Sunday, November 18, 2007

Private costs of cigarette smoking

From Andrew Leigh I learned of a remarkable new paper by W. Kip Viscusi and Joni Hersch on the private costs of cigarette smoking. Only the summary is available at the link though you can download the whole paper if you have a university email address.

This is important research and first-rate applied economics with huge policy implications.

The private costs of cigarette smoking are the costs a smoker bears themselves as a consumer – mainly here the purchase costs plus the costs of ill-health and increased mortality.

There can be contrasted with the external costs of smoking which society bears as a consequence of smoking – these are mainly the extra health care costs - beyonf those borne by thye individual - that society picks up as a consequence of smoking and the very real costs of passive smoking.

These external costs are significant even if tax yields from excise charges on smoking exceed their implied monetary valuation of the externalities. Try telling a wife that her lung cancer death is offset in value by the extra taxes that her smoker husband contributed to the public purse. Moreover since a Pigovian tax is directed at marginal damages there is no reasohn for supposing that efficiency requires tax bills to equal non-internalised costs.

But in straight numerical terms - adapting a human capital viewpoint - the private costs of smoking are much more significant than these externality costs. It is the reason that I emphasise the role of private costs in my own arguments against smoking. Of course if you follow the libertarian view that people rationally weigh up all the costs of smoking and then smoke as freely-chosen, rational decision you won’t buy an emphasis on private costs at all. But I have never believed that smoking involves acts of rational choice.

A more accurate description would be that smoking is an act of adolescent folly that smokers continue to pay for over the rest of their lives because of the addictive character of nicotine.

I follow Andrew in quoting the abstract of the Viscusi and Hersch paper.

This article estimates the mortality cost of smoking based on the first labor market estimates of the value of statistical life by smoking status. Using these values in conjunction with the increase in the mortality risk over the life cycle due to smoking, the value of statistical life by age and gender, and information on the number of packs smoked over the life cycle, produces an estimate of the private mortality cost of smoking of $222 per pack for men and $94 per pack for women in 2006 dollars, based on a 3% discount rate. At discount rates of 15% or more, the cost decreases to under $25 per pack.

The figures suggest that the single packet of cigarettes purchased for about $12 would end up costing a consumer around $222 in total at reasonable discount rates once health costs are factored in. The estimated cost per packet is about 10 times the cost of earlier estimates by Sloan that have been discussed before on this blog. I’ve always found a cost in the order of $20 per packet to be about right. Ignoring discounting then if, as Sloan et al assume, a male loses 4.4 years of life as a smoker compared to a comparable person who does not smoke and this life is valued at $100,000 per year then the total cost is $440,000. If one smokes a packet of cigarettes per day over 40 years then 14,600 packets are consumed altogether which works out at about $30 per pack.

The much higher figures of Viscusi and Hersch arise because they account for the fact that smoking doesn't just kill people at the end of their statistical lives. Significant numbers of younger people die too as a consequence of smoking – this recognition accounts for about half the difference in these costs estimates from much lower earlier estimates.

Viscusi has in the past been criticised by some for minimising the impact of smoking by emphasising the role of external costs of smoking (such as passive smoking) which are small relative to other costs. He argues in 'Smoke Filled Rooms' that smokers know about the consequences of smoking and internalise costs, pay enough cigarette taxes to far more than cover their personal health costs and hence should be left alone. I wonder if he now revises his views. He should - it strains rationality to believe that men are internalising accurately costs of $222 per pack when they buy a packet of fags.


Fredrik Eich said...

No doubt a wife of a smoker would indeed find it a little vexing if after being told that she had less than a year to live that "her lung cancer death is offset in value by the extra taxes that her smoker husband contributed to the public purse".
Especially if she were familiar with the 2003 Enstrom and Kabat study "Environmental tobacco smoke and tobacco related mortality in a prospective study of Californians, 1960-98” which would hardly reassure her that there was a proven measurable elevated risk to her health from her husbands smoking habit.
Doubtless if her husband, as well as being keen on the risky activity of smoking, enjoyed risky mountaineering and liked to do a bit of risky skiing every weekend – there would be "huge policy implications" for these activities too – policies that many of the worlds professional tobacco control advocates are so mysteriously keen on.

Fredrik Eich

hc said...

Yes Fredrik, But these latter activities don't impact directly on her.

Fredrik Eich said...

I am not quite sure if word 'directly' is being applied equally to the risk involved in all three activities. There is no area of human activity that does not involve risk. It is highly unlikely (although not impossible)that a husband while actively skiing would directly break a leg of a wife who was passively watching him engage in this activity. But one can not assume that all broken legs of all wives of skiing husbands are directly as a result of husbands' skiing activities. And given the conclusions of the study I mentioned above it is highly unlikely (although not impossible that the cancer question would have been directly caused by anyone smoking including her husband. Therefore her death is not offset in value of extra taxes because her death should not even come in to this equation.
But, then, the risky activity of skiing is not a trillion dollar global gold mine - as it is, sadly, for people who enjoy the risky activity of smoking.
Fredrik Eich