Motivation and background

In Thailand, approximately 40% of the population falls into the combined categories of obese and overweight, while in the US, 40% of its population is classified as strictly obese, according to Anocha Aribarg, Professor of Marketing at the University of Michigan. Professor Anocha shared some empirical results from her research, co-authored with Yiqi Li and Matthew Osborne on “Quantifying the Heterogeneous Impact of FDA Changes to the Nutrition Facts Label” at Sasin Research Seminar. As obesity has become a global epidemic, the US FDA introduced the back-package Nutrition Facts label (NFL) in 1994 to provide information that helps consumers make informed choices about the packaged food they purchase. In 2016, the FDA required manufacturers to update the design and information provided by the NFL to reflect new scientific information. The mandate was supposed to take effect between January 1, 2020, and January 1, 2021. The required changes include the display of calories per serving in a more prominent and larger type font, the inclusion of added-sugar information, and the removal of calories from fat shown on the label. The intention is to emphasize the negative impact of added sugar and the fact that the type of fat consumed matters more than its amount. Serving sizes for different product categories were also altered to better reflect the average portion an average person consumes in one sitting.  

Literature Review

Professor Anocha gave an overview of the different types of nutrition labels in different countries. Notably, while the US FDA paid attention to the back-of-package label, most countries focus on front-of-package labels. In 2014, the traffic light label was created in England and Ireland to indicate the energy, fat, saturated fat, sugar, and salt content in packaged food. In 2017, France and Belgium introduced the nutri-score label, which combined all ingredients into a composite food grade from A to E, indicating the overall food quality. Both labels are voluntary. Dubois et al. (2021) ran a field experiment in France among 60 supermarkets with 1,200 product categories, testing the effectiveness of different labels, including traffic light and nutri-score labels. The study reveals that the nutri-score label works better than all other labels. In addition, the nutri-score label works better in a product category that has high variance or a lot of uncertainty in perceived product healthiness. In 2016, Chile issued mandatory stop-sign warning labels on foods and drinks exceeding limits for sugar, sodium, saturated fat, or calories. Research by (Araya et al. 2022) shows that the label works for product categories where people do not expect to see unhealthy options, such as cereal as opposed to chocolate or cookies. Professor Anocha pointed out that fewer research has studied the impact of NFL or nutrition information. Russo et al. (1986) find that positive nutrition information, such as vitamins, does not affect consumer’s choices. However, consumers respond to negative information, such as added sugar, in their field experiment. Kiesel and Villas-Boas (2012) suggest that some nutrition labels (e.g., low fat) can backfire, as it may convey that the food is tasteless. Finally, research shows that some consumers respond more to labels than others, including those who are more motivated, older, female and educated (Bhawra et al. 2022).  

Data

The purpose of the empirical analysis is to study the impact of the label change on both the demand and supply sides using store sales data and consumer purchase data from the US syndicated data companies, Circana and Syndigo. Overall, the findings suggest a more prominent impact of the label change on the demand side than on the supply side. The new FDA food labels have significantly influenced consumer purchasing behavior with effects varying across product categories (yogurt, ice cream, soda, juice, potato chips, cheese). Specifically: Overall Demand Shifts:

• Increased overall demand for yogurt and juice
• Decreased overall demand for cheese

Nutrient-Specific Reactions:

• Yogurts and ice creams with higher-added sugar saw reduced demand, reflecting growing consumer aversion to sugar.
• With the calories removed from fat on the package label, ice cream, cheese, and potato chips with higher fat content experienced increased demand.

Manufacturer Responses:

• Evidence of manufacturer adjustments in pricing or reformulating products in response to the new labels is limited, suggesting little strategic response from manufacturers.

In addition, Professor Anocha explored heterogeneous responses to NFL at the household level in the yogurt category. The findings reveal that the impact of the label change on decreased demand for high-added sugar and increased demand for fat is highest for overweight, as opposed to either healthy or obese people. However, the survey also asked people about their perceived weights in addition to their actual weights. Interestingly, “a lot of people do not live in reality… some people who are overweight think they are slightly overweight, and many people who are obese think that they are slightly overweight,” Professor Anocha commented. As a result, when using perceived weight as a criterion, people who perceive themselves to be very overweight turn out to be those who responded most to the information about added sugar and calories from fat. These results imply that people behave in the way we expected (i.e., the labels should have more impact on people with heavier weight) but only based on their perceived weights. In summary, while the labels successfully guided consumer behavior, it did not appear that manufacturers made an effort to correct potential changes in their product demand. Future research could explore how consumer perceptions of their own health drive the heterogeneous effects of nutrition information and NFL.