Abstract

Chronic diseases (e.g. stroke, heart diseases, and diabetes) – often driven by unhealthy diets – account for a fast growing share of mortality and morbidity in developing countries. They impose not only a disease but also a substantial economic burden to the countries concerned. Chile is facing an enormous challenge to curb rising rates of obesity and diet-related ill-health. Much of this disease burden is affecting the poor more than the rich within Chile. Making healthier choices more affordable - and unhealthier options more expensive - might appear as a promising option to tackle the problem. After all, using price incentives has been shown to work very effectively in the efforts to reduce tobacco and alcohol consumption, where tax policies have long been applied. Yet when it comes to using economic incentives to promote healthier diet choices, we are so far left with preciously little, convincing evidence on whether such tax or subsidy policies would in fact have the desired results. This is because the bulk of the “evidence” is based on hypothetical modelling of such policies, rather than on the evaluation of actual examples of cases where they have been implemented in the real world. Chile is one out of a handful of countries world-wide that have now - in September 2014 - implemented a price incentive for healthier diets. More specifically, the Chilean Government has increased the excise tax on sugar sweetened beverages (SSBs), while reducing the tax on what it considers as non-SSBs. This is a great and extremely rare opportunity to first of all evaluate whether this policy has worked over a comparatively short period of time (one year). Conducting such an evaluation is challenging and requires the use of methods that produce credible impact evidence, even though the policy has not been implemented as a randomised experiment (which is what would have happened if evaluating the effectiveness of a new drug). This is the case in many areas of public health, however, and there are methods out there that have been, and can be, applied. The other challenge is that suitable data is not readily at hand in many countries, including in Chile, to conduct such an analysis. There is, however, high quality, detailed household purchasing data routinely collected by certain commercial market research companies, and we strongly believe this is the only data that currently allows us to assess the policy impact. Such data is exceptionally fine-grained, providing information on essentially all take-home purchases of beverages made by a nationally representative set of households over several years before and after the Chilean policy entered into force. Once we have evaluated whether the policy has ‘worked’, in terms of reducing ‘unhealthy’ beverage purchasing, we then also want to know whether this short term effect translates into meaningful longer term health effects. This can only be done via complex modelling of the longer term consequences, and the team has the expertise and track record to develop such models. In addition, the Chilean Government is keen to explore further refinements to this policy. In order to directly inform such deliberations, we will use our model to also analyse a range of related tax or subsidy policies, again aimed at promoting healthier diets, which appear at least promising on the basis of existing evidence. We have established a suitable, two-way communication channel with the senior policymakers in the Ministry of Health that will help ensure that the kind of evidence we produce will answer to the knowledge needs of the Government, without unduly interfering with the research. Overall, this research will contribute vital evidence not only to the Chilean government but also to many other low- and middle-income countries also facing the challenge of diet-related ill health and eager to understand the actual scope for fiscal policies in the promotion of healthier diets and the prevention of chronic diseases.