Friday 11 November 2011

Costs of Smoking: Auld edition

Chris Auld makes the very sensible point that we cannot tally smokers' costs to the public purse by simply adding up the cost of treating smokers; rather, we need to compare the total lifetime costs of smokers with the total lifetime costs of non-smokers. I'd add that we'd also have to adjust for that smokers would likely be greater health risk takers even absent smoking. That's at pretty strong variance with the procedure that New Zealand's Ministry of Health used.

But Auld raises the more interesting question: what happens if leading a healthy lifestyle winds up costing the government more because you spend longer as a superannuitant and you wind up dying of prolonged aging-related diseases rather than the quicker and cheaper vice-related diseases?
If healthy behaviors wind up increasing lifecycle health care costs, we should either subsidize less than we otherwise would, or perhaps even tax, healthy behaviors. Healthy behaviors in this scenario benefit the person exhibiting the behavior but impose costs on everyone else, and this logic demands that we discourage healthy behavior relative to whatever policies we would otherwise have enacted.
This argument does not sit well with me. I dislike the argument that we should penalize smokers because of the health care costs of treating smoking-related illnesses, and I dislike the argument that we should penalize non-smokers because of the health care costs of treating smoking-related illnesses. These external effects smack of standard pecuniary externalities—externalities that operate through market mechanisms and do not require policy to fix, although they are not conventional pecuniary externalities per se. The perverse aspects of arguing that we need to control externalities which are artifacts of government programs is illustrated more forcefully by studies which treat the cost of investigating, arresting, and imprisoning illicit drug users as external costs of drug use itself! (“We are arresting you for this doobie.” “Why?” “Because the costs of arresting you are a negative externality!”)
Cost of drug use figures that count the costs of enforcing prohibition regimes are best deemed police agitprop.


  1. Totally of task...whats ur take on this...

  2. Eric - I read Chris Auld's original article on the Globe and Mail Economy Lab (linked from your link) and thought "yup, that makes sense."

    My favourite quote from Chris's Globe and Mail piece:

    "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."

    I liked it so much I wrote on it here:

    If you look you'll see that the Globe and Mail post got hundreds of comments, mine got 60. People just couldn't deal with the cognitive dissonance: smoking is bad for you, but it might end up saving the health care system money in the long run.

    I'd be interested in knowing the differences between alcohol and tobacco. My sense is that alcohol addiction has very large societal costs in terms of things like domestic violence, marital breakdown, etc etc - but somehow because they're inflicted on the drinker and the drinker's family, as opposed to strangers and the health care system, and they're harder to quantify (what's the cost of the fear that tonight he'll be drinking and he'll start shouting at the kids? what's the cost to a child of seeing mom passed out on the floor?) they attract less policy attention.

  3. Bob Jones once made two observations about smoking..

    The first was that smokers seemed to be a very hardy breed up to somewhere past age 40, and then they increasing succumbed to smoking related health issues.

    His suggestion was to use these two factoids to the benefit of all, ie, by age three children should have a pack a day habit and from about age 20 people should be progressively weaned off the habit by age 30 or thereabouts.


  4. I ran through the numbers from an Australian government estimate of the costs of smoking a while back taking into account the decrease in aged pension expenses. The results depended a lot on how much of the "net production costs" of Tobacco use you estimate is an externality. At 0% the correct "tax" rate ends up being a subsidy of 18c per cigarette, while at 100% it ends up being an actual tax of 16c per a cigarette (noticeably less than the current rate).

    It should also be pointed out that in 2004 (which is when the government estimate of costs was done) the cheapest cigarettes where apparently about 45c each, 22c of which was tax)

    Pointing this sort of thing out is an interesting thought experiment that you can suggest to people: "If the cost of healthy eating to society was actual a net-negative would you support taxing it?" If not then it's obvious that the "Cost to the taxpayer" reason for taxing tobacco is poorly a rationalisation.

  5. @Anon: now posted. Canterbury provincial holiday; was out with the kids this morning.

    @Frances: Thanks for the pointer. I tend to think your explanation 3 gets most of it, but perhaps adding in a bit of voter distinction between deserving and undeserving sick people.

    There are tons of papers on the social cost of alcohol use. Most of them deem not only the cost imposed on family as being social, which is a hard case to make given any kind of Coasean considerations (and, further, once we go that way, where do we stop?), but also the costs drinkers impose on themselves. So one crappy NZ study counted $700 million spent by drinkers on alcohol as being a social cost - a sixth of their final figure; an equally bad Australian study did the same thing.

    A more recent Australian study tried to quantify the kinds of costs you're thinking about, but has some awful howlers. For instance, the average difference in reported happiness between those knowing a nasty drinker and those not knowing a nasty drinker is monetized and ascribed entirely to booze rather than giving any consideration to that the kinds of folks who know a lot of heavy drinkers probably have other differences from the kinds of folks who don't.

    @JC: Ha! I'll not start Ira, despite that strong endorsement.

  6. @TimP: Sorry you were caught in the spam filter. Will have to peruse the figures.

  7. Eric: Thanks for those references and the reply. Another reason to be cautious about those estimates - alcoholism is quite often a form of "self-medication" - i.e. alcohol is being used as a substitute for Prozac or other prescription anti-depressants. So the best comparison would be between long-term users of alcohol and long-term users of comparable prescription drugs.

    "Most of them deem not only the cost imposed on family as being social, which is a hard case to make given any kind of Coasean considerations"

    I don't know how many people have noticed that Becker's Rotten Kid Theorem is just an application of the Coase theorem. Ted Bergstrom has some nice stuff on the limitations of the Rotten Kid theorem(which suggests interesting parallel limitations to the Coase theorem). Rotten Kid Theorem says household members will act so as to maximize utility of (altruistic) household head. Doesn't always hold.

  8. Didn't know that anybody took the Rotten Kid other than as a Coase theorem application.

    Agree on morbidity with and self-medication for depression, among other things. We attempted to correct for this in a reverse engineering and correction of a NZ alcohol costs study. The authors weren't happy.