Skip to main content

How to reduce saturated fat consumption? The effect of taxes versus behavioural interventions

Wednesday 11 October 2023, 1400-1500
Hybrid: Charles Thackrah G.29 OR Online - book below
David Palma, CDR, University of Leeds



An excessive consumption of Saturated Fats (SAF) can lead to weight gain and increased risk of heart disease. Despite these negative consequences, consumption of SFA in the UK is considerably higher than dietary recommendations, making its reduction an important public health goal. But changing eating habits is not easy. In this study, we measure and compare the effectiveness of three different possible interventions to reduce SFA consumption: (i) providing general dietary advice, (ii) providing general advice and personalised food-shopping advice, and (iii) taxing high-SFA food items. We also test the effect of a combination of some of these interventions. We do this by modelling people’s food shopping behaviour using a Multiple Discrete-Continuous econometric model, and later predicting their behaviour under different scenarios. We estimate the model using panel data from 111 adults’ supermarket loyalty card, all of whom had raised cholesterol, in Oxfordshire, UK. We find that personalised shopping advice has the biggest impact, followed by taxing, and finally the general advice. However, effects are mostly additive, so a scenario with all three interventions is the most effective way (among the tested approaches) to reduce SFA intake.

The Speaker

David is a lecturer in business analytics and applied artificial intelligence at the Centre for Decision Research, in the Leeds University Business School. He is an engineer and Ph.D. whose main area of expertise is the mathematical modelling of human decision making. His research interests include econometrics, including discrete and discrete-continuous choice models; machine learning; decision support systems; and research software development. He also has experience as a freelance consultant in the private and public sector, mainly in the area of demand modelling and forecasting. David’s peer-reviewed publications have been cited over 300 times, and he is one of the two developers of Apollo, a popular statistical software package.