The effectiveness of fat taxes and thin subsidies in improving diets

Policy makers have become increasingly concerned about the obesity ‘epidemic’ and other aspects of diet, such as consumption of too little fruit and vegetables and too much saturated fat. Together these impose substantial costs on the National Health Service as well as lost productivity to the economy through days of work lost and early retirement due to ill-health. The Government’s recent Foresight Report estimates the cost of overweight and obesity alone to be £7b per year and extrapolates it to rise to £45.5b by 2050. They propose a ‘system-wide’ set of policy interventions to tackle the problem and recognise that taxes on unhealthy foods and nutrients or subsidies on healthy foods or nutrients could be a part of the system. Taxes on unhealthy foods have in the past been dismissed as having an unfair burden on poorer segments of society and for being ineffective. Recent research in Denmark suggests however that it may be possible to combine a tax on the saturated fat component of foods with a subsidy on fruit and vegetables, thereby promoting healthy eating without imposing an excessive burden on either households’ or the government’s wallet. Models will be developed to assess the impact of different combinations of taxes and subsidies on nutrient demand in England by socio-economic group, Government Office Region and income. The burden of the tax/subsidy combinations on different socio-economic groups will also be calculated. The objective will be to find whether certain combinations of tax and subsidy are particularly effective, especially to those social groups whose health outlooks are poorest, and whether these can be achieved without placing a financial burden on the poorer groups.

Foresight (charged by Government to create challenging visions of the future to ensure effective strategies now) has called for a system-wide approach to tackling obesity and other nutrition-related ill-health; fiscal interventions could be an important element of the system. Past proposals for taxes on unhealthy foods have been dismissed as regressive and ineffective, but there is potential for tax/subsidy combinations that overcome this problems. Models will be developed to simulate the impact of different combinations of taxes and subsidies on foods and nutrients, on nutrient demand and the fiscal burden, by socio-economic group, age, Government Office Region and income. Demand modelling, will use the Quadratic Almost Ideal Demand System (QAIDS) and, in order to examine the distributional impacts of taxes and subsidies, it will follow an approach which enables the parameters of the model to vary according to demographic characteristics of the households. The model will be estimated using data from the Expenditure and Food Survey (EFS). Prices will be derived from unit values by assuming all households surveyed in the same week in the same Government Office Region (GOR) face the same prices. The impacts of a given tax or subsidy on food choice and expenditure will be computed and the impacts of the policy on nutrient and calorie intake will be obtained by converting changes in food demand to changes in nutrient and calorie demand using a matrix of food composition coefficients. The distribution of these effects will be examined across, income levels, regions, ages and social class.