COY0055
Written evidence from Thijs van Rens
Tax interventions to support and fund an obesity strategy
The case for taxes in the fight against obesity
Evidence previously reviewed by the committee revealed the urgency to stop the epidemic increase in obesity, particularly among children, and the central role of diet for obesity. Are taxes an appropriate instrument to achieve the change in diets that is necessary to slow down or stop the obesity epidemic?
There is little doubt among policy makers and academics alike, that (sufficiently high) taxes are effective at changing consumption patterns, and the UK, like almost all other countries have a long history of so-called sin taxes, mostly on alcohol and tobacco. The component of the Government’s Childhood Obesity plan that was best received by experts, was the introduction of a ‘sugar tax’ on sugar-sweetened beverages.
There are at least three reasons, or externalities, why dietary choices cannot and should not be left as a matter of personal responsibility. The first and most obvious reason is that the NHS is free to individual users, so that consumers do not have an incentive to take into account the effect of their weight on health care costs. The public health care system makes the case for an obesity tax particularly compelling in the UK, as opposed to -say- the US.
Individual consumers do, of course, care about their health, but -and this is the second externality, or ‘internality’ as it is sometimes called- there is strong evidence that they discount the long-term effects of their dietary choices because of the difficulty of self-control (O’Donoghue and Rabin 2006, Haavio and Kotakorpi 2011). Such issues are more severe among the poor (Mani et al. 2013), whereas poor consumers are also more price sensitive (Gruber and Kőszegi 2004), so that a tax intervention is likely to reduce health inequality.
The third reason for wanting to tax dietary choices that lead to obesity is that the environment, in which consumers make choices about food, is very far from neutral. A lot has been written about obesogenic environments (see e.g. Swinburn, Egger and Raza 1999), and the committee has previously made proposal to intervene in many of its elements, including school food, location of take-aways, restrictions on advertising and labelling. Prices are an important environmental factor determining purchases, and tax increases are often passed through to prices 100% (Benzarti et al. 2017). Healthy food, particularly fresh fruit and vegetables, have steadily become more expensive relatively to unhealthy and junk food (Capacci, Mazzocchi and Shankar 2012), so that one might think of taxes on unhealthy food (or subsidies on healthy food) as correcting a distorted environment.
All of these reasons are particularly true for children, for whom the personal responsibility argument is even less convincing.
What to tax?
A tax on sugar-sweetened beverages, as proposed by the committee and now adopted by the government, is a sensible first step towards a fiscal strategy against childhood obesity. Consumption of added sugars, at over 20% of calories consumed, is well above recommended levels (no more than 5% of total calories) in the UK, particularly but not only among children (Azais-Braesco et al. 2017). About a quarter of those added sugars are consumed in the form of sugar-sweetened beverages, and more than that for children (Dubois, Griffith, and O'Connell 2017).
But a tax on sugar-sweetened beverages by itself should not, of course, be expected to have a large effect on obesity. Three quarters of added sugar are consumed in other forms than sugary drinks, and 80% of calories consumed are not in the form of added sugar. In addition, there is some evidence that higher prices of SSB make consumers substitute towards other sugary food and drink, particularly among high sugar consumers (Briggs et al. 2013, Dubois, Griffith, and O'Connell 2017).
Taxing a much wider range of foods is needed, both because of the scale of the problem, and in order to avoid substitution towards alternatives that are equally bad or even worse for health. Such a tax could take the form of a tax on calories, or sugar and/or fat content, or a tax on food products that contain high amounts of these ingredients. Generally, nutrient-based taxes may be less prone to avoidance (Griffith, O'Connell and Smith 2016), but may be more difficult to implement and explain to consumers.
International experience is almost exclusively with various versions of a tax on sugar, sugary drinks or sugary foods (Mytton, Clarke and Rayner 2012,), most recently in Catalonia (El Pais, 2 May 2017).[1] These tax interventions are often broader than the UK tax on sugar-sweetened beverages, including for instance chocolate (Norway), confectionary (Finland, Australia), biscuits (Australia), or drinks containing added sweetener (France). While it is difficult to establish a causal link from such taxes on obesity, the evidence supports the view that such taxes encourage a healthier diet (Alagiyawanna et al. 2015), particularly when revenues are used to fund other aspects of an obesity prevention programme (Kim and Kawachi 2006).
As a preliminary conclusion, there is substantial room to expand on the sugar taxes, or more generally health taxes in the UK. If the UK wants to be a front-runner on this issue, plans for expanding the tax measures need to be put in place now.
The case against tax interventions
Opponents of health taxes often point towards potential job losses (e.g. Oxford Economics 2015). But these taxes have winners as well as losers. The levy on sugar-sweetened beverages, for example, is likely to generate some job loss in the soft drinks industry but may create new jobs in the dairy industry (Richardson and van Rens 2016). The overall impact on employment may well be positive (Powell et al. 2013). For taxes that are primarily aimed at raising revenue, the gains tend to be smaller than the losses: the tax generates a ‘deadweight loss’ because it distorts individual decision in the economy, leading for instance to a suboptimal supply of labour in the case of income taxes. However, health taxes are Pigovian taxes, primarily aimed at changing behaviour. They raise revenue as a (welcome) by-product. Thus, there is no deadweight loss and the total effect of welfare is expected to be positive.
An important concern are the distributional effects of health taxes. Since poorer households tend to consume more unhealthy foods, a health tax is regressive: it affects the poor more than the rich (Mytton, Clarke and Rayner 2012, Leicester and Windmeijer 2004). The flipside is that the health effect of the tax is progressive: the health of poorer households is expected to improve more as a result of a tax on unhealthy food, particularly because they also tend to be more price-sensitive. Nevertheless the regressive nature of a health tax may undermine public support, and it is important to address both health and income inequality.
Inequality in obesity and diet
It is well documented that there is large inequality in obesity, with children living in the most deprived communities being twice as likely to be obese or overweight as those in the least deprived areas (PHE 2015).
Inequality in diet and obesity is partly due to differences in preferences between families of different income and education levels. Handbury, Rahkovsky and Schnell (2017) document large nutritional disparities by SES even across households that shop in the same stores. A similar conclusion is reached by Allcott, Diamond and Dube (2018), who find very small increases in healthy eating when households move to a different neighbourhood with better access to healthy food, or when a new supermarket opens in a neighbourhood. It is important to note that ‘preferences’ is used broadly here, and it is quite possible that household purchasing decisions can be changed with education, information or awareness campaigns.
However, preferences are not the whole story. In our own work on the US, we show that prices of healthy food relative to unhealthy food are higher in poor counties (Pancrazi, Vukotic and van Rens 2018). If inequality in diet were driven only or mostly by preferences, we would expect households in poor areas, where the demand for healthy food is low, to pay lower, not higher, prices for healthy food. The fact that we observe the opposite is clear evidence that inequality in diet is due at least in part to the environment. Using a structural model, we estimate that approximately 60% of inequality in diet is due to environment, and 40% to ‘preferences’.
It is worth noting that whether inequality in obesity is driven by preferences or environmental factors, it will respond to taxes. The recent experience with a tax on sugar-sweetened beverages in Mexico is of particular interest. Data for the second year after the introduction of the tax have recently become available and show a sustained (and even somewhat augmented) reduction in purchases of taxed beverages. More importantly, the data show that low-SES households, which purchase the largest quantities of these beverages, also experienced the largest reduction in purchases (Colchero et al. 2017).
Income-dependent taxes and subsidies
Taxes are a blunt instrument, but not quite as blunt as is sometimes assumed. It is possible to correct the distributional effects of a health tax, at least in part, by making the tax income-dependent. Income-dependence is quite common for direct taxes (income tax), but much less so for indirect taxes (VAT, excise, including health taxes). The reason is simply that it is almost impossible to determine the income of consumers at the point of sale, where indirect taxes are levied.[2]
Much easier than making indirect taxes directly income-dependent, these taxes can be conditioned on other observables that are correlated with income, in particular on location. This is particularly relevant for health taxes because observed differences in prices of healthy food contribute to health inequality. There are no examples of such location-specific health taxes to date, but this is an important avenue to consider in order not to undermine public support for the tax, which can otherwise easily be portrayed as ‘unfair’ (Nestle 2015).[3]
Using an estimated model of consumption choices over healthy and unhealthy food for US households, we show that it is possible to almost completely eliminate inequality in diet due to price differences. This can be done with a tax on unhealthy food in combination with a small subsidy on healthy food, which is dependent on the average income in the county where the store is located, see figure 1 (Pancrazi, Vukotic and van Rens 2018). The tax policy shown in the figure is revenue neutral: the subsidy on healthy food is funded by the tax on unhealthy food. As a result, households buying food in relatively affluent areas will mostly pay tax, whereas households in the poorest counties will be net recipients. This tax intervention offsets the differences in relative prices faced by households shopping in poor and rich areas and provides incentives for stores and supermarkets to supply healthy food in all areas. The policy is specifically aimed at reducing inequalities and can be combined with a tax on unhealthy food (or even a subsidy on healthy food) that applies in all areas.
Implementing taxes as part of comprehensive obesity strategy
It is almost impossible to estimate how high an optimal health tax should be. There are two reasons for this. First, while the evidence that taxes are effective is overwhelming, the strength of the effect of a given tax on behaviour is very hard to predict (Gortmaker et al. 2011). Estimating price elasticities of demand is notoriously difficult and imprecise, and this is even more true for the cross-elasticities, i.e. measuring which other foods consumers substitute towards and how much. Second, while it has been estimated that smokers in the US more or less pay for their own increased health care costs through excise taxes on tobacco, whereas taxes on alcohol are too low to recuperate the external costs of alcohol consumption (Manning et al. 1989), there is no similar calculation for obesogenic foods anywhere in the world.
A practical solution to this problem is to introduce health taxes gradually. There is some evidence that even a low tax can be more effective than one would expect based on the demand elasticity, because it signals to consumers what is healthy and what is not (Brockwell 2013). Perhaps more importantly, once a tax is in place, it is easier to raise it if its effect is smaller than desired. This approach also allows to monitor any undesired effects of the tax while they are still small. Proper evaluation by an independent government body is required to prevent a continuing lobby either for or against the tax. Practical implementation is made easier by the fact that the less effective a tax is at changing behaviour, the more revenue it generates, so that if the tax fails to induce a better diet directly, at least it makes it possible to continue the fight against obesity by funding other interventions.
Another reason to use the revenues of an obesity tax to fund other interventions against obesity is that, while income-dependent taxes can go some way towards targeting a tax at those who need it most and mitigating its adverse side effects (regressive distributional effects), such targeting cannot be very precise. Other interventions can complement the tax to in reaching specific target groups, such as fruit and vegetable vouchers for pregnant women as in the US Special Supplemental Nutrition Program for Women, Infants, and Children.
Finally, health taxes are an important component of a comprehensive strategy against obesity simply because they raise revenue, whereas most other interventions cost money. Seeing the tax revenue spent on a good cause raises and maintains popular support for the tax (Nestle 2015).
References
Alagiyawanna A, Townsend N, Mytton O, Scarborough P, Roberts N, Rayner M. (2015). Studying the consumption and health outcomes of fiscal interventions (taxes and subsidies) on food and beverages in countries of different income classifications; a systematic review, BMC Public Health, 15(887)
Allcott, Hunt and Diamond, Rebecca and Dubé, Jean-Pierre (2018). The Geography of Poverty and Nutrition: Food Desserts and Food Choices Across the United States, Stanford University Graduate School of Business Research Paper No. 18-6
Arantxa Cochero, M & Rivera-Dommarco, Juan & Popkin, Barry & Ng, Shu Wen. (2017). In Mexico, Evidence Of Sustained Consumer Response Two Years After Implementing A Sugar-Sweetened Beverage Tax. Health affairs, 36(3)
Azaïs-Braesco, Véronique, Diewertje Sluik, Matthieu Maillot, Frans Kok and Luis A. Moreno (2017). A review of total & added sugar intakes and dietary sources in Europe, Nutrition Journal, 16(6)
Benzarti, Youssef, Carloni, Dorian, Harju, Jarkko and Kosonen, Tuomas, (2017). What Goes Up May Not Come Down: Asymmetric Incidence of Value-Added Taxes, NBER Working Paper 23849
Bødker, Malene & Pisinger, Charlotta & Toft, Ulla & Jørgensen D.M.Sc, Torben. (2015). The rise and fall of the world's first fat tax. Health Policy. 119(6)
Briggs, Adam D M, Oliver T Mytton, Ariane Kehlbacher, Richard Tiffin, Mike Rayner, Peter Scarborough (2013). Overall and income specific effect on prevalence of overweight and obesity of 20% sugar sweetened drink tax in UK: econometric and comparative risk assessment modelling study, BMJ, 347:f6189
Brockwell, Erik (2013). The Signaling Effect of Environmental and Health-Based Taxation and Legislation for Public Policy: An Empirical Analysis, CERE Working Paper
Capacci, Sara & Mazzocchi, Mario & Shankar, Bhavani, 2012. "The regional price of junk foods relative to healthy foods in the UK: indirect estimation of a time series, 1997-2009," 86th Annual Conference, April 16-18, 2012, Warwick University, Coventry, UK 134720, Agricultural Economics Society
Currie, Janet and Enrico Moretti (2008). Short and Long-Run Effects of the Introduction of Food Stamps on Birth Outcomes in California, in Making Americans Healthier: Social and Economic Policy as Health Policy, R.F Schoeni, J House, G Kaplan, and H Pollack (eds.), New York: Russell Sage
Dubois, Pierre & Griffith, Rachel & O'Connell, Martin, 2017. "How well targeted are soda taxes?," CEPR Discussion Papers 12484, C.E.P.R. Discussion Papers.
El Pais, 2 May 2017, https://elpais.com/elpais/2017/05/01/inenglish/1493645507_575034.html
Gortmaker SL, Swinburn BA, Levy D, Carter R, Mabry PL, Finegood DT, Huang T, Marsh T, Moodie ML (2011). Changing the future of obesity: science, policy, and action, Lancet, 378(9793), pp.838-47
Griffith, Rachel, Martin O'Connell and Kate Smith (2016). Sweetening the sugar tax? IFS Observations
Gruber, Jonathan and Botond Kőszegi (2004). Tax incidence when individuals are time-inconsistent: the case of cigarette excise taxes, Journal of Public Economics, 88(9-10), pp. 1959-1987
Haavio, Markus and Kotakorpi, Kaisa, (2011), The political economy of sin taxes, European Economic Review, 55, issue 4, p. 575-594.
Handbury, Jessie, Ilya Rahkovsky and Molly Schnell (2017). Is the Focus on Food Deserts Fruitless? Retail Access and Food Purchases Across the Socioeconomic Spectrum, working paper
Herman DR, Harrison GG, Afifi AA, Jenks E (2008). Effect of a targeted subsidy on intake of fruits and vegetables among low-income women in the Special Supplemental Nutrition Program for Women, Infants, and Children, American Journal of Public Health, 98(1), pp.98-105
Kim, Daniel & Kawachi, Ichiro. (2006). Food Taxation and Pricing Strategies to “Thin Out” the Obesity Epidemic. American Journal of Preventive Medicine, 30, pp.430-7
Leicester, Andrew and Frank Windmeijer (2004). The 'fat tax': economic incentives to reduce obesity, IFS Briefing Note 49
Mani, Anandi, Sendhil Mullainathan, Eldar Shafir, Jiaying Zhao (2013). Poverty Impedes Cognitive Function, Science, 341(6149), pp. 976-980
Manning WG, Keeler EB, Newhouse JP, Sloss EM, Wasserman J (1989). The Taxes of Sin: Do Smokers and Drinkers Pay Their Way? JAMA, 261(11), pp.1604–1609
Mytton, Oliver T, Dushy Clarke, Mike Rayner (2012). Taxing unhealthy food and drinks to improve health, BMJ, 344(e2931)
Nestle, Marion (2015). Soda Politics: Taking on Big Soda (and Winning), Oxford University Press
O'Donoghue, Ted & Rabin, Matthew, 2006. Optimal sin taxes, Journal of Public Economics, 90(10-11), pp. 1825-1849
Oxford Economics (2016). The economic impact of the Soft Drinks Levy, report for the British Soft Drinks Association
Pancrazi, Roberto, Marija Vukotic and Thijs van Rens (2018). Inequality in Diet and Obesity, Work in progress, University of Warwick
PHE, Public Health England (2015). Sugar Reduction: The evidence for action, publications gateway number 2015391
Powell, Lisa M., Roy Wada, Joseph J. Persky, Frank J. Chaloupka (2014). “Employment Impact of Sugar-Sweetened Beverage Taxes”, American Journal of Public Health, 104(4), pp.672-677.
Richardson, Ben and Thijs van Rens (2016). Case against soft drink levy is sugar coated, The Conversation
Swinburn, Boyd, Garry Egger, Fezeela Raza (1999). Dissecting Obesogenic Environments: The Development and Application of a Framework for Identifying and Prioritizing Environmental Interventions for Obesity, Preventive Medicine, 29(6), pp. 563-570