• Reduce text

    Reduce text
  • Restore text size

    Restore text size
  • Increase the text

    Increase the text
  • Print


Impact of a "fat tax" on the purchasing behaviour of French households

Obesity is a major national and international public health concern. In France, 38.5% of men and 28% of women are overweight, and 14.5% of male and female adults are obese. In view of the economic repercussions of the development of obesity, governments are trying to change dietary behaviours. Of the measures that have been envisaged, the introduction of a "fat tax" would consist in taxing foods with high calorie, fat and sugar contents. A research team at the INRA Nutrition and Social Sciences Research Unit (ALISS) in Ivry-sur-Seine has studied the impact that such a measure might have on the purchasing behaviour of French households.

Updated on 06/14/2013
Published on 04/05/2012

Two levers are traditionally activated to exert an effect on dietary behaviour: nutritional information and price intervention. The taxing of products harmful to health, like a "fat tax", would have the dual advantage of reducing consumption through a negative elasticity-price effect, and generating tax income that could fund public health campaigns (support for healthy foods, incentives regarding physical exercise, etc.). In the context of this study, the "fat tax" corresponded to a rise in the VAT rate levied on certain products which would cause an overall rise of 10% in their price.

To measure the impact of a tax on high-calorie products, the food demands of households were modelled (Almost Ideal demand system), based on data regarding the quantities, expenditure on and prices of all foods taken from the TNS Worldpanel and concerning 5000 households. The study period retained (1996-2001) was before the first French National Nutrition and Health Programme (PNNS), so that it was possible to measure precisely the impact of such a tax without any nutritional information being supplied. All the participants recorded their food purchases, either directly through the reading of barcodes on the products, or using a code diary supplied to them so that they could record their purchases of fresh products bearing no barcodes. The characteristics of the products purchased were broken down into 32 nutrient groups, which included: energy, plant and animal proteins, carbohydrates, lipids, cholesterol, alcohol, fibres and vitamins.

The initial results of the simulations performed using this model demonstrated that a reduction in purchases correlated to a rise in prices. But this behaviour needed to be qualified further as a function of the social class of the household and the different food groups concerned. For example, wealthier households limited their purchases of fish, dried fruits, dairy products, cheese/butter/cream, fatty and sweet products and bottled water when their prices rose. The same applied for poor households, but to a lesser extent. Inversely, wealthier households seemed to be less sensitive to rises in the prices of fruits, vegetables and alcoholic beverages than poor households, where rises in their prices caused a reduction in purchasing.

Another aspect to be taken into account in a context of price rises is the effect of replacing purchases by other products. For example, the study revealed a correlation between a drop in the price of fresh vegetables and a reduction in the purchase of potatoes and fatty and salty foods. Similarly, a rise in the price of non-alcoholic drinks (excluding fruit juices) caused a reduction in the purchases of potatoes, prepared dishes and fatty and salty foods, which suggests that the application of a tax on sodas might induce a marked reduction in their purchase.

These findings emphasise the importance of choosing the right products to tax in order to achieve the effect that has been targeted. The optimum reduction in calorie intakes could be achieved by taxing prepared dishes, the cheese/butter/cream group and fatty and sweet foods. Applied to these three groups of products, a "fat tax" would cause a 3.4% reduction in the purchase of calories by wealthier households and a 3.6% reduction in poor households.

More specifically, the taxing of prepared dishes is that which would generate the most marked effects: both a reduction in purchasing and positive nutritional effects (reduction in the purchase of sodium, increase in the purchase of beta-carotene and vitamin D). A reduction in the purchase of saturated fatty acids and cholesterol, and globally the number of calories purchased, would be achieved differently depending on the social category of the household: by the taxing of cheese/butter/cream in wealthier households and by an increase in the price of fatty and salty products in poor households.

Although the impact of a "fat tax" would indeed reduce purchases of calories and saturated fatty acids, as recommended by the PNNS, these results should be qualified: a "fat tax" is also harmful to the consumption of nutrients that are essential to health (vitamins, calcium, magnesium, potassium and phosphorus); it only exerts weak effects on the weight of individuals in the short term; and finally, it engenders costs in terms of well-being that are much higher in poor households than in wealthier households.


Allais O., Bertail P., V. Nichèle (2010). The Effects of a Fat Tax on French Households’ Purchases: A Nutritional Approach, American Journal of Agricultural Economics, 92:228-245.