See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/4726321

Markets: Gift Cards

Article  in  Journal of Economic Perspectives · February 2007

DOI: 10.1257/jep.21.2.227 · Source: RePEc

CITATIONS READS

32 641

1 author:

Jennifer Pate
Loyola Marymount University
17 PUBLICATIONS   139 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Jennifer Pate on 04 February 2015.

The user has requested enhancement of the downloaded file.

Journal of Economic Perspectives—Volume 21, Number 2—Spring 2007—Pages 227–238

Markets
Gift Cards

Jennifer Pate Offenberg

This feature explores the operation of individual markets. Patterns of behavior

in markets for specific goods and services offer lessons about the determinants and
effects of supply and demand, market structure, strategic behavior, and govern-
ment regulation. Suggestions for future columns and comments on past ones
should be sent to James R. Hines Jr., c/o Journal of Economic Perspectives, Department
of Economics, University of Michigan, 611 Tappan Street, Ann Arbor, MI 48109-
1220.

Introduction

The Mobil Oil Company introduced the first retail gift card that recorded value on
a magnetic strip in 1995. In under a decade, such gift cards replaced apparel as the
number one item sold during the Christmas season (Deloitte and Touche, 2004).
Figure 1 shows how total sales of gift cards have increased since 1997, exceeding
$60 billion in 2005 and heading for $80 billion in 2007. The average American
consumer purchases seven gift cards annually, spending approximately $250 in total,
according to a ValueLink (2004) consumer survey. The National Retail Federation
(2005) estimates that the end-of-the-year holiday season is responsible for roughly
30 percent of annual gift card sales; that is, holiday sales of gift cards reached over
$18 billion in 2005. The average consumer spent $88 on gift cards over the 2005
holidays, approximately 16 percent of their gift budget. Consumers can now buy gift

y Jennifer Pate Offenberg is Assistant Professor of Economics, Loyola Marymount University,

Los Angeles, California. Her e-mail address is 具joffenberg@lmu.edu典.
228 Journal of Economic Perspectives

Figure 1
Gift Card Timeline
(annual gift card sales in billions of dollars)

90
est. 80
80
est. 72
70
63
Billions of dollars

60 55

50
43.5
40 37
32
30 24.5
19
20
12
10
negligible
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Source: Bain and Co. (2003) and TowerGroup (2005).

cards from third party “kiosks” like Quickgifts.com, which offers online purchasing
from 122 companies. Gift cards used to be offered primarily by large chain store
companies, but more and more local businesses are joining the trend.
A gift card is similar to the formerly standard paper gift certificate, but has
the added benefit that purchases are automatically deducted and that the card
is entirely transferable. The focus of the discussion here is on gift cards offered
by retailers. There is another form of gift card, commonly called a bank card,
that holds a cash value and acts as a debit card, and these cards are available
directly from banks or credit card companies like Visa. These cards typically
have handling fees ranging between $4.95 and $8.95, depending on the stored
value. They do not appear in the data on gift cards discussed in this paper, but
do offer their own issues— especially with regards to the loss associated with
giving a bank card instead of a cash gift, since these cards are even closer to cash
than retail gift cards.
This study will discuss the reasons for the strong surge in the gift card market.
It will then consider the value of gift cards as an intermediate option between two
alternatives: purchasing a physical gift, which could possibly be returned or ex-
changed, versus giving cash. Empirical data on the resale price of gift cards from an
Internet auction website provide information on the value that recipients place on
gift cards suggesting that the difference between the cost of a gift card to the giver
and its value to the recipient is substantial, although perhaps not quite as large as
the parallel gap involved in physical gifts.
 Jennifer Pate Offenberg 229

What Changed?

Paper gift certificates have been available for decades, but have now been
largely replaced by plastic gift cards with either a barcode or magnetic strip for
recording the buying power of the card and deducting purchases. Dick Outcalt of
Outcalt & Johnson Retail Strategists calls gift cards a “win–win–win situation.” That
is, a gift card is a win for the retailers; a win for the gift-giver because it “relieves the
pressure” of deciding on an appropriate gift; and “a win for the recipient because
they can choose what they want and when they want to buy it” (Bolt, 2004).
The new magnetic strip technology is one major factor driving the rise of
gift cards. For retailers, the new technology created the ability to account for
sales electronically and greatly reduced the cost of offering a gift-certificate-like
product. For consumers, the magnetic stripe technology makes gift cards easily
transferable, convenient, and lower maintenance than the traditional gift
certificate.
A second factor driving the growth of gift card sales is the rapid spread of
chain stores. Gift cards can now be purchased, given, and redeemed nationally
as companies like Wal-Mart and Home Depot continue to add hundreds of new
store locations every year, reaching into the outer regions of all 50 states. As the
costs to firms to incorporate gift cards into their business have declined, the
benefits have become more apparent. In addition to securing guaranteed sales,
stores may see gains in brand awareness and an increase in customer loyalty, as
consumers may make several visits to redeem their card. According to a 2004
ValueLink survey, in 55 percent of purchases involving the use of a gift card,
consumers spent more than the card value. Also, an estimated 10 percent of gift
cards are never redeemed (Maranjian, 2004), and this amount is almost pure
profit to the firm. For example, in 2005, Home Depot reported $52 million of
additional income from gift cards for which “the likelihood of redemption by
the consumer is remote.”
In addition, gift cards have allowed firms to manage their reported earnings
by manipulating the timing and rate at which unredeemed gift cards are written
off from liabilities (Berner, 2005). These benefits have led companies to in-
crease their marketing of gift cards. For example, some stores, like Target, sell
novelty gift cards that play holiday tunes at the push of a button, while other
cards are shaped like candles with glowing lights. Since gift cards are worthless
until value is added at the register, shoplifting of the cards from stores is not an
issue, allowing them to be placed predominantly at front entrances for prime
advertising.
Finally, although gift cards were once considered a lazy present, they have
become increasingly popular with both gift-givers and gift recipients. A 2005
National Retail Federation survey found that two-thirds of American consumers
planned to buy at least one gift card during the 2005 holidays, while over
52 percent of consumers listed gift cards as their preferred gift to receive. Gift
cards may have also flourished as a result of changes in the composition of our
230 Journal of Economic Perspectives

society. Historically, a considerable amount of time may have been devoted to

carefully selecting appropriate presents—time which, for married couples,
would typically have been performed by a housewife. However, household
characteristics have changed. Since 1969, the proportion of wives working
year-round full-time has increased from 17 percent to 39 percent in households
with children. For married couples without children, 60 percent of wives are
now employed full-time (McNeil, 1998). Although the average hours worked
per week has declined, households with two income-earners generally lack the
time to spend hours shopping. Gift cards fit naturally into a convenience-driven
lifestyle, are easy to buy and give, and help givers stay within their holiday
budget.

The Welfare Loss of Gifts

A gift involves a cost for the giver and a benefit for the recipient. Most of us
know from experience that it is quite possible for the gift-giver to spend more on
the gift than the gift is worth to the recipient. This section discusses some evidence
on the deadweight loss of gift giving and raises some considerations that should
affect the extent of this deadweight loss, including issues of how cash gifts are
regarded and how well the gift-giver understands the preferences of the recipient.

The Deadweight Loss of Christmas

Waldfogel (1993) used surveys of Yale undergraduates to explore the gap
between the price of a good to the giver and the value to the recipient. He
memorably called this gap the “deadweight loss of Christmas” and estimated that
“between a tenth and a third of the value of holiday gifts is destroyed by gift-giving”
(p. 1336). Waldfogel’s argument was that recipients of gifts would have received
higher welfare from the cash equivalent of the gifts they received.1
Waldfogel’s (1993) survey also showed that the size of the gap was smaller for
gifts from friends and “significant others,” and larger for gifts from extended
family. A plausible explanation of this finding is that when a gift-giver knows more
about the preferences of the gift recipient, the value of the gift to the giver will be
closer to the value for the recipient.
Subsequent replies to Waldfogel’s (1993) article pointed out the absence of a
calculation for the sentimental value of gifts, with List and Shogren (1998) finding
a welfare gain to gift-giving after incorporating sentimental feelings. This finding
suggests an upward bias in the originally estimated deadweight loss for physical
gifts. With gift cards, however, it is difficult to convey sentimental feelings; thus they
are more straightforward to measure in terms of value.

1
Analyses of government programs like food stamps often point out that recipients would benefit if
these programs were replaced with cash (for theoretical discussion and references, see Offenberg,
2005).
 Markets: Gift Cards 231

When Cash Gifts are Inappropriate

Cash gifts are traditionally seen as inappropriate. A common rationale for
nonmonetary gifts is that both thought and effort are important. A 2004 Deloitte
and Touche survey estimated that gift-givers prefer gift cards over cash by a 2-to-1
margin. Webley, Lea, and Portalska (1983) found that consumers rated gift certif-
icates closer to presents than cash, in terms of acceptability. Thus, a gift card might
be seen as more thoughtful than a cash gift and also a manner of avoiding the
traditional cash stigma.
While this appeal to the traditional stigma of cash gifts has some truth, it is
not obvious why a gift card should be preferred to cash. After all, with either a
cash gift or a gift card, the exact amount spent on the recipient is revealed,
except that in one case the funds are restricted to a specific purpose: purchases
from a certain store. Gift cards offer considerable convenience, but simply
giving cash would offer even more convenience to a gift recipient—along with
flexibility on where to spend the gift.
Gift-givers might actually prefer the restrictions of a gift card because it makes
the gesture more memorable than a cash gift. Perhaps a gift card creates a kind of
shared bond in which the giver can gain satisfaction from imagining the recipient
going to the store and the recipient can appreciate the gift when shopping at the
store. If people enjoy giving gift cards and other people appreciate receiving them,
then the restriction to a certain store should be viewed as part of the value of the
gift, rather than a limitation on the gift.
However, if the thought is what counts in getting a nonmonetary gift, then
Prendergast and Stole (2001, p. 1794) ask, “[Couldn’t] the donor simply
communicate information to the recipient by telling her what he would have
purchased and then making an appropriate transfer of money . . .” Applying
their logic to gift cards, this proposal is equivalent to giving a cash gift and
including a note that the recipient might want to spend the money at [insert
name of store here].

Corresponding Preferences and Welfare Loss

If consumers are perfectly informed of their own preferences, then a
gift-giver will be unable to produce a higher utility-increasing gift for the
recipient than cash. If the giver instead chooses a gift card, then the more the
giver understands the consumer’s preferences, the more appropriate store they
choose. For example, a consumer who routinely spends $100 at Target should
not suffer any welfare loss from receiving a $50 Target gift card (given certain
assumptions over consumer preferences), because the gift effectively frees up
$50 of cash to be spent on whatever the recipient desires. Thus, to the extent
that the giver understands the recipient’s preferences towards Store X, the
welfare loss can be reduced or eliminated.
It may also be the case that recipients are imperfectly informed of their own
preferences, giving the donor an opportunity to offer a gift card that creates value.
232 Journal of Economic Perspectives

One example is a gift card for a store that the recipient is unaware of, but that fits
their personality or style. This possibility would help explain why people give gift
cards, as they may be attempting to achieve this outcome.

Evidence from the Resale of Gift Cards

Gift cards are bought and sold, at a discount, through Internet auction sites
like eBay.2 The existence of this secondary market demonstrates a gap between the
face value of the card and the value to the recipient. I collected information from
2,002 completed sales of gift cards by at least 500 unique sellers on eBay.com from
February to June 2004. In these data, auctions of gift cards from 31 stores took
place, ranging from electronic stores to discount “supercenters.”3 Most readers will
recognize these store names because they are all national chains. The average
company in the sample operates around 1,000 locations in 45 states. The average
company has been open 52 years, and the average store for that company offers
over 55,000 square feet (1.28 acres) in floor space. The dominance of nationwide
firms in the sample suggests that chain stores were an important factor in the rapid
growth of gift card sales.
These data provide evidence on how recipients perceive the cash value of gift
cards and on how this cash value varies with the stored value of the card and the
type of store.

Cash Equivalence
The “face value” of the gift card is the amount of the store’s credit currently
stored on the card. The average discount of a card on eBay is calculated as the
difference between the average face value of the gift card being sold and the
average sale price, divided by the average face value. Sellers accept significantly less
cash for their gift cards, at a loss of approximately 15 percent of the initial value on
average. Defining welfare loss as the difference between how much a gift-giver
spends and the ultimate value to the recipient, a gift card’s discount on eBay is an
estimate of the welfare loss to giving that gift card. Although the highest bidder’s
willingness to pay for the gift card is not revealed in an ascending auction, the
appropriate value to use when calculating the welfare loss to gift-giving is the
recipient’s willingness to accept, as it measures the direct monetary loss between

2
Gift card sellers on eBay face several restrictions, with the most limiting rule being that a seller can only
sell at most one gift card per week. This restriction prevents mass sales by any particular seller.
3
The 25 stores in the sample with 20 or more observations were: Abercrombie & Fitch, American Eagle,
Banana Republic, Best Buy, Bloomingdales, Circuit City, Express, GAP, Home Depot, JCPenney’s,
K-Mart, Lowe’s, Macy’s, Nordstrom, Office Max, Old Navy, PetSmart, Sears, Staples, Starbucks, Target,
Tiffany & Co., Toys “R” Us, Victoria’s Secret, and Wal-Mart. For a summary table showing for each store
the average face value of the gift, the average sale price, the average discount, and information on
number of stores, store size, how long the firm has been in operation, and sample size, see the Appendix
Table that appears with this article at 具http://www.e-jep.org典.
 Jennifer Pate Offenberg 233

Figure 2
The Ten “Best” and Five “Worst” Gift Cards
(according to sale price as a percentage of credit value)

Smaller Welfare Loss

Home Depot
Lowe’s
Office Max
Circuit City
Starbucks
Old Navy
Wal-Mart
Target
Staples
Best Buy

Tiffany & Co
Victoria’s Secret
Abercrombie & Fitch
American Eagle
Express

60% 65% 70% 75% 80% 85% 90%

Sale price as % of credit value

giver and recipient. Figure 2 shows the ten stores with the smallest discounts and
the five stores with the largest discounts.
A majority of the auctions offered “free” shipping, which in effect means that
shipping is therefore an additional cost incurred by the seller and should be
included as part of the cost. Absent in this 15 percent figure are the costs associated
with selling on eBay, which may add an additional 5 percent of funds lost in the
transfer.4 Taking the sales price and the selling costs together, a fair estimate would
be that gift cards have a cash value 20 percent lower than their face value to
recipients, which is similar in magnitude to Waldfogel’s (1993) estimate of the
deadweight loss of presents.
Another issue potentially affecting the value of the card is that many stores
have policies that specify the number of months a consumer has to spend the value

4
Since these data include an auction’s starting bid and sale price, an approximation of average selling
cost per auction can be calculated. The average fee was approximately 5 percent of the sale price, with
a maximum of 19.25 percent. The fee percentage was also negatively correlated with the sale price of the
item, at a value of – 0.34, suggesting that sales of lower-valued gift cards may incur a slightly higher
proportional fee than higher-valued credit sales.
234 Journal of Economic Perspectives

of the card before certain “maintenance” fees will be assessed or the card expires.
Most states have passed laws making it illegal for gift cards to expire or for
companies to charge these fees. For example, California banned gift card expira-
tion dates in 1997. In some states this battle continues. In Massachusetts a 2003 law
extended the minimum time before gift cards may expire from two to seven years.
A 2006 Ohio law banned expiration dates of less than two years, though gift cards
may still expire at that time. In these data, the original purchase date of the gift card
being auctioned is unavailable, so it is impossible to calculate how the value of the
card might be affected by expiration dates or maintenance fees. However, a recent
survey examining gift card policies listed 18 stores as having the best policies with
no deductions or expiration dates, including Best Buy, Circuit City, Gap, Home
Depot, JCPenney, Lowe’s, Nordstrom, Old Navy, PetSmart, Sears, Starbucks, Tar-
get, and Wal-Mart (Montgomery County Division of Consumer Affairs, 2005). Since
together these stores make up over 67 percent of the dataset, gift card expiration
policies should not cause much or any bias in the analysis.
Is this estimate of a 20 percent gap between the cost of gift cards and the value
to the recipient likely to contain a predictable upward or downward bias? On the
one hand, if a substantial number of people are aware of the secondary market for
gift cards and those choosing to sell their gift cards are the recipients who value
them least, then many recipients value their cards more highly (or at least too
highly to sell them), and the 20 percent estimate would be biased upward. On
the other hand, some people would be willing to sell their gift card at a
discount, but do not do so because they are unaware that a secondary market
exists. Doing a recent small survey of Yahoo! users, I found that less than 39
percent of these Internet-experienced consumers knew gift cards could be sold
for cash (N⫽103). A survey of the general population would likely yield a
smaller percentage. In addition, some of those recipients willing to sell their gift
cards at a 20 percent discount would presumably also have been willing to sell
at a larger discount, if that had been the market price. This suggests that the
welfare loss associated with giving gift cards could have been greater, especially
if the recipient had been unable to trade the gift card subsequently. Without
knowing the percentage of gift card sellers relative to the total gift cards given,
the percentage of the public that is aware of this market, and the elasticity of
demand and supply of gift cards to this market, this estimate for welfare loss
from gift cards must remain an approximation.

Store Characteristics
The more potential uses of a gift card, the less welfare loss it should create. The
data was pooled, according to the type of store, into ten categories shown in Figure
3. Gift cards for home improvement and discount stores have the smallest welfare
loss; gift cards for jewelry and apparel stores have the largest loss. A similar pattern
is apparent in Figure 2 provided earlier, where gift cards from a home improve-
ment store (Home Depot) and a discount store (Wal-Mart) are among the least
 Markets: Gift Cards 235

Figure 3
Welfare Losses of Gift Cards by Store Type
(according to sale price as a percentage of credit value)

95%
Sale price as % of credit value

90%

85%

80%

75%

70%

65%

60%
nt re es p ics tore tore tore re el
e me t sto p pli e sho tron y s t s t s y sto ppar
v n s u ffe Elec To tmen e el r A
pro iscou ffice Co r
P
Jew
e im D O pa
m D e
Ho

discounted, while clothing and jewelry stores show the most discounting from face
value (Express and Tiffany & Co.).
Gift cards from stores that offer more general use items or larger product
variety tend to sell with smaller discounts, resulting in smaller welfare losses.
Wal-Mart, one of the top sellers in terms of the cash equivalence of their gift cards,
offers a number of possible choices, from completing car repairs to buying sports
equipment and even groceries. This reasoning may also apply to the potential
sellers of Wal-Mart gift cards. If gift card recipients can spend them without
difficulty, then supply will decrease and the price will rise, all else equal. Therefore,
if a certain store characteristic like product variety increases the demand for gift
cards, decreases the supply of gift cards, or both, then stores with greater product
variety will sell for higher prices.
Regression analysis, using the discount from face value as the dependent
variable, confirms that traits of stores influence the size of the discount. A greater
number of locations, locations in more states, the availability of shopping and
purchasing online, and greater product variety as measured by larger retail space all
result in a smaller gift card discount. The number of years in which a store had
been in existence had a positive and significant effect on the size of the discount
(equal to a negative impact on a gift card’s sale price), suggesting that “newer”
retailers may be more popular with the eBay bidder population.5

5
For specific regression results, as well as comments on the data and alternative specifications, see the
appendix to this article available at 具http:/www.e-jep.org典.
236 Journal of Economic Perspectives

The Size of Gift Cards

Giving a smaller amount in the form of a gift card is less likely to create a
welfare loss. If a consumer had budgeted $20 for purchases at Store X and then
received a gift card valued at $20 for that store, then the income can simply be
shifted to other goods and no welfare loss arises. If a gift-giver is already planning
to spend a small amount on someone, then a gift card or cash gift would likely
create the same impact on the recipient.
This insight suggests that larger-valued gift cards are more likely to create
significant welfare losses. Conversely, smaller-valued gift cards would then be
less likely to be sold and the secondary market should be characterized by a
disproportionate number of higher valued gift cards. Since eBay fees are
proportional to the value of the item, selling costs should have a consistent
effect on the selling decision across all values of gift cards and should not
disproportionately affect smaller valued gift cards. However, there is also a fixed
cost of time incurred by sellers that may affect the decision of whether to sell a
gift card.
These data support the hypothesis that smaller-valued gift cards are less
disruptive to a consumer’s consumption preferences, creating little or no
welfare loss. Figure 4 displays the distribution of gift card values over the
sample. A ValueLink (2003) survey found the average amount spent per gift
card purchase to be $48 for men and $37 for women, making the average gift
card purchased in the United States roughly $42. The average gift card sold on
eBay has a face value of $192. Gift cards valued up to and including $20 account
for just 6.5 percent of the sample, while cards valued up to $40 account for only
17.1 percent of the auctioned gift cards. Smaller valued gift cards are less likely
to be sold on eBay, supporting the hypothesis that smaller-valued gift cards are
less likely to create welfare losses.

Conclusion

Gift cards are booming in popularity. However, gift cards lack the sentimental
value of a physical present and have similar properties of cash gifts, such as
revealing the total amount spent. Moreover, gift cards can produce welfare losses
between the giver and the recipient. This loss in the transfer from giver to recipient
is estimated at 15 percent of the original amount spent, increasing to 20 percent
when the eBay seller transaction costs are included. If the projected $80 billion is
indeed spent on gift cards in 2007, then a welfare loss of 20 percent would be equal
to $16 billion. Of course, the sample is censured, because it only consists of sellers
willing to part with their gift cards. Also, the existence of a secondary market for gift
cards likely reduces the size of the welfare loss by providing an opportunity for
recipients to exchange cards for cash, an opportunity which has long existed for
 Jennifer Pate Offenberg 237

Figure 4
Distribution of Gift Card Values

400 100%

360 90%

320 Frequency 80%

Cumulative %
280 70%

Cumulative %
240 60%
# of cards

200 50%

160 40%

120 30%

80 20%

40 10%

0 0%
5 100 200 300 400 500 600 700 800 900 1000
Value stored on gift card

physical goods received as presents. If Waldfogel’s (1993) estimation of the welfare

loss from physical gifts at 10 –30 percent is correct, then at an average welfare loss
of 15–20 percent, gift cards have a welfare loss that is comparable—although
perhaps not worse—than physical presents.
The evidence presented here suggests some advice for those considering a gift
card. First, givers who are less aware of the recipient’s preferences should rely more
on gift cards from general-purpose stores that offer greater product variety like
Target or Home Depot, since gift cards from these retailers are less likely to create
welfare losses. Second, smaller-valued gift cards are less likely to produce welfare
losses than higher-valued gift cards. Finally, for higher-valued gifts and when the
gift-giver is imperfectly informed of the recipient’s preferences, a cash gift will
generally provide a larger increase in the recipient’s utility than a gift card.
Gift-givers planning on giving a gift card might want to bear in mind the possible
benefit of a cash gift with a note to the recipient suggesting that the money could
be spent at [insert name of store here]—to add the thought that counts.

y I would like to thank Jack Barron, Dan Kovenock, Tim Cason, Stephen Martin, David
Offenberg, Colin Cameron, and James Konow for guidance. I also thank James Hines,
Timothy Taylor, and Michael Waldman for comments. This research is based on one chapter
of my Ph.D. dissertation, and I gratefully acknowledge the Purdue Research Foundation for
their support.
238 Journal of Economic Perspectives

References

Bain and Co. 2003. “U.S. Gift Pre-Paid Market lease. http://www.nrf.com/content/default.asp?
Size by Sales in Dollars from 1997 through 2001 folder⫽press/release2005&file⫽giftcards1105.
and Forecast for 2002.” Credit Card Management, htm&bhcp⫽1.
15(12): 36. Offenberg, Jennifer P. 2005. “Welfare Impli-
Berner, Robert. 2005. “Gift Cards: No Gift to cations of Gift Cards and the New Market for
Investors.” Business Week Online, March 3, 2005. Store Credit: Theory and Evidence from Internet
Bolt, Kristen. 2004. “Retail Notebook: Plastic Auctions.” http://myweb.lmu.edu/joffenberg/
Makes Perfect Gift This Holiday Season.” Seattle research.html.
Post-Intelligencer, November 13, 2004. Offenberg, Jennifer P. 2006. “Seller Reputa-
Deloitte and Touche. 2004. “Holiday Shop- tion as a Determinate of Sale Price in Online
ping: It’s in the Cards.” Deloitte Consumer Hol- Auctions: Theory and Evidence from Gift Card
iday Survey Findings. http://www.prnewswire. Sales.” http://myweb.lmu.edu/joffenberg/research.
com/cgi-bin/stories.pl?ACCT⫽109&STORY⫽/ html.
www/story/11-09-2004/0002399718&EDATE⫽. Prendergast, Canice, and Lars Stole. 2001.
Home Depot. 2005. Annual Report 2005.
“The Non-Monetary Nature of Gifts.” European
http://ir.homedepot.com/reports.cfm.
Economic Review, 45(10): 1793–1810.
List, John, and Jason Shogren. 1998. “The
TowerGroup. 2005. “U.S. Gift Card Market to
Deadweight Loss of Christmas: Comment.” Amer-
See Rapid Sales Growth through 2007.” Tower-
ican Economic Review. 88(5): 1350 –5.
Group Press Release. http://old.towergroup.com/
Maranjian, Selena. 2004. “Careful With Those
Public/presscenter/pressreleases/040604.htm.
Gift Cards!” The Motley Fool. November 29,
2004. http://www.fool.com/News/mft/2004/ ValueLink Survey. 2003. “Twice as Many
mft04112902.htm. Americans Purchased Gift Cards in 2003 vs.
McNeil, John. 1998. “Changes in Median 2004.” ValueLink Press Archive. http://www.
Household Income: 1969 –1996.” U.S. Depart- valuelink.info/news/2003PressArchive.htm.
ment of Commerce, Current Population Report, ValueLink Survey. 2004. “U.S. Annual Survey
Special Studies P93-126. Percentages of Gift Card Spending in Six Con-
Montgomery County Division of Consumer sumer Behavior and Spending Categories.”
Affairs. 2005. “Gift Cards 2005: Many Good Credit Card Management, 17(8): 10.
Retail Cards. Any Good Bank Cards?” http:// Waldfogel, Joel. 1993. “The Deadweight Loss
www.montgomerycountymd.gov/content/ocp/ of Christmas.” American Economic Review, 83(5):
consumer/gift_cards_report_2005.pdf. 1328 –36.
National Retail Federation. 2005. “Gift Card Webley, Paul, S. Lea, and R. Portalska. 1983.
Sales to Surge Again this Holiday as Popularity “The Unacceptability of Money as a Gift.” Journal
Increases.” National Retail Federation News Re- of Economic Psychology, 4(3): 223–38.
 Markets: Gift Cards A1

Appendix

Regression Evidence on Gift Card Discounts and Store Traits

To test how certain store characteristics affect the magnitude of a store’s gift
card discount, a regression was estimated with the dependent variable taking the
value of ln(Face value) – ln(Sale price). In this case, the “Sale price” is calculated
as the winning bid plus the cost of shipping (if applicable). Data regarding different
store characteristics were collected to examine which factors may be responsible for
the variance across store gift cards, and the size of their effect. The explanatory
variables include the number of locations of each store and the number of states,
to test for national brand awareness. Stores with more locations and/or in more
states should be better known across potential bidders, which should increase
demand. The average number of acres per store is included as a proxy for product
variety, under the assumption that stores occupying more retail space offer greater
product variety. To test for generational brand awareness, the age of the company
entered the equation. Finally, to measure the effect of convenience, a dummy
variable representing the availability of online sales is also included in the analysis.
Regression coefficient estimates for each of these variables, appearing in
Appendix Table 2, indicate that all but one of the above characteristics have a
negative and significant impact on the size of a gift card’s discount.
Although not a focus of this article, information regarding other auction
characteristics like the seller’s reputation or feedback score and payment restric-
tions were collected and controlled for in the analysis. For descriptive statistics and
an in-depth analysis of how these different auction characteristics affect sale price,
see Offenberg (2006).
An alternate test was conducted with the dependent variable ln(Sale price)
regressed on ln(Face value), along with the other variables. The coefficient for
ln(Face value) was not significantly different from 1 (p-value ⫽ 0.265). The R2 term
for that regression was 0.99, and the values of the auction coefficients simply
change sign, while their level of significance remains the same.
 A2 Journal of Economic Perspectives

Table A1
Summary Statistics for Gift Card Auctions

Average Average Average Average Years

face sale discount Number of store size of
value price (%) locations/states (acres) service Count

Abercrombie & Fitch $ 177.09 $ 140.12 21.85% 366/49 0.229 16 164

American Eagle 109.45 84.61 23.03% 849/50 0.094 27 42
Banana Republic 147.29 119.79 18.21% 462/46 0.128 26 37
Best Buy 179.49 156.63 13.34% 698/48 0.735 21 187
Bloomingdales 188.39 147.21 22.78% 36/12 5.186 132 33
Circuit City 218.44 195.88 12.08% 617/45 0.689 55 67
Express* 161.41 120.99 25.47% 884/47 0.140 24 48
GAP 108.38 92.85 15.49% 1,390/50 0.275 35 66
Home Depot 305.28 276.98 10.83% 1,900/50 2.502 26 260
JCPenney’s 108.43 91.32 16.07% 1,017/50 2.664 92 44
K-Mart 71.08 60.59 16.07% 1,479/49 2.778 105 26
Lowe’s 252.65 226.62 11.02% 1,100/48 2.755 58 137
Macy’s 88.17 72.90 17.67% 423/34 4.079 146 30
Nordstrom 200.54 179.57 12.20% 151/27 3.444 103 26
Office Max 170.46 152.25 11.19% 945/49 0.459 16 24
Old Navy 51.84 45.00 12.40% 907/48 0.344 10 20
PetSmart 117.75 95.02 17.81% 726/48 0.528 17 32
Sears 157.72 136.39 14.89% 900/50 3.139 111 61
Staples 107.12 91.53 12.87% 1,442/50 0.482 18 54
Starbucks* 78.99 67.15 12.39% 8,909/50 0.034 33 109
Target 111.57 98.65 12.65% 1,330/47 2.869 42 104
Tiffany & Co 1483.40 1187.16 20.34% 55/23 0.184 167 33
Toys “R” Us 96.26 82.16 15.50% 927/50 1.034 56 76
Victoria’s Secret 106.23 84.51 20.99% 1,001/48 0.108 22 56
Wal-Mart 154.99 138.12 12.62% 3,066/50 3.412 42 219
All 192.29 163.35 15.05% 1,125/45 1.277 51.7 2002

Note: Stores with less than 20 observations do not appear in this table.
* Indicates that products are not available online.

Table A2
Gift Card and Store Characteristic Estimates
(Dependent variable: ln(Face value) ⫺ ln(Sale price))

Store characteristic variables Estimate Standard error

Number of store locations ⫺0.0013* 0.0002

Locations in number of states ⫺0.114* 0.032
Size of average store (in acres) ⫺0.95* 0.143
Years of service 0.018* 0.006
Available online ⫺5.846* 1.336
R2 0.19

* Significant at the 1% level, based on 2,002 observations.

View publication stats