The effect of different food and beverages tax schemes on consumer purchases in a virtual supermarket

In recent years, fiscal policies to improve diet have received considerable attention. Although the focus of most tax initiatives has been on sugar-sweetened beverages, taxes targeting a wider range of “unhealthy” foods and beverages may have more beneficial effects on overall diet quality and health than taxation on sugar-sweetened beverages alone. Therefore, the aim of the present randomized controlled trial is to measure and compare the effect of different tax schemes on consumer purchasing in a virtual supermarket. Participants will be randomly assigned to one of the following conditions: a sugar-sweetened beverages tax condition, a nutrient profiling tax condition, or a control condition.

Our hypothesis is that less sugar-sweetened beverages will be purchased in both tax conditions compared to the control condition, and that the nutrient profiling tax condition will affect healthiness of the total shopping basket and total purchased energy.

One time point. Participants will visit the virtual supermarket once and complete a questionnaire afterwards.

In the virtual supermarket, participants will be randomly assigned to one of the following conditions:

(i) Sugar-sweetened beverages tax condition. In this condition, prices of sugar-sweetened beverages are taxed on a scheme similar to the UK Soft Drinks Industry Levy, where beverages containing 5 to 8 grams of sugar per 100 milliliter are taxed €0.21 per liter and beverages containing more than 8 grams of sugar per 100 milliliter are taxed €0.28 per liter.
(ii) Nutrient profiling tax condition. In this condition, taxation of energy-dense, nutrient-poor foods and beverages is based on the Nutri-score (nutrient profiling scheme). Food and beverages with a label “D” or “E” are classified as “unhealthy” and taxed at a 20 percent level based on previous scientific evidence.
(iii) Control condition. In this condition, regular prices will be used. The control condition will also be used for the purpose of another randomized controlled trial evaluating the effect of a meat tax on consumer purchases (registration in NTR will follow).

To reflect a realistic situation in which taxes are introduced, participants in the experimental conditions will be informed about the taxation before entering the Virtual Supermarket. Participants in the control condition will not receive such a notification.