The Efficiency of Gift Giving: Is It Really Better to Give than to Receive?Published: December 15, 2004 in Knowledge@Wharton
Now that the holiday season is in full throttle - and panic has set in among those who fret over finding the right gifts for family members, friends and perhaps even co-workers - it's worth considering a not particularly festive question: Just how efficient is gift giving? And what do the recipients of consumer purchases actually think about the gifts they get?
Joel Waldfogel, professor of business and public policy at Wharton, sheds some light on this question in a new paper titled, "Does Consumer Irrationality Trump Consumer Sovereignty?" According to Waldfogel, who looks at the issue as an economist and not as a holiday spirit-quasher, consumers' own purchases generate between 10% and 18% more value, per dollar spent, than items received as gifts. In other words, consumption choices made by others tend to generate less satisfaction than one's own choices.
While this may seem evident on one level, scholars who study behavioral theory have collected "an impressive body of evidence that consumer behavior in a large number of laboratory and real-world contexts is not fully rational," Waldfogel states. For example, "failures of rationality" have been identified in the area of retirement savings - where some people don't accurately forecast what their savings needs will be and thus don't put aside enough money for their post-retirement years - and in the insurance area, where some home-owners will forego flood insurance even though they live in flood-prone vicinities, figuring that bad weather happens to others, not them.
Even in the area of consumer choice, the subject of Waldfogel's paper, prior research shows that people have trouble predicting - and remembering - what they like. One study found that individuals "making one snack choice each week selected differently than subjects choosing simultaneously for three weeks in advance. The latter group chose more varied bundles, incorrectly predicting their future preferences," writes Waldfogel in his paper. Another study showed that in trying to remember their preferences, people's "recollections are not based on the entirety of an experience; rather they tend to be based on the best or worst parts, as well as the last parts of experiences," the paper notes.
These studies, says Waldfogel, raise serious questions about whether consumer choice would be reliable "even in the most straightforward contexts."
Red Sweater vs. Blue Sweater
To test whether one can assume rationality in any sphere of consumer behavior, Waldfogel looked at a simple economic transaction - the purchase of consumption items such as clothing, books and CDs. The idea was to see whether individuals make better consumption choices for themselves than others do.
He chose as his laboratory the Christmas/Hanukkah gift-giving season during which "individuals receive many 'experimental' items chosen by others, and their valuations of these items can be compared against their valuations of a 'control sample' of their own purchases," the paper notes. Gifts in the experiment were mainly from friends, significant others and immediate and extended family members.
Waldfogel conducted a survey of 1044 gifts and 538 own purchases for 202 college students at three universities in early 2002, soon after the holiday season. He collected information about the relationship between the giver and the recipient as well as a description of the item bought. The sample included 500 instances in which a respondent both purchased and received the same kind of item (e.g. different CDs). Items ranged from clothes, books, electronics and shoes to cosmetics, videogames, appliances and jewelry.
Based on that survey, Waldfogel concluded that individuals valued their own purchase at an average of 18% more, per dollar spent, than they valued items they received as gifts (not counting sentimental value.) Those who came closest to giving the most valued gifts were grandparents and significant others, presumably because they are in closer contact with the recipient or have a better ability to find out specific preferences. "Our results support economists' deference to consumer sovereignty in a simple current consumption choice contest," Waldfogel writes.
He suggests that his research will be of interest to several audiences. "For example, retailers might want to know how they can help gift purchasers buy things that don't turn out to be unwanted. The answer could be a gift certificate or a wish list such as the one offered on Amazon where individuals can enter in items they want at any given time." Wedding registries are the same idea; the bride and groom select certain items at certain prices, thereby guaranteeing that friends and family are choosing desired gifts.
"The basic distinction in the paper is the one between the valuation I place on things I buy for myself and the things that you buy for me," says Waldfogel. "Part of that is due to the fact that I buy different things than you do. But even if you do a comparison within items - e.g. comparing CDs I buy versus ones you buy for me - there is still a differential. It's somewhat smaller, but it is still significant."
Of course the fact that the process of gift giving destroys value for the recipient ignores the benefits that such a transaction may provide for the giver, says Waldfogel. "Abolishing the act of giving, or encouraging people to give cash instead, might be good for the recipient in some narrow sense, but it might be lousy for the giver." As his paper puts it: "Gift giving may allow givers to demonstrate their keen understanding of recipient preferences, impose reciprocal obligations on recipients ... or provide social signals ..."
Also, even for the recipient, there is some sentimental value associated with receiving the gift, Waldfogel says. "In my study, I asked the recipient to exclude that. But if the sentimental value could only get transmitted by this choice of gift, then asking people to give something else, like cash, means the sentimental value would not be transmitted. So again, there would be a loss.
"That said, however, if the sentimental value can be transmitted either with the blue or the red sweater, and you like red, then red is better."
The Christmas Grinch
Waldfogel wrote a paper in 1993 entitled, "The Deadweight Loss of Christmas" in which he asked participants in a study how much they valued items they received as gifts. He found that 13% of the value of what gift givers gave was destroyed through the process of choosing the wrong things. "This paper didn't have any questions about items that people had bought for themselves," says Waldfogel, adding that the paper generated a sizeable amount of controversy, with some respondents going so far as to label him a "grinch." In addition, a number of subsequent studies challenged his results by suggesting that recipients value gifts above the prices that givers pay for them.
"In that 1993 paper I was comparing how much people valued their gifts against a 100% benchmark, and I showed that they valued the gifts at 87% of what the givers had paid for them. I also looked at how much the recipients would have been willing to pay for the gifts themselves," says Waldfogel. "In some cases the answer was 'nothing.' I then averaged the answers and came up with 87%. But I would fault all the papers, including my own, for comparing people's evaluation of gifts to the wrong benchmark. The right one for comparison is how well a person would have done buying for him or herself, not comparing the gift to this 100% benchmark."
As he notes in his current paper, "in evaluating whether individuals are better at choosing their own consumption bundles than others, it is important to compare consumer valuations of items chosen by others against an appropriate benchmark of their valuation of their own purchases. Existing gift studies assume that own purchases have consumer valuations equal to their prices. While consumer theory implies that rational, maximizing consumers value their last unit purchased at the price paid, it is reasonable to expect average own purchases to be valued above the price paid....
"We expect yields on our own purchase to exceed 100%. What about gifts? If consumers are perfectly informed, and presumably better informed than givers, then it is impossible for givers to do better than recipients at choosing the recipients' consumption bundles," the paper states. "On the other hand - and especially given the evidence on consumer irrationality - it is possible that givers know more about recipients' preferences, and goods, than do recipients. All of this suggests that gift givers more familiar with recipient preferences will choose items more highly valued by recipients, per dollar spent."
Waldfogel says that while results of his current paper don't challenge the "documented irrationality of consumers in other contexts, our results do suggest a limit on the reach of the behavioral critique of rationality." Individuals may be so bad at making [certain] consumption choices that "paternalistic interventions" could improve their decisions - for example, the intervention of an experienced investment advisor who could help senior citizens better invest their retirement money. "But other choosers do not seem to outperform individuals for simple current consumption choices. Rationality may be bounded, but irrationality is bounded as well. The results identify a sphere of decision-making - current consumption choices among familiar objects - where consumer sovereignty is warranted," Waldfogel writes.