A substantial tax on sugar sweetened drinks could help reduce obesityBMJ 2013; 347 doi: http://dx.doi.org/10.1136/bmj.f5947 (Published 31 October 2013) Cite this as: BMJ 2013;347:f5947
- Jason P Block, assistant professor
- 1Obesity Prevention Program, Department of Population Medicine, Harvard Medical School/Harvard Pilgrim Health Care Institute, Boston, MA 02215, USA
Public health researchers, advocates, and policy makers have increasingly proposed taxes on sugar sweetened drinks as a way to reverse the rising consumption of such drinks—a trend believed to be a major contributor to the global obesity epidemic. In a linked article (doi:10.1136/bmj.f6189), Briggs and colleagues have modelled the expected impact of such a tax on energy intake and the prevalence of obesity and overweight.1
According to recent national data from the United States, 12-19 year olds consumed an average of 1260 kJ per day (301 calories) in sugar sweetened drinks.2 Adults over 20 years consumed 850 kJ per day.3 Consumption is lower in the United Kingdom—448 kJ per day for 4-18 year olds and 682 kJ per day for 19-64 year olds in one survey—but still substantial.4 In response, several European countries, including Hungary, Finland, and France, have recently levied or increased existing taxes on these drinks.5 In the United States, 34 states and the District of Columbia have food taxes that affect sugar sweetened drinks; 23 of these have taxes specifically targeting these drinks.6
The question remains—do these taxes have positive effects on public health? Even without an effect on consumption, a tax could raise much needed public funds for a variety of health needs. Nevertheless, the success of such a tax should be defined by its effect on the consumption of sugary drinks, energy intake, and, most importantly, rates of obesity and overweight. Briggs and colleagues’ article makes an important contribution to this topic.1 Using data from four nationally representative UK surveys of dietary purchases, the price of drinks, and body weight, they estimated that a 20% tax would lead to an average per capita reduction of 16.7 kJ per day (95% credible interval, 11.3 to 21.7). This change is small, but, were it to be applied across the entire UK population, the authors predict a 1.3% reduction in the prevalence of obesity, just over 180 000 people, with an additional reduction of 104 000 overweight people.
Not surprisingly, the reduction in energy intake from a tax would be greatest in those who consume the greatest number of sugar sweetened drinks—16-29 year olds (56.3 kJ). People over the age of 50 years, who consume very few of these drinks, would have no significant change in energy intake. People on lower incomes would have slightly smaller reductions than those with higher incomes, partly because of their substitution of high fat milk for sugar sweetened drinks.
To arrive at these estimates, the authors used a comprehensive econometric approach. This calculated the association between the prices and purchases of sugary drinks; the effect of altered purchasing on energy intake; and the dynamic impact of reduced energy intake on prevalence of obesity and overweight, while accounting for probable reductions in metabolic rate during weight loss. They also allowed for possible substitution by other drinks when the tax comes into force, projecting some increase in the consumption of juices and milk. The small energy deficit remained despite these substitutions.
The association they reported between prices and purchases (elasticity of −0.81 to −0.92) was in line with other estimates.7 However, two modelling studies in the US, using similar methods to those of this study, found substantially higher reductions in energy intake from a 20% tax on sugary drinks. These differences could be explained by overall higher consumption of these drinks in the US than in the UK. One study estimated a 101.7 kJ per day per capita reduction from such a tax; the other estimated a 142.4 kJ per day reduction.8 9
So, can we be confident that a tax on sugary drinks would work as intended? In agreement with these earlier studies, Briggs and colleagues show that the tax would be no panacea for preventing obesity. An approach that deals with all contributors to obesity and weight gain, from healthcare to marketing and the environment, would be needed to reverse the obesity epidemic.10 Even without a large effect, however, if high enough, such a tax is likely to help. Any measurable reduction in average per capita energy intake, even by a minimal daily amount, translates to meaningful reductions in rates of obesity and overweight across a large population. Concerns about such a tax being regressive remain but seem limited—Briggs and colleagues found that the financial effects of the tax would be small at any income level. Furthermore, they found no statistical differences between income groups.
This study provides evidence that a 20% tax on sugary drinks can work. The more pressing question is whether policy makers can implement a tax this high. Existing taxes in the US and Europe are low, typically far less than 10%. These low rates, although generating important revenue, have never been shown to alter obesity rates.11 Denmark, the one European country with a high tax, recently repealed it, citing concerns about citizens shopping for goods in neighbouring countries without a tax.12
We need more countries to implement high taxes and measure the results, using quasi-experimental study designs. A proposal by the Academy of Medical Royal Colleges calling for the UK to pilot and monitor the effect of a one year 20% tax on sugary drinks would be a good start. It would provide definitive data about the viability and success of such a tax.10 Other European countries currently considering taxes, including Romania, Ireland, and Belgium, could become sites for investigation.5 Econometric modelling studies are important and helpful but provide projections rather than measures after actual policy change. We now need policy makers to act and provide opportunities for real world evidence on a 20% tax on sugar sweetened drinks.
Cite this as: BMJ 2013;347:f5947
Competing interests: I have read and understood the BMJ Group policy on declaration of interests and declare the following interests: None.
Provenance and peer review: Commissioned; not externally peer reviewed.