Obesity is often considered a public health crisis in rich countries that might be alleviated by preventive regulations such as a sugar tax or limiting the density of fast food outlets. This paper evaluates these regulations from the point of view of equity. Obesity is in many countries correlated with socioeconomic status and some believe that preventive regulations would reduce inequity. The puzzle is this: how could policies that reduce the options of the badly off be more equitable? Suppose we distinguish: (1) the badly off have poor options from (2) the badly off are poor at choosing between their options (ie, have a choosing problem). If obesity is due to a poverty of options, it would be perverse to reduce them further. Some people in public health say that preventive regulations do not reduce options but, I shall argue, they are largely wrong. So the equity case for regulations depends on the worst off having a choosing problem. It also depends on their having a choosing problem that makes their choices against their interests. Perhaps they do. I ask, briefly, what the evidence has to say about whether the badly off choose against their interests. The evidence is thin but implies that introducing preventive regulations for the sake of equity would be at least premature.
- distributive justice
- political philosophy
- public health ethics
Statistics from Altmetric.com
Very many adults and children are fatter than they ought to be, according to medical advice.i In the opinion of some, several countries face an obesity epidemic that could slow or perhaps reverse the rise in life expectancy.1 While no doubt some of the claims have been hyperbolic,2 there is nonetheless a near-consensus that obesity is a public health problem if not a crisis. However, there is no broad consensus on what ought to be done about it. Some disagreements are about what measures would effectively reduce obesity; others are about what costs in money, welfare and freedom would be worth paying and who should pay them.
This paper is about one type of policy response to obesity, ‘preventive regulations to counter obesogenic environments,’3 and it is specifically concerned with evaluating them from the point of view of equity. Preventive regulations include taxes on sugary drinks or fat; bans or limits on ingredients such as trans fats, salt and sugar; and planning regulations to limit the density and siting of fast food outlets, for instance to prevent their being near schools. Some of these measures have been adopted, while others are waiting in the arsenal of public health. The key relevant feature of these measures for this paper is that, at least on the face of it, they reduce choice because they increase the money or time costs of obesity-promoting consumption without decreasing such costs for healthy consumption.
When it comes to the effects on equity, a common picture is of obesity striking unequally. Obese people are more likely to be poor and less well educated.4 Let us simplify and take the badly off to be people who do not have access to much money and the worst off to be those with least access to money. Inequalities in health are often thought to be inequitable when they are unfair or unjust, linked to some social indicator (such as money) and are to some degree avoidable.5 Presuming obesity is bad for health, the inequality in obesity then looks inequitable because it seems unfair that the worst off are less healthy in a way that is at least to some degree avoidable.
Here is a public health equity case for preventive regulations. Obesity is a problem that disproportionately affects the worst off. The obese would benefit from becoming thinner. Preventive regulations would make people thinner and would be more equitable than health promotion, which tends to benefit the more educated middle class.6 7 Preventive regulations would reduce the inequality between worst and better off and so make outcomes more equitable. In its concern for equity, this argument might be seen as a return to the roots of public health in social justice.8
Here are two examples of the equity argument. On behalf of a tax on sugary drinks, Brownell and Frieden write: ‘But the poor are disproportionately affected by diet-related diseases and would derive the greatest benefit from reduced consumption’ (p. 1806).9 Franck et al make a similar point in arguing for taxes on junk food: ‘However, low-income populations consume more junk food than do high-income ones, and they are generally at higher risk of obesity and chronic diseases. It follows that low-income individuals might be more likely to change their consumption behaviors and experience long-term health benefits’ (p. 1951).10
As against this public health case, one might make the counterargument that preventive regulations, especially those raising the monetary price, would remove options from the worst off. The worst off are the very people who have the fewest options. The regulations would therefore be inequitable. More generally, the problem with poverty is a lack of money, and the solution is to give people some (p.275).11 Trying to control people’s diets is middle class busybodying, which may or may not be a return to the roots of public health.
This counterargument potentially raises a serious difficulty for the equity case for preventive regulations. This paper brings out the conditions under which that equity case could succeed, or fail. The paper unfolds like this. I set out the scope of the argument, where what are within scope are policies that in fact reduce choice in a sense to be specified and applied. In assessing the equity of genuinely choice-reducing policies, we need to distinguish these two ideas: (1) badly off people have poor options and (2) badly off people have problems in how they choose between their options. The paper uses this distinction to claim that reducing choice for the worst off is inequitable unless (necessary condition) they are poor at choosing between their options. However, this condition is not sufficient for an equity gain. A further necessary condition is that the reduction in choice would make the worst off better off. The most plausible welfare argument is one favoured by a class of paternalists called ‘means paternalists’, which holds that without the reduction in choice, people would pick options that are not what they themselves most prefer. Setting out these conditions gives some tasks to people who favour the preventive regulations on grounds of equity, although I do not think they fully realise that these are their tasks, which may explain why they have not been done properly. I also adduce some admittedly limited empirical evidence, with a conclusion not particularly encouraging for those who favour preventive regulations.
Some further initial point. This paper focuses on adults; children, who often have choices made for them, raise complications beyond this paper. Also, the focus is on obesity and not other problems, such as tooth decay. Next, preventive regulations that aim to reduce obesity might inadvertently reduce equity for the simple reason that they do not work but, for the sake of this paper, assume that preventive regulations would actually reduce obesity. Lastly, the paper ignores externalities such as health system costs, which are anyway largely irrelevant given the focus on equity.
Preventive regulations and choice sets
Above I said that preventive regulations, on the face of it, reduce choice. Let me explain ‘reducing choice’ and then why appearances may deceive. A choice set is the set of all options from which one can choose. Choice is reduced when people have a worse choice set than before. What makes one choice set better than another is not always easy to determine, but it should be reasonably uncontroversial to say that a choice set would be better if it had the same options as before plus some good ones or it has different options that the chooser prefers.
Preventive regulations certainly could change options in a way that reduces choice. A soft drink tax removes the option of a cheaper soft drink and substitutes a more expensive soft drink. In itself, this change worsens a choice set. Reformulation or bans on ingredients could also increase prices. Planning regulations could raise time costs. In principle, the class of preventive regulations that could reduce choice is both large and broader than tax alone.
When preventive regulations do reduce choice, I will go on to claim, one must rely on paternalism to make the equity case for them. One must show that something has gone wrong with how people choose rather than with the choice set they have. This point is curiously overlooked by writers who claim that badly off people are unfree to refuse cheap unhealthy products.ii They may be unfree, but they become even less free if those products are made harder to get, with no improvements in other options.
However, while preventive regulations may look as though they must reduce choice, in fact they need not. Some people claim that preventive regulations would actually improve the choice sets of the worst off (and perhaps others) and they could be right. Here is a representative example. According to Mary Cheh, a Washington DC councillor, ‘The soda tax would help low-income residents by encouraging them to choose healthier beverages and by creating more grocery stores and jobs in low-income neighborhoods’ (quoted p. 118 12). It is the part of this statement about stores and jobs that is about increasing choice; the part about choosing healthier beverages looks as if it speaks more to problems with choosing. Other examples of potential improvements to choice sets might be planning regulations that make it possible for healthy food outlets to open, since they would otherwise be squeezed out by the unhealthy ones, or taxes and bans on certain ingredients that force companies to reformulate their food at no cost in consumer appreciation or money, as might have been true of New York City’s trans fat ban.13 If these effects occur, the regulations could count as improving or at least not worsening the choice set of the worst off. In that case, they would be out of the scope of this paper.
It is an empirical question what preventive regulations would do to choice sets and the answer may well vary according to the type of regulation and the market it regulates. I am nonetheless doubtful that preventive regulations would in general leave choice sets no worse. Why would the choice-enhancing effects occur? Suppose that without a planning regulation banning some fast food outlets, the worst off would have the options of cheap unhealthy food or expensive healthy food. What seems to be envisaged is that by preventing cheap unhealthy food, cheap healthy food will somehow become available. Why should it? What is it that makes healthy food expensive, such as short shelf life, that would disappear in the event of a planning regulation? Or consider reformulation. If food companies could reformulate to make healthier food that tasted just as good and cost no more, why are they not doing it already? It is possible that their corporate lives are just too easy in an uncompetitive market or perhaps they are too nervous of the competition to try reformulating first. However, it seems unlikely in developed countries that the market for food and drink has widespread failures of competition that could be improved by preventive regulations. In short, those who argue that the preventive regulations will improve choice sets have an implicit or explicit model of the economy that on the whole seems implausible, although improvements may be possible in some cases for some regulations.
It is, though, especially tricky to assess the effects on choice sets of regulations that raise revenue to be spent on the worst off. One cannot treat such revenue-raising regulations as simply reducing choice, although one must observe that to the extent that a regulation does raise revenue, it fails to reduce the targeted consumption since people have to buy the regulated product in order for revenue to be raised. However, revenue-raising makes an equity assessment complicated because the effects on equity depend on who pays, who receives and how much the recipients value what they get from the revenue. Take a proposal to levy taxes and spend the revenue on such benefits as free cookery classes and eating advice. If the worst off consume and pay the tax, they pay for taxed food and drink but they get advice. If they would rather have the package of untaxed food and drink but no advice, they would be worse off by their own preferences. If the better off consume and pay the tax but the worst off switch to untaxed food and drink, then the worst off get advice that is free to them but they still have to consume what they would otherwise rather not, such as expensive healthy food and drink. They may or may not then be better off. And those are only two possibilities. So you can see why assessing the equity effects of a revenue-raising tax would be difficult.
What then is in the scope of this paper are those regulations that reduce choice, a set that is likely to include many of the taxes, bans and planning limits already mentioned. Notice that this set is broader than tax alone and that choice need not be reduced only by the monetary effects. What are out of scope are policies to control obesity that do not reduce choice. These policies could include subsidies for healthy food or gyms, or improvements to public transport, or making stairs and water more available in public places. These policies would (leaving aside the costs of funding them) generally improve choice sets. Another policy that would improve the choice sets of the worst off and may even help with obesity would be income transfers or general improvements in socioeconomic conditions.14 Also out of scope are policies to change how people choose between their options while leaving those options unchanged. This category includes the classical nudging, of Richard Thaler and Cass Sunstein, which by stipulation leaves choice sets unchanged while trying to influence decisions through choice architecture.15 It also includes policies on information, whether to provide it, as with food labelling, or suppress it, as with advertising bans. All these policies are out of scope so this paper should not be read as an argument that nothing should be done about obesity.
Equity and obesity: the vital distinction
Preventive regulations might reduce the obesity of the worst off and therefore appear a gain in health equity, but they are often characterised as a nannyish restriction on choice. To this objection, many public health advocates retort that people at the bottom do not really have a choice.16
To sort through the arguments over choice, we need this vital distinction between:
Badly off people have poor options and
Badly off people have problems in how they choose between their options.
Although these ideas are often run together in the claim that the badly off have no choice, they not only differ, they have very different implications for assessing preventive regulations. With this distinction, we will be able to see what premises are needed to assess the equity of preventive regulations, and we will have a heuristic with which to think through the many factual, explanatory and value claims that are made about obesity and its prevention.
To forestall some criticism, let me make it clear at the outset that this distinction does not imply that all badly off people must be in one category if they are not in the other. Some people may choose obesity-promoting options because they reasonably see them as best, while others because of a problem with how they choose. Or people may switch at different times, sometimes making reasonable choices, sometimes not. Nor do I say that poor options and poor choosing are causally independent because bad options may affect the ability to choose well.17 18 ,iii However, they are conceptually independent.
It is possible, indeed, likely, that people who are badly off make their choices in different ways and that interventions in their choices would be good for some and bad for others. We get to that point later, with a discussion of distribution. However, let us initially trace out the implications if people do choose well, then if they do not.
Obesity is only one contributor to less than perfect health. Health is only one aspect of welfare or, more generally, ‘choice-worthiness’, and it does not have overwhelming ‘lexical’ priority over the other aspects. These claims seem too obvious to require much defence. Assuming they are correct, obesity could be the outcome of choosing well, on any plausible view of what it is to choose well.
Even well-off people could be obese as the outcome of reasonable choosing, given certain tastes. Still more could people who struggle to afford healthy food and drink or to find the time, energy and resources to exercise. In general, when circumstances change, it is reasonable to choose differently. If struck by appendicitis, one might reasonably consent to an appendectomy although one otherwise has no good reason to consent to being cut open. When it comes to a lack of money, what one might choose if one had more options may not be what one would—entirely reasonably—choose when one has fewer. Healthy food is often more expensive in time and money than unhealthy food. A capsicum, not a meal in itself, can cost $4 when a carton of chips would cost only $2. Fresh fruit and vegetables require more time to prepare and, because they perish, to shop for than processed foods.3 The money and time cost may be too high for badly off people, especially single parents, trying to hold down jobs and feed their families.19 The explanation for the greater obesity of the badly off could then lie in their choosing well from poor options.
Suppose poor options were the whole explanation for disparities in obesity. What might follow, and what would not? What would follow, but only with extra premises, would be two key points made in the literature about obesity. First, badly off obese people should not be blamed, or made to pay the costs that arise from being obese. Second, in being traceable to the social cause of low income and wealth, the inequality in obesity would be a health inequity. However, what would not follow is that preventive regulations, even ones that really did disproportionately reduce the obesity of the worst off, would be a gain in equity.
A standard account of health equity takes it to be the absence of avoidable socially caused or linked inequalities in health. This account raises several questions relevant to this paper that are perhaps more familiar in political philosophy than in the social sciences. One is about equality and giving priority to the worst off.iv If the only way to reduce a socially caused or linked inequality were to worsen the position of the worst off, which would be more just or otherwise better? Another question is about the relation between health equity and equity more broadly.v If inequity in health were reduced by improving the health of the worst off, but at a price in some of the non-health goods of the worst off, under what conditions would this reduction be unjust? Obviously, more detail is going to be needed, but I will assume here that health is not so ethically important that it is lexically prior to all other goods. I also assume that improving the position of the worst off is more important than decreasing inequality, that is, inequality should not be reduced by levelling down that worsens the position of the worst off.
Let us return to the possibility that obesity is due to poor options and not poor choosing. I take it as a datum that if poor people have no problem in how they choose between their options but do have a problem with the options from which they can choose, it must be at least in one way unjust simply to reduce options they might want to take. When the problem is a poverty of options, restricting them further would be perverse. It would still be perverse even if restricting them reduced inequalities of health.
What I am taking as a datum can be explained in many different conceptions of equity. Suppose equity is particularly concerned with the welfare of the worst off and suppose the introduction of preventive regulations changes the consumption of the worst off, as its proponents intend. Then the worst off would consume what they prefer less, which would reduce their welfare, and this would be inequitable. (To bring in a philosophical point, this claim does not presuppose that welfare is constituted by the satisfaction of preferences. If a person’s preferences can be and are in some sense incorrect, then they have a problem in their choosing and belong in the next section.) Suppose a conception of equity focuses on opportunity for welfare. Then reducing the options of the worse off would again be inequitable because it would reduce opportunities they might value. If nonetheless some conception of health equity favours the restriction, then so much the worse for that conception. By hypothesis, the reduction in inequality comes merely from worsening the options of the worst off. In that case, the health equity gain may be outweighed or, as is quite possible, health equity has been wrongly conceived, perhaps both fetishising health as a good independent of its contribution to welfare and overvaluing it at the same time.
‘When the problem is a poverty of options do not restrict them further’: this point is important to remember when reading debates about whether preventive regulations would be regressive. A regulation is regressive if it causes poorer people to spend disproportionately more of their incomes than richer people. Suppose poorer people buy more soft drinks than richer people. A flat tax on sugary drinks that did not change anyone’s consumption would then be regressive. However, the tax would not be regressive if poorer people ceased to buy soft drinks because now none of their income would be spent on the drinks or the tax. Some considerable effort goes into trying to establish whether antiobesity and, indeed, other public health regulations would be regressive. However, a tax might not be regressive and yet make outcomes more inequitable.vi Suppose the poor rank these options from best to worst:
Cheap healthy food.
Cheap unhealthy food.
Expensive healthy food.
Expensive unhealthy food.
Before the tax, option A is not available so the poor take option B, buying cheap unhealthy food. After the tax, option B turns into option D, but the tax does not make option A available, so the poor switch to option C, buying expensive healthy food. The tax would not be regressive in the sense that the poor would be disproportionately paying the tax because, by hypothesis, they would not be paying the tax. It would, however, put them into a position that they preferred less than before the tax or, put another way, make them worse off as judged by their preferences. Suppose now that richer people always preferred option C, expensive healthy food, to option B, cheap unhealthy food. The tax, in removing option B, has no effect on them. Then the tax would make the richer no better off and the poorer worse off, which is an increase in inequity (and levelling down because it would make some worse off and no one better off), and yet it need not be a regressive tax, which shows that whether a tax is regressive does not settle judgements of equity.
To bring some threads together, if the explanation for obesity lies in badly off people optimising in the face of poor options, it would be inequitable to reduce options. Preventive regulations would reduce options, a few cases aside (see the section on choice). Preventive regulations would then be on the whole inequitable. The equity case for preventive regulations therefore hinges on obesity being the result of problems in choosing. Some writers in public health will not like this way of putting things. They blame the ‘obesogenic environment’, and they believe that blaming poor choosing is a trick from the corporate playbook designed to deflect attention away from the real causes.20 Nonetheless, if poor people had no problems with how they choose, preventive regulations would make things worse for them.vii
Problems with choosing and paternalism
Let us then now consider the problems with choosing that people might have and how one might take them as the basis for constructing a paternalistic case for preventive regulations. Many writers say that we do not choose in fully rational ways, that our rationality is ‘bounded’ and our choices are the outcome of biases.viii In the specific context of obesity, problems in choosing are said to arise from: being uninformed or misinformed; the marketing of processed foods; the use of fat, sugar and salt, which some consider addictive; our apparent ‘mindlessness’ where our consumption is, without our awareness, heavily influenced by external cues such as portion size; a lack of attention to healthy diets, affected by stresses such as a lack of time or money; wishful thinking and self-deception (‘it won’t happen to me’); and failures of self-control.ix
One the basis of this list, many writers have claimed that paternalism need not interfere with autonomy because the behaviour interfered with is not autonomous anyway. Be that as it may, this list does not speak to the first task of paternalism, which is to show that interventions improve welfare. Suppose a soft drink tax causes people to choose water instead of soft drinks. Why would water be better? The reason cannot be merely that people would otherwise choose soft drinks irrationally or non-autonomously. That does not show the soft drinks were the wrong choice or bad for people. The obvious reply is that soft drinks are less healthy but that reply just pushes the question further back. Health is not everything, so on the basis of what criterion can it be said that the health benefits of water outweigh the non-health benefits of soft drinks?
I will assume that the most defensible criterion of welfare is to be found in a leading form of modern paternalism, namely means paternalism. Means paternalists think it is permissible under certain conditions for the state to intervene paternalistically to correct for what it sees as people’s mistakes in trying to reach their own ends.21 The means paternalist’s general criterion for welfare is then people’s own ends, while among writers on obesity, a common criterion is the fulfilment of long-term goals.13 22–24 The scope for paternalism arises because, it is said, people choose what makes them unhealthy, and their choices conflict with their long-term goal to be healthy. What must be being assumed, but is often not explicitly stated, is that the long-term goal is more important so that the benefits from fulfilling it outweigh the benefits of the unhealthy consumption. If the argument is a means paternalist one, it must be that the person interfered with must herself see her long term goal as the more important. The means paternalism case for preventive regulations then goes like this. In the case of unhealthy consumption, people prefer all-things-considered to consume healthily yet act inconsistently with this long-term goal because they are irrational. Preventive regulations could in principle make people better off according to this criterion despite restricting their choices by discouraging them from consuming in a way they themselves would all-things-considered rather not.
I agree that preventive regulations could make people better off. But would they?
Distribution problem and scaling
All policies tend to be good for some people and bad for others and preventive regulations are no exception. On the welfare criterion suggested here, they could be good for people who would otherwise choose against their own goals but could be bad for those who do not. The question of whether to implement preventive regulations is therefore in part one of distributive justice.
In a particularly helpful article, Johannes Kniess accepts that some would lose from preventive regulations. Nonetheless, he favours them on the grounds that distributive justice should give priority to the worst off and that the regulations would indeed help the worst off.25 I accept that distributive justice should give priority to the worst off, but I think Kniess was too quick to claim that these regulations would help the worst off.
In essence, Kniess’s argument is that many people are ‘non-voluntary overeaters’, by which he means people who overeat as the result of some failure in reasoning. He writes that it is ‘reasonable to assume that… their proportion will be greatest among the least advantaged members of society, reflecting inequalities in education and information’(p.891),25 and he cites the social gradient in obesity as confirming evidence. He then says, ‘it is more plausible to think that the eschewal of liberty-restricting public health interventions would be particularly disadvantageous for the worst-off members of society’. As against Kniess, let me make the following points. First, he does not ask himself whether non-voluntary overeaters would be better off for being prevented from overeating. They might not be. Second, even if non-voluntary overeaters are disproportionately among the worst off, most of the worst off could nonetheless be harmed by preventive regulations. One must remember to count people who do not overeat or who voluntarily overeat. Third, an alternative explanation for the social gradient in obesity is, as we saw, that the worst off have fewer options rather than that they have a reasoning problem. It cannot be said too often that ‘poor people choose unhealthy options’ does not entail ‘poor people behave poorly’.
What we need then is evidence about the extent to which obesity is or is not the outcome of badly off people choosing against health goals that they regard as more important than the goals they are in fact fulfilling by their consumption. I would like to emphasise certain elements in that sentence. If we are assessing a policy, we need evidence of extent that can be applied to a population; small or non-representative samples are insufficient. If we are going to use the means paternalist welfare criterion, we need evidence that people choose against goals that they themselves consider more important.
I do not think an overall body of evidence exists. At least, none of the obesity paternalists whose work is cited in this paper refers to any such evidence. In fact, they do not even raise the question of how many people’s consumption is contrary to their goals. It is as if they assume that obesity-promoting consumption just must be in all or most cases contrary to a more important long term goal.x To be sure, there is plenty of evidence about what preventive regulations and other interventions do to obesity, broken down by class, income and so on, and this evidence allows conclusions to be drawn about the health equity effects of interventions on certain conceptions of health equity. However, ‘health equity’, as we saw, could go up while the position of the worst off goes down, in which case, as I said, so much the worse for that conception of health equity. There is also plenty of evidence about the determinants of consumption choices but that also does not show whether the people who make these choices go against goals that they regard as more important.
I would like to offer one piece of evidence. The starting point is smoking, the subject of its own preventive regulations. It is often argued that most smokers want to quit and very many try to; evidence of quit attempts is evidence that smokers would be better off not smoking according to their own opinion; antismoking measures such as cigarette taxes are tailored to affect smokers only, so the measures make most smokers better off without burdening non-smokers.26 Let us assess for comparison the preventive regulations against obesity. Unlike preventive regulations against smoking, which primarily affect smokers alone, many of the regulations to reduce obesity are not tailored to affect only the obese. They affect everyone who comes within their scope, and unlike the smokers in the argument, such evidence (which is from the USA) as there is implies that while many people say they would like to lose weight, the majority in all income groups are not trying to. Among them, those with least money and education are the least likely to attempt to lose weight.xi
Thus, we have grounds to think that preventive regulations would be against the interests of most badly off people and therefore inequitable. These grounds are by no means decisive. More research from different fields would be needed to establish how far among the badly off obesity is down to a lack of options or to problems with their choosing that lead them to act against the balance of their preferences. The evidence may come that preventive regulations would benefit the worst off; this paper is not a doctrinaire argument that people’s choosing must be perfect or even good. Nor, to reiterate, am I commenting on measures besides preventive regulations to reduce obesity; this paper is not an argument that the status quo that has produced obesity should be left untouched.
Some may think that if the conclusion is that we should have no regulations without really solid research, it is just the same old delaying tactic as used by tobacco and alcohol firms and climate change deniers. The enemies of public health unreasonably insist on certainty when we already have good grounds to act. But we do not yet have good grounds for preventive regulations in the case of badly off adults. For the reasons set out, preventive regulations could worsen the position of the worst off on the most plausible criterion of their welfare, and there are some grounds to believe it would. We do not have the evidence to believe preventive regulations would be no worse for equity on a defensible view of what equity is. So the conclusion is that, as far as equity and adults are jointly concerned, preventive regulations are premature.
Mike King, Geoff Kemp, Kathy Smits, Steve Winter, two anonymous referees and audiences inAuckland and Dunedin
↵i ‘Fatter’ is a word that some may find offensive. I draw their attention to the reclamation of this word, for some good reasons, in the field of Fat Studies. See Rothblum ED.27 Fat studies. The Oxford handbook of the social science of obesity.174. Some further points that deserve to be in a note: (1) I acknowledge that Body Mass Index is not a wholly reliable measure of body fat and (2) I do not engage with the question of how bad being obese is for health; I just assume it is for this paper. If obesity were not bad for health, reducing obesity would not be a good reason for the preventive regulations this paper discusses.
↵ii A distinguished figure in public health, Michael Marmot, has a more complicated point with a similar conclusion. As I understand him, he regards the low price of alcohol as a force beyond individual control, which it is, and therefore as a constraint on freedom, which it might be. But a lower price for goods increases options compared with a higher price so it seems less of a constraint on freedom than the higher price. It is another question, of course, whether the low price improves welfare. See Marmot’s discussion of alcohol pricing in Marmot M.28 The health gap: the challenge of an unequal world: Bloomsbury Publishing. Pp.69, 75.
↵iii In which case, improving their options might improve the quality of their choosing. Rebonato R.29 A critical assessment of libertarian paternalism. Journal of Consumer Policy 37: 357–396. at 388.
↵iv The classic discussion is Parfit D.30 Equality and priority. Ratio 10: 202–221.Parfit. See also the essays in Clayton M and Williams A.31 (2002) The ideal of equality: Palgrave Macmillan. Applied to equity, Peter F.32 Health equity and social justice. Journal of applied philosophy 18: 159–170.
↵v See the vast ‘equality of what?’ debate started in Sen A.33 Equality of what? Tanner lecture on human values. Tanner Lectures. Stanford university. In the context of health, see Daniels N.34 Just health: meeting health needs fairly: Cambridge University Press.; of obesity, Barnhill A and King KF.12 Ethical agreement and disagreement about obesity prevention policy in the United States. International journal of health policy and management 1: 117.
↵vii This section provides a further lesson. It is regrettably common to bundle up together ‘choice’, ‘responsibility’, ‘freedom’ and opposition to preventive regulations into a package often labelled (but only by people who do not like it) ‘neoliberalism’. Examples include Baum F and Fisher M.6 Why behavioural health promotion endures despite its failure to reduce health inequities. Sociology of health & illness 36: 213–225. This section shows how the ideas come apart. One can reasonably believe that if poor options cause obesity, the worst off are not to blame, should not be held responsible, acted responsibly, and should not have their choice further restricted by preventive regulations.
↵viii Thaler and Sunstein must be the best known exploiters of these ideas in making a case for a form of paternalism.
↵ix The evidence is cited in these pro-paternalist writings: Offer A.36 The challenge of affluence: self-control and well-being in the United States and Britain since 1950, Oxford; New York: Oxford University Press, Skipper RA.37 Obesity: Towards a system of libertarian paternalistic public health interventions. Public Health Ethics 5: 181–191, Womack CA38. Ibid.Public health and obesity: When a pound of prevention really is worth an ounce of cure. 222–228, Kniess J.25 Obesity, paternalism and fairness. Journal of medical ethics: medethics-2014–1 02 537. The importance of these problems has been challenged both in general and in the case of obesity. See Rebonato R.39 Taking liberties: a critical examination of Libertarian paternalism, Houndmills, Basingstoke, Hampshire; New York, NY: Palgrave Macmillan, Rebonato R.29 A critical assessment of libertarian paternalism. Journal of Consumer Policy 37: 357–396, Sugden R.40 Do people really want to be nudged towards healthy lifestyles? International Review of Economics 64: 113–123. (Rebonato TL 198–9, Critical Analysis LP 393–4 Sugden, Better off As judged 117).
↵x Thaler and Sunstein simply cite obesity as an example of failure in choice. Thaler RH and Sunstein CR.15 Nudge: improving decisions about health, wealth, and happiness, New Haven: Yale University Press., p.7. They make no attempt to show that obesity is contrary to people’s own preferences.
↵xi See the figures using US National Health and Nutrition Examination Survey data in Seward H.41 Socioeconomic status and weight loss behaviours. Sociology. Virginia Commonwealth University, 81. See esp. pp35-6.
Contributors I wrote this.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.