Direct and indirect impact of food taxes and subsidies on food consumption and population health

Obesity and the implications of high bodyweight are among the greatest challenges facing health systems worldwide, including in the UK. Implications arise from an increased burden of preventable non-communicable diseases (NCD's), including diabetes, cardiovascular diseases and several cancers. The number of obese people in the UK has recently been forecast to increase by 11 million by 2030 with associated combined medical costs of £1.9-2bn per year. Given the scale and urgency of the problem there is a need for effective measures to slow this trend, and food consumption and diets are one of the key areas for health policy to focus.

This fellowship project will provide an assessment of the effectiveness of food taxes and subsidies in changing food consumption towards healthier diets and thereby improving population health outcomes. It will build on existing evidence but will bring in new theoretical insights and use innovative data. Consumption of unhealthy foods can be reduced by increasing their price through taxes, whereas the consumption of healthy foods can be promoted by decreasing prices through subsidies. Taxes and subsidies therefore offer a potentially powerful policy tool to affect food consumption, diets and health. However, the empirical evidence shows that taxes have to be relatively high to have a meaningful effect and the evidence on the effectiveness of subsidies is small with mixed findings. This research project incorporates novel aspects by testing the linearity of consumers' response to changes in prices. First, it tests whether consumers are more responsive to taxes compared to subsidies. Secondly it will test whether the relative response to taxes or subsidies changes when the scale of taxes and subsidies change.

Another aspect that current evidence often fails to incorporate, and what this study will explicitly address, is the indirect effects of taxes and subsidies, mainly towards the consumption of other foods or beverages. For example an increase in the price of beef may reduce its consumption but at the same time increase the consumption of, now relatively cheaper, chicken which happens to be a healthier alternative. However, substitution could happen instead towards unhealthier foods. Also, consumers may still continue buying the now higher priced food but reduce the quantity of all other foods they consume in order to continue affording it, including eating less of healthy foods such as vegetables. Similar effects can occur when subsidies for healthy foods are introduced. These indirect effects can therefore reinforce or undermine the impact of food taxes or subsidies and are important to consider.

The research will include two empirical studies. The first will use econometric models to analyse changes in food consumption due to price changes using a large nationally representative household data set from the UK from 2012-2013. The second empirical study will collect and analyse primary data from an experimental study where the response in consumption decisions to various scenarios of taxes and subsidies on a range of foods will be tested.

For an appropriate policy response to the growing obesity problem, a detailed knowledge of the effectiveness of price-based policy measures is required. Although food taxes and subsidies are currently being considered in developed countries, the growing prevalence of obesity and non-communicable disease in parallel to under-nutrition in less developed countries suggests that soon the governments of these countries will also be seeking effective solutions. Thus this project has the potential to have an impact reaching beyond the UK and developed world.

The project will be carried out by health economists from the London School of Hygiene & Tropical Medicine and will involve collaborations with researchers from the University of Cambridge, the School of Oriental and African Studies and the University of Bologna.

The fellowship project aims to enhance the existing evidence on the effectiveness of taxes and subsidies in influencing consumers towards healthier food consumption, and thus on health outcomes. Specific objectives are:
Obj 1. To test if, in the context of food consumption, people change consumption patterns less when subsidies are involved compared to their reaction to taxes.
Obj 2. To test whether consumers' response can be assumed to be linear with respect to large and small taxes and subsidies.
Obj 3. To test whether the indirect (cross-price and income) effects undermine, reinforce or do not affect the direct impact (own-price effect) of taxes and subsidies on food consumption.
Obj 4. To analyse whether taxes and subsidies have further indirect effects on non-food consumption patterns.
Obj 5. To explore the importance of other determinants of demand, such as habits and preferences, in consumer decisions.
Obj 6. To test whether the above effects vary across different socio-economic groups or other sub-populations defined by the quantities of foods consumed or the BMI measure.
These objectives will be achieved through implementation of two empirical studies. The first will use a highly disaggregated household panel micro-data set from Kantar World Panel UK and apply econometric models based on the Quadratic Almost Ideal Demand System (QAIDS). The QAIDS model will be extended to enable exploration of differential response mechanisms with respect to price increases and decreases, for example by implementing a two-stage approach where the first stage discriminates between the two situations.
The second study will employ discrete choice experiment methodology to test the consumers' response to food taxes and subsidies. The experimental setting allows relaxing neo-classical assumption of consumer rationality and test the (in)direct impact of various scenarios of taxes and subsidies and explore the role of other attributes that affect consumption choice.

The research carried out during the fellowship will benefit following people and institutions:
a) Myself: through the knowledge and skills acquired during the fellowship I will be able to continue developing to become an independent researcher in health economics and establish my research platform. In the long-run I want to be a leader in the field and producing high-quality research, teach and supervise others' research activity.
b) LSHTM will benefit from this research through a more diversified research portfolio, and high-quality research outputs.
c) LCIRAH. By continuing to stay as a member of the LCIRAH core group the centre will benefit from this project through presentations in its seminars, conference, and publication of a policy brief. The project fits under LCIRAH's overarching research question on how to achieve sustainable food and agriculture systems which promote health and well-being for all people and will thus add to the research its staff and members are conducting.
d) Academia. This research will introduce novel aspects which have not been explored before to address the shortcomings of and refine current evidence in the assessment of effectiveness of food taxes and beverages on health outcomes. It will offer new methodological approaches and will bring new insights into consumer behaviour towards food and is likely to spark new ideas and instigate further research.
e) Health policy makers. The project aims to provide evidence-based answers that help in deciding on and/or designing price based policies to influence food or beverage consumption habits. The scientific papers and other research output (for example a LCIRAH policy brief) from the project will include clearly written policy suggestions that health policy makers can benefit from.
f) Health charities, lobby groups and other interest groups who wish to influence health policy regarding reductions in obesity and NCD's. These groups will benefit as their work can be based on peer-reviewed and published research findings.
g) Ultimately the beneficiaries of this research are the people whose health and life quality can be improved by effective policies through reduction in obesity and NCD prevalence rate. Healthier people mean healthier workforce and thus savings at both private and national level from lower medical costs and reduced work absence which will eventually boost economic growth for the country. Furthermore, resources saved from lower prevalence of obesity and NCD's can be used elsewhere to improve other public services.
h) In the longer run it is not just the UK and developed countries that benefit but also developing countries where the growing issue of double burden of disease means that these governments will be seeking effective solutions including these tested in developed countries.