352 The Marketing Book
would rise reflecting less efficient use of resour-
ces or less economical raw material purchases’
(Morris and Morris, 1990, pp. 45 and 101).
In general, the following caveats apply
when using the formula for price elasticity:
The formula is based on a ceteris paribus(i.e.
‘all other things being equal’) assumption,
which implies that the onlyvariable that affects
changes in sales volume is the change in price;
prices of competitors, incomes, preferences,
etc. are assumed to be constant. This
assumption is obviously questionable,
particularly in the case of oligopolistic markets,
where interdependence among suppliers is
likely to lead to reactions when prices are
changed (e.g. price wars in the case of price
reductions).
A distinction needs to be made between
market(or primary) price elasticity (i.e. that
relating to the market demand) and brand
price elasticity (i.e. that relating to the
particular brand under consideration). For
example, while the demand for an overall
product category may be inelastic, demand for
particular brands withinthis category may be
elastic. Moreover, price elasticity can vary
substantially across different product brands
within a particular class. For example, ‘the
sales of small market share brands tend to be
more price sensitive than those of brands with
larger market shares’ (Nagle, 1987, p. 79),
while ‘price cuts by higher quality tiers are
more powerful in pulling customers up from
lower tiers, than lower tier price cuts are in
pulling customers down from upper tiers; i.e.,
customers “trade up” more readily than they
“trade down”’ (Dolan and Simon, 1996, p. 87).
What these considerations also imply is that,
depending on how broadly one definesthe
market, different elasticity estimates may
result.
From the price elasticity formula, it is not
immediately apparent that the value of isnot
the same at all prices; for example, an
elasticity estimated around a price of, say,
£2.00 will not be the same as an elasticity
estimated around a price of, say, £10. A
constant elasticity at all prices within a given
range is very much the exception rather than
the rule. Indeed, a typical mistake that decision
makers make is to assume that the price
elasticity of demand is equal to the slope of
the demand curve (i.e. the price response
function); this is notthe case even when a
linear demand function (i.e. of the form q = a- b ×p, where a, b > 0) is involved. For more
details, as well as a description of the properties
of the isoelastic(i.e. constant elasticity) price
response function, see Simon (1992).
While also not obvious from the elasticity
formula, ‘because customers differ in the
amount of value they attach to a product or
service and in their ability to pay for an item,
their elasticities also differ. One customer may
respond very little to fairly large changes in
price while another reacts strongly to a
relatively minor price change for the same
product’ (Morris and Morris, 1990, p. 46).
Indeed, differences in price elasticity are a
major basis for segmenting a market with a
view of customizing prices (see, for example,
Simon, 1989).
Since individuals tend to be more sensitive to
the prospect of a loss than to the prospect of
a gain (Kahneman and Tversky, 1979),
computations of price elasticity based on price
increases of a certain magnitude may produce
different results than those based on price
decreases of the samemagnitude.
Price elasticity does not remain constant over
time, since ‘the percentage change in a
product’s sales is usually not the same in the
long run as in the short run’ (Nagle, 1987,
p. 77). Factors such as inventory building,
substitute awareness, ‘lock-in’ contracts, new
product introductions and price expectations
all combine to introduce differences in the
magnitudes of short- versus long-term price
elasticities. For a discussion of the dynamics of
price elasticity for different types of products,
see Simon (1979), Shoemaker (1986), Kucher
(1987), Lillien and Yoon (1988), and Parker
(1992).