ADVERTISING AND MARKET SHARE MOBILITY*
| Date | 01 June 1974 |
| DOI | http://doi.org/10.1111/j.1467-9485.1974.tb00185.x |
| Published date | 01 June 1974 |
| Author | W. Duncan Reekie |
Scottish
Journal
of
Political Economy
Vol.
XXI,
No.
2,
June
1974
ADVERTISING AND MARKET SHARE
MOBILITY*
W.
DUNCAN REEKIE
I
INTRODUCTION
The nature of the relationship between advertising and competition has
been the subject of considerable scrutiny in recent years. Nonetheless,
despite the intensity of empirical investigation considerable disagreement
persists and
no
writer has yet produced
an
undisputed and definitive state-
ment
on
the topic. This paper presents the results of one recent investigation
in
the area and provides a further piece
of
evidence
in
support of the view
that it is dangerous to make broad generalisations about advertising and
competition.
The paper
is
organised as follows.
In
Section
I1
some important recent
literature is briefly reviewed. Explanations are provided as to how this
study differs from previous investigations. Section
III
elaborates
on
the
theory of advertising and market share stability. Section
IV
describes the
data, sample and variables employed to test the theory that advertising and
competition as measured by share mobility are associated. The results of
testing the model and the statistical difficulties encountered
in
the testing
process are detailed in Section
V.
I1
ADVERTISING
AND
COMPETITION
Advertising is unnecessary in a situation of perfect competition with
homogeneous products and perfect information. It can
be
argued, however,
that advertising has the ability to change the position and shape of a
firm’s
demand curve.
In
other words, it
can
increase a
firm’s
market power by
further separating its market from those of its competitors. Conversely,
Stigler
(1961)
has suggested that in conditions of non-complete information
and
so
in departures from perfect competition, advertising is one method
of reducing the costs of consumer search and
so
improving the market in
information.
The nature of the relationship between advertising and market power
thus cannot
be
determined
on
the basis of
a
priori
reasoning alone. The
foregoing arguments have widely differing welfare implications and
empirical examination, although as yet unable to provide indisputable
support for either view, is possibly the area more in-need of development.
(a)
Seller
Concentration Ratios
Early studies of the subject typically measured the extent of the market
power with which advertising may be associated by the degree of seller
*
This study
was
financed
by
a
grant
from
the
Advertising Association.
143
144
W.
DUNCAN
REEKIB
concentration. Other measures used have included the levels and con-
sequences
of
entry barriers and the stability of market shares. Telser (1964)
made one of the first attempts to ascertain what relationship exists between
advertising and market concentration. He compared advertising intensity
(i.e. advertising
:
sales ratios) with market concentration
data
for forty-two
industries. The correlation he discovered between the two was
'
unimpres-
sive
'.
A later study based
on
a differently constructed sample of fourteen
industries, however, found a significant and strong statistical association
between advertising intensity and concentration (Mann, Henning and
Meehan, 1967). This resulted in a controversy, with both sets of authors
defending
their
respective conclusions and the statistical
bases
on
which
they rested (Symposium
on
Advertising and
Concenttation.
1969).
At least
two
points were left largely unconsidered by both sets of
protagonists. One was the nature
of
the relationship between the
two
variables and entry barriers (discussed below). The second
was
the fact
that inter-industry differences in advertising depend
on
several other factors,
in
addition to market concentration. The consequence of omitting these
other factors is, as Doyle (1968) pointed out, that Telser's conclusion
is
not
a
particularly surprising
one.
On
the supply side
of
an industry these factors
include the level
of
innovation and the extent to which produds are manu-
factured with
a
view to producing relatively tangible or intangible consumer
satisfactions.
On
the demand side advertising may vary with the demand
for knowledge of the product, which in turn is dependent
on
a wide variety
of
variables.
An
attempt was made to meet these latter criticisms in
a
study
of
thirty-
seven markets within the British pharmaceutical industry (Reekie, 1970).
The study analysed pharmaceutical advertising at the level of the therapeutic
submarket.
In
economic terms these are the segments of the total pharma-
ceutical market between which the cross-elasticity of demand for competing
products is very low. It is in these areas, not the total market, that study of
market power is most meaningful.
It is probable
that
conditions
on
both the supply and demand side of
each of these markets were very similar to those in every other market.
The
firms
participating in each were all members of the pharmaceutical
industry. production techniques in each were not dissimilar, and scale
economies were largely absent. The customers operating in each market
were all general practitioners and
so
would all participate to a more or less
similar extent
in
each market. Some other dissimilarities between the sub-
markets were, of course, inevitable (e.g. the level of innovation
and
the
state of technology in each) and an attempt was also made to take these
variables into consideration.
The results obtained suggested that there is
no
relationship between
market structure and advertising. However, although some
of
the difficulties
involved
in
making valid intermarket comparisons were overcome, the
exercise related
overall
to only one industry, pharmaceuticals. and
so
the
right to draw generalised conclusions was forfeited.
To continue reading
Request your trial