Obesity Prevention in the Supermarket—Choice Architecture and the Supplemental Nutrition Assistance Program

As obesity-related health care costs in the United States reach nearly $210 billion annually, policymakers, health experts, and antihunger advocacy groups debate whether the government should pay for sugar-sweetened beverages (SSBs) and other unhealthy foods with food assistance programs. The Supplemental Nutrition Assistance Program (SNAP), formerly known as the Food Stamp Program, is the largest federal food assistance program; it provides 44 million low-income Americans with $66 billion in food assistance annually. A significant portion of this spending is on SSBs. A recent US Department of Agriculture analysis of sales from a national supermarket chain found that soft drinks were the number one purchase by SNAP households.¹ Although health and nutrition groups advocate restricting the purchase of SSBs with SNAP benefits, antihunger groups object that these restrictions are stigmatizing and discriminatory. To move beyond this stalemate, we propose a SNAP “choice architecture” policy that would balance the major health risks of consuming SSBs and other unhealthy foods with freedom of choice by making unhealthy foods less convenient and less visible at the point of purchase.

CHOICE ARCHITECTURE AND PUBLIC HEALTH

“Choice architecture” refers to the context in which people make choices. Behavioral science research has shown that people’s choices can be highly influenced by features like accessibility, salience, colors, information, and ordering of items. Subtle aspects of choice architecture can have significant effects on public health; if bottles of soda are easily accessible and highly salient (e.g., located at the checkout counter rather than the middle of a supermarket aisle), people are more likely to choose them. Having a background in which people make choices is inevitable, and so the question is how to design the best choice architecture, not whether to have one. Good choice architecture can help advance public health by influencing small health-related decisions that cumulatively lead to chronic disease.

One example is restaurant calorie labeling, required by the Affordable Care Act, which places calorie information at the point of purchase rather than buried in a pamphlet or Web site. Another example is state and local policies that restrict the placement of tobacco products (e.g., behind the counter placement, elimination of “power walls”) to discourage sales of cigarettes. Both calorie labeling requirements and tobacco placement restrictions increase the likelihood of healthier behaviors—consuming fewer calories or not smoking—while maintaining people’s freedom to make the less healthy choice.

SUPERMARKET CHOICE ARCHITECTURE

Supermarkets and other food retailers have long enlisted choice architecture to maximize sales and profits. For example, buying milk is one of the most common reasons for going to the grocery store, and it is purposely placed at the back of the store so that shoppers have to travel farther through the store to get it, increasing the likelihood they will make other, unplanned purchases. More generally, supermarkets are designed with the knowledge that most shoppers prefer to move counterclockwise around the perimeter of the store and make only short trips up and down the aisles.² The end of the aisle (endcap) displays face in three different directions, and products located in endcap or other free-standing displays account for 40% of all supermarket sales.² Almost half of all supermarket sales of chewing gum, candy, and soft drinks are selected in the checkout aisles.³

Existing supermarket choice architecture is designed to sell products, but it comes at a steep cost to Americans’ health. Endcap and checkout counter displays are designed to encourage impulse purchases and are stacked with SSBs, candy, salty snacks, and sweetened baked goods.⁴ Research has shown that higher exposure to these marketing techniques is associated with a higher body mass index.⁴ A study of a UK supermarket showed that carbonated beverages located in end of the aisle displays had 51% higher sales than did sales of the same beverages located in the middle of aisles, even after accounting for number of display locations, price, and weekly promotions.⁵ Despite widespread knowledge about the risks of an unhealthy diet, many Americans purchase SSBs, sweets, and unhealthy snacks at high rates in the supermarket.¹ More than half of adult’s and children’s calories from SSBs are consumed in the home,⁶ and reducing supermarket purchases would likely reduce consumption.

HEALTHY CHOICE ARCHITECTURE AND POLICY

By contrast, placing healthy items, such as fruits and vegetables, in prominent positions and placing unhealthy items in less prominent positions lead to healthier choices.⁷ In 2014, the US Department of Agriculture concluded that product placement is a promising approach for promoting healthier food choices by SNAP participants and recommended further research in this area (bit.ly/2sJE16t). But there is no reason to delay implementation of empirically supported policies to replicate research that scientists and the grocery industry have already conducted.

Ideally, of course, supermarkets would change the choice architecture on their own, and some have taken steps to do so. But evidence about retailer practices makes it reasonable to assume that most supermarkets are unlikely to move in the direction of healthier choice architecture without being required to change. We believe that a SNAP choice architecture policy should be implemented that would restrict placement of certain accessory food items, such as SSBs, desserts, candy, and salty snacks, from endcap, free-standing, and checkout counter displays in stores participating in the program. Although these items would remain available for purchase in the store aisles, they would be out of view from the highest traffic areas; thus both adults and children would see them less, leading to fewer unplanned purchases.

SNAP benefits are redeemed at approximately 260 000 food stores nationwide, and 81% of benefits are spent at supermarkets and superstores. National and state policies that restrict the placement of SSBs and other unhealthy items in stores participating in the SNAP program would have a positive impact on the US food shopping environment and could significantly reduce unhealthy purchases among households that shop at these stores, whether they are participating in the SNAP program or not. From the perspectives of both public health and individual liberty, SNAP choice architecture policy is a win–win. Choice architecture does not discriminate: SNAP and non-SNAP households would have similar access to healthy and unhealthy choices. Choice architecture policy would also complement other established programs, particularly the SNAP education program, by helping to reduce unhealthy, impulsive purchases in the supermarket.

ANTICIPATED ARGUMENTS

A choice architecture policy might lead some stores to not participate in SNAP because the anticipated loss of profit from SSBs and other foods would be greater than would the loss of SNAP customers. To reduce this risk, particularly for smaller stores, the US Department of Agriculture might consider a minimum choice architecture requirement—such as for SSBs and candy—and provide financial incentives to encourage stores to participate in other categories. Larger supermarket chains and superstores could—and should—find creative ways to promote healthy choices while maintaining profits.

Some opponents of SNAP choice architecture might claim that the policy represents government intrusion on personal freedom and puts undue burdens on supermarkets and other food stores. However, use of choice architecture would not regulate the attributes of the food or beverage items; it would preserve people’s ability to purchase the products that they like.

CONCLUSIONS

No single food policy can be a panacea, but healthy choice architecture in the supermarket is an easy and cost-effective way to reduce a substantial proportion of SSB and other unhealthy food purchases among low-income families. Preferably, movement in this direction would be voluntary. To the extent that it is not, we should move forward with this policy, which would improve Americans’ health while fully preserving freedom of choice.