The effect of implicit and explicit taxes on the purchasing of ‘high-in-calorie’ products: A randomized controlled trial

Public health taxes on less healthy food and beverage products have been shown to be effective in various settings. However, it is unclear if observed reductions in the quantity of taxed products purchased is a result of price increases due to the tax or the accompanying messaging and if the effects are influenced by the level of support for such taxes within the population. 941 adults residing in Singapore were randomized and asked to shop in one of four versions of a fully functional on-line experimental grocery store: 1) no tax control; 2) implicit tax showing only post-tax prices (i.e., 20 % higher than control prices) on high-in-calorie products; 3) fake tax showing pre-tax prices and a label falsely indicating that the price includes a 20 % tax on high-in-calorie products; and 4) explicit tax showing the same label as in 3) and an actual 20 % price increase applied to the high-in-calorie products. The proportion of high-in-calorie products purchased was 14 % in the control arm. We were unable to reject the null hypothesis of no effect in the implicit tax arm compared to control (0.08, 95 % CI −3.31 to 1.77) or in the fake tax arm compared to the control (2.59, 95 % CI −5.04 to 0.00) but observed a statistically significant 3.35 percentage point decrease (95 % CI −6.01 to −0.5) in the explicit tax arm compared to control. We were unable to reject the null hypothesis of no effect in any of the outcomes related to diet quality. Individuals who support the tax showed greater responsiveness to the explicit and fake taxes compared to those who do not (price elasticities of demand of −1.38 and −0.51 respectively). Results suggest that reductions in the proportions of high-in-calorie products purchased may be largely attributable to explicit messaging rather than to price increases. However, even when effective, policymakers should recognize that changes in purchasing patterns may not improve diet quality and that results may not generalize to other areas where levels of support differ.