“Of course, most economists would disagree vehemently [that taxing unhealthy behaviors is a good thing]. Raising taxes does tend to reduce consumption and, where consumption generates large negative externalities (costs borne by uninvolved parties) can even be efficient: Pigovean taxes (taxes proportionate to those external costs) can push us closer to socially-optimal outcomes. But, there is no inefficiency caused by people choosing to live lifestyles they view as preferable despite the health costs.
If I decide to enjoy a shorter life rather than eek out a miserable existence without wonderfully-marbled steaks, a beer or several, or even the occasional cigar, zero inefficiency is induced thereby.
…what evidence there is suggests that to the extent smoking induces a “fiscal externality,” the sign of the effect is wrong: smokers pay more in cigarette taxes than they ever cost the public purse. They die earlier of cheaper diseases and collect less in superannuation than do non-smokers. And, as a 10% increase in cigarette taxes correlates with a 2% increase in obesity, one wonders whether increased cigarette taxes consequently require further increases in taxes on fatty foods.
Crampton supports the idea of “De gustibus non est disputandum,” we should not criticize individuals’ preferences.