Are Incentives Corrupting? The Case of Paying People to be Healthy.

Written by Dr Rebecca Brown

Financial incentives are commonplace in everyday life. As tools of states, corporations and individuals, they enable the ‘tweaking’ of motivations in ways more desirable to the incentiviser. A parent may pay her child £1 to practice the piano for an hour; a café offers a free coffee for every nine the customer buys; governments offer tax breaks for homeowners who make their houses more energy efficient. Most people, most of the time, would probably find the use of financial incentives unobjectionable.

More recently, incentives have been proposed as a means of promoting health. The thinking goes: many diseases people currently suffer from, and are likely to suffer from in the future, are largely the result of behavioural factors (i.e. ‘lifestyles’). Certain behaviours, such as eating energy dense diets, taking little exercise, smoking and drinking large amounts of alcohol, increase the risk that someone will suffer from diseases like cancer, heart disease, lung disease and type II diabetes. These diseases are very unpleasant – sometimes fatal – for those who suffer from them, their friends and family. They also create economic harms, requiring healthcare resources to be directed towards caring for those who are sick and result in reduced productivity through lost working hours. For instance,the annual cost to the economy of obesity-related disease is variously estimated as £2.47 billion£5.1 billion and a whopping $73 billion (around £56.5 billion), depending on what factors are taken into account and how these are calculated. Since incentives are generally seen as useful tools for influencing people’s behaviour, why not use them to change health-related behaviours? Why not simply pay people to be healthy?

A number of schemes using financial incentives to reward healthy behaviours have been tried out, including offering grocery vouchers to pregnant women in exchange for not smoking; variable cash rewards for successfully participating in a weight loss program; shopping vouchers for women to breastfeed their children; payments to adhere to anti-psychotic medication, and many others. These schemes vary along numerous dimensions, including the form and value of the reward (e.g. cash, vouchers, gifts), the payment schedule, the target behaviour, the recipient, and the identity of the individual or organisation offering the reward. There may also be different methods used to ensure the recipient has ‘earned’ the reward. For instance, payment for quitting smoking may be dependent upon passing a carbon monoxide breath test to show that no cigarettes have been smoked.

Many people have an intuitive, hostile response to the idea of some or all of these schemes. There seems to be something about paying people to take care of their health that doesn’t sit well with many of us. A number of factors could be driving such a response: perhaps people simply think incentives won’t work, and are just a waste of precious resources. We might worry that incentives will put undue pressure on people to behave in certain ways – coercing them to act contrary to their preferences. Some might think that it is simply none of the state’s business how someone chooses to live their life, and so it is wrong for the state to interfere to seek to change lifestyles in this way. Or it might be argued that incentives are unfair: that those who already lead healthy lifestyles won’t receive a reward for doing so, whilst those who fail to live healthily will now have their ‘bad behaviour’ rewarded with the potential to be paid to stop.

A number of these criticisms depend upon empirical evidence. For instance, the claim that incentives won’t work (the evidence on which is mixed). However, some of these criticisms incorporate contentious moral claims, such as the idea that the people who receive incentives are undeserving.

A criticism I find particularly interesting comes from the political philosopher (and BBC Public Philosopher) Michael Sandel. Sandel worries that when we start to pay people to do things that are socially valuable, we commodify them. This means that we encourage people to view them as only instrumentally valuable: as something we can exchange or use as a bargaining chip in order to get other things we care about. Sandel argues that there are some things which become corrupted once we start to view them in this way. He uses examples such as friendship and baby selling to illustrate situations where we might feel squeamish about subjecting things to monetary exchange. We are right to worry about such transactions, Sandel argues, since they expose the subjects of the transaction to monetary valuation and exchange, commodifying and corrupting them. Moreover, Sandel worries that this corruption is irreversible, and that once we start to treat thing as if it were acceptable to buy and sell them, we can’t go back to treating them with the respect we did before they were commodified.

Whilst Sandel’s account holds some appeal, I’m not sure we should take it as a persuasive argument against the use of incentives to encourage healthy behaviours (something Sandel himself discusses in his book What Money Shouldn’t Buy, in a chapter on ‘Bribes to Lose Weight’). Part of this rests on Sandel’s empirical claims, specifically, that subjecting certain things to monetary exchange corrupts them in some way. In the case of health incentives, this might imply that paying someone to quit smoking, or breastfeed their child, or lose weight, causes them to value not smoking/breastfeeding/weight loss only insofar as those behaviours earn them money, rather than valuing them as a part of a healthy lifestyle. Yet it is not clear that people really will come to value those activities only instrumentally. The opposite effect might arise – the offer of an incentive to breastfeed could well signal to the recipient that this behaviour is important and worthwhile. Thus, offering incentives could reinforce values rather than corrupt them. Similarly, Sandel doesn’t establish that, if corruption does happen, it is irreversible. Both the claim that incentives corrupt, and that this corruption is irreversible are empirical claims, and ones with a contentious evidentiary basis.

Sandel also doesn’t really explain why we should care about the corruption of certain things – such as the view that we ought to value health-promoting behaviours for the sake of health rather than cash reward – beyond an assertion that this is likely to be less sustainable (i.e. if you start paying people to be healthy, rather than encouraging them to be healthy for the sake of health alone, then you will have to keep on paying them). First, this is another empirical claim that is not (thus far) established. Second, and more importantly, I’m not sure we do have to value health (and various other things) for its own sake. I care about being healthy precisely because it permits me to lead the kind of life I want to lead – go to work, enjoy leisure activities, avoid discomfort. I might also need a certain amount of money to be able to engage in such a lifestyle. Sandel seems to be guilty of perfectionism (in the philosophical sense) when he suggests I should value things in certain ways for certain reasons.

A final way in which I think the corruption argument seems rather unconvincing is that it uses the wrong comparison when it proposes that incentives create a worse situation. The worry seems to be that people should adopt healthy behaviours (quit smoking / lose weight / breastfeed) because those behaviours have important health benefits and express a particular attitude towards the importance – and value – of health. But the problem is that the people targeted by incentives are already not valuing those behaviours  in the way Sandel wants them to, because they are not engaging in healthy behaviours. Thus, the corruption worry compares a situation where people lose weight because they want to have long healthy lives with a situation where they lose weight because they want to earn some money. But this is not the right comparison. The right comparison is between people not losing weight, and people losing weight because they want to earn some money. In this case, it doesn’t look like a better motivation has been replaced with a worse one, but only that a new motivation has been created.

Arguments like Sandel’s that warn us of the dangers of ‘market creep’ and the dangers that commodification might bring, sound initially appealing. However, we need to think carefully about how plausible and substantiated they are. It might well be the case that there are good reasons not to use money to motivate desirable behaviours, including healthy behaviours. But we shouldn’t assume that our initial distaste is based on sound arguments. Rather, we should scrutinise those arguments and consider how well supported they are by both available empirical evidence and philosophical reasoning.

An extended discussion on this subject, ‘Social values and the corruption argument against financial incentives for healthy behaviour’ has been published in the Journal of Medical Ethics (2016) Vol. 43 No. 3 pp. 140-144.

  • Facebook
  • Twitter
  • Reddit

3 Responses to Are Incentives Corrupting? The Case of Paying People to be Healthy.

  • Andrews says:

    Perhaps this is not clearly articulated under Sandel’s pen but one of the better arguments against the “commodification” of certain valuable behaviours is that they present a risk of hindering the motivational-epistemic structure that underpins *ideal* adaptation of actions to reasons. The idea is this. Ideally you’d like people to be sensitive and responsive to reasons so as to turn these reasons into motives toward values (i.e. those constitutive of the reasons they are sensitive or responding to). And ideally you’d want to have this process to preserve the subject’s epistemic access to those values throughout. This is because you don’t just want people to change behaviour; you’d want them to change behaviours for the good reasons, which here means for reasons which they acknowledge as their own and in virtue of which they are motivated to change behaviour.

    Now if you severe the motivational-epistemic link between reasons and action, you run the risk of estranging people from their own behaviour, in the sense that they are no longer able to identify a single source of their motivation and target of their best rational judgement. And if this happens at the scale of society you run the risk of estranging people to each other’s behaviours, to the effect that people are no longer able to come together on what really should motivate them according and be the target of their best (collective) rational judgment. In short, you run the risk of making people less and less able to understand the moral grounds for certain beahaviours.

    Of course I can hear you say that we need empirical evidence in favour of the claim that commodification severes the ideal motivational-epistemic structure between reasons and actions; or that in certain circumstances (i.e. when the people are not yet aware of the reasons in the appropriate way to see them as reasons for actions with a motivational potential), we are fine with a non-ideal, optimal surrogate for that structure, and we don’t need to look for a better one as long as the one we have suffices for reliably bringing about good outcomes. Perhaps you will be right. Nonetheless at the end of the day, the question boils down to this: Do you want to have a single unified moral system for society or several? If you want to have one, and if this moral system elevates the kind of motivational-epistemic structure mentioned above as something which everyone should aim for, you have to admit that commodification is at best a temporary stage (perhaps an acceptable one from a consequentialist point of view which does not value this kind of motivational-epistemic structure) on our way toward a more desirable moral standing.

  • Matthew Cote, MD says:

    I commend Andrews desire to preserve a motivational-epistemic cohesive ethical structure, as a societal ideal. However, the evidence that we do have clearly indicates that there are individuals, aware of the adverse health consequences of their decisions, who continue to pursue the unhealthy behavior choices (overeating, smoking, etc). They do this because either they don’t care about the health effects, there is someother perceived gain that out weights the perceived health risk, or they lack the ability to change (through deficiency in either internal and/or external circumstances). Certainly a motivational incentive of payment may be inductive to some in each category to reconsider. The bottom line goal in all circumstances would seem to be the improvement of the individua’ls health in regards to minimizing adverse health outcomes that not only will affect the individual, but also directly affect society at large through healthcare expenditures, either resulting in a general decreased quality of life for everyone in the society (to the extent that one accepts a premise of a decrease in personal financial resources resulting in a lower happiness potential) as well as societal loss of the individuals productivity and social contribution, or the improvement thereof when the individual improves their health and thereby avoids bad health outcomes. What is the ethical obligation of the individual to society in this regard? If the end is a validation of an individual’s autonomy to choose the good or bad for themselves without additional (commodifying) inputs, then what is the ethical obligation of others to be obligated to dealing with the resulting negative consequences (higher taxes to cover healthcare costs for example) except as a degree of loss of their own (resource) autonomy? It would be ideal, as Sandel desires, for individuals to choose behaviors solely for their health benefits – but they demonstrably do not. People are not always rational in their choices despite knowing all the facts,
    Regarding the commodification concerns, in some ways, its seems that commodification is already occurring with the concept of health in general, independent of any individual’s health related choices. Of course, I am merely an emergency room doctor, so my knowledge of these things is probably inconsequential.

  • Andrews says:

    I wonder if the author intends to reply to comments or if this post just acts as a pointer to the article of hers published in the Journal of Medical Ethics.

    O Internet, you shapeless flux…