As a patron of both (perhaps too regular a patron), I know that Starbucks and McDonalds have both recently introduced calorie information to their menus in the UK as part of the government’s Public Health Responsibility Deal. New York passed legislation in 2009 that requires restaurants with more than 15 outlets to print such information on its menus.
But does displaying calorie information on menus result in more healthy meal choices? Not according to recent research from Sweden.
The results of that research, based on a field experiment in a Swedish factory restaurant, found that easily accessible nutritional information on menus does not lead to statistically significant changes in meal choice, or in average calorie consumption.
Far more important was the location of the meal on the menu – meals at the bottom of the menu were 118 times less likely to be selected than those at the top. The authors concluded that the ingredients mattered more than labelling – chicken and turkey sold more than red meat dishes or vegetarian dishes.
Yet there were limitations to the study – the restaurant chosen disproportionately served men and blue-collar workers, whose behaviour may not be representative.
Other research on food product labelling suggests nutrition information does lead to healthier meal choices. In a laboratory experiment in Switzerland, participants in the study – regular shoppers – were asked to select products from an online food catalogue for their family for a week. They were then given nutrition education and asked to reselect products from another catalogue, this time containing products with nutritional information on the front of the packaging. The labelling varied in terms of its content and style.
In that study, all labelling lead to a nutritional improvement in the shoppers’ baskets. The study found that simple information, such as red or green dots, was better than complex information, like calorie numbers.
Yet the Swiss study lacks a suitable control group: shoppers without nutritional education. More nutritional baskets may be the result of education, rather than the labelling.
More healthy people may respond more positively to signals about the nutritional content of food, because of the type of person they are or the level of health education they have received, rather than the introduction of the labelling.
What are the implications for policy makers?
First, food labelling alone is unlikely to significantly alter consumer behaviour. So don’t expect a collapse in sales of Praline Mocha with Whipped Cream.
Second, not all food labelling has the same effect. Mandatory schemes using simple cues are likely to work better than voluntary schemes where complex or confusing information is presented. Simply listing calories may therefore be pointless.
Third, health education is important. Otherwise the labelling is unlikely to be interpreted correctly.
Finally, governments may want to take menu regulation further and force restaurants to put the healthiest choices first. Of course, it’s all relative: the menu may only contain a Big Mac and a Bacon Double Cheeseburger. Don’t expect adzuki bean sprouts at McDonalds any time soon.