It depends on your perspective I suppose, and if you think that regulating/taxing anything is paternalistic and believe that everyone is rational, not addicted, truly knows the health effects, etc. then I agree there would be a percentage of the population where consumer surplus would be taken into account and who enjoy smoking (although this population who says they enjoy smoking today may not say that X years later).
I think you might be mistaken on several counts:
1) not all taxes are paternalistic (e.g. pigouvian taxes), but tobacco taxes almost definitely are. wikipedia:
Paternalism is action that limits a person’s or group’s liberty or autonomy and is intended to promote their own good.
rather, the debate is about whether or not paternalism is justified (in this instance). In both this comment and the grandparent I’ve implicitly assumed that paternalism can be justified, but you’re right that there are libertarian arguments that paternalism is essentially always wrong.
You might also enjoy Robin’s recent writing on paternalism, though it doesn’t directly bare on this argument; unsurprisingly he concludes it is primarily about status.
2) People don’t have to be fully rational for their decisions to have information about welfare. If full rationality were required then no decision ever would satisfy! While it seems clear that humans are not 100% rational, there is still some logic to people’s actions. Indeed, models of rational addiction have been around for decades; it is definitely not true to say that addiction invalidates any inference about consumer welfare. See for example Becker and Murphy (1988).
3) ‘truly knows the health effects’ is again not required. For example, if people had noisy but unbiased estimates of the health impacts, some would over-consume and some would under-consume, and on average, tobacco taxation would not be welfare enhancing. If lack of knowledge is the issue, the appropriate response is to provide the information, not to tax it.
4) Consumer surplus should be taken into account for everyone regardless of whether or not on the whole smoking is optimal. It is an element in the cost-benefit calculation (probably the largest on one side of the equation). It might be larger or smaller for different people, in different circumstances, etc., but that is something that must be estimated and taken into account, not simply ignored.
I think you might be mistaken on several counts:
1) not all taxes are paternalistic (e.g. pigouvian taxes), but tobacco taxes almost definitely are. wikipedia:
rather, the debate is about whether or not paternalism is justified (in this instance). In both this comment and the grandparent I’ve implicitly assumed that paternalism can be justified, but you’re right that there are libertarian arguments that paternalism is essentially always wrong.
You might also enjoy Robin’s recent writing on paternalism, though it doesn’t directly bare on this argument; unsurprisingly he concludes it is primarily about status.
2) People don’t have to be fully rational for their decisions to have information about welfare. If full rationality were required then no decision ever would satisfy! While it seems clear that humans are not 100% rational, there is still some logic to people’s actions. Indeed, models of rational addiction have been around for decades; it is definitely not true to say that addiction invalidates any inference about consumer welfare. See for example Becker and Murphy (1988).
3) ‘truly knows the health effects’ is again not required. For example, if people had noisy but unbiased estimates of the health impacts, some would over-consume and some would under-consume, and on average, tobacco taxation would not be welfare enhancing. If lack of knowledge is the issue, the appropriate response is to provide the information, not to tax it.
4) Consumer surplus should be taken into account for everyone regardless of whether or not on the whole smoking is optimal. It is an element in the cost-benefit calculation (probably the largest on one side of the equation). It might be larger or smaller for different people, in different circumstances, etc., but that is something that must be estimated and taken into account, not simply ignored.
Tobacco taxes are pigouvian under state sponsored healthcare.
Just wanted to write that I’ve seen your comment, I’m on the road and don’t have time to respond well now but will try in the next week or so.