A SIMPLE PROOF OF SWEEZY'S ‘KINKED‐DEMAND’ CONJECTURE
| Published date | 01 February 1992 |
| Author | R. Rothschild |
| DOI | http://doi.org/10.1111/j.1467-9485.1992.tb00605.x |
| Date | 01 February 1992 |
Scuiiirh
Journal
of
Polrircal
Erononry.
Vol. 39,
No.
I.
February
1992
r
1992
Scollish
Economic
Society
A SIMPLE PROOF
OF
SWEEZY’S
‘KINKED-DEMAND’ CONJECTURE
R. ROTHSCHILD*
University
of
Lancaster
I
INTRODUCTION
Although
P.
M. Sweezy’s ‘kinked demand curve’ (Sweezy,
1939)
no longer
enjoys its early status in the literature
of
the theory
of
oligopoly, and certainly
not at the higher levels, the model retains its appeal for the authors
of
textbooks and for students. This fact is both curious and ironic. It is curious
because it is by now generally held that, from the perspective
of
‘good’ theory,
the
original model suffers from
a
number
of
quite serious flaws;
it
is ironic
because economic analysis has produced few models which appear better able
than Sweezy’s
to
confirm
a
widespread intuition and to rationalise casual
empiricism.
The original formulation
of
the Sweezy model is easily stated.
’
The industry
consists
of
N
firms,
N>
1,
each producing
a
variety
of
a
differentiated
product. Firms choose price noncooperatively on the conjecture that once the
common price has been established, rivals will match deviations in the
downward but not in the upward direction. Consequently, each firm anticipates
that
at this level any unilateral change in its own price will result in
a
reduction
in profits. In the simplest case, in which both firms and their conjectures are
identical, the perception that rivals’ responses will be
asymmetric
upward and
downward (ie. ‘kinked’) ensures that the common price will be stable, not only
for
given costs but also for small changes in costs in either direction.
The criticisms
of
Sweezy turn on both theoretical and empirical issues. At the
theoretical level, the principal argument is that insofar as firms must be
expected
to
hold their asymmetric conjectures at any given price the location
of
the ‘kink’ is indeterminate and hence unpredictable. At the empirical level
the criticism is that there appears to be little evidence that firms do in fact form
the postulated conjectures. Moreover, an implication
of
the analysis-that
prices in oligopolistic markets should be more stable than those in
monopolies-is not confirmed by empirical research.
*
I
wish to thank an anonymous referee for a number
of
very perceptive and useful
comments on an earlier draft.
’
For
an excellent treatment of the genesis and detail
of
the Sweezy analysis, see Reid
(1
98
1).
Date of receipt
of
final manuscript: 25th February
1991.
69
To continue reading
Request your trial