In his inaugural Economy Lab post, Chris Auld does a superb job of demolishing the idea that a healthy lifestyle substantially decreases demand on the health care system. He calls it the "zombie argument" because:
it has been repeatedly shot, stabbed, and poked at with sharp sticks, but it won't just die.
People still believe that getting people to stop smoking, say, or to lose weight, will lead to long-term reductions in health costs.
Yes, a trim non-smoking 50 year old probably costs the health care system less money than an obese, chain-smoking 50 year old. But as Chris cogently observes:
All people — and I do not mean to shock anyone — die some time, even including people who live very healthy lifestyles. Preventing someone from dying of a smoking-related illness only means that they will die of a non-smoking related illness. The effect of smoking on lifecycle health care costs is the difference between costs which are incurred if the person smokes and the costs which would be incurred if the person doesn't smoke. Whether improvements in lifestyle increase or decrease lifetime health care costs depends in a complicated manner on how a healthy lifestyle affects length of life and health care costs at any given age. Whether smoking or other unhealthy behaviors increase or decrease health care costs is an empirical question.
The question that Chris Auld's article raised in my mind was: why? Why is it so hard to convince people that public health measures, like reducing smoking, might lead to increased health care costs in the long-run? I can think of a number of explanations.
1. Policy makers are stupid. That's not a good economic explanation; the basic premise of economics is that people are (usually) not (that) stupid. But it might be right.
2. Policy makers aren't stupid, but they want to raise money by suing tobacco companies.
3. Policy makers are paternalistic. Unhealthy behaviours are bad, either because they are morally reprehensible (the sin of gluttony), or because unhealthy behaviours make people unhappy. The zombie argument – "quit smoking for the good of the health care system" – is trotted out in order to induce people to behave better. Or in order to justify our moralistic views.
3a. Because people believe unhealthy behaviours are bad, the idea that these behaviours could be beneficial – that they could, for example, reduce health care spending – causes cognitive dissonance. Thus people refuse to believe it.
4. Policy makers have high discount rates. The person who quits smoking today will soon be at lower risk of suffering a heart attack – and at increased risk of living long enough to get Alzheimer's disease. But if a policy maker has, say, a 10 percent discount rate, $100,000 in health care expenses in 20 years time is equivalent to less than $15,000 in health care expenditures today. So present costs matter far more than future costs.
5. Policy makers only care about certain costs. Some types of illness place large burdens on Canada's publicly funded health care system – cancer, for example, which is largely treated in hospitals. Other forms of illness place burdens on informal caregiving networks. If you or someone you love is diagnosed with Alzheimer's for example, it will generally be up to you to provide home care, or arrange and pay for at least part the cost of long term residential care. If quitting smoking reduces cancer treatment costs by $1 and increases long-term care costs by $1.50, the publicly funded health care system might still have saved money.
I don't know what the right answer is. But I did find Chris Auld's exploration of the issue well worth reading.
Also worth reading in Economy Lab: Tammy Schirle's inaugural post "We are the 30%". Chris and Tammy, welcome to the E-Lab team!
Update: A number of people in the comments below made the argument "healthy people are more productive for more years, generating more tax revenue" – and thus more health care funds.
The fiscal sustainability of health care depends upon health care expenditures per capita relative to productivity per capita.
Smoking has two impacts on per capita productivity.
First, it decreases the productivity of smokers while they're alive. For example, smokers get colds more often. This effect tends to decrease per capita productivity.
Second, smoking kills off smokers, so smokers live less long. Because older people are, on average, less productive than younger people (sad, but true, search the medical literature), this effect tends to increase per capita productivity.
The effect of smoking on per capita productivity is, therefore, ambiguous.
Thus the effect of smoking on the resources available to pay for health care costs is theoretically ambiguous – as is its impact on those costs.
Worthwhile Canadian Initiative 
Take a few slow, deep, breaths.
Nicotine-infused ones, if that helps.
And mellow out guys.
“Nicotine-infused ones, if that helps.”
You’re just saying that so you don’t have to pay for my CPP, OAS and health care. 🙂
Bob: “I suspect for most of us, the value of the foregone years for Joe the Smoker in Montreal is precisely zero.”
(This reminds me of the discussions on immigration policy.) I think for the average Canadian the value of the foregone years for Joe Stranger must be >0. If this is false why do we (appear to) care about famines, tsunamis, forest fires, HIV treatment, lupus, cystic fibrosis, plane crashes, train wrecks, foreign dictators, or any other front page news story?
At a bare minimum, I think to support your claim you would need find enough people for whome the value is negative to offset the anonymous kidney donor.
sigh
Frances, in complete honesty, your argument with CPP can be levied at your Carleton Pension which is also DB (with Additional Voluntary Contributions, it is not DC. I had this debate with Nick in a different thread a while ago, Carleton puts its info online and in public). DB Pensions are a form of life annuity and therefore life insurance. By design they involve pooling of risk and reward. CPP is no different in this way than any other DB plan.
Bob:
Please stop using American concepts. The CPP benefit is set by legislation, just as a pension formula is set by the employer. Pension formulas in multi-employer plans can be changed. There is no legal recourse under legislation to the Consolidated Revenue Fund for a CPP deficit. A CPP deficit has to be met with a benefit cut (50% loss of indexing) or a rate increase up to 9.9% of pay.
Recourse to sponsor funds, either employer or government is a well-known concept. Single-employer plans operate this way, their forulas cannot be reduced, the sponsor has to make up a deficit. CPP is actually more conservative than employer pensions.
Your government pension plan arguments are simply baseless. Take a look at the Public Service Pension Plan, the one for the Public Service of Canada. I did, I have applied for PS jobs and had interviews.
Frances:
The case for anti-smoking is actuarial. Actuaries divide the population into two groups, smokers and non-smokers. Smokers, based on experience in diverse lines of private insurance and in government programs, have shorter lives and Higher claims and costs. The people who live longer and are allowed to catch anything they can get still make lower claims. We have already allowed them to be as expensive as they want to be and still they cost less.
Public Health is intensely actuarial and empirical.
Chris Auld’s argument is the same as arguing that we shouldn’t build more expensive fire-proof homes because the initial cost is higher. The house has a finite life anyway. The real world says we are all better off if we build better homes.
There is a deep misunderstanding here of economic predictions vs. actuarial predictions. Actuaries start from evidence, not theory.
“Economists tend to write (and think) in a utilitarian framework where the benefits of being alive = the happiness enjoyed while being alive. So I’m using happiness in a technical way as a metaphor for the value of life. ”
I guess what I’m getting at is that aside from this happiness, people who are healthy also tend to produce goods and services, pay taxes, etc. Ie, does someone with x+3 years of healthy lifespan have a higher NPV to society than someone with x years?
Ah, the dismal science.
I believe the “stupid” policy makers have it right. When considering what might be good or bad for society one not only needs to take costs into account but one also needs to consider something like a social welfare function. The social welfare function would be a sum current and future utilities for all persons. There are two different scenarios we can consider, a world with and without smoking. I think that the social welfare function in the world with smoking lies unambiguously lower (is worse) than the social welfare function in the world without smoking because people die off early and their utilities are not in the sum. True, some people actually enjoy smoking to the point where maybe they’re glad to have done it even taking into account if they die early. But given the number of people who regret it and the externalities involved I think society believes that it is a bad and that social welfare is lower with smoking.
There is also the cost side which would affect the resource constraint. As noted we can’t really tell whether on net there’s an increase or decrease in costs to smoking. But it seems that this consideration is orders of magnitude smaller than the effect on the social welfare function to be not even worth considering. Incidentally, it’s also why we care (or should anyways) about people suffering from mental illness, drug addictions and other kinds of illnesses.
ml –
On your utilitarian calculus: It’s not obvious whether a utilitarian should try to maximize the sum of utilities or average utility. The two have quite different implications, as you can see in the example you give.
Andrew F:
“does someone with x+3 years of healthy lifespan have a higher NPV to society than someone with x years?”
After reading your comment, I spent a gloomy half hour contemplating whether or not the NPV of my existence, i.e. the difference between the happiness I generate and the resources I consume, is greater than zero. And even if it’s true now, will it be true in 10, 20, 30, 40 (if I live that long) years time?
Thank you ml for your very wise post.
ISTM that we should try to optimize average utility. Optimzing the sum rather than the average implies we are happy with one person or a small group’s social utility dominating the sum, in other words an unequal world. If you care at all about equality, “social justice”, fraternity or anything like that you should go for the average.
Frances: At this point many people turn to religion.
Frances, I wonder because also in the back of my mind (somewhat related to this topic) is the issue of life extension. Oddly, many people are opposed to life extension because they somehow believe that living a long time is awful. But if you couch it in terms of delaying or reversing the effects of age-related diseases (ie, extending healthy lifespan), people are generally in favour.
It seems to me that if you can extend life expectancies by 10 years without extending the period of decline at the end of life (quality of life is more like a rectangle with a short, sharp decline at end of life rather than a longer slope down), that would be a very worthwhile endeavour for society, even if it came at the result of increased healthcare costs.
I didn’t intend to trigger a melancholy reverie. I expressed it in terms of NPV to help get my point across, but I think the social welfare function is a better way of expressing it. I certainly wouldn’t use strict social NPV analysis as a basis for morality! We are more than the sum of our discounted cash flows.