European Journal of Marketing

The effect of odd pricing on demand

Philip Gendall Judith Holdershaw Ron Garland
Article information:
To cite this document:
Philip Gendall Judith Holdershaw Ron Garland, (1997),"The effect of odd pricing on demand", European Journal of
Marketing, Vol. 31 Iss 11/12 pp. 799 - 813
Permanent link to this document:
http://dx.doi.org/10.1108/03090569710190541
Downloaded on: 13 February 2016, At: 20:31 (PT)
References: this document contains references to 32 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 2513 times since 2006*
Users who downloaded this article also downloaded:
Philip Gendall, (1998),"Estimating the effect of odd pricing", Journal of Product & Brand Management, Vol. 7 Iss 5 pp.
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

421-432 http://dx.doi.org/10.1108/10610429810237754
Nicolas Guéguen, Patrick Legoherel, (2004),"Numerical encoding and odd-ending prices: The effect of a contrast in discount
perception", European Journal of Marketing, Vol. 38 Iss 1/2 pp. 194-208 http://dx.doi.org/10.1108/03090560410511186
Keith S. Coulter, (2001),"Odd-ending price underestimation: an experimental examination of left-to-right processing effects",
Journal of Product & Brand Management, Vol. 10 Iss 5 pp. 276-292 http://dx.doi.org/10.1108/10610420110401838

Access to this document was granted through an Emerald subscription provided by emerald-srm:472570 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service
information about how to choose which publication to write for and submission guidelines are available for all. Please
visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of
more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online
products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication
Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct at time of download.

 The effect of odd
The effect of odd pricing on pricing on
demand demand

Philip Gendall, Judith Holdershaw and Ron Garland

Massey University, Palmerston North, New Zealand 799
Received March 1996
Introduction Revised August 1996
The practice of odd pricing in retailing is so prevalent that its efficacy is
generally taken for granted. However, there is little reported evidence to
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

support the underlying assumption of odd pricing; namely, that prices set just
below the nearest round figure produce higher than expected demand at that
level. This paper reports the results of a study designed to test this assumption
by examining the effect of odd pricing on demand for a range of products at
different price levels.
There is no general agreement on the definition of odd prices, and they are
sometimes referred to as magic prices, charm prices, psychological prices,
irrational prices, intuitive prices or rule-of-thumb prices (Boyd and Massy, 1972;
Dalrymple and Thompson, 1969; Gabor, 1977; Kreul, 1982; Monroe, 1990;
Rogers, 1990; Sturdivant, 1970). However, their general characteristic is that
they are set just below the nearest round figure; for example, $4.99 instead of
$5.00, or $19.95 rather than $20.
There is little agreement about the origin of odd pricing, but evidence of it
can be traced back more than 100 years (Schindler and Wiman, 1989). One
theory is that odd pricing originated after fixed pricing became the norm in the
USA, shortly after the end of the Civil War. It was only after even, or round,
prices became established that odd pricing, as it is now known, could emerge as
a common retailing practice. Before that time, prices were determined by
haggling between consumers and retailers (Georgoff, 1971).
The standardization of currency in America may also have had an effect.
Imported British goods underwent a currency conversion of the British pound
sterling into dollars, often giving British goods an odd price ending. As a result
of the quality attributed to British goods, odd prices became associated with
superior products. In the late 1800s American retailers would often attach an
odd price ending to domestic goods because of the high quality image
associated with odd priced imported products (Georgoff, 1971).
Another commonly cited explanation for the introduction of odd pricing is
that it arose as a measure to help combat theft by employees (Harper, 1966;
Högl, 1988; Sturdivant, 1970; Twedt, 1965). This is widely believed to have
begun earlier this century when Macy’s New York department store introduced
99 cent sales. Odd prices were adopted to force salespeople to issue change and
thereby make it difficult for them to pocket the customer’s payment without European Journal of Marketing,
Vol. 31 No. 11/12, 1997, pp. 799-813.
recording a sale (Kreul, 1982; Rudolph, 1954). The idea was considered novel by © MCB University Press, 0309-0566
 European consumers and consequently had a positive effect on sales. This convention was
Journal subsequently adopted by retailers around the world (Gilmour, 1985).
of Marketing Changes in retailing methods and technology mean that the conditions
which stimulated the emergence of odd pricing no longer exist. Nevertheless,
31,11/12 retailers’ use of the technique of odd pricing is now extremely common
(Schindler and Wiman, 1989).
800
Prevalence of odd pricing
The noticeable characteristic of odd pricing is the sheer prevalence of this
practice in comparison to even pricing, and in particular, the dominance of
prices ending in the digit 9.
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

In 1948, an analysis of 3,025 retail store advertisements in newspapers in 37

American cities revealed 64 per cent of prices ended in odd digits (Rudolph,
1954). Another early general observation of retail food prices showed that prices
ending in 9 were most popular, with prices ending in 5 being second in
popularity (Printers Ink, 1954; Twedt, 1965). In fact, the 9s and 5s often
accounted for 80 per cent or more of retail prices (Friedman, 1967).
A more recent extensive analysis of scanner data from a major American
supermarket chain revealed that over 80 per cent of the store’s prices ended in
the digit 9 (cited in Schindler and Wiman, 1989). Högl (1988) reported a similar
situation in Germany, where most supermarket prices lie just below a Deutsche
Mark amount (i.e. 99 Pfennigs, DM4.99). The same pattern was also found in
New Zealand in a study which estimated that around 87 per cent of advertised
prices used odd endings, with around 60 per cent of these prices ending in the
digit 9 (Holdershaw, 1995).

Rationale for odd pricing

The rationale for odd pricing is that it creates greater than expected demand at
these prices. In other words, odd pricing is assumed to produce a “kink” in the
expected demand curve for the product concerned. This is illustrated in
Figure 1.
The usual explanation proposed for this effect is that by setting a price at,
say $39.95 rather than $40.00, an illusion is created which makes the product
seem much cheaper to consumers than the nearest (and higher) round figure.
This illusion of cheapness then triggers an enhanced buyer response (Boyd and
Massy, 1972).
Brenner and Brenner (1982) suggest that this phenomenon is the result of a
biological constraint, namely, consumers’ limited capacity for storing directly
accessible information. They believe that consumers exposed to price
information store only the more valuable parts of the message they receive, the
first digits of a number. For instance, when a price is $299, the digit 2 is more
significant as information than the first 9, which in turn is more significant
than the next 9. Thus consumers will recall that the price is $200, then maybe
that it is $290, but rarely that it is $299. The reason offered for not rounding the
three digits up to $300 is based on memory processing time. Rounding up
 Price The effect of odd
pricing on
demand
$40.00

$39.95 801
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

$35.00

1.0 2.0 3.0 4.0 5.0 6.0 Figure 1.

“Kinked” demand curve
Quantity

involves an additional decision compared with storing the integer part of the
number.
However, while the existence of a price illusion is a plausible explanation for
the effect of odd pricing, there is little evidence to support it. A study by
Georgoff (1971) found that, although a price illusion may occur for certain
products among certain groups of consumers, any net effect on sales was weak
or confounded by situational and intervening variables. A later study by
Lambert (1975) suggested that lower price illusions were associated with odd
prices under some circumstances, but Dodds and Monroe (1985) found no
evidence of differences in consumers’ perceptions of quality, value, or
willingness to buy for products priced at odd and even prices.
Other explanations for the effect of odd pricing are that it suggests to
consumers that goods are marked at the lowest possible price (Harper, 1966), or
that consumers have become conditioned by retailers to expect odd prices. A
study by Gabor and Granger (1964) provides some support for the latter
explanation. For one of two products tested in this study, consumers’ purchase
intentions were higher when presented with an odd price than with the next
lowest price point.
Thus the cause of the odd pricing effect, if it does in fact occur, is debatable.
 European Effect of odd pricing on demand
Journal The earliest documented study into the effect of odd pricing on demand was
of Marketing conducted in the 1930s by a large American mail order company. The company
suspected that the effectiveness of its odd-cent pricing resulted from “habit and
31,11/12 inertia” (Ginzberg, 1936). To investigate this assumption, even pricing was used
for a representative sample of items in several regional issues of one of the
802 company’s catalogues. The usual odd cent prices were used in its other
catalogues. The company was able to account, with a reasonable degree of
certainty, for any variables which may have influenced demand, other than
price, by a detailed review of sales activities in the preceding and present
period.
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

The results of this experiment were inconclusive. Even pricing greatly

increased the sales of some items, cut the sales of other products in half, and left
the sales of some items unchanged. An executive of the firm estimated that the
sales losses were about equal to the sales gains.
In another American example, a department store chain successfully shifted
from odd to even pricing (Dalrymple and Thompson, 1969). The store, which
had traditionally used a 95 cent price ending on many items, observed no
adverse sales effects after it changed to even dollar pricing. Nevertheless,
although apparently satisfied with the new even pricing policy, the store
continued to use odd pricing in budget departments and on sale merchandise
(Dalrymple and Thompson, 1969).
More recently, a positive effect of odd pricing on sales when popular brands
of margarine were discounted and advertised as weekly specials has been
reported (Nagle, 1987). Unit sales for one brand increased by 194 per cent when
its price was discounted from 83 cents to 63 cents, but increased by 406 per cent
when discounted to 59 cents. A second brand showed a similar effect, with a 65
per cent increase in unit sales when discounted from 89 cents to 71 cents and a
222 per cent increase when discounted to 69 cents. Unfortunately, without
examining the results together with other variables such as promotional
activity and competitors’ activities, it is not possible to conclude that odd
pricing alone was responsible for the disproportionate increase in sales when
prices were discounted to a number ending in 9. Nevertheless, these findings are
consistent with those of a study by Schindler and Warren (1988) in which
subjects were asked to select items from a simulated restaurant menu. The
authors concluded that pricing an item just below a round number can increase
its likelihood of being selected to a greater extent than would be expected on the
basis of the few cents involved.
By contrast, a New Zealand study reported by Holdershaw (1995) revealed
no difference in the purchase rates of two samples of consumers exposed to a
mail order catalogue which differed only in the pricing of the 42 products
included. Half of the catalogues featured even dollar prices, the other half
showed the same products at slightly cheaper odd prices with 95 cent price
endings. The proportion of consumers ordering was 9.8 per cent from the
 sample which received the even-priced catalogue, and 10.3 per cent from those The effect of odd
who received the odd-priced catalogue. pricing on
Overall, the findings of research into the effects of odd pricing on demand demand
have been mixed and inconclusive. In some instances odd prices appear to have
increased demand for items, while in other instances demand was reduced, and
for other products no effect was observed. Clearly, the efficacy of odd pricing
remains unproven. 803
In summary, there is widespread use of and belief in odd pricing, and no
shortage of anecdotal evidence to support this practice. Perhaps surprisingly,
however, there is little empirical evidence of its effectiveness. As Holloway
stated, “It is interesting that a strategy so widely used and accepted by
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

academicians has so little proof behind it” (cited in Schindler and Wiman, 1989,
p. 107).

Method
The research reported in this paper involved estimating a demand curve for
each of six different products and investigating whether the expected odd
pricing effect occurred. The process used was similar to the Gabor-Granger
technique of pricing research, which involves showing respondents a product,
then determining their purchase intentions at a number of prices. This
technique produces the familiar downward sloping demand curve, thus
enabling a measure of price sensitivity to be made at each point tested (Gabor
and Granger, 1964).
In this study, however, instead of asking respondents to give their purchase
intentions, the Juster scale was used to elicit their purchase probabilities at
different prices. The Juster scale, illustrated in Figure 2, is an 11-point purchase
probability scale which has been successfully used to predict consumer
purchase rates for a range of items including durables, services and fast-
moving consumer goods. Since its development in the 1960s, the Juster scale has
consistently been shown to be a better predictor of consumer purchases than
verbal buying intentions (Clawson, 1971; Day et al., 1971; Gabor and Granger,
1972/73; Gruber, 1970; Heald, 1970; Isherwood and Pickering, 1975; Juster,
1966).

10 Certain, practically certain (99 in 100)

9 Almost sure (9 in 10)
8 Very probable (8 in 10)
7 Probable (7 in 10)
6 Good possibility (6 in 10)
5 Fairly good possibility (5 in 10)
4 Fair possibility (4 in 10)
3 Some possibility (3 in 10)
2 Slight possibility (2 in 10) Figure 2.
The Juster purchase
1 Very slight possibility (1 in 10)
probability scale
0 No chance, almost no chance (1 in 100)
 European The sample for the study consisted of 300 respondents who were either mostly
Journal or jointly responsible for their household’s shopping, selected by mall intercept
of Marketing in a Palmerston North shopping centre in September 1994. The response rate
was 49 per cent, and the sample was randomly divided into three subsamples of
31,11/12
100, for reasons described subsequently.
Six products were tested in this study: a block of cheese, a frozen chicken, a
804 box of chocolates, a hair dryer, an electric kettle and a blender. Two criteria were
used to select these products. First, products with reasonably broad appeal were
chosen so as to minimize the chance of a very high proportion of zero purchase
probabilities. And, second, products were selected because their normal retail
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

price coincided with four “critical” price levels identified by retailers. These
“critical” price levels are under $10, $20, $50 and $100[1]. Retailers believe that
these price levels are critical in the sense that exceeding them will result in a
disproportionately high loss of sales, and that the effect of odd pricing at these
levels is particularly strong.
Each respondent was shown a series of 18 showcards featuring these six
products at three different prices, and asked to estimate their purchase
probability for each product at each price. One of the three showcards for each
product included a price slightly above an even price point (for example,
$10.10), and one a price slightly below the same price point (for example, $9.90).
These two prices were presented to all 300 respondents and served the purpose
of creating top and bottom “anchor” points of the demand curve for the product
concerned. The third showcard included either the appropriate even test price
(for example, $10.00) or one of two odd test prices (for example, $9.99 or $9.95).
Each of these three “test” prices was presented to a subsample of 100
respondents.
For two products at the $20 or dearer level, instead of testing the 99 cent odd
price ending, which is less commonly used for higher prices, the slightly higher
5 cent price ending was tested (for example, $20.05). Although a 5 cent ending
is not defined as “odd”, including this price provided another point on the
demand curve and an independent test of the effect of varying the price by five
cents around the even price concerned. The products and corresponding price
points tested are shown in Table I.

Bottom Test prices Top

Product anchor price ($) T1 T2 T3 anchor price ($)

Cheese 4.90 4.95 4.99 5.00 5.10

Frozen chicken 5.90 5.95 5.99 6.00 6.10
Chocolates 9.90 9.95 9.99 10.00 10.10
Table I. Hair dryer 19.90 19.95 20.00 20.05 20.10
Products and prices Kettle 49.90 49.95 50.00 50.05 50.10
tested Blender 90.00 95.00 99.00 100.00 110.00
 The six products were presented to respondents three times, in the same order. The effect of odd
The first time, all of the lowest prices were presented, then all of the highest pricing on
prices, and, finally, one of the three test prices for each product. demand
The test prices were assigned to respondents in such a way that 100
respondents were presented with each test price. To overcome the effects of
order bias within subsamples, test prices were rotated among the three
subsamples of respondents in the manner illustrated in Table II. Thus, 805
respondents in Sample 1, for example, viewed the following test prices: cheese
$4.95, chicken $6.00, chocolates $9.99, hairdryer $19.95, kettle $50.05, and
blender, $99.
This process produced 300 purchase probabilities for each top and bottom
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

“anchor” price and 100 purchase probabilities for each even and odd “test”
price, for each of the six products involved. These purchase probabilities and
their associated price points provided the means for estimating a demand curve
for each product.

Product 1 2 3 4 5 6

Low anchor price * * * * * *

Sample 1 T1 T3 T2 T1 T3 T2
Sample 2 T2 T1 T3 T2 T1 T3
Sample 3 T3 T2 T1 T3 T2 T1
High anchor price * * * * * *
Note: *high and low anchor prices were presented to all respondents Table II.
T1 = test price 1, T2 = test price 2, T3 = test price 3 Rotation of test prices

Results
Before estimating the demand curves, the age-sex distributions of the three
subsamples and the average purchase probabilities for both anchor prices for
each subsample were examined for evidence of differences which might
confound the effect of odd pricing. First, the data were weighted so that the age-
sex distributions of the three subsamples were the same. However, this process
had only minimal effect on the estimated purchase probabilities. Therefore, the
results reported here are based on unweighted data.
Next, the anchor point purchase probabilities for each of the subsamples
were compared. This revealed evidence of systematic bias in these purchase
probability estimates. In other words, in one subsample, the purchase
probabilities were consistently lower than average, while in another subsample
they were consistently higher. To eliminate this bias the anchor price purchase
probabilities were averaged, then a linear transformation was applied to the test
price purchase probabilities. This transformation involved recalculating the
test price probabilities in proportion to the positive or negative “shift” of the
subsample anchor points. This process is illustrated in the Appendix.
 European The demand curves shown in Figures 3 to 8 were estimated using these
Journal transformed data, by joining the purchase probabilities for the top and bottom
of Marketing anchor points through those for the even prices tested. For two products, the
hair dryer and the kettle, the probabilities associated with the 5 cent price
31,11/12 ending appeared to lie on the demand curve, consequently the curve was
formed through these points as well. Similarly, the demand curve for the
806 blender was formed through the probability for the $95 price point for the same
reason.
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

Price Key
$5.10 3.88
$5.00 4.49
$4.99 5.02
$4.95 4.96
$4.90 5.27
$5.10

$5.00
$4.99

$4.95

$4.90

Figure 3. 1.0 2.0 3.0 4.0 5.0 6.0 7.0

Demand curve – cheese
Purchase probability

To reiterate the assumption underlying odd pricing; if odd pricing “works”,

greater than expected demand would occur at these prices. That is, the
purchase probabilities for the odd prices tested would lie well to the right of the
estimated demand curves, producing a kink in the demand curves at these
points.
As can be seen from Figures 3 to 8, for all six products tested, plotting the
average purchase probabilities for the top and bottom anchor points and the
corresponding transformed even price probability produced a concave,
downward sloping demand curve. This gives some reassurance about the
validity of the method used as a means of testing the effect of odd pricing on
demand.
In total, ten odd prices were tested; five 95 cent endings, three 99 cent endings
and two whole dollar odd prices, namely, $95 and $99. For nine of these ten
 Price Key The effect of odd
$6.10 4.25 pricing on
$6.00 4.18
$5.99 4.88 demand
$5.95 5.49
$5.90 5.54
$6.10
807

$6.00
$5.99
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

$5.95

$5.90

Figure 4.
1.0 2.0 3.0 4.0 5.0 6.0 7.0 Demand curve – frozen
chicken
Purchase probability

prices there was evidence of an odd pricing effect, demonstrated by the

purchase probabilities at these prices lying to the right of the estimated demand
curve. Though none of these odd pricing effects is statistically significant on its
own[2], the likelihood of nine differences out of ten occurring in the same
direction by chance is less than 1 per cent. Thus, these results provide strong
support for the assumed effect of odd prices on demand.
Price Key
$10.10 1.84
$10.00 2.22
$9.99 2.52
$9.95 3.10
$9.90 3.11
$10.10

$10.00
$9.99

$9.95

$9.90

Figure 5.
1.0 2.0 3.0 4.0 5.0 6.0 7.0 Demand curve –
chocolates
Purchase probability
 European Price Key
Journal $20.10 2.06
$20.05 2.13
of Marketing $20.00 2.31
31,11/12 $19.95 2.76
$19.90 2.92
$20.10
808
$20.05

$20.00
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

$19.95

$19.90

Figure 6.
Demand curve – hair 1.0 2.0 3.0 4.0 5.0 6.0 7.0
dryer
Purchase probability

For the five products tested with odd price cents endings (i.e. 99 cents or 95
cents), all eight odd price points tested produced greater than expected demand.
The likelihood of this finding occurring by chance alone is less than 0.5 per cent.
Sensitivity to odd pricing was greatest for the three products tested at the
$10 and under level; cheese, frozen chicken and chocolates. Five of the six odd

Price Key
$50.10 2.95
$50.05 3.05
$50.00 3.19
$49.95 3.82
$49.90 4.09
$50.10

$50.05

$50.00

$49.95

$49.90

Figure 7. 1.0 2.0 3.0 4.0 5.0 6.0 7.0

Demand curve – kettle
Purchase probability
 Price Key The effect of odd
$110 1.44 pricing on
$100 1.63
$99 2.19 demand
$95 1.99
$90 2.92
$110
809

$100
$99
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

$95

$90

1.0 2.0 3.0 4.0 5.0 6.0 7.0 Figure 8.

Demand curve – blender
Purchase probability

price points tested at this level showed noticeably greater estimated demand.
The sixth odd price tested (cheese at $4.95) revealed only a weak odd pricing
effect compared to that at $4.99.
Overall, the effect of each of the odd prices tested on demand for cheese,
frozen chicken and chocolates was similar for the 95 cent and 99 cent endings.
That is, neither odd price ending produced consistently greater estimated
demand than the other.
Two of the three electrical appliances, the hair dryer tested at the $20 level
and the kettle tested at the $50 level, produced very similar demand curves, and
for both estimated demand was greater than expected at the 95 cent odd price
ending. However, the increase in estimated demand was less noticeable at these
two price levels than for most odd prices tested at lower price levels. For the
remaining electrical appliance, the blender, a shift to the right of the estimated
demand curve occurred at $99, but no odd pricing effect was detected at $95.

Discussion
According to economic theory, quantity demanded will increase when price is
reduced (except in certain special cases). Thus, an odd price of, say, $9.99 is
expected to produce greater demand than a slightly higher even price of $10.00.
The odd pricing assumption is that such odd prices result in greater than
expected demand, resulting in a “kink” in the traditional downward sloping
demand curve at these points.
This study provides support for this assumption. Purchase probabilities for
nine of the ten odd prices tested fell to the right of the estimated demand curve
 European for the product concerned and, for the cents-ending odd prices, the effect was
Journal even more marked, with all eight prices tested producing greater than expected
of Marketing demand. Individually, the differences between expected and actual purchase
probabilities were not significant, but the overall effect is very unlikely to have
31,11/12 occurred by chance.
The greatest sensitivity to odd pricing occurred for the low-priced grocery
810 products tested. The explanation for this enhanced sensitivity to odd pricing for
such products may be consumers’ greater price awareness for these regularly
purchased items, or perhaps the fact that the relative price differential between
the odd and even cents prices tested is greater for lower-priced products.
In this study there was no obvious difference in the effect on demand of 95
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

cent and 99 cent endings. The implication of this finding is that retailers should
use 99 cent price endings and benefit from the extra four cents revenue from
each sale. This conclusion supports the predominant use of prices ending in the
digit 9, and is one which has already been reached by a major New Zealand
supermarket retailer. This supermarket chain moved from 99 cent price endings
to 95 cent price endings when one and two cent coins were withdrawn from
circulation in New Zealand. Subsequently, the firm returned to 99 cent price
endings to achieve a higher gross margin for each item. No adverse effect on
sales occurred as a result.
The effectiveness of odd prices ending in the digit 9 appears particularly
marked at the $100 level. Many retailers believe that $100 is an important
psychological price point and that prices set above $99 (but close to $100) result
in a disproportionate loss of sales. This study tends to support this conclusion,
to the extent that a test price of $95 produced no odd pricing effect, whereas the
effect at $99 was quite marked.
What cannot be determined from this study is how the effect of odd pricing
occurs. Perhaps odd prices create an illusion of cheapness, perhaps consumers
have been conditioned to expect odd price endings, or perhaps there is some
other explanation. Some researchers (Schindler and Wiman, 1989; Schindler,
1991) have suggested that odd pricing may have several effects on consumers,
some positive, some negative. For example, the perception of an odd-price
ending as a low price might be offset by an association of odd-ending prices
with lower quality items. Thus, further research is needed to determine the
separate effects of odd pricing on demand.
Reasons for the previous lack of evidence to support the assumption that odd
pricing has a positive effect on demand are not clear. One possible explanation
is that the empirical investigations which did not find support for this
assumption were conducted in real shopping situations, where consumers are
subjected to in-store influences other than price. The effect of odd pricing per se
may be less in practice than in an experimental situation where price is the
focus of attention.
There are also some limitations to this study which need to be acknowledged.
First, the results are based on purchase probabilities, collected in an
experimental setting, not on actual purchase behaviour. Second, respondents
 were asked to evaluate the same product at different price levels, whereas in The effect of odd
practice they would be confronted with a number of different brands in the pricing on
same product field. Finally, the range of products and price levels tested was demand
relatively small. Consequently, it is difficult to say how generalizable this
study’s results may be to other products and other price levels.
An alternative method of experimentation which would more closely
approximate a true buying situation would be to test a range of brands within 811
different product categories, featuring varying odd and even prices, rather than
testing a single brand in each category. By including a wider range of products
and a wider spread of anchor and reference prices, most of the limitations of the
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

current study would be overcome.

Three decades ago, Twedt (1965) stated:
Wouldn’t it be interesting if a practice that may have started as a safeguard against petty theft
persists today as a major imperfection in the price-making process?
Experimentation is clearly called for to determine whether the popularity of odd numbers,
and the “9 fixation” in particular, really represent “magic numbers” that promote sales. Or are
they only “sticky prices” that hinder scientific pricing decisions and optimum profits? (p. 55).
Despite its limitations, this study provides empirical support for the
assumption underlying odd pricing and for the common practice of setting
retail prices which end in 99 cents or $99.

Notes
1. These “critical” price levels were determined in preliminary discussions with a sample of
27 retailers throughout New Zealand. These represented seven Palmerston North-based
outlets and 20 national retail chains. In each case the respondent interviewed was
personally involved in implementing pricing policy.
2. Based on t-tests of the difference between observed purchase probability for each test price
and purchase probability calculated from the estimated demand curves at these prices.

References
Boyd, H.W. and Massy, W.F. (1972), Marketing Management, Harcourt Brace Jovanovich, Orlando,
FL.
Brenner, G.A. and Brenner, R. (1982), “Memory and markets, or why are you paying $2.99 for a
widget?”, Journal of Business, Vol. 55 No. 1, pp. 147-58.
Clawson, C.J. (1971), “How useful are 90-day purchase probabilities?”, Journal of Marketing,
Vol. 35, pp. 43-7.
Dalrymple, D.J. and Thompson, D.L. (1969), Retailing - An Economic View, The Free Press, New
York, NY.
Day, D., Gan, B.C., Gendall, P.J. and Esslemont, D.H.B. (1971), “Predicting purchase behaviour”,
Marketing Bulletin, Vol. 2, pp. 18-30.
Dodds, W.B. and Monroe, K.B. (1985), “The effect of brand and price information on subjective
product evaluations”, Advances in Consumer Research, Vol. 12, pp. 85-90.
Friedman, L. (1967), “Psychological pricing in the food industry”, in Phillips, A. and Williamson,
O.E. (Eds), Prices: Issues in Theory, Practice, and Public Policy, University of Pennsylvania
Press, Philadelphia, PA.
Gabor, A. (1977), Pricing: Principles and Practices, Heinemann Educational Books Ltd, London.
 European Gabor, A. and Granger, C.W.J. (1964), “Price sensitivity of the consumer”, Journal of Advertising
Research, Vol. 4, December, pp. 40-4.
Journal
Gabor, A. and Granger, C.W.J. (1972/73), “Ownership and acquisition of consumer durables: report
of Marketing on the Nottingham Consumer Durables Project”, European Journal of Marketing, Vol. 6,
31,11/12 pp. 234-48.
Georgoff, D.M. (1971), Odd-Even Retail Price Endings, Michigan State University Press, Ann
812 Arbor, MI.
Gilmour, J. (1985), “One cent less doesn’t make sense”, Australian Business, 20 March, p. 34.
Ginzberg, E. (1936), “Customary prices”, American Economic Review, 26 June, p. 296.
Gruber, A. (1970), “Purchase intent and purchase probability”, Journal of Advertising Research,
Vol. 10 No. 1, pp. 23-7.
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

Harper, D.V. (1966), Price Policy and Procedure, Harcourt, Brace and World, New York, NY.
Heald, B.I. (1970), “The relationship of intentions-to-buy consumer durables with levels of
purchase”, British Journal of Marketing, Vol. 4, pp. 87-97.
Högl, S. (1988), “The effects of simulated price changes on consumers in a retail environment -
price thresholds and price policy”, Esomar Congress Proceedings, Lisbon.
Holdershaw, J.L. (1995), “The validity of odd pricing”, unpublished MBS Thesis, Massey
University, Palmerston North.
Isherwood, B.C. and Pickering, J.B. (1975), “Factors influencing individual purchases of motor
cars in Great Britain”, Oxford Bulletin of Economics and Statistics, Vol. 37, pp. 227-49.
Juster, F.T. (1966), “Consumer buying intentions and purchase probability”, Occasional Paper 99,
National Bureau of Economic Research, Columbia University Press.
Kreul, L.M. (1982), “Magic numbers: psychological aspects of menu pricing”, Cornell Hotel and
Restaurant Administration Quarterly, Vol. 23 No. 1, pp. 70-5.
Lambert, Z.L. (1975), “Perceived prices as related to odd and even price endings”, Journal of
Retailing, Vol. 51, Fall, pp. 13-22, 78.
Monroe, K.B. (1990), Pricing: Making Profitable Decisions, 2nd ed., McGraw-Hill, New York, NY.
Nagle, T.T. (1987), “Psychological aspects of pricing: influencing price perception and evaluation”,
The Strategy and Tactics of Pricing, Prentice Hall, Englewood Cliffs, NJ.
Printers Ink (1954), “Psychology makes cents in pricing”, Editor’s reply, 29 October, p. 45.
Rogers, L. (1990), Pricing for Profit, Basil Blackwell, Cambridge, MA.
Rudolph, H.J. (1954), “Pricing for today’s market”, Printers Ink, 28 May, pp. 22-4.
Schindler, R.M. (1991), “Symbolic meanings of a price ending”, Advances in Consumer Research,
Vol. 18, pp. 794-801.
Schindler, R.M. and Warren, L.S. (1988), “Effect of odd pricing on choice of items from a menu”,
Advances in Consumer Research, Vol. 15, pp. 348-52.
Schindler, R.M. and Wiman, A.R. (1989), “Effects of odd pricing on price recall”, Journal of
Business Research, Vol. 19, pp. 165-77.
Sturdivant, F.D. (Ed.) (1970), Managerial Analysis in Marketing, Scott, Foresman and Company,
Glenview, IL.
Twedt, D.W. (1965), “Does the ‘9 fixation’ in retail pricing really promote sales?”, Journal of
Marketing, Vol. 29 No. 4, pp. 54-5.

Appendix. Linear transformation of purchase probability data

An example of the linear transformation performed on data for the product frozen chicken is as
shown in Table AI.
 First, the mean purchase probability for the top and bottom anchor points was calculated: top The effect of odd
anchor price ($6.10), mean probability = 4.25; bottom anchor price ($5.90), mean probability =
5.54. The linear transformation was then made using the following equation: pricing on
demand

Price ($) Purchase probability Subsample

813
5.90 4.35 1
5.90 5.56 2
5.90 6.70 3
6.00 3.18 1
5.95 5.60 2 Table AI.
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

5.99 5.71 3 The purchase

6.10 3.44 1 probabilities for each
6.10 4.64 2 subsample prior to
6.10 4.68 3 linear transformation

Original purchase Transformed purchase

“Test” price ($) x y a b probability probability
Table AII.
6.00 0.81 1.19 0.10 0.10 3.18 4.18 Transformation of
5.95 –0.39 –0.02 0.50 0.05 5.60 5.49 “test” price purchase
5.9 –0.43 –1.16 0.11 0.09 5.71 4.88 probabilities

pt = po + s
pt = po + x + (y – x) * a/(a + b)
where:
pt = transformed “test” price purchase probability;
po = original “test” price purchase probability;
s = shift, either positive or negative, of each “test” price point;
x = mean purchase probability of the three top anchor price subsamples minus the individual
subsample purchase probability;
y = mean purchase probability of the three bottom anchor price subsamples minus the
individual subsample purchase probability;
a = differential between the top anchor price and the test price;
b = differential between the bottom anchor price and the test price.
Figures derived from the transformation calculations are presented in Table AII.
 This article has been cited by:

1. Eric Cardella, Michael J. Seiler. 2016. The effect of listing price strategy on real estate negotiations: An experimental study.
Journal of Economic Psychology 52, 71-90. [CrossRef]
2. Yong (Eddie) Luo, Veronica Wong, Ting-Jui Chou. 2015. The role of product newness in activating consumer regulatory
goals. International Journal of Research in Marketing . [CrossRef]
3. Matthew S. Lewis. 2015. Odd Prices at Retail Gasoline Stations: Focal Point Pricing and Tacit Collusion. Journal of Economics
& Management Strategy 24:10.1111/jems.2015.24.issue-3, 664-685. [CrossRef]
4. Peter G. Roma, Steven R. Hursh, Stanton Hudja. 2015. Hypothetical Purchase Task Questionnaires for Behavioral Economic
Assessments of Value and Motivation. Managerial and Decision Economics n/a-n/a. [CrossRef]
5. Jan Wieseke, Anika Kolberg, Laura Marie Schons. 2015. Life could be so easy: the convenience effect of round price endings.
Journal of the Academy of Marketing Science . [CrossRef]
6. Charlotte Gaston-Breton, Lola C. Duque. 2015. Utilitarian and hedonic promotional appeals of 99-ending prices. European
Journal of Marketing 49:1/2, 212-237. [Abstract] [Full Text] [PDF]
7. Jungsil Choi, Yexin Jessica Li, Priyamvadha Rangan, Promothesh Chatterjee, Surendra N. Singh. 2014. The odd-ending
price justification effect: the influence of price-endings on hedonic and utilitarian consumption. Journal of the Academy of
Marketing Science 42, 545-557. [CrossRef]
8. Eli Beracha, Michael J. Seiler. 2014. The Effect of Listing Price Strategy on Transaction Selling Prices. The Journal of Real
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)

Estate Finance and Economics 49, 237-255. [CrossRef]

9. Alexander C. Larson, Rita L. Reicher, David William Johnsen. 2014. Threshold effects in pricing of high-involvement
services. Journal of Product & Brand Management 23:2, 121-130. [Abstract] [Full Text] [PDF]
10. Vytautas Dikcius, Kristina Katkuviene. 2014. Perception of odd prices in a Post-Soviet country: an impact of the ideology of
planned economy. Journal of Business Economics and Management 15, 356-368. [CrossRef]
11. Franz Hackl, Michael E. Kummer, Rudolf Winter-Ebmer. 2014. 99 Cent: Price points in e-commerce. Information Economics
and Policy 26, 12-27. [CrossRef]
12. Brian R. Kinard, Michael L. Capella, Greg Bonner. 2013. Odd pricing effects: an examination using adaptation‐level theory.
Journal of Product & Brand Management 22:1, 87-94. [Abstract] [Full Text] [PDF]
13. Sandrine Macé. 2012. The Impact and Determinants of Nine-Ending Pricing in Grocery Retailing. Journal of Retailing 88,
115-130. [CrossRef]
14. Sabine Kleinsasser, Udo Wagner. 2011. Price endings and tourism consumers’ price perceptions. Journal of Retailing and
Consumer Services 18, 58-63. [CrossRef]
15. Robert E. Carter, David J. Curry. 2010. Transparent pricing: theory, tests, and implications for marketing practice. Journal
of the Academy of Marketing Science 38, 759-774. [CrossRef]
16. Kenneth C. Manning, David E. Sprott. 2009. Price Endings, Left-Digit Effects, and Choice. Journal of Consumer Research
36, 328-335. [CrossRef]
17. George J. Avlonitis, Kostis A. Indounas. 2007. An empirical examination of the pricing policies and their antecedents in the
services sector. European Journal of Marketing 41:7/8, 740-764. [Abstract] [Full Text] [PDF]
18. Christine Harris, Jeffery Bray. 2007. Price endings and consumer segmentation. Journal of Product & Brand Management
16:3, 200-205. [Abstract] [Full Text] [PDF]
19. Jianping Liang, Vinay Kanetkar. 2006. Price endings: magic and math. Journal of Product & Brand Management 15:6, 377-385.
[Abstract] [Full Text] [PDF]
20. Sajeev Varki, Sanjiv Sabherwal, Albert Della Bitta, Keith M. Moore. 2006. Price‐end biases in financial products. Journal of
Product & Brand Management 15:6, 394-401. [Abstract] [Full Text] [PDF]
21. George J. Avlonitis, Kostis A. Indounas. 2006. Pricing practices of service organizations. Journal of Services Marketing 20:5,
346-356. [Abstract] [Full Text] [PDF]
22. Thanos Skouras, George J. Avlonitis, Kostis A. Indounas. 2005. Economics and marketing on pricing: how and why do they
differ?. Journal of Product & Brand Management 14:6, 362-374. [Abstract] [Full Text] [PDF]
23. Keith S. Coulter, Robin A. Coulter. 2005. Size Does Matter: The Effects of Magnitude Representation Congruency on Price
Perceptions and Purchase Likelihood. Journal of Consumer Psychology 15, 64-76. [CrossRef]
24. Marcus T. Allen, William H. Dare. 2004. The Effects of Charm Listing Prices on House Transaction Prices. Real Estate
Economics 32:10.1111/reec.2004.32.issue-4, 695-713. [CrossRef]
25. Gianluigi Guido, Alessandro Peluso. 2004. Consumers' perception of odd‐ending prices with the introduction of the Euro.
Journal of Product & Brand Management 13:3, 200-210. [Abstract] [Full Text] [PDF]
 26. John Ashworth, Bruno Heyndels, Carine Smolders. 2003. Psychological taxing in Flemish municipalities. Journal of Economic
Psychology 24, 741-762. [CrossRef]
27. Keith S. Coulter. 2003. The effects of congruent/incongruent magnitude representation on explicit and implicit knowledge
of prices. Journal of Product & Brand Management 12:5, 293-306. [Abstract] [Full Text] [PDF]
28. Keith S. Coulter. 2002. The influence of print advertisement organization on odd‐ending price image effects. Journal of
Product & Brand Management 11:5, 319-334. [Abstract] [Full Text] [PDF]
29. Keith S. Coulter. 2001. Odd‐ending price underestimation: an experimental examination of left‐to‐right processing effects.
Journal of Product & Brand Management 10:5, 276-292. [Abstract] [Full Text] [PDF]
30. Robert A. Bartsch, Vivian I. Paton. 1999. The Presence of Odd Pricing in the Texas State Lottery1. Journal of Applied Social
Psychology 29:10.1111/jasp.1999.29.issue-11, 2394-2409. [CrossRef]
31. Philip Gendall. 1998. Estimating the effect of odd pricing. Journal of Product & Brand Management 7:5, 421-432. [Abstract]
[Full Text] [PDF]
Downloaded by UNIVERSITY OF MELBOURNE At 20:31 13 February 2016 (PT)