Longitudinal study of the effects of price and promotion incentives on purchases of unhealthy foods: evidence for restricting food promotions.

Objectives: Taxes and restrictions on promotions have recently been proposed as policy instruments to reduce consumption of unhealthy foods. The objective of this study is to add to the limited evidence on the comparative effectiveness of price changes, price promotions and volume promotions in changing household purchasing of unhealthy foods, using biscuits, crisps and savoury snacks as examples.

Design: Longitudinal regression analysis of consumer microdata.

Setting: Secondary data on itemised household purchases of biscuits, crisps and savoury snacks from 2006 to 2012.

Participants: Sample of 3024 households in Scotland.

Main outcome measures: Changes in the number of calories (kcal) purchased in the product category by a household caused by changes in the price for the product category, any temporary in-store price promotions and any temporary in-store volume promotions. Changes are measured at the mean, median, 25th percentile and 75th percentile of the household purchasing distribution for the full sample. Subgroup analyses were conducted by household income band and for households with and without children.

Results: Between product categories, the scale of purchasing response to incentives varies significantly. Within product categories, the mean calories (kcal) purchased by a household are more responsive to any volume promotion than to price or any price promotion for all product categories. As the volume of items purchased increases, households are less responsive to price, less responsive to any volume promotion and more responsive to any price promotion. Statistically significant differences are observed between household income groups in their response to price and promotion incentives within the biscuits category only. In cases where statistically significant differences are observed, households with children are more responsive to promotion and price incentives than households without children.

Conclusions: For all product categories analysed (biscuits, crisps and savoury snacks), household purchasing is most responsive to any volume promotion. Therefore, assuming the response of consumers to incentives remains constant following legislation, the most effective policy instrument to reduce the calorie intake from these products may be a ban on volume promotions.