Dinesh K. Gauri is a professor and Walmart chair in the department of marketing at the Sam M. Walton College of Business at the University of Arkansas. He is also the executive director of retail information at the Walton College. His research and teaching interests include retailing, pricing, marketing analytics, shopper marketing, e-commerce and social media marketing. He advises for various companies in these areas and is a recognized leader in marketing.
To improve food security and access to a healthy diet for low-income households, the U.S. government provides economic assistance for food purchases through the Supplemental Nutrition Assistance Program (SNAP). The program aids more than 40 million low-income individuals, about 1 in 8 people in the U.S., purchase food. Research has highlighted benefits of the program, such as lifting families out of poverty and providing long-benefits to children.
SNAP aims to address a fundamental problem – the lack of resources available to buy food. Rather than provide food directly or restrict purchases to certain food items, SNAP provides low-income households additional resources to purchase food that fits their preferences and needs. The program, therefore, relies on retailers in the private sector to provide participants with efficient access to food.
SNAP benefits are modest, at around $7.57 per person per day, on average in 2022. Benefits are determined based on the U.S. Department of Agriculture’s Thrifty Food Plan, which calculates the cost of groceries needed to provide a healthy diet on a limited budget, assuming all meals and snacks are prepared at home. However, since participants can make their own decisions about their food purchases, SNAP households’ purchases may not align with the Thrifty Food Plan recommendations. Understanding how SNAP participants spend their money – and why – has important implications for retailers.
Considering food insecurity concerns, one may expect SNAP households to prioritize staple food items over discretionary ones, such as desserts, snacks and beverages. However, research indicates that these sweetened, hedonic foods make up a larger proportion of a typical SNAP household’s shopping cart compared to non-SNAP households. While these studies show that SNAP households purchase more hedonic, unhealthy food, previous research has not examined the motivations underlying the behavior.
Why do SNAP households purchase more unhealthy food? Are SNAP households more likely to buy unhealthy food items when they use the electronic benefit transfer (EBT) cards? Or do SNAP households spend more on unhealthy food because they consider such food items to be better value for money? Or is their predilection for unhealthy food caused some deep-seated craving for food caused by their past food insecurity? Our research addresses these questions through three separate studies. We investigated whether SNAP households buy more hedonic, sweetened food items, compared to non-SNAP households, and if so, what psychological mechanisms drive SNAP participants to purchase more unhealthy food.
The first study explored if SNAP households purchased more unhealthy food items, and if so, if they were more likely to do so when using Electronic Benefit Transfer (EBT) cards or cash. The second study tested whether SNAP households experienced stronger food cravings or underestimated the unhealthiness of hedonic food items compared to non-SNAP households, and if so, whether it influenced their food purchases. The final study analyzed if SNAP participants consider high calorie foods to be better value for the price.
We analyzed point-of-sale data to understand shopping behavior differences between SNAP households and non-SNAP households. Consistent with prior studies, we found that, regardless of whether they paid by EBT card, credit card or cash, SNAP households’ food baskets had more calories per ounce and higher unhealthiness ratings compared to non-SNAP households. These results suggest SNAP households do not just purchase unhealthy foods because they have extra resources. Rather, their shopping behavior may be influenced by other factors.
To test the findings, we simulated grocery shopping to understand the impact cravings and unhealthiness perceptions have on food purchases. Compared to non-SNAP households, SNAP households reported stronger food cravings and underestimated the unhealthiness of hedonic food. Cravings and unhealthiness perceptions explain the differences in purchase behavior between SNAP and non-SNAP households, even after controlling for income, education and age.
In every case, SNAP households spent more money on hedonic food than non-SNAP households did. This difference in spending was not influenced by price evaluations but by food cravings and unhealthiness perceptions. We do not find any evidence for the claim that SNAP households’ unhealthy food purchases are driven by their attempt to maximize energy density or calories-per-dollar.
Our results suggest that SNAP households spend just as much on groceries as non-SNAP shoppers. Considering that almost one in eight shoppers in the United States are enrolled in SNAP, this sizable customer segment exhibits purchase patterns that differ from non-SNAP households. Retailers would benefit from understanding the behavior and motivations of SNAP households to better serve their customers.
As retailers venture into healthcare, our research suggests an interesting opportunity for companies to further improve health outcomes. SNAP households buy more unhealthy food, partly due to a biased understanding of what makes food unhealthy. Companies that provide healthy food should develop advertising campaigns that more effectively promote healthy options. Through targeted advertising, companies could also identify patterns and gravitate shoppers toward healthier options over time. By embracing such socially responsible strategies, companies can capture excess returns.
A common perception is that healthy food items are more expensive compared to unhealthy alternatives. This is generally true on a price per calorie basis, but research shows healthy foods are usually less expensive when measured by price per edible weight or price per average portion size. These metrics are generally not accessible to consumers in grocery stores, which usually just display total price and price per unit. Research has shown that consumers are more likely to purchase healthy food if prices were lower. The same may be true if retailers displayed the price per edible weight or price per serving to enable consumers to make better comparisons.
Our research shows SNAP households’ purchases are driven by a combination of food cravings and biased perceptions of unhealthy foods. Interventions targeting these root causes would be most effective in reducing unhealthy decisions and improving SNAP households’ diets. Educating low-income households about what makes food healthy or unhealthy could reduce bias and lead to better health outcomes.
Manoj Thomas (Cornell University) and Yu Ma (McGill University) also assisted in the study.
