Pigou and Paternalism

Here is a question from the final exam for my public finance course:

A typical person’s demand for potato chips is given by p=5-q where q=the number of packages of chips purchased, and p is the price of chips in dollars per package. The marginal cost of producing potato chips is $1 per package. The potato chip market is competitive, so p=MC. Dr. Economides estimates that each package of potato chips consumed creates marginal damage of $0.50. 

1. Illustrate these facts with a diagram.
2. Define the phrase “Pigouvian tax”
3. Show how a Pigouvian tax could be used in this example to improve efficiency.
4. Calculate the revenue raised by the Pigouvian tax. Show on your diagram, and calculate, the efficiency gain created by the tax.
5. Opponents of the tax on potato chips take a careful look at Dr. Economides’ study. It turns out that the only people harmed by potato chip consumption are potato chip eaters themselves, as potato chip consumption is associated with bad skin, weight gain and depression. Does this strengthen or weaken the argument for taxing potato chips?

Solutions to parts 1 to 4 are here. The interesting part is question 5: if potato chips only harm potato chips eaters, does that strengthen or weaken the case for taxation? Vote here to express your opinion (only the first 200 answers will be counted).

"Pigouvian taxes" are imposed to make people take into account the full impacts of their actions. The classic example is a carbon tax. Whenever people produce or consume carbon-based fuels they harm other people, from those who breathe in exhaust fumes to future generations affected by climate change. A carbon tax makes people recognize the full cost of their consumption and production choices, thus creates incentives for people to change their behaviour.

The potato chip example, however, is different: the chip eater in the exam question isn't harming other people. The bad skin, weight gain, and depression is experienced by the chip eater himself. It is not obvious that a Pigouvian tax is needed to make chip eaters take into account the harms that they causing - why not just tell people that potato chips cause bad skin, and if people think the chips are worth the zits, let them munch away? They're only harming themselves.

Interestingly, a few of my students thought that this additional information about who is harmed by potato chip eating strengthened the case for imposing Pigouvian taxes. 

Why is that?

One possibility is that the students were inadequately taught, or simply didn't study the material.

Some students disputed the basic premise of the question, the idea that potato chip eaters are only harming themselves. Bad skin, weight gain and depression, they argued, are harms to others, because we have a public health system. Yet the starting point for the analysis is that potato chips create damage of $0.50 per package. If chip eaters take any of any of that damage into account when making chip purchasing decisions, the case for taxing chips weakens. Also Canada's public health system often does not cover skin creams, anti-depressants and counselling, so the link between bad skin and depression on the one hand, and health care spending on the other is not obvious. 

What interested me about this response was how "health" becomes a lens through which public policy issues are viewed, and a justification of policy choices. Perhaps, though, the students were just mislead by the wording of the question. Guessing that the specific details about bad skin, weight gain and depression must matter in some way, they figured that the question must be asking about health care. 

Finally, it seems that some students really don't believe that people are rational decision-makers, fully taking into account the long-term effects of their consumption choices. Even when people are only harming themselves, they support Pigouvian taxes on paternalistic grounds, to stop people from harming themselves.

Today's Chronicle of Higher Education has an article arguing that mainstream economics is dominated by a "free-market fundamentalism," one tenet of which is that people are rational decision-makers. While this may be true, it's far from obvious that mainstream economists are successful proselytizers, bringing others around to their way of viewing the world. 

Update: 62 responses to the survey so far, 87% have responded "weaken". 

Comments

[Nick Rowe]

My immediate thought was: "do the chip eaters know eating chips causes zits?" Then I realised that even in the extreme case where none of them know this, all this means is that the case for the tax is neither strengthened nor weakened.

[Greg Ransom]

Ummm, the $0.50 number was pulled out of thin air?

The voodoo of the "Dr." before the guys name doesn't make the $0.50 any more of a fact about reality than if a wild monkey somehow generated to number.

Did any student call BS on the notion apt hat a made up "fact" is somehow a fact because it is made up by a "Dr."?

A famous expire meant of a "Dr." in a white lab coat instructing people to shock people comes to mind ....

[Greg Ransom]

Does anyone realize that the leading free market thinker of the last 100 years rejected this picture?

"mainstream economics is dominated by a "free-market fundamentalism," one tenet of which is that people are rational decision-makers."

Mainstream economics in the U.S. is dominated by Democrats, who make up a majority in the discipline.

There is a lot of muddle and myth about economics and the profession packed in the sentence.

[Min]

"it's far from obvious that mainstream economists are successful proselytizers"

How hard do you try? I can't say that I have ever seen an economist, or anybody, trying to convince people that they are rational. ;)

[Frances Woolley]

Greg: "Ummm, the $0.50 number was pulled out of thin air?" "Did any student call BS on the notion apt hat a made up "fact" is somehow a fact because it is made up by a "Dr."?""

Yup, as was the demand function and the stuff about potato chip markets being perfectly competitive. "Dr Economides" is actually the name of a grad school drinking buddy.

Excessive realism makes exam questions far too difficult. Students - being rational and self-interested - just want questions that are reasonably simple and straightforward, and ones that examine the material taught in the class (as opposed to stuff that wasn't covered.)

Nick, exactly.

[K]

Frances: "Nick, exactly."

Is that right? If you impose the tax then the chip eaters will bear the full harm twice. Once from the tax (that they know about) and then again from the zits (that they don't). So they will consume the correct amount of chips *and* you will effect an additional random wealth transfer. So the tax is bad.

[DavidN]

I would guess the majority of students who answered ‘strengthens’ fall into the first category (‘students were inadequately taught, or simply didn't study the material’; I’m presuming the students are freshman or second year?). For the students who fall in the second category you can always add a statement along the lines of ‘assume all premise(s) of question are true’. For the third category, I will give some marks for ‘thinking outside the square’ but not full marks. I would give an extra mark if a student answers along the lines of ‘strengthens unless it can be shown people exhibit systematic cognitive bias towards consuming potato chips in which case it weakens the case’.

Btw, I like this exam question.

I don’t have an issue with the content of the typical freshman econ course (though I wouldn’t mind seeing more eclecticism) but I do believe not enough emphasis is given to distinguishing between positive and normative economics which is a problem. If most of the students who take econ 101 take no other econ course and during the whole of the course the difference between positive v normative economics is given no more than a couple of sentences for definition sake, and they’re being taught out of Mankiws book then I’m not surprised there is a general impression that "mainstream economics is dominated by a ‘free-market fundamentalism’".

I just don’t think most undergraduate students are sophisticated enough to grasp the subtle difference between a policy which is ‘optimal efficiency wise’ (which is the position often being taught in class) and a policy which should be supported on normative grounds due to morality, ethics, socio-political factors etc. etc.

[DavidN]

hence the belief that ‘economics is inherently ideological’.

[DavidN]

*impression (instead of belief)

[Greg Ransom]

A part of my explanation - building on Coase - for the pathology which is "economic science" is that much of what is ordained as "science" has been driven by what is easy for professors to teach & test, e.g. Samuelson's textbook was great for teaching to engineers & graded by TAs and even grading machines.

"Excessive realism makes exam questions far too difficult. Students - being rational and self-interested - just want questions that are reasonably simple and straightforward, and ones that examine the material taught in the class (as opposed to stuff that wasn't covered.)"

[reason]

I think the answer to 5 must be - you've never cleaned buses or cinemas.

[Michael]

Since the information in #5 states "the only people harmed by potato chip consumption are potato chip eaters themselves" I would write an answer that says the case for using a Pigouvian tax is weakened. But the actual wording of the question is: "Does this strengthen or weaken the argument for taxing potato chips?" So I had to ask myself "does the professor mean to include non Pigouvian taxes?" Here in New York and other parts of the US there are proposals to tax soda and other sugary drinks. The parallel between potato chips and soda led me to think that maybe the professor is referring to "sin taxes." Thus, although the tax is not Pigouvian, the case for it still exists (is strengthened). Perhaps I would have known the professor's meaning had I gone to lecture, but, alas, I overslept that day.

[reason]

Michael,
good point. Both revenue and paternalism are perfectly valid reasons for a tax. If the effect is more concentrated, it may make the paternalism argument stronger not weaker.

[K]

Michael,

Assuming the same quantity of harm, the case for the tax is reduced. In both cases, the tax will result in the optimal amount of chip consumption. But in the case where the harm is to the consumer, there is an additional effect of an unjustified wealth transfer from chip eaters to the public. Including the resulting possible income effects means the tax is less justified.

[Michael]

K,

The health care economics textbook I used last semester has an entire chapter on "health capital." The analysis is parallel to "human capital." Could it be that the inefficiency you identify is simply an investment cost of health capital? As a parent of a teenager I would like her to reduce her consumption of chips and soda both today and in the long run. Today I am trying to modify her habits with respect to her long-run health not just her current behavior. Perhaps this is one of economic sciences' limits: it takes preferences as given, I want to modify them.

[Marc Labbé]

....as potato chip consumption is associated with bad skin, weight gain and depression. Does this strengthen or weaken the argument for taxing potato chips?

If a close of you is depressive, you will be affected too...

Then revise your argumentation

[Frances Woolley]

K "Is that right? If you impose the tax then the chip eaters will bear the full harm twice. Once from the tax (that they know about) and then again from the zits (that they don't). So they will consume the correct amount of chips *and* you will effect an additional random wealth transfer. So the tax is bad."

Economic analysis tends to focus on efficiency. The way this question is structured, it turns out that, on the margin, the reduction in harm outweighs the reduction in chip eating enjoyment, so the tax increases efficiency.

Yet, as you rightly point out, there is also a redistributive effect - and for things like carbon taxes, the redistributive impact can be very large relative to the efficiency impacts. But because economists tend to ignore these redistributive impacts we're left scratching our heads unable to figure out why people don't like our carbon tax proposals.

Michael: "So I had to ask myself "does the professor mean to include non Pigouvian taxes?" "

K gives a good answer to your concern. It's a ceteris paribus thing. *All else being equal,* if people recognize harms that they themselves suffer, there a weaker case for taxing potato chips. Taxation is actually next term, and in that course I tell students that, if you're going to remember anything from this course, its: broad base, low rate.

reason: "you've never cleaned buses or cinemas" ;-) True. That would strengthen the case for soda taxes, too.

DavidN " policy which should be supported on normative grounds due to morality, ethics, socio-political factors etc. etc."

We do some formal normative econ in the course, e.g. social welfare functions, social indifference curves, the utilitarian case for income redistribution. People typically don't connect with it - perhaps because it's taught in much the same way as consumer theory is taught, and students typically don't enjoy that, either. It is something I need to find a better way of teaching.

[Frances Woolley]

Marc "If a close of you is depressive, you will be affected too..."

Exam questions very often taken the form "True, false, or uncertain: If A, then B." The aim of the question is to see if students understand the nature of the relationship between A and B. A good number of students will generally reply "Not A." Even if "Not A" is a perfectly reasonable position to take - potato chips make buses hard to clean, depression hurts other people - it doesn't address the issue at hand, which is "If A, then B."

What's sometimes called "exam technique" is often just simple logic.

Michael - "Perhaps this is one of economic sciences' limits: it takes preferences as given, I want to modify them. " As I noted in the post, this is a very common position. With regards to your own children perhaps you have a right, even a duty, to shape their preferences. But do you have a right to modify or disregard mine? Perhaps I smoke 'cause I'm hoping for an early death, and I need to cling to something. So why take it away from me?

[Michael]

Frances,

I do not have the right to modify your preferences, but I do believe that through democratic processes the community has that right. I cannot imagine a society in which the community is not modifying and disregarding preferences. With democratic processes we humans are simply gaining some control over our social ordering. Ah, but this takes us far away from economics.

[Frances Woolley]

Michael : "I cannot imagine a society in which the community is not modifying and disregarding preferences."

And it's always been our most important social institutions that have modified preferences (and vice versa). Historically that was family, church, state. Now it's Walt Disney and McDonalds. But when students have got their basic values from watching Family Guy, it's really hard to even have a discussion about morality and ethics. People tend to like the idea of rights, but then the question becomes "what do we do when rights conflict?" When my right to eat potato chips conflicts with your right not to pay for health costs associated with self-inflicted harms. It's one thing to talk about introducing values etc into economics, it's less easy to do it when faced with a class of 60 (or 200) students, all of whom have different value systems.

[Bob Smith]

"I do not have the right to modify your preferences, but I do believe that through democratic processes the community has that right."

Perhaps, but countries like Canada are not merely democratic societies, we're LIBERAL (small "L") democratic societies, which usually take as their starting point the proposition that people should be free to do whatever they want unless there's some offsetting harm to the broader community.

I suppose it's a fair intellectual critique of economics that, as a social science, it starts from that proposition (although it's not wholly surprising, given the intellectual origins of economics in the liberal enlightenment). That being said, given the liberal democratic nature of countries like Canada and the US, that's hardly an unreasonable starting point for a social science. Moreover, I suspect that many of the people who would make that critique are somewhat selective in terms of the choices that they think community can constrain through the democratic process (i.e., the same people who are keen on having the "community" decide to tax potato chips likely wouldn't be so keen on having the "community" decide to tax, say, abortions).

[K]

Michael: "Perhaps this is one of economic sciences' limits: it takes preferences as given, I want to modify them."

Some people, and especially kids, exhibit hyperbolic discounting. They are simply unable to coherently weight future consequences in their choices. It's a case of severe time inconsistency and it's hard to see it as anything other than irrational. Rather than modifying her preferences, I would see what you are doing as helping her become a more rational decision maker (not neglecting her *future* preferences). But certainly rationality is probably the most pervasively wrong assumption in economic modeling. And not coincidentally, also the most critical in producing parsimonious, elegant and mathematically tractable models.

Frances:  "we're left scratching our heads unable to figure out why people don't like our carbon tax proposals."

Of course, you are supposed to pay the revenues to all those who are harmed by the emissions, but if you just paid it out equally to all Canadians you might find overwhelming support. The evil-doers will be paying their fair share *and* Canadians would be grossly overcompensated since most of the harm is elsewhere in the world.

"The way this question is structured, it turns out that, on the margin, the reduction in harm outweighs the reduction in chip eating enjoyment"

I see that now. You make a good case for sin taxes. But it would be better if you could get the revenues back to chip eaters. Since giving them the money would defeat the purpose (unless you could deceive them as to the relationship to chip consumption which might not be too hard since they were pretty easy to deceive about the health effects of chips), maybe you could attach a Lipitor and a little tube of zit cream to each bag of chips. Or if, in fact, they suffer from hyperbolic discounting, just pay them in the future.

Bob: "A sin tax on abortion"

That's disgusting, Bob. EeeeW. Apart from that I agree with you.

[Kelvin]

Mike Moffatt (citing other people) seems to lean toward: yes, that is an externality, or at least if we're talking about time-inconsistent preferences. Or, at least, if a Pigovian tax creates time-consistent behaviour on inconsistent individuals, and causes increased total welfare, then the case for imposing them is not different than imposing such a tax to increase total welfare by removing the disutility caused upon others.

http://economics.about.com/b/2010/03/18/can-you-impose-an-externality-on-yourself.htm

[Frances Woolley]

K "Or if, in fact, they suffer from hyperbolic discounting, just pay them in the future." interesting.

Kelvin: "Or, at least, if a Pigovian tax creates time-consistent behaviour on inconsistent individuals, and causes increased total welfare, then the case for imposing them is not different than imposing such a tax to increase total welfare by removing the disutility caused upon others."

If a student gave a response along these lines, and correctly noted that this doesn't strengthen the case for Pigouvian taxation, but just it leaves it unchanged (harms imposed on your future self are just as worthy of correction as harms on other people) they would receive full marks.

[Frances Woolley]

Terry McGarty comments on this post here http://terrymcgarty.blogspot.com/2011/12/pigou-and-potato-chips.html.

Terry, even if 80% of the costs are borne by non-chip eaters through health care expenses, the case for Pigouvian taxation is weaker than it would be if 100% of the costs were born by non-chip eaters.

Also, as Chris Auld has argued elsewhere, what matters is the *net* impact of smoking, obesity etc on expenditures. Just think of the savings to the Social Security system when chip eaters drop dead at 64.

[Min]

Frances Woolley: "Even if "Not A" is a perfectly reasonable position to take - potato chips make buses hard to clean, depression hurts other people - it doesn't address the issue at hand, which is "If A, then B."

"What's sometimes called "exam technique" is often just simple logic."

That depends on what you mean by 'If A then B'. If it means material implication, as it does by simple logic, then "not A" says that it is true. (Anything follows from falsehood.) And, since anything follows from falsehood, the students who answer "not A" are pointing out the weakness in the argument.

Let's look at the statement in question in Q. 5: "It turns out that the only people harmed by potato chip consumption are potato chip eaters themselves, as potato chip consumption is associated with bad skin, weight gain and depression." Exam technique is a game, and, since I know the game pretty well, I realize that the phrase, "It turns out" means that what follows is to be assumed in the answer. Students who don't pick up that clue, or those who don't care (and are willing to risk a bad grade on this question), will take the statement that the only people harmed are the chip eaters, not as an assumption or fact, but as a conclusion to a bad argument. They then will criticize the conclusion. Logic, as you say. :)

English not being logic, "if A then B", often means "given A then B" or "assuming A then B". With that meaning, then "not A" is irrelevant. The truth of A is assumed. I read "It turns out" as "Let us assume". Suppose that the phrase were, "They claim" ("they" being the opponents of the tax). Then "not A" is quite relevant. Now, the question could have made the assumption explicit. For instance, "If the harm were limited to chip buyers, would that strengthen or weaken the argument for the tax?" Yes, that would make the game easier. But what do you want to test? :)

[K]

Frances: "interesting"

Yeah, but weird: selling chips with embedded 30 year bonds? Presumably people would just adjust their bond holdings. Unless there's evidence that chip eaters (and smokers) don't own bonds. Or maybe people are time inconsistent for lung cancer but not for dollars??? In which case it wont work at all. Seems to me like people just find ways to repress thinking about bad things like dying. In which case the best policy is to shove it in their face like we do with gruesome pictures on packages of cigarettes. Maybe we need to put really gross pictures of morbidly obese people on bags of chips. Enjoy!

[reason]

"(i.e., the same people who are keen on having the "community" decide to tax potato chips likely wouldn't be so keen on having the "community" decide to tax, say, abortions)"

That may be true, but surely taxing abortions would be much preferable to having them criminalised (then pro-choice people could switch to raising may to pay for them).

[reason]

"Also, as Chris Auld has argued elsewhere, what matters is the *net* impact of smoking, obesity etc on expenditures. Just think of the savings to the Social Security system when chip eaters drop dead at 64."

I think the full analysis of this is difficult. Maybe they drop dead at 64 after having had a stroke at 53.

[Bob Smith]

Indeed, and taxing chips would be much preferable to having them banned, but I can understand why chip eaters might not really appreciate that distinction. (Plus, I don't want to have to buy my doritos from the Hell's Angels). The ability to visualize a worse alternatives doesn't relly make a bad policy better.

[Min]

I wrote: "Now, the question could have made the assumption explicit. For instance, "If the harm were limited to chip buyers, would that strengthen or weaken the argument for the tax?" Yes, that would make the game easier. But what do you want to test? :)"

Perhaps you wanted to ask something else, like: "If the chip buyers bear much of the harm, does that strengthen or weaken the argument for the tax?" :)

BTW, my answer would be, it would affect the size of the tax, not the argument. ;)

[Bob Smith]

"I think the full analysis of this is difficult."

Sure it is, but I believe that a number of valiant attempts have been made. In any event, even if you don't think chip eaters impose a net cost on the fisc, that cost is likely to be significantly less than just the gross cost of treating chip related illnesses.

[reason]

Bob Smith,
ah yes, but we live in an imperfect world. Sometimes a bad policy is an improvement.

[Frances Woolley]

Min: "Students who don't pick up that clue, or those who don't care (and are willing to risk a bad grade on this question), will take the statement that the only people harmed are the chip eaters, not as an assumption or fact, but as a conclusion to a bad argument. They then will criticize the conclusion."

In general, yes, this is something that one encounters frequently when marking exams.

But wouldn't the "conclusion to a bad argument" reasoning led students to answer "it weakens the argument for the tax, because clearly Dr. Economides' study is totally bogus"? Whereas in fact a number of students wrote that the specific harms (bad skin etc) strengthened the argument for taxing chips.

On your suggestion for rewording: The question asked was: "It turns out that the only people harmed by potato chip consumption are potato chip eaters themselves, as potato chip consumption is associated with bad skin, weight gain and depression. Does this strengthen or weaken the argument for taxing potato chips?" Your suggested rewording is ""If the harm were limited to chip buyers, would that strengthen or weaken the argument for the tax?" Perhaps your wording is better - it's direct, and less wordy, and my students frequently complain that my exam questions are too wordy.

My worry with your suggested wording is that the students would be at a loss to figure out what "the harm" refers to. I guess you could say "If the 0.50 marginal damage were limited to chip buyers..." But again, I don't think they'd understand what that meant.

[Bob Smith]

"ah yes, but we live in an imperfect world. Sometimes a bad policy is an improvement"

Sure, if were had a starting point where chips were illegal, I'd agree.

[Frances Woolley]

Min: "They then will criticize the conclusion" - so I guess the argument is that if harm to chip eaters = 0.50, then total social harm is 0.50+x, where x is the health care costs, so the case for taxation is strengthened.

So there's this kind of health care costs "multiplier" which - like the multipliers cost benefit analysts know and love - can be used to blow up any effect to get you the results that you want.

[kevin quinn]

I'm not sure I understand Nick's comment. Are you saying that only in the case where no one knows the link is the case for the tax unchanged? I agree with that. But the more people know, the weaker the case,no? If all know, the WTP schedule is the Marginal net benefit - net of the .50 marginal cost of more weight or zits - and so a tax would distort, right?

[Min]

@ Frances Woolley:

I don't know about the best wording of the question, but I do like giving context, since that is closer to the kind of question students will face in real life. :) Perhaps something like this:

Opponents of the tax on potato chips take a careful look at Dr. Economides’ study. It indicated that the potato chip eaters themselves were harmed, as potato chip consumption is associated with bad skin, weight gain and depression, and the $0.50 is his estimate of the harm to potato chip eaters. Does this additional information strengthen or weaken the argument for taxing potato chips?

[Frances Woolley]

Kevin, don't have time to really think things through (am busy marking the question about the surgery on Prof Woofey's dog) but your reasoning looks correct to me.

Min - that's saying that the damage to chip eaters is $0.50, leaving open the question of whether or not there is additional damage over and above the original $0.50 estimate. The way I asked the question says the total damage is $0.50, and that harm mostly falls on chip eaters. But then again, I'm biased, so probably not the best judge.

[kevin quinn]

Frances: I think this is related to the last paragraph of your post. When I teach commons goods, I go through a little model of traffic congestion. I show them that if the road were privately owned, the owner would maximize toll revenues by charging a fee equal to the Pigouvian tax. Then I say: or we could have the state charge the toll and send everyone a check at the end of the year equal to the total toll revenues divided by the number of people who use the road. Many are sure that in the latter case, people would have no incentive to change their behavior - they get the money back, on average, after all. When I explain that the marginal addition to the check they get at the end of the year when they get on the road is the toll/N, where N is a big number, and so is negligible, they don't budge. But here's the thing, if people are like my students, and we did institute such a congestion tax with a rebate, THEY WOULDN'T CHANGE THEIR BEHAVIOR and the policy would fail!

[Franz]

"Strenghten" actually makes sense through the health care channel. If public health care is inefficient, say it costs the public $1 net to offset the damage and that's what happens given the law, the social cost is increased.
But that violates the assumption that the non-monetary cost is $0.50, which is not lifted in Q5; Q5 is only about the incidence.

I guess there's no non-behavioral/non-paternalistic way out.

[K]

Kevin: "send everyone a check at the end of the year equal to the total toll revenues divided by the number of people who use the road"

There's something wrong here. So I just have to use the road once to get an equal share of the congestion tax? Surely the harm is roughly both inflicted *and* borne in proportion to how much someone uses the road. So no tax required. Or maybe non-users bear an indirect cost in which case the revenues should be distributed to everyone. But I can't see a case for equally among drivers.

[Linda]

Fran , I admire your willingness to subject an examination question to outside scrutiny; I'm not sure I would have the courage to do so! Interesting experiment, though, demonstrating just how difficult it is to create questions that work in confining students to the words on the paper.

[Frances Woolley]

Kevin - would you mind posting that question? I'm trying to think of a simple way of modelling the fact that that toll revenues are maximized by charging a fee equal to the Pigouvian tax.

I have one that's a bit similar with carbon taxes, drivers and cyclists, revenue going back to drivers and/or cyclists, and a political equilibrium. I love the question, but it's very complicated, and takes a long time to go through.

Franz, yeah, but people tend to forget the assumptions stated at the beginning by the time they get to Q5.

[Frances Woolley]

Linda - well, having scrutinized someone else's question just a few weeks ago, it only seemed fair to subject mine to the same treatment. I'm glad to see, though, that WCI readers are mostly answering 'weaken', suggesting the question wasn't too bad.

[kevin quinn]

Frances: sure.

There is a highway that goes from point A to point B directly. There are many circuitous local routes between A and B that take 45 minutes and never get congested. On the highway a trip takes 15 + (1/60)T minutes, where T is the number of trips being taken, so one more trip adds one second to the trip time of all the other drivers. What is the Time saved by using the highway as a function of T? (30- (1/60)T
Graph this function of T. How many people will use the road if there is no charge? (30-(1/60)T* = 0, so T*=1800.) What is the marginal external cost of a trip as a function of T? ((1/60)T). Graph it on the same diagram. What is the optimal number of trips? (MB=MC where 30-(1/60)T** = (1/60)T**, so T**=900.) As Benjamin Franklin knew, Time=Money. To make things simple, let's say all drivers are alike and value time at $1/ minute. So your diagram now tells you the dollar value of a trip on the highway as well as the dollar value of the external cost. Suppose you own the road. How many trips will be taken if you charge a $30 toll. (O) If you charge nothing (1800). If you charge $15? (900). Can you see that the dollar value of a trip function is the(inverse) demand function for trips?
Suppose that your costs are all fixed costs, so you maximize profits by maximizing revenues. Remember how to maximize revenues with a linear demand curve? What toll would you charge to maximize revenues and how many trips will be taken? ($15,900).

[Frances Woolley]

Kevin, thanks, nice question.

[kevin quinn]

Damn, I left off the punchline:

What is the Pigouvian tax that will correct this externality? (At the optimal T=900, the marginal trip imposes a cost of 1 second on each of 900 drivers, for a total of 15 minutes. With our assumption about the value of time, then, the Pigouvian tax is $15, just what the greedy road-owner would charge to maximize profits!)

[Frances Woolley]

Kevin, would you expect students to work this out on their own? At what level? I mean it's easy enough see that people use the highway until the time saved from doing so equals zero once you think about it, but can students work that out without you going through an example first?

[Bob Smith]

Kevin quinn: "But here's the thing, if people are like my students, and we did institute such a congestion tax with a rebate, THEY WOULDN'T CHANGE THEIR BEHAVIOR and the policy would fail!"

That reminds me of a story that one of my professors at UofT used to tell from his early days of teaching in the 1970's. The big issue of the day was rising gas prices, and he'd have a big discussion with his class about the need (or not) to subsidize or otherwise regulate gas prices. In his early days, say, 1973, he'd make the case that rising gas prices wasn't the end of the world, that people would change their behaviour (i.e., drive less, drive smaller cars, take public transit, etc.) and that there was no need to subsidize or regulate the price of gas (and that it might be inefficient to do so) - and his students would go nut: "people need to drive", "it's an essential good", "it's a neccesity", "what about the poor? Won't anyone think of the children?". And so it went for a couple of years.

Then, one year, he got up and made the same argument, that there was no need to regulate or subsidize the price of gas, people would adjust their behaviour, yada, yada, yada, expecting the usual objections. But no, his students just looked at him like he was a moron: "Well, duh, sir, everyone knows that!". While his intuition might have eluded his earlier classes, in the intervening years, his new crop of students had lived that intuition such that it was a self-evident truth.

I wonder how many of your students would change their answer after a couple of years of paying road tolls.

[DavidN]

Kevin: I have to agree with Bob Smith. Can you trust what your students say to you in class? Sure, your students probably believe they won’t change their behaviour but as all economists know, talk is cheap. You can only trust revealed preference.

Regardless, it sounds like your class hasn’t taken intermediate micro. Charging a toll changes relative prices. Getting a rebate increases shifts the budget line out. Effect is ambiguous without more information.

[Frances Woolley]

Eric Crampton has some interesting reactions on his Offsetting Behaviour blog: http://offsettingbehaviour.blogspot.com/2011/12/pigovean-paternalism.html

[Ryan]

You said "Canada's public health system often does not cover skin creams, anti-depressants and counselling, so the link between bad skin and depression on the one hand, and health care spending on the other is not obvious." Actually, counselling for depression is covered by the Canadian health care system. You may want to correct this in your above article, because as it is this is a pretty classic example of the stigma surrounding mental illness. Mental illness is as real as any other illness, and psychiatrists have very really medical degrees too.

Obviously your students should have answered "weaken." A weaker argument doesn't mean there's no argument however.

[Frances Woolley]

Ryan - "Actually, counselling for depression is covered by the Canadian health care system"

There is some coverage - your family doctor will give you as much counselling as he/she has time to squeeze in and a prescription for pills - but it's a drop in the ocean.

[Min]

Frances Woolley: "The way I asked the question says the total damage is $0.50, and that harm mostly falls on chip eaters. But then again, I'm biased, so probably not the best judge."

Oh, I changed the meaning more than that. ;)

But if Dr. Economides estimated different costs for different groups, why not say what the different estimates were?

[Determinant]

Michael : "I cannot imagine a society in which the community is not modifying and disregarding preferences."

And it's always been our most important social institutions that have modified preferences (and vice versa). Historically that was family, church, state. Now it's Walt Disney and McDonalds. But when students have got their basic values from watching Family Guy, it's really hard to even have a discussion about morality and ethics. People tend to like the idea of rights, but then the question becomes "what do we do when rights conflict?" When my right to eat potato chips conflicts with your right not to pay for health costs associated with self-inflicted harms. It's one thing to talk about introducing values etc into economics, it's less easy to do it when faced with a class of 60 (or 200) students, all of whom have different value systems.

Wow, that's pessimistic. It's also not my experience. Most people I know can tell satire when they see it. The state still does a lot of ethics-building primarily through the education system.

Second, "church" historically was and still is a source of ethics but ethics as a form of practical theology is as varied in its results as any other theology is. Different churches have different theologies and different histories. Canada as a society of immigrants long had a pattern of religious plurality and the only consistent thing that can be said about churches is that for every position taken by one church another church can be found that disagrees entirely and takes the opposite position.

As an extreme example, for the Roman Catholic Church's very conservative stance on human sexuality there is the United Church which has GLBT ministers and recently considered adopting the "Our Whole Lives" sex ed curriculum from the United Church of Christ in the US. Our Whole Lives puts anything done in school sex ed classes to utter shame.

Lastly, a debate on pharmacare seems like another thread and the tangent on depression is not helpful. I have a lot of experience in this area and I would appreciate it if this tangent went away before it gets insulting.

[Jacques René Giguère]

When we talk about how polluters could pay a tax corresponding to the damage done or buy some offsetting behavior from somebody else ( tree planting in exchange for gas emitted on a plane ride), have you notice the reaction? Many of my students ( plus essentially all of the ordinary public) react with disbelief or horror: "You are letting the polluters get away with it!"

[Min]

"Many of my students ( plus essentially all of the ordinary public) react with disbelief or horror: "You are letting the polluters get away with it!"

Indeed. Lock'em up. Just like we lock up the bankers. ;)

[Greego]

Frances: "Excessive realism makes exam questions far too difficult. Students - being rational and self-interested - just want questions that are reasonably simple and straightforward, and ones that examine the material taught in the class (as opposed to stuff that wasn't covered.)"

Reality is not a linear function. The problem that Greg raised cannot be solved by simplification of the question, it invalidates the question completely.

"Both revenue and paternalism are perfectly valid reasons for a tax."

Of course! There's nothing at all sadistic about using government force to stop other adults from doing things that you don't approve of.

[Determinant]

Speaking of sadistic, whatever shall I do with the revenues from my Common Bawdy House?

[Phill]

I'm confused.

"Finally, it seems that some students really don't believe that people are rational decision-makers, fully taking into account the long-term effects of their consumption choices."

This is sarcasm, right?

[Phill]

Phill from the past, this is Phill from the future:

Yes, indeed I think he was making a funny, rendering these comments null and void.

[DavidN]

Greego: ‘There’s nothing at all sadistic about using government force to stop other adults from doing things that you don’t approve of.’ - It may or may not be sadistic but it certainly is human nature. The day you can guarantee all human beings will stop interfering with activities of other human beings the day I will jump on the libertarian bandwagon.

[Alex]

Finally, it seems that some students really don't believe that people are rational decision-makers, fully taking into account the long-term effects of their consumption choices

Because some of their friends smoke. Taxes on cigarettes are an example of a policy that does, in fact, respond to smokers' irrational preference for lung cancer by asking them to respond rationally to prices, and that does, in fact, reduce demand for cigarettes. The UK has repeatedly put up taxes on tobacco since the 1950s, until something like 70% of the price of a packet is tax, and the demand for it has fallen dramatically.

Obviously it's part of a policy-mix, but you'd have to be a professor to miss that one, and you'd have to be an economist to complain that your students are stupid because they didn't.

Further, since when does illness not have non-medical costs to society? Sick people take time off work, and if they get sick enough they stop working entirely, and in the case of depression, they very often do so by committing suicide and thus denying the economy the entire net-present value of their future potential income stream and all the investment that has been made in their human capital.

Suggestion: make your premises empirically true. Unless of course your aim is to train your students to believe any amount of epicycles if authority asks them to...

[Min]

Frances Woolley: "But then again, I'm biased, so probably not the best judge."

I hope that you don't think that I implied any bias on your part. :)

[Frances Woolley]

Alex: " you'd have to be an economist to complain that your students are stupid because they didn't."

I did not complain, or even suggest, that my students were stupid. I said "Finally, it seems that some students really don't believe that people are rational decision-makers."

If you think that means my students are stupid, then you're a stronger believer in rational decision-making than I am.

To be clear: if a student said "it makes no difference" and supported their answer with the argument about people are irrational and can't resist potato chips, they would receive full, or close to full, marks. The reasoning matters, not the answer - that's why I mark at least some of the answers myself, instead of having someone else mark the papers according to a rigid, pre-determined answer key.

Furthermore, let's not lose track of the basic point: Even if illness has some external costs that are borne by society at large, the case for corrective taxation is weaker than it would be *if all of the costs were external costs.*

I am not arguing that there is no case for taxing potato chips. I'm arguing that the case is weaker than it would be otherwise, that is, if all harms were external harms.

B.t.w., be careful with throwing around phrases like "very often" with relation to suicide. Millions of Canadians live with depression, and deal with it in countless different ways. It's not even obvious that depression makes you less effective on the job. Sure, depressed people are lousy at sales, but in jobs where the ability to think of everything that could possibly go wrong is a valuable skill, mild depression can be a positive asset (I read a study once that suggested tax lawyers fall into this category). But let's stop this discussion about depression - it's beside the point, and Determinant asked us to earlier.

[Frances Woolley]

Min: "I hope that you don't think that I implied any bias on your part. :)" - no, not at all.

[Ryan]

Frances - yah, I'm not objecting to the point that as you said "the case for corrective taxation is weaker than it would be *if all of the costs were external costs." My objection is that you identify mental illness as one of those conditions without external costs. There's a reason you chose depression and not diabetes for your example. Prejudice against mental illness is one of the worst bigotries in society today, and I think it behooves all of us to not unintentionally perpetuate it.

It's true that some people (such as Abraham Lincoln, Winston Churchill, Ernest Hemingway) have managed turn their disability into a source of strength. Most don't. There are 500,000 Canadians unemployed due to mental illness right now in Canada, and it is the second leading cause of disability and premature death (http://www.camh.net/news_events/key_camh_facts_for_media/addictionmentalhealthstatistics.html).

Again, I'm not objecting to your central point. I'm objecting to you using depression to make it. If were one of your students and I suffered from depression, I would have been deeply offended by it.

Cheers.

[Frances Woolley]

Ryan, would you rather be having an open discussion about mental illness right now, or would you rather be talking about the answer to some trite and inoffensive question about Christmas lights or pretty gardens or clearing snow off sidewalks?

[Determinant]

Sigh. As a diabetic I am well aware of the false perceptions of diabetes that circulate. It happens with a lot of diseases, diabetes' problem is that there are two types that really should have different names, they are quite different diseases under the hood.

When I said this tangent on Depression on mental illness is unhelpful, I meant it. There are lots of things to be said about it and they are not going to get a fair hearing in as a tangent on a thread about an economics exam.

Back to Pigouvian taxes and government nudges, I have had two interviews for two different positions at Health Canada. A standard question is "What is the mission, values and/or goals of Health Canada?", the answer to which can be found on their website. One explicit goal is to lower Canada's rate of smoking from its current 18% down to 9%. Finance and the CRA takes the lead on taxes while Health Canada is in charge of advertising and other 'soft' initiatives. It is very much part of a 'Pigou Plus' policy suite that Alex mentions.

[Frances Woolley]

Determinant: "There are lots of things to be said about it [depression] and they are not going to get a fair hearing in as a tangent on a thread about an economics exam."

Agreed.

[Patrick]

Something I don't understand, and maybe it's better asked of Eric Crampton since he brought it up: if polluting water and killing fish is a technological externality, then why isn't polluting the people (e.g with cigs? I think I'm asking if labour in an advanced economy (which typically skilled) is in some sense a common pool resource ... but I'm not sure. I suppose we don't pay the fish their MP, but then if is there is a such a thing as a Keynesian multiplier, degrading the labour pool with smoking such that it makes us all poorer relative to the no-smoking alternative reality might qualify ... I dunno. Too many moving parts. I've probably descended into the nonsense abyss.

[reason]

"Sure it is, but I believe that a number of valiant attempts have been made. In any event, even if you don't think chip eaters impose a net cost on the fisc, that cost is likely to be significantly less than just the gross cost of treating chip related illnesses."

Not necessarily - look at my example - there is lost income involved not just direct costs.

[reason]

"Of course! There's nothing at all sadistic about using government force to stop other adults from doing things that you don't approve of."

a. In this case we aren't stopping them
b. I thought we were talking about children
c. Sadistic? Its sadistic to stop people littering? Or spitting on the pavement? Yeah, I get pleasure looking at their tortured faces.

[reason]

For Greego

P.S. Intolerant, or illiberal, was the word you were looking for.

P.P.S. This is exercise in teaching people how to think, not in teaching them about the real world.

[Frances Woolley]

Patrick: "then why isn't polluting the people e.g with cigs?"

Polluting people with cigarettes is an externality if the smoker is polluting others, it's not an externality if the only person harmed is the smoker him or herself (ignoring, for the moment, the harm to people who care about the smoker and are sickened by the thought of something happening to him).

Eric's point (I think) is that public finance for the cost of medical treatment for the smoker just takes money from one person and gives it to another. Sure some people are better off and some people are worse off as a result, but nothing is gained or lost in aggregate. It's the difference between re-arranging the deck-chairs and throwing some of the deck-chairs overboard.

On the loss of productive labour argument - I'm thinking this is some kind of a variant on the lump of labour fallacy, but need to have more coffee in order to think this through.

[Bob Smith]

"On the loss of productive labour argument - I'm thinking this is some kind of a variant on the lump of labour fallacy, but need to have more coffee in order to think this through."

Agreed, generally I would have thought of foregone income as a private loss of the chip eater (and unfortunate side effect of being sick or dead) and fully internalized. On Reason's account, a larger population leads to higher GDP per-capita. Empirically that isn't true (at least not at the levels of population we typically observe - at the limits with very small populations, you could see that argument). Reducing the size of the population shouldn't, inherently, make a country poorer on a per-capita basis.

I suppose there is a social loss in the sense that we're all joint venturers with the fisc. which takes a portion of our income through taxes, but my recollection is that the studies I'm thinking of took into account lost tax revenue. In any event, once we're talking about people dying at 65 versus 85, that's not neccesarily a huge issue one way or another, since, past 65, people are often either living on government support or by drawing down their savings rather than engaging in productive activities (indeed, early deaths may be preferable for the fisc. to the extent that it triggers taxes on deemed dispositions of savings, RRSPs, etc. all at once on death(at the top marginal tax rate) rather than spread out over 20 years.

[reason]

Bob Smith
????
More income => more taxes

[reason]

My example
Stroke at 53 before dying at 64 => (probably) 11 years not paying in or paying less in. Given that most health costs are incurred in the first few and the last few years of life SURELY that makes a difference to the net calculation.

[reason]

Look I'm not saying anything one way or the other empirically, I'm just saying that a priori nothing is clear.

[reason]

And your GNP/population argument ignores human capital investment.

[Bob Smith]

"More income => more taxes"

Not neccesary, in a progressive tax system, the fisc will collect a lot more tax if I realize all my accrued gains or collapse all of my RRSP in one year(say, because I die), than it will if realize those gains, and collapse that RRSP over, say, 10 or 20 years (notwithstanding that I may earn additional income or gains on those savings over the course of that period. And, of course, they realize those taxes sooner.

[Bob Smith]

"And your GNP/population argument ignores human capital investment."

Not really, human capital is an inherently private asset, and one that's non-transferable. I.e., the fact that you or I may have a great deal of human capital doesn't inherently make others in society better off.

It's only when we use our human capital to provide valuable goods and services, that society benefits, but that just brings us back to the income point - the benefit of our human capital is largely internalized by its owners. Moreover, unless we're talking about a very small population (or a health cost that reduces the population by a signicant degree - say the Black Death), others will still be able to access those goods and services from other suppliers (i.e.,if we killed off 10% of Canada's population, and 10% of its tax lawyers, people would still have access to the same services previously provided by the 10% of the tax lawyers we killed off).

I think to be able to tell as human capital story, where the population is so small (or the health impact is so enormous) that the loss of one or two skilled individuals would deny the population access to a particular good or service. You could probably tell this story at a micro level in parts of rural Canada (say, aboriginal reserves in Northern Ontario), but I wouldn't think it would be generally true.

Moreover, keep in mind, when we're talking about health related issues, much (though not all) of the lost life occurs at the end, when arguably human capital has less value (because their owners are retired and not using it, or it is outdated, or both).

[Patrick]

Let's not muddy the waters with health care issues or sentimentality. Imagine we don't care about the suffering of others and imagine some vice that doesn't incur any additional health care costs to the public system. Say it just makes you less productive than you would be otherwise and you die younger than you would otherwise.

The early death thing seems to boil down to lower population than otherwise. Not sure what the effect of that is. I suppose it depends. Foregone old sick people may not matter much. Foregone babies probably matters a lot if the baby boom is any indication.

But lower productivity *in aggregate* would seem to be unambiguously a bad thing thought, even if the cost is born entirely by the individual. Their foregone income is foregone consumption is my foregone production is my foregone income. So, I'm poorer because they are less productive, no?

[Bob Smith]

So, I'm poorer because they are less productive, no?

But why would productivity be lower if the population is smaller? Sure, some high productivity people may die earlier, but so would low productivity people (unless we believe that there is an income based difference in chip consumption), so it's not clear why a smaller population leads to lower productivity. National income may be smaller, but why would individual income change? Certainly, it's hard to see a theoretical argument for it, and empirically it clearly isn't true (i.e., there are lots of small high productivity countries and large low productivity countries).

Now, if you could tell a story about how eating chips is concentrated amongst the high productivity people, maybe there's story there, but that's a hard story to tell vis-a-vis potato chips or cigarettes (quite the contrary, consumption of those goods is probably inversely related to income - it certainly is for cigarettes, and in the modern era obesity is disproportionately a disease of the poor). Even, then, it isn't clear that the loss of the high productivity people results in lower incomes for the others. Average incomes may fall, but the low-productivity people never had the "average" income (in the "high: and "low" universe, no one ever had the "average" income), they had low incomes, so it isn't clear that they're worse off.

[Declan]

Given how many people I've known who've done horrible things to their future selves due to hyperbolic discounting, but would never be willing to knowingly do the same harm to others, it seems that we'd need more information about the preferences of the typical potato-chip eater in order to answer one way or the other.

But I guess the students were meant to automatically assume that people care more for their future selves than they do for others? In fact, based on some of the comments above, such as Frances at 10:55, it seems there is a universal, unnecessary to be stated assumption that the typical person has 0 care for the impact of their actions on others. No surprise that the study of economics continues to correlate with unproductive, self-interested behaviour I guess.

[Patrick]

Bob, I think I said I wasn't sure about smaller pop.

I am probably not being clear, but I am trying in my muddled way to figure out the aggregate effects relative to the alternative world with no hypothetical human capital squandering vice. A voice in my head keeps telling me that we're all worse relative to the alternative. It's sorta like the agument against unions.

[Bob Smith]

Declan: "But I guess the students were meant to automatically assume that people care more for their future selves than they do for others? In fact, based on some of the comments above, such as Frances at 10:55, it seems there is a universal, unnecessary to be stated assumption that the typical person has 0 care for the impact of their actions on others."

I don't think that the answer to Frances questions depends on that assumption at all. She is telling students that the only harm from pototo chips is borne by potato chip eaters. It doesn't matter whether you think people care deeply about the welfare of others or not all, it they aren't affected one way or the other, that shouldn't change the result. Now, you could question the plausibility of Frances statement that the cost of potato chips is borne solely by chip eaters (i.e., by talking about health care costs or the emotional costs on your family from your dying young), but given that statement you don't need to make any assumptions about the preferences of chip eaters vis-a-vis the well being other others, the only well-being that matters.

Moreover, if people do care about the well-being of others, than arguably the case for the tax on chips disappears entirely, because chip-eaters will be internalizing the costs imposed on others (and will only eat chips until the marginal social benefit from doing so equals the marginal social cost). I.e., if my utility is harmed by the increased cost I impose on you from eating chips, than that's in effect an internal tax on my chip eating. Frances statement that the cost of chip eating is borne solely by chip eaters might be based on the assumption that the chip eaters fully internalize the harm they impose on others. That seems unlikely (for reasons discussed below), but I suppose it could be the case.

Declan: Given how many people I've known who've done horrible things to their future selves due to hyperbolic discounting, but would never be willing to knowingly do the same harm to others.

Sure, but harming yourself by smoking, drinking, etc. is a somewhat different beast from harming others by making them smoke, drink, etc (not the least of which is that you would probably go to jail for forcing a beer down someone's throat). In any event, it surely isn't that strong an assumption to say that people care MORE about their well-being (and their future well being) than they do about the well being of others, particularly in this context where the others concerned are (at least if we're talking about health care costs or reduced tax revenue) people they don't know, and the harm is not readily apparent and is hidden in government finances 20 years down the road. Moreover, to the extent where talking about future harm to others, even if we care about their well-being (say, the others are family members), the same hyperbolic discounting which causes us to give less weight to our future well being would presumably also cause us to give less weight to their future well-being as well.

[Eric R]

Three points occur to me reading the post and the comments following it.

1)Assuming that the costs of eating the chips is entirely internal, then rationale for a pigouvian tax does weaken, however depression and obesity are both linked to less economic productivity by workers, so even in the absence of a public health care system, there is reasoning to impose a tax.

2)OHIP, at least, does cover treatment for clinical depression,and weight related illnesses, which means that the public health-care system would suffer from the effects of excessive chip consumption. these diseases, even in a private care system, would also raise demand for health-services, increasing price, which could be another reason to favour the tax.

3) The problem is that the question is counter-intuitive to many people. If the 50 cent damage was entirely internalized, there's no reason to assume that this would translate to obesity or depression, as those would be signs of excessive consumption. If we assume that the damages are not significantly harmful in the long-term, or that the consumer has a rational reason for discounting future well-being (terminal illness?), then there are no broader social implications to the problem. This picture, of course, is very different from reality,, where humans do show a strong discounting bias to present enjoyment vs. future costs.

Students just have to remember that economic problems like this ones are though experiments, they don't apply to reality. In reality, there would be no way to determine the marginal damages of a given act, since this would necessarily vary by context. This is not to mention the problem of trying to place valuation on human quality-of-life. People who say that the case for a Pigovian tax strengthens are in effect arguing that the marginal costs are not internal, which contradicts the premise of the question.

[Declan]

"It doesn't matter whether you think people care deeply about the welfare of others or not all, it they aren't affected one way or the other, that shouldn't change the result."

...

"Moreover, if people do care about the well-being of others, than arguably the case for the tax on chips disappears entirely"

Are you arguing against me, or yourself? At any rate, you (in your second comment quoted) are reiterating my point, that if people care about others sufficiently, then the case for the tax (in the event that the harm falls on others) disappears, hence the case for the tax strengthens in the situation where the harm comes to the eater directly but the eater no longer recognizes this harm due to hyperbolic discounting.

"Sure, but harming yourself by smoking, drinking, etc. is a somewhat different beast from harming others by making them smoke, drink, etc"

Precisely. All I'm saying is that many people feel as you do (I'm assuming here that you believe harming others is 'worse' than harming yourself, if not then I guess we'll just have to agree to disagree). But the question assumes implicitly that people don't feel this way and in fact are the reverse, entirely discounting the costs of making others smoke and drink, but fully internalizing the costs of doing it to themselves. That is quite inconsistent with my observations of people I've known.

"Frances statement that the cost of chip eating is borne solely by chip eaters might be based on the assumption that the chip eaters fully internalize the harm they impose on others."

That wasn't how I read the question. My understanding was that "the only people harmed by potato chip consumption are potato chip eaters themselves, as potato chip consumption is associated with bad skin, weight gain and depression." In other words, direct harm to the eater, not indirect harm via the harm caused to others, transmitted to the eater via empathy. Indeed, if it was possible to interpret Frances' statement in the manner you suggest, then the whole question appears nonsensical - if we're treating harm to others and to ourselves as equivalent, then how could moving the harm around change anything? If your point were valid, then surely Frances' 'correct' answer is in fact incorrect, and her students deserve a re-marking.

"it surely isn't that strong an assumption to say that people care MORE about their well-being (and their future well being) than they do about the well being of others"

Speak for yourself. And thanks for helping to prove my point. Less argumentatively, I'd say that the validity of this assumption depends strongly on a) the context, and b) the particular person we're talking about. Basing exam questions on an assumption that all people are purely self-interested in all situations (without stating this assumption explicitly) seems absurd to me.

[Peter T]

As well as the assumption of rationality, there are two other unstated ideological positions framing the question - individualism and liberalism. Maybe the students who gave the "wrong" answer did not subscribe to both these ideologies? Fair enough if they did not - even in Canada surely not everyone believes that people should be free to do what they like so long as it does not harm others, and individualism is very contestable on reliable neuroscience alone, never mind ideology.

What did you say you were giving an exam in?