather feel they are making their own life choices.

Which type of food subsidy?

In an experiment, shoppers who received a subsidy for fruits and vegetables boosted spending on these items—even when given a choice to apply the discount to baked goods instead. Receiving information about this choice before shopping further increased spending on fruits and vegetables, especially when the shoppers picked how to apply their subsidy before heading to the store.

Shoppers understand that carrots are healthier than Twinkies, he says. When they have both the power and the time to make a decision between the two, they are more likely to make the choice that works for their future selves. Additionally, once people make the healthy choice and are bound to it, an inherent desire for consistency will make them more likely to stick to that choice when it’s time to go shopping.

Inherent patience levels also affect food choices, Brownback, Imas, and Kuhn find. They surveyed all shoppers on their time preferences—whether they were more likely to want fast rewards in life or to allow for delayed gratification.

Patient shoppers were more likely to buy and consume a larger share of fruits and veggies and more likely to apply the subsidies to healthy foods when given a choice, the researchers find. Conversely, shoppers who were less patient tended to purchase and consume fewer fruits and veggies and were less likely than the patient shoppers to choose the healthy subsidies.

The upshot: there could be a way to apply both incentives and insights from behavioral science. If program administrators were to survey benefit recipients on their time preferences, for example, they could better anticipate issues related to impatience and more effectively target subsidies using this knowledge.

Some consumers might be encouraged to consider their choices ahead of time. The researchers report that 20 percent of SNAP benefits are spent on the day of disbursement, and more than half by the end of that week. To be sure, this pattern may be due more to economics than anything else. But if administrators were to contact less-patient recipients with reminders that their subsidies were arriving days or even hours before the actual disbursement, this could create a deliberation period of sorts ahead of the shopping trip, allowing them more time to consider the benefits of a healthier choice.

Other factors: Marketing and communications

People who work directly in local food service note that there are other issues, beyond individuals’ choices and patience levels, that affect demand. Many of the stores opening in underserved areas may not have the right products or pricing for the community. Or if they do, they might not necessarily know how to brand the items correctly.

“A lot of these stores that are opening up are undercapitalized, and they’re not really good,” says Liz Abunaw, founder of Forty Acres Fresh Market, a small mobile grocer and food delivery service offering lower-cost fresh, whole foods to neighborhoods on Chicago’s West Side.

She points to the example of a Whole Foods that opened in 2016 in the Englewood neighborhood of Chicago, only to close six years later. “It’s hard to overcome a brand perception that you are super expensive in a neighborhood that is more price conscious,” she says. Englewood, on the city’s South Side, isn’t technically a food desert but has a median annual household income of approximately $25,000, which is close to the US federal poverty line for a three-person household.

Even if residents are told that a store known for higher prices will have affordable options, Abunaw says, “they do not believe it, or they think that, if prices are going to be lowered for a certain neighborhood, the food quality will also be lower.”

“It’s not enough to simply tell people you should eat healthy because it’s good for you.”
— Liz Abunaw

She notes that the Englewood Whole Foods had an uphill battle it may have eventually won when it came to pricing and perception, and says it’s possible it all “just fell apart” during the pandemic. Despite the store’s efforts to connect with the community through events including live music nights and wine tastings, pricing was a key issue. When the store shut its doors, then-mayor Lori Lightfoot publicly questioned whether previous mayor Rahm Emanuel had made the right choice for this community when he helped bring it to the neighborhood, noting that it was simply too expensive for many residents. Whole Foods did not respond to a request for comment.

The store was eventually replaced by a Save A Lot, whose prices were cheaper. But concerned residents called the store’s quality and cleanliness into question. Swinging from an upscale brand to what Abunaw calls “the wrong brand” is not a strategy primed for lasting success, she says.

Dawson, of Dion’s Chicago Dream, adds that a general lack of cultural understanding can hurt some ventures. “A lot of times when we look at the supply and the demand, you’re talking about a supply chain that does not represent underserved communities culturally,” he says. “You don’t have any high-level execs that represent them and the offerings. It is not culturally sensitive.”

Abunaw and Dawson both say stores that open in low-income communities need to understand and communicate with local residents—both for their own survival and to align with the greater policy goal of attracting residents to healthier food options.

Better diets across the spectrum

There’s good reason for diets in low-access areas to be a policy priority. A diet lacking in proper nutrition can lessen lifetime earnings and aggravate inequality, according to Brownback, who notes that poor childhood diets often lead to negative adult health outcomes.

Poor nutrition can also result from inequality. “Public policy and economic practices have created these areas that have low access to foods,” notes Detroit-based food activist Malik Yakini in an article presented by the T. Colin Campbell Center for Nutrition Studies. He considers the phrase “food deserts” to be offensive because it “does not speak to that intentionality.” The USDA dropped the term in 2013 and since then has described areas as “low-income and low-access,” which it says more accurately reflects what it statistically measures.

Poor diets are indeed 25 percent more prevalent among lower-income Americans, according to Brownback, Imas, and Kuhn. But people of all backgrounds and income levels can struggle with maintaining a healthy diet. Indeed, data from the CDC indicate that fast-food purchases increase along with household income.

And the explosion of the drug semaglutide, marketed as Ozempic or Rybelsus (for the treatment of type 2 diabetes) or Wegovy (for weight loss), has demonstrated the struggle that people across all economic classes—including wealthy celebrities—can have with diet. Some news reports indicate that this drug may shift eating behavior in part by eliminating the mental “food noise” that can lead to intense cravings for ultraprocessed, high-calorie items.

A 2023 study led by Richard Lopez of Worcester Polytechnic Institute emphasizes the importance of connecting all shoppers to their future selves. The researchers surveyed a group of about 350 people of varying backgrounds on their food cravings and consumption, as well as on their resistance to certain foods. People with a better connection to their future selves had stronger cravings for foods such as fresh fruits and vegetables, while those who had “higher Consideration of Future Consequences scores” reported a greater resistance to foods such as cookies, white bread, chips, and french fries. This dovetails with the findings from Brownback, Imas, and Kuhn that having the time to consider the longer run, rather than short-term enjoyment, may have health benefits.

Changing how we communicate about food might also help shift behaviors. Companies such as McDonald’s, PepsiCo, and Kraft Heinz spend billions of dollars creating effective marketing campaigns for convenient, tasty foods full of fat and sugar. But eaters are also innate advertisers, notes research from Booth’s Ayelet Fishbach and Instagram’s Bradley Turnwald, a former postdoctoral scholar at Booth. They find that people—whether online or in person—use far more positive emotions when describing food such as burgers, tacos, and macaroni and cheese than broccoli, asparagus, or cauliflower. And in turn, those enthusiastic food descriptions inspire other people to want to consume them, just as advertising does. When more emotional and enthusiastic language is used to describe healthier foods, people tend to be more motivated to eat them.

Recommended Reading

Abunaw agrees. “There’s a stigma that vegetables don’t taste good, and I don’t think that that’s true,” she says. “But it’s not enough to simply tell people you should eat healthy because it’s good for you. It’s more, ‘Try this. It’s good.’ So how do you make healthy food fun? Because it can actually be a lot of fun.”

Creating healthy eaters

Both Forty Acres and Dion’s Chicago Dream are tiny compared with grocery chains and food-bank systems. In February, Dawson went to a Washington, DC, conference to represent his organization as part of the Biden administration’s Challenge initiative, which has so far pledged about $9 billion in its aim to “end hunger and build healthy communities” across the US by 2030. Nevertheless, Dawson has had a difficult time getting government funding.

But both organizations are confronting the challenge of healthy eating—and providing anecdotal evidence that a focus on raising demand is hard but worthwhile. At Forty Acres, for example, boxes are curated for each customer, and on the basis of customer feedback, staffers will remove items the buyer does not like. This creates repeat business and enthusiastic patrons, says Abunaw. Moreover, getting to know customers’ tastes, as well as backgrounds and culturally relevant preferences, can help her suggest new, healthy foods—such as bok choy to people who already love cabbage.

“There’s a whole grocery experience that you can put around people that over time will influence their tastes,” she says, noting that Forty Acres serves people who have a range of incomes, and that her approach has worked with her customer base at all levels. “It’s exposure. It’s respecting people where they’re at.”

Dawson touts a similar goal of community involvement and connection. Dion’s Chicago Dream employs health workers to conduct biweekly conversations with its clientele and gather feedback. It also partners with the American Diabetes Association; recently, Dawson says, the association conducted a focus group with some of Dion’s clientele, and about 75 percent self-reported lower insulin levels since joining the program.

CountyCare, a Chicago-based Medicaid program, allows some residents with health issues to receive free food from Dion’s Chicago Dream for up to 26 weeks. When the program period ends, Dawson says, some recipients express that they miss the deliveries and are even annoyed that they need to get on a long waiting list for the regular program if they want to continue the service.

Now, he finds, these consumers are happy to add fresh fruits and vegetables to their diet, but they need to wait or in some cases travel farther in order to get them. “Prior to getting our box, there was no demand from these recipients,” Dawson says. “But the beautiful thing about that is the exposure created a feeling and an acceptance of what could be. And so, as a result, now it goes from a demand issue to a supply issue.”

More from Chicago Booth Review

More from Chicago Booth

Your Privacy
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.