The economics of Christmas presents: how Michael Sandel gets beaten by his own straw man
10 December 2013
SUGGESTED
Uncategorized
9 December 2013
Uncategorized
Tax and Fiscal Policy
19 November 2024
The pre-Christmas period has begun, and with it, the hunt for Christmas presents. Do you find this irritating? Yes, you do. Trust Michael Sandel, the author of What Money Can’t Buy: The Moral Limits of Markets, on this one. [I’m assuming that as a reader of this blog, you are either an economist, or steeped in economic thinking for other reasons.]
Sandel asserts that economists find the notion of giving someone a present deeply troubling, and even frustrating. That is because economists, on Sandel’s account, believe that everybody is a super-rational, coldly calculating, utility-maximising robot. Since everybody knows their own preferences best, choosing a gift for someone else must necessarily involve inefficiencies. The recipient could have attained a higher utility level if they had simply been given the corresponding amount of money in cash (or a voucher), leaving it up to them what to buy with it. Within standard microeconomic assumptions, Sandel claims, gift-giving can never be superior to cash. It can at best be equally good, if the giver buys exactly what the recipient would also have bought.
Sandel then seriously goes on to explain the errors in this thinking, as if he really expected anyone to disagree with him on this matter. Straw-man bashing is fine as long as it serves a rhetorical purpose, but it gets weird as soon as you start believing in your own straw men, as Sandel apparently does. Since I have never met an economist who mistakes Christmas gifts for an inefficient attempt to maximise consumer surplus, I will not comment on Sandel’s reasoning. It is noteworthy, however, that if this straw man came alive and began to fight back, Sandel would probably lose the fight.
Even in the most narrowly conceived utility-maximisation framework, Christmas presents need not be inefficient. If necessary, one could quite easily integrate Christmas presents into a standard utility function (not that I’d recommend doing so).
In order to do this, one could, first of all, incorporate the assumption that on occasions like Christmas, most people enjoy being surprised. Technically, this would mean including a ‘surprise factor’ in the recipient’s utility function. For a cash gift, that surprise factor would be set to zero; for a very predictable gift, it would take a low value. Total utility derived from a gift would then be a combination of the utility derived from the good itself, and the utility derived from being surprised. The surprise element could be modelled in such a way that it can also work the other way round: a nasty surprise would subtract further from the already low utility associated with a poorly chosen gift.
One could also model interpersonal differences in surprise-affinity, that is, the ratio at which people are willing to trade off consumption utility for surprise utility. A teenager who desperately wants the latest computer game would be at the lower end of that spectrum. They do not want to be surprised, they just want their computer game. Under those conditions, a cash gift or a voucher need not be such a bad idea. Elderly people, meanwhile, who already have everything that is genuinely useful, could be on the other end of the spectrum.
Secondly, one could model the fact that giver and recipient possess different sets of information about product markets. A gift-giver cannot know your preferences nearly as well as you do, but they can nevertheless have an informational advantage over you: they may spot potential gifts that you would never have discovered for yourself. I don’t know about you, but the Christmas presents that I receive are often better than what I would have bought for the cash equivalent, simply because they contain things I was not even aware existed. So one could quite easily model a situation in which a gift provides greater utility than a cash sum, and still remain entirely within the confines of the microeconomics textbook. Nobody knows your preferences as they currently stand as well as you do, but somebody else might have a good guess about your preferences as they will be once you have seen their gift. Gift-giving would then be an entrepreneurial activity – it would be about discovering somebody’s latent preferences, rather than maximising utility on the basis of their already known preferences. The more disparate the sets of information which giver and recipients possess, the greater the potential for that effect.
In short, even in the unlikely event that you took every page in your microeconomics textbook at face value, that would not be an argument against joining the Christmas present fray. So do join it, and go on a utility-maximisation spree. I hope Michael Sandel finds a decent economics book under his Christmas tree this year.
Sandel asserts that economists find the notion of giving someone a present deeply troubling, and even frustrating. That is because economists, on Sandel’s account, believe that everybody is a super-rational, coldly calculating, utility-maximising robot. Since everybody knows their own preferences best, choosing a gift for someone else must necessarily involve inefficiencies. The recipient could have attained a higher utility level if they had simply been given the corresponding amount of money in cash (or a voucher), leaving it up to them what to buy with it. Within standard microeconomic assumptions, Sandel claims, gift-giving can never be superior to cash. It can at best be equally good, if the giver buys exactly what the recipient would also have bought.
Sandel then seriously goes on to explain the errors in this thinking, as if he really expected anyone to disagree with him on this matter. Straw-man bashing is fine as long as it serves a rhetorical purpose, but it gets weird as soon as you start believing in your own straw men, as Sandel apparently does. Since I have never met an economist who mistakes Christmas gifts for an inefficient attempt to maximise consumer surplus, I will not comment on Sandel’s reasoning. It is noteworthy, however, that if this straw man came alive and began to fight back, Sandel would probably lose the fight.
Even in the most narrowly conceived utility-maximisation framework, Christmas presents need not be inefficient. If necessary, one could quite easily integrate Christmas presents into a standard utility function (not that I’d recommend doing so).
In order to do this, one could, first of all, incorporate the assumption that on occasions like Christmas, most people enjoy being surprised. Technically, this would mean including a ‘surprise factor’ in the recipient’s utility function. For a cash gift, that surprise factor would be set to zero; for a very predictable gift, it would take a low value. Total utility derived from a gift would then be a combination of the utility derived from the good itself, and the utility derived from being surprised. The surprise element could be modelled in such a way that it can also work the other way round: a nasty surprise would subtract further from the already low utility associated with a poorly chosen gift.
One could also model interpersonal differences in surprise-affinity, that is, the ratio at which people are willing to trade off consumption utility for surprise utility. A teenager who desperately wants the latest computer game would be at the lower end of that spectrum. They do not want to be surprised, they just want their computer game. Under those conditions, a cash gift or a voucher need not be such a bad idea. Elderly people, meanwhile, who already have everything that is genuinely useful, could be on the other end of the spectrum.
Secondly, one could model the fact that giver and recipient possess different sets of information about product markets. A gift-giver cannot know your preferences nearly as well as you do, but they can nevertheless have an informational advantage over you: they may spot potential gifts that you would never have discovered for yourself. I don’t know about you, but the Christmas presents that I receive are often better than what I would have bought for the cash equivalent, simply because they contain things I was not even aware existed. So one could quite easily model a situation in which a gift provides greater utility than a cash sum, and still remain entirely within the confines of the microeconomics textbook. Nobody knows your preferences as they currently stand as well as you do, but somebody else might have a good guess about your preferences as they will be once you have seen their gift. Gift-giving would then be an entrepreneurial activity – it would be about discovering somebody’s latent preferences, rather than maximising utility on the basis of their already known preferences. The more disparate the sets of information which giver and recipients possess, the greater the potential for that effect.
In short, even in the unlikely event that you took every page in your microeconomics textbook at face value, that would not be an argument against joining the Christmas present fray. So do join it, and go on a utility-maximisation spree. I hope Michael Sandel finds a decent economics book under his Christmas tree this year.
4 thoughts on “The economics of Christmas presents: how Michael Sandel gets beaten by his own straw man”
Comments are closed.
SHARE
Surely an additional point about gift giving is that it also gives a reward to the giver of the gift?
The problem in general with Sandel – aside from basic errors like this – is that he seems to believe that everything which is not provided by markets must be provided by the state, forgetting that this is a) false and b) ignores the effects of state intervention in distorting behaviours and c) state provision of goods creates moral problems of its own. Alas, it’s not surprising he gets away with it as this is the attitude of his readers who love having their prejudices confirmed
If you give cash as a present, it’s obvious you haven’t spent much trouble — just cash. Whereas if you attempt to give something specifically for the donee’s benefit, of course you may make mistakes but at least you’ve taken some trouble — and that knowledge may well also gratify the donee. Of course the aim is mainly to benefit the donee not to maximise the cost to the donor. (One of my brothers always argues that people should give something that they themselves value, not something that they think the donee will value!)
Or one could point out that christmas is not about gifts unless your an economist. And that showing appreciation for another human being could be expressed in other ways. Small tokens “gifts” could be given at anytime one feels compelled. I feel I have been tricked. I want life not an economy.
Your criticism of Michael Sandel surprises me. The whole purpose of his book is to question whether money can take the place of (or be used to quantify) human values. Of course he doesn’t believe this nonsense of utility-maximization in gift-giving. Read the chapter again, please.
If you have never met an economist who mistakes Christmas gifts for an inefficient attempt to maximise consumer surplus, you should introduce yourself to Joel Waldfogel at the University of Pennsylvania. It is Waldfogel’s 1993 article in the American Economic Review (The Deadweight Loss of Christmas) and his 2009 book (Scroogenomics: Why You Shouldn’t Buy Presents for the Holidays) that is clearly cited by Sandel as the source of this ridiculous notion. (Wow – economist Waldfogel has been plugging this for at least 16 years!)
I think it as obvious to Sandel as it is to the rest of us that the objective of gift giving has little to do with utility-maximization and more to do with a communication between two human beings, incorporating, as you rightly point out, elements of surprise, information-sharing, etc.