Researchers: How Does Posting Calories Affect Behavior?

Research shows that posting calorie counts at restaurants has a big impact on consumption but none on revenue.

February 01, 2011

| by Stanford GSB Staff



A customer cradles her drink at a Starbucks in central London. (Reuters photo by Andrew Winning)

A new study released Jan. 6 examined consumer behavior before and after calorie counts were posted, and determined that when restaurants post calories on menu boards, there is a reduction in calories per transaction. Based on transaction data provided by Starbucks, researchers from Stanford GSB found that calorie-posting in New York City in 2008 led to a 6% reduction in calories per transaction. According to the study, beverage choices at Starbucks are unaffected by calorie posting. However, calorie posting leads consumers to buy fewer food items, and to switch to lower calorie food items.


To facilitate the analysis, Starbucks gave the researchers access to transaction data from Starbucks locations in three major cities — New York, Boston, and Philadelphia — from January 2008 to February 2009. Starbucks began posting calorie information in its New York City stores in April 2008, so the study looks at changes in consumer behavior before and after calorie-posting. To control for seasonal changes in purchasing behavior, the authors compare the New York changes to observed behavior in Boston and Philadelphia (where there was no calorie posting) over the same time period.

The main findings of the study are described in the working paper “Calorie Posting in Chain Restaurants,” and can be summarized as follows:

  • Calorie posting at Starbucks led to a 6% reduction in calories per transaction, from 247 to 232 average calories per transaction.
  • Almost all of the effect is related to food purchases. Average beverage calories per transaction did not change substantially, while average calories from food per transaction fell by 14%, of which 10% is due to people buying fewer items and 4% is due to people buying lower-calorie food items.
  • For those consumers who averaged more than 250 calories per transaction, calories per transaction fell by 26%.
  • The calorie reduction persisted for at least 10 months after calorie counts were first posted.
  • Overall, Starbucks revenues were not affected by the calorie-posting requirement. However, for Starbucks stores located within 50 meters of a competitor, calorie-postings led to an increase in Starbucks revenue.

Opponents of calorie-posting laws have argued that nutritional information is already available (for example, on restaurants’ websites, or on placemats or brochures in the store) and that most consumers don’t make decisions based on calorie counts. However, the researchers: Bryan Bollinger, a PhD candidate, and Phillip Leslie and Alan Sorensen, both associate professors of economics and strategic management at Stanford GSB, find that posting calorie counts on menu boards does affect consumer behavior.

The research team does acknowledge that a 6% reduction in calories is too small to have a major effect on the nation’s waistline. However, even a small benefit from calorie posting would exceed the low cost of posting, making it a worthwhile policy. The long-run effects may also be more dramatic, especially if it encourages restaurants to offer more low-calorie items. The study also presents preliminary evidence that this is indeed happening in New York City.

Obesity rates in the United States are the highest in the world, and have been steadily increasing over the past 15 years. One approach to dealing with the problem has been to enact laws requiring restaurants to post calorie content of their offerings, such as the 2008 New York City law requiring all restaurants with 15 or more outlets to post calorie information as prominently as prices on their menus and menu boards. The requirement is also included as part of the health care legislation currently being debated in Congress.

For media inquiries, visit the Newsroom.

Explore More