I will admit that I returned a few Christmas gifts this year. There's usually a couple in my pile each year that I find aren't quite what I wanted. This year only three things begged to be brought back: Two movies and a shirt.

Now, before anyone criticizes me, let me state that I actually liked both movies that I returned. The problem with the gift was not so much the content but the format. You see, I'm a widescreen whore, but the DVDs I got were full screen. I simply returned those and plan to get their widescreen counterparts as soon as a sale comes up (and pocket the difference...).

The shirt was just, well, flannel. Not that there's anything wrong with that.

Anyway, I started thinking about all this after learning about the term deadweight loss at the Freakonomics blog (they have a fairly humorous post about Brett Favre's contribution to deadweight loss this season). This sounds like kind of a harsh term, and maybe it is.

Basically, deadweight loss is the difference between the amount a giver pays for a gift and the amount that a recipient would have paid had she bought the gift herself. This does not imply anything about the amount of appreciation or desire for the gift. For example, say you really, really, really wanted a new GPS, but just couldn't justify the expense of $200-600 to buy one, but might be willing to buy one at $150. So you put it on your Christmas list, and either your parents, a cartel of good friends or your significant other end up getting you one. The giver(s) was willing to pay more for the gift than you were – and you are happy they were!

For those who want to read more, Stephen Dubner links to a paper about deadweight loss by Joel Waldfogel in his Freakonomics blog post I mentioned above. The math is a little beyond me, but it's interesting to read about the surveys of econ students¹, not to mention some of the conclusions and suggestions he makes. Here are some of them, with my commentary.

Commentary on Deadweight Loss

Non-cash gifts typically have deadweight loss

Logically, this makes sense. Whether you are getting a really great gift or a really crappy one, chances are that the person who bought it paid more than you would have for it. For crappy gifts, this is obvious: You wouldn't have bought it at all, so any money spent on it is too much. For good gifts, you probably would have bought it, but given that you had not bought it yet and instead asked for it for Christmas (or didn't ask for it and were surprised) means that you were not willing to pay for it at any given known price, and therefore the giver was willing to pay more for it than you were.

There are exceptions to this, of course. At least one exception that M$L readers probably care about is really good sales. Revisiting the GPS example above, say a GPS system that you've had your eye on goes on 50% sale without you seeing the ad. Your mom/friend/sig-other buys it at this sale price, and then on Christmas they brag to you about the great deal they got on your gift.² Realizing that it's below the $150 threshold you would have paid, you rejoice at the news of the money saved – and perhaps given to you in the form of another gift.

Cash gifts are immune (usually) to deadweight loss, but gift cards are not

Cash has no deadweight loss because the recipient can use the cash in exactly the way she wants. There is no opportunity for the descrepancy that is required to create a deadweight loss, unless somehow you paid more for the cash than the value of the cash itself. This could happen if you had to pay to cash a check or had to use an ATM not owned by your bank, but hopefully nobody reading this blog has to do either of those things.

Gift cards, however, can (but don't always) have deadweight loss for one of two reasons, depending on the type of card. Store-specific cards can have deadweight loss simply because they are store-specific – you can't buy that book in the bargain bin at Borders with a Barnes & Noble gift card, and so either you spend your own money to purchase it at the sale price or use the gift card to get it at the higher price. Generic gift cards (such as Visa prepaid debit cards or American Express gift cards) almost always have some kind of purchase or activation fee, and may not be usable everywhere, especially in the case of Amex cards. Many gift cards also have fees associated with inactivity or non-use by a specific date.

10-33% of value is lost when giving gifts

I sort of skimmed through the statistical analysis. The conclusion is that, on average, the deadweight loss from giving non-cash gifts is one-tenth to one-third of the value of the money used to purchase them. Stated another way, on average the recipients of gifts value those gifts about 10-33% less than the givers.

Of course, this does not mean that it's not worthwhile to give gifts. I'll talk more about that in my conclusion below.

The better you know someone, the smaller the loss

This is probably somewhat inuitive, but how well you know somebody is a big factor in the amount of deadweight loss for a particular gift. Think about it: Are you more likely to get a really great gift from your Aunt Maude who you haven't seen in 10 years or from your fianceé? If you said Aunt Maude, maybe you shouldn't make any down payments on a reception hall quite yet.

Government "gifts" have deadweight loss too

Waldfogel goes on briefly to mention deadweight loss in the context of government entitlement programs. Basically citing another study, he states that handouts that "most resemble cash" such as food stamps and rent supplments, have little or no deadweight loss. However, other programs, such as public housing, Medicare and Medicaid, show deadweight losses that are comparable with the gift-giving losses described above, and sometimes lower. He does not make any value judgments about this loss (to his credit, since it's an economics paper and not a policy or think tank paper) – so I will.

The most disturbing thing to me is that this is the loss in value of just the handout itself. This doesn't account for the amount of money spent to administer the programs, collect taxes and so forth. Of course, you will never see any mention of this type of loss on any official report about the successes of government programs.

Conclusion

Deadweight loss is simply the name given to the difference between the value of a gift as perceived by the giver and receipient. While it's interesting to think about (and wonder how much value the gifts I give have lost), please don't construe any of this to mean that giving gifts is not worthwhile.

Instead of thinking of deadweight loss as, well, a loss, you could think of it as a premium. This premium pays for the smile on the face of a friend who happily accepts and appreciates your gift, or the lovin' you get from your significant other later. These types of emotional and intangible goods the giver receives are not present if the recipient buys the item for herself, and so it makes sense in this context that the giver would pay more.

As for the study, I would have liked to see some mention about deadweight loss with regard to charitable gifts – both gifts given to charities (which are often cash, but not always) as well as gifts given by charities.

Notes

1. One wonders how valid an economic survey whose sole participants are econ students can be.

2. I realize this might sound strange or even horrific to some people. However, my family does this frequently for birthday, holiday and any other kind of gifts given. It's not uncommon during Christmas gift exchanges to hear sporadic exclamations of, "I only paid [insert absurdly small amount] for that!" or "That was [insert insanely high percentage] off!" Getting the best deal of the season is a badge of honor.